Stand-alone Panels and Terminal Controllers

Up- to-date information on functional and friendly stand-alone panels and the need to improve airterminal con trollers so that both can be effectively integrated in to single sys terns

hen the book on the 20th century is closed a scant decade from now, a strong case will be made that the most profound technological development of the century has been the digital computer. Although one seldom sees a substantial impact of the technology in the controls of buildings today, the situation is rapidly changing. Development of digital controls for building systems lagged behind other industries, but the last several years have seen enormous growth in the power and economy of direct digital control (DDC) systems. Today, DDC systems are fast beBy THOMAS HARTMAN, PE, The Hartman Co., Seattle, Wash.

coming the leading innovative force in building design. So many DDC improvements have been released recently that some have suggested manufacturers should pause to let the rest of the industry digest all that is new before embarking on further developments. Such a pause in development at this time is most inappropriate because in reality the building controls industry, like Robert Frosts traveler, still has many promises to keep and miles to go. In this article, we will review the reasons DDC controls, as they are generally available today, have become so successful. We will also look at why some of the promises of extending DDC to terminal control have yet to be realized. Finally, we will review suggestions that have

been offered to ensure that the promise of extending DDC control to terminal HVAC units and lighting systems is realized. D D C systems today Users, and some suppliers, are still working with DDC systems that were designed a decade or more ago. Many of these people have difficulty understanding the changes that have taken place in DDC systems over the last few years. From a hardware perspective, systems are much simpler. They are easier to install, start up, calibrate, operate, and troubleshoot. Today, operators with basic skill levels and training can maintain modern DDC system applications in-house. Or users can sign extended warranty contracts that,
41

Heating/Piping/Air Conditioning , November 1990

DDC controls

when procured competitively, cost as little as $4 to $8 per DDC system point per year (it used to be that $50 to $100 per DDC point per year was considered by many to be a reasonable range for DDC maintenance costs). The costs of systems themselves have fallen dramatically. A decade ago, designers often used $1000 per point to estimate the installed cost of DDC systems. In some recent competitively procured retrofit applications, installed DDC system costs have been quoted below $200 per point. Meanwhile, the performance of DDC systems continues to improve. Programming features make it easier than ever for an operator to create, document, or adjust control programs, view the value and status of system points, and add new points or calibrate existing ones. In the last few years, the real success story in the HVAC control industry is the development and refinement of DDC stand-alone panels. They have become the platform for almost all recent DDC improvements. By understanding the performance attributes that have made stand-alone DDC panels successful, we can better appreciate where the industry is now and what the focus should be for further improvements. Panel hardware Stand-alone panels make up the nucleus of a modern DDC system. How the stand-alone panel performs is the single most important factor determining the overall performance of a DDC system. A basic stand-alone panel typically has the capacity to connect directly to about 50 total input and output points, and most panels also have the capacity for expander boards or slave I/O multiplex boards that can increase the point capacity almost indefinitely. The limiting factor in expansion is usually the amount of panel memory available to support applications programs, trend logs, displays, etc., for the points connected. When procured compet42

itively, stand-alone panels typically cost the end user about $2500 to $5000 (uninstalled) for a 50point configuration. There are several approaches to configuring stand-alone panels in the industry today. One approach adopted by manufacturers employs a fully configured stand-alone panel on a single printed circuit board. Fig. 1 is an example of a single board panel. The idea is that a DDC system composed of fewer parts is easier and more economical for the manufacturer to make, the distributor to stock, the contractor to install, and the user to maintain. Indeed, our firms experience is that single board panel configurations often provide the lowest first cost. To make certain the panel serves a variety of applications, most single board panels permit DDC connection points to be configured with great flexibility. Each point can typically be analog or digital, but output points and input points are generally not interchangeable. The downside to the single board approach is that any component failure requires complete board replacement. Although plug-on terminations and fast reload procedures make panel replacement nearly as fast as replacing a single component, it can be expensive. So

2 Example of a modular DDC stand-alone panel.

1 Example of a DDC stand-alone panel on a single board.

long as failures are few and manufacturers or third-party firms provide economical board repair, the single board approach will be successful. A second approach is to build a panel from modules. Separate modules are inserted into a mother board to configure each input or output, and only those modules that are required for the actual I/O points in each panel are installed. Power supplies, communication controllers, and other functions are also modularized by some manufacturers. An example of a modular panel is shown in Fig. 2. Some advantages exist for the modular panel because it can be configured to meet the exact I/O combination required. However, because the sum of the individual modules tends to be more costly than the single board panel, the real advantage of the modular approach is the enhanced ability to isolate and repair failures quickly and inexpensively. How much of an advantage this can become depends on the costs of troubleshooting and replacing a single module compared to replacing the whole panel.
November 1990

Heating/Piping/Air Conditioning

A third approach to stand-alone panel hardware is to make a compromise between the panel on a board and the modular panel. Manufacturers adopting this approach wish to use advantages from both the other approaches. The semi-modular panels, as shown in Fig. 3, have both advantages and disadvantages compared to the others when various features are considered, but the differences among all the approaches are certainly not significant. In fact, our firm has never found any compelling reason to specify one of these types of panels over another. The most effective means of determining the right panel for a given application is to require a long-term extended warranty proposal along with the DDC system price. The system cost added to the present value of the extended warranty provides the best picture of the true system cost over time. Whether or not the user intends to sign an extended warranty agreement, a warranty proposal from the vendor in a competitive environment is probably the best estimate of the true cost of maintaining the system. And in the final analysis, cost is the primary reason for selecting one hardware configuration over another. Panel function Stand-alone panels typically contain all the capabilities needed to provide DDC control, including coordinating communications among panels, executing control programs, PID loop control, storing trend log values and status, etc. Operator interface devices can all be disconnected without affecting automatic control in true standalone panels because all real-time functions are contained at the panel level. Important progress has been made in the last few years with stand-alone panel technologies. Though manufacturers have adopted a variety of philosophies regarding operating features, a number of features have been uni-

versally recognized and incorporated in stand-alone panel designs. Recent improvements have been advanced in: l Reliability- A few years ago, stand-alone panels had a reputation for being sensitive to powerline problems. Elevators and emergency generators were but two of a myriad of reasons (a better word might be excuses) given for unexplained DDC system crashes and other failures. In recent years, the reliability of stand-alone panels has improved remarkably. Today, it is not unusual to find large systems operating several years without a failure serious enough to require even reloading a single panel. l Panel memory- M e m o r y poses the same problems to DDC system manufacturers as it does to computer manufacturers in other

to utilize valuable software features fully. l Programming languagesNot so long ago, it was widely held by vendors that canned programs with simple input parameters were all a building operator needed (or more to the point, all the building operator could handle) to implement control strategies in buildings. Users demands for more functional systems have at last been heard. Though there are several different viewpoints about the form applications programs should take to be most functional, the power and versatility of control languages in most products have been significantly upgraded in the last few years. l Panel-to-panel communications- Although users have long supported the distributed control

3 Example of a semi-modular DDC stand-alone panel.

industries-they have a tough time telling how much is enough. In DDC systems, point databases, applications programs, and trend logs all compete for available memory. Early stand-alone panels were notorious for their limited memory. Users were frustrated when needed features were severely limited by the amount of memory available. Newer panels supply much improved memory or memory upgrade capabilities that permit users

concept of stand-alone panels, they also demand that a DDC system perform as a single integrated entity and not as a series of small separate systems. When a program is written in one panel that requires point information from another, the operator shouldnt be required to set up the network required for the information to be shared between the panels. A fast, fully automatic communications network is an important ingredient for suc43

Heating/Piping/Air Conditioning. November 1990

DDC controls

cessful stand-alone panel configurations. Recent product releases and updates confirm that the industry understands the need to provide such networks. l Graphic displays and trend logs- Graphic displays of real-time point data and historically trended data vastly improve an operators ability to utilize the ever-growing supply of information available from a DDC system. Only a few years ago, many systems had very limited graphics features or graphics that were so slow they were impractical. Now, manufacturers appear to have focused on graphic features that really work and are easy to set up as well. Needed improvements With all the improvements of the last few years and the continuing trend to lower DDC system prices, one might wonder what more is needed. But for many who have recently started up full DDC systems (a full DDC system is one that con-

trols the entire system, including the terminal units that directly supply occupied areas), a whole new series of problems is threatening to renew the old complaints that systems are complicated, have slow response, and lack flexibility. To see what the problem is and what can be done about it, lets look at the relatively new DDC products that have made full DDC systems possible. Terminal controllers The relatively recent development of terminal controllers offers new opportunities to HVAC designers and system operators. Terminal controllers typically offer limited point capacity (8 inputs and 8 outputs or less) and usually have no expansion capability. Uninstalled, a terminal controller panel sells for $150 or more, depending on manufacturer and configuration. Because terminal controllers

4 Example of a DDC terminal controller board.

controller with enclosure.

have been designed with the idea of controlling small actuators and devices, some come with only digital outputs, in which case two physical outputs are employed to control each analog output device. Terminal controllers are almost always single board products. An example of a terminal controller board is shown in Fig. 4, Figs. 5 and 6 show typical terminal controllers with enclosures. Typically, terminal controllers are configured on a separate trunk that is supported by one of the stand-alone panels. The communication of this trunk is generally slower than the stand-alone panel trunk and doesnt employ peer-topeer communications. Depending on the manufacturer, 30 to several hundred terminal controllers can be, in theory at least, attached to a single stand-alone panel. Fig. 7 shows how terminal controllers are usually configured in DDC systems today. Terminal controllers have several problems that can cause considerable headaches for DDC system operators who have grown accustomed to the power, flexibility, and simplicity of standalone panels now available. The problems terminal controllers can cause are: l Lack of consistency-Fig. 7 shows that in a full DDC system, input and output points may be connected to either a stand-alone panel or a terminal controller. However, many DDC manufacturers see the terminal controllers as being application specific and put tight constraints on the type of points that can be connected to terminal controllers and how they are accessed. For the configuration in Fig. 7, assume temperature sensors are attached to several stand-alone panels and to the terminal controllers. Because of inherent differences between the stand-alone panel and the terminal controller, the operator may have to employ two very different techniques to connect the sensors, define their point database, calibrate them,
l

44

Heating/Piping/Air Conditioning

November 1990

manually override them, program them, or display their values, depending on how they are connected to the system. Some full DDC configurations more closely resemble two distinct DDC systems: one consisting of

fices is supplied by a single variable air volume (VAV) box. Traditional controls would place a single space thermostat in one office to operate the box damper and reheat coil with the hope that all offices would have consistent heating and cooling

Printer

Host computer Modem

1 DDC svstem I component level 4 Operator interface

Stand-alone panel

Terminal -controller
Device

points
7 Typical

lnput/output points

Input/output ppoints o

DDC system architecture. loads and thus would require the same HVAC response to remain comfortable. Building engineers know all too well that this type of control is unsatisfactory. Sooner or later an occupant in one of the rooms without the thermostat will complain about comfort. The building engineer has no satisfactory means to solve such problems. Can the thermostat be moved without offending that occupant? Will a rebalance of the outlets to send more or less flow to the offending space be anything more than a short term fix? These are situations designers have not had the tools to resolve, and they result in operating problems that are almost never entirely solved. With a functional full DDC control system, the opportunities to provide a comfortable, high quality environment and at the same time improve the energy efficiency of the system are much improved. DDC system temperature sensors are very low cost items, making it feasible to install a temperature sensor in every office. To control lighting and the HVAC terminal unit, each

stand-alone panels and another consisting of terminal controllers. This makes the operators task of remembering the different operating rules and procedures and how or where they apply most difficult. Manufacturers who have developed techniques that make their stand-alone panels simple and straightforward to operate appear to have underestimated the need for consistency of operation to meet these goals in DDC systems consisting of stand-alone panels and terminal controllers together. l Lack of function- The problem of consistency noted above would not be so serious if one only wished to replace the simple control functions of present terminal control devices. But replacement with no added function is not a compelling reason to upgrade to DDC terminal controls. The reason most designers and users desire to move up to full DDC systems is to provide more functional control at the terminal level to solve problems inherent with mechanical systems. Fig. 8 illustrates a common control problem. A group of small of-

office might also employ a pushbutton or occupancy sensor to notify the DDC system that the space is occupied. If a pushbutton is the occupancy device, occupancy may be assumed to be continuous for the day when activated during normal business hours or for an agreed fixed time if activated outside business hours. An occupancy sensor offers further control by providing a signal only when someone is in the room. A short delay in switching to unoccupied condition after the room is vacated permits stable operation. The occupancy signal controls lighting (for each office) and the VAV terminal box that serves all three offices. If only one office is occupied, then the box operates to satisfy that office. When multiple offices are occupied, the box operates to satisfy the average of the occupied offices unless an office exceeds high or low space temperature limits; in that instance, the box acts to bring the office temperature within limits, then returns to averaging control. The box need have no set minimum air flow. If all rooms are unoccupied, the box may shut off entirely so long as all offices are within suitable unoccupied temperature limits. When one or more space is occupied, the minimum air flow is set by the DDC system to provide the required outside air to the zone-based on the percentage of outside air being supplied to the box at the time. Such an operating sequence, together with additional features such as warm-up and night purge cycles, can lead to comfort and air quality levels for the occupants that are vastly superior to those provided by traditional terminal controls. The marginal DDC system cost to implement better terminal control is increasingly compensated for by energy cost savings. Further energy savings are possible by employing terminal regulated air volume (TRAV) control strategies for the central fan (see HPAC, July 1989). The prob45

Heating/Piping/Air Conditioning . November 1990

DDC controls

lem facing designers is that very few of the terminal control products on the market today have sufficient functions to accomplish effectively the sequence described above. Of those that can, most require extraordinary databases or program manipulations to do so, making such sequences difficult to implement and difficult for an op-

not practically be accomplished at all in many terminal controllers. ROM-based backup programs in terminal controllers are a good idea. ROM programs allow terminal units to be started and tested before the DDC system has been installed. ROM programs ensure that compressors or fans in certain types of terminal units will not be

bersome process because each point that is to be unbundled must be redefined in the stand-alone panel. Still, in some instances the unbundled points cannot be operated the same as panel points. Also, unbundled terminal controller points add to the point count of the stand-alone panel, which can severely restrict the number of terminal controllers a single stand-alone panel can serve. Controller development There are some lessons learned during development efforts of stand-alone panels that may help solve many problems the industry is now experiencing with terminal controllers. First, manufacturers should be encouraged to further develop their terminal control products so that, to the greatest extent possible, all points in the DDC system, whether connected to stand-alone panels or to terminal controllers, can be defined, calibrated, programmed, and overridden, with the same parameters and by the same simple processes. A flexible programming language should be available to write programs in terminal controllers in the same form as those in stand-alone panels. However, ROM-based backup programs should be preserved for startup and fail-safe control. Finally, it would be worthwhile for manufacturers to consider adding an area of memory in standalone panels to store the databases of all connected terminal controllers. This would simplify DDC system backup and reload procedures, Loading a stand-alone panel would automatically load all the terminal controllers attached to it. The stand-alone panel could monitor the state of all attached terminal controllers, reload any that lose their data, and alert the operator to any that are having difficulty maintaining their databases. Manufacturers sometimes build in limitations to terminal devices because the margin of profit is less
November 1990

@ Occupancy sensor
0 Temperature sensor A VAV box

4
8 HVAC terminal unit serving multiple offices. Such situations represent potential problems for traditional control svstems but can be handled effectively and efficiently with full DDC
control.

erator to support, l Lack of flexibility- When manufacturers began designing terminal controllers, it appeared the primary goal was to replace existing pneumatic controllers with a DDC product whose only advantage was remote monitoring. This unambitious goal has resulted in the industrys being deluged with terminal controllers whose application software is preprogrammed with simple control sequences in read-only memory (ROM). Many of the products available today not only lack the functional capacity to solve the design problem outlined above but also lack the flexibility for the operator to make even simple changes or adjustments in their sequence of operation once they have been installed. Any change the operator wishes to make that is not a preprogrammed option can46

damaged by short cycling if the unit is operated without a program. However, ROM-based programs should be limited to providing backup for startup and default conditions or executing very simple control strategies. A flexible application program language that follows the same basic rules as the programming language in standalone panels is needed to provide effective terminal controller programs. Some manufacturers have built in the ability to unbundle terminal controller points and operate them as if they were points connected to the stand-alone panel to which the terminal controller is connected. There are two problems with the unbundling approach that make it unworkable for typical terminal control applications. In most systems, unbundling is a cum-

Heating/Piping/Air Conditioning

on these products and they do not want them to compete with their more expensive stand-alone panels. This is a short-sighted view because the potential market for full DDC systems will expand enormously as the two products are better integrated into efficiently configured DDC systems. Opportunities ahead As more functional and versatile terminal controllers are developed, the opportunities for improving the comfort and economy of HVAC systems will open to entirely new areas. The recent ASHRAE standard for ventilation has been criticized because it does little to ensure that outside air is actually delivered to each of the buildings occupants. Using methods similar to the terminal control example earlier, full DDC systems can ensure on a zone-by-zone basis that

adequate outside air is supplied to every occupant and at the same time improve the comfort and energy efficiency of each building. Meanwhile, the development of DDC and associated technologies is continuing to reduce the cost of applying terminal control products. Electric quasi-modulating techniques provide good modulation characteristics at low cost for terminal reheat and other terminal modulation requirements. The costs of standard temperature, humidity, air flow, and occupancy sensing devices have all decreased substantially in the last few years. The prospects for expanding building DDC systems to terminal control is indeed very promising. But to be certain the promises regarding opportunities for full DDC are kept, users must encourage DDC system manufacturers to improve the function and flexibility of

their terminal control products and to develop operational features more consistent with the standalone panels to which they are connected. Once such powerful products are available, designers can use the added power to improve building comfort, air quality, and cost efficiency up to the levels owners and occupants now demand. Next month we will explore options in programming languages for DDC controls today. We will discuss the advantages and disadvantages of each basic approach to programming DDC panels and suggest where the future may lead system manufacturers in this critical DDC system feature. a
The author appreciates DDC manufacturers cooperation in providing photographs to help the descriptions in this article. The photographs have been chosen to be representative of the industry and are not endorsements of any kind.

. . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

DDC CONTROLS TODAY

PROGRAMMING LANGUAGES FOR DDC SYSTEMS

An examination of the current and future state of programming languages and the various options available for programming DDC panels

By THOMAS HARTMAN, PE, The Hartman Co., Seattle, Wash.

Star Trek movie finds Capt. Kirk, through no A fault of his own, mysteriouslv transported back to the primitive present. To correct the situation and return to his proper place in time, he solicits the use of a computer. When shown a powerful computer, one of Kirks lieutenants immediately tries to initiate a dialog by commanding Computer! After a moment of uncomfortable silence, the operator points out the mouse, which Kirks lieutenant raises to his mouth and again tries to command the computer verbally. I was reminded of that scene recently at a presentation of a new DDC (direct digital control) product. The vendor was nervously presenting what he thought were advanced programming features to a group of experienced DDC users. The questions and criticisms flew fast and furiously. Finally the exasperated rep asked, Well, how do you want to program DDC systems? One of the participants responded immediately, We want to tell the system what we want it to do, and we want it to understand

and do it. I never had a chance to ask that fellow if he got this idea from the movie, but I suspect it wouldnt make any difference. Virtually everyone who uses computers believes they could be vastly more friendly to use than they are. DDC operators are no exception. Until systems operate as the Star Trek crew expected, the pressure for improvements will continue to be very strong indeed. In the last few years, the DDC industry has learned a great deal about performance and the elements of successful programming languages. A number of ideas have been tried and many have been very successful. In this article we will examine the current state of programming languages for DDC systems, what the various options are for programming DDC today, and the benefits and drawbacks of each. Finally, we will look to the future and suggest a path for continued improvement of DDC system programming languages.

The beginning
In the early days of computerbased building control, most programming languages offered very few features and even less flexibility. As a result, the notion developed that the system operator

should be more a specialist in computers than in HVAC systems. Many systems were supplied with programs written at the factory. These programs were provided in a low-level assembly type language that allowed the operator only a few of what we call hooks into the system. For example, typical programs permitted the operator to define occupancy schedules and adjust certain set points, but the sequence of control was fixed and could only be changed by recompiling the program, which usually had to be done off site. Problems that could not be adjusted away with the built-in hooks required elaborate schemes to correct. Operators who were very knowledgeable in the operation of the computer could sometimes adjust certain database parameters, fooling the system into performing more satisfactorily. However, users became very frustrated by the inflexibility of these systems. Building control problems that seemed quite simple and straightforward often required elaborate measures and a computer specialist to solve. A flurry of activity took place by manufacturers, users, and the building design community to improve the success of computer-based building control systems. Some of the initial solu-

Programming languages

tions proposed (such as tri-services specification) failed because they tried to treat the symptoms and not the cause. But when the dust settled early in the 8Os, two new approaches to building control programming were being offered that allowed operators who were not skilled in computers to support building automation systems more effectively than ever before.

Line programming

One of the approaches to improved programming capabilities offered by some manufacturers was the ability to write sequences of operation in standard line-program formats. Line programming had been employed for many years in the general computing industry. The formats of these DDC languages look very much like the high-level general computing languages (such as BASIC) except CALCULATE MAXIMUM, MINIMUM AND AVERAGE SPACE TEMPS" that certain additional functions are added to permit the language to STMAX = MAX(ST1 ,ST2,ST3,ST4,ST5) STMIN = MIN(ST1 ,ST2,ST3,ST4,ST5) issue start and stop commands, STAVE = AVE(ST1,ST2,ST3,ST4,ST5) control outputs to PID (proporCALCULATE SUPPLY AIR SETPOINT BASED ON AVERAGE SPACE TEMP tional/integral/derivative) algorithms, tie into occupancy schedSASP = 65 - 3*(STAVE-STOBJ) ules, etc. Some languages are comADJUST SUPPLY AIR SETPOINT FOR PROJECTED HIGH OUTDOOR TEMP piled and some interpretive; but all offer similar flexible control capaSASP = SASP - (PHT-50)/5 bilities, and they can be easily deADJUST SUPPLY AIR SETPOINT FOR COLD DAY MODE OPERATION veloped and altered on site by the system operator. A sample line proIF CDM = ON THEN SASP = SASP + 2 gram in the form our firm typically ADJUST SUPPLY AIR SETPOINT FOR HIGH OR LOW SPACE TEMPS employs to calculate the supply air set point for simple fan systems is IF STMAX > 74 THEN SASP = SASP - (STMAX-74)*3 shown in Fig. 1. IF STMIN < 70 THEN SASP = SASP + (70-STMIN)*3 Some line-programming languages contain all the features of powerful high-level languages and LEGEND: include formatting features that make programs written in these SASP SUPPLY AIR TEMPERATURE SETPOINT languages quite readable. Some STMAX MAXIMUM ZONE SPACE TEMPERATURE newer releases also include addiSTMIN MINIMUM ZONE SPACE TEMPERATURE tional features, such as full-screen STAVE AVERAGE ZONE SPACE TEMPERATURE STOBJ SPACE TEMPERATURE OBJECTIVE (CALCULATED ELSEWHERE IN PROGRAM) editors and on-line error checking, ST1 -ST5 SPACE TEMPERATURES IN AREAS SUPPLIED BY AIR SYSTEM permitting operators to view, edit, PHT DAYS PROJECTED HIGH OUTSIDE AIR TEMP (CALCULATED ELSEWHERE IN THE change, and debug virtually any PROGRAM) CDM COLD DAY MODE (LOGICALLY DETERMINED ELSEWHERE IN THE PROGRAM) control sequence quickly and easily. NOTE: ..... ARE COMMENTS THAT ARE FOR THE BENEFIT OF THE OPERATOR, AND ARE IGNORED BY THE PROGRAM It is important to note, however, that there are wide variations among line-programming languages. Some are crude, inflexible, 1 Line program for calculating supply air temperature set point.

and very difficult to use, though their suppliers still claim they provide high-level line programming. As with other features of DDC systems, equals in languages do not exist. Users and consultants should make themselves knowledgeable about the features of any programming language to be supplied with a system before it is purchased. The primary advantage to line programming rests in its power and flexibility. Line-program functions have already proved themselves in solving diverse general computing problems. With the addition of a few special functions for building control, system designers and building operators find they have the tools they need to develop just about any control sequence(s) for

particular HVAC system control requirements. Another advantage of some line programs is that they are self-documenting. When a designer or operator determines that a program change is necessary, a printout of the program provides an accurate and fairly readable description of the new control sequence. And because line programs are in the form of general computing languages, many operators have already had some experience with this type of language at school or home. The primary disadvantage cited for line programming is that some operators have trouble understanding and writing line programs without specific training. Indeed, some of the line languages have ro!rr/lllrNlorl ,OK 7.1

Programming languages

long lists of rules regarding their use, and operators are often frustrated when they cannot easily make occasional program changes. However, line-based languages with fewer rules, which permit the use of comments and special formatting, are usually supported successfully by building operators. Another disadvantage of line programs is that software development for typical projects can become time consuming by requiring entire programs to be rewritten for multiple systems even though they all may operate very much the same. Fortunately, certain copying features and editing aids mitigate this disadvantage in the more advanced line program languages.

act functions provided by the function blocks included with the systern, the programming effort is very easy. The disadvantage of the function-block approach is that whenever sequences are required that do not match available function

point of a simple fan system as space conditions change. However, the line program in Fig. 1 also includes outdoor weather factors and could easily be changed to accommodate different relationships or additional factors-all in this one

TEMPERATURE LOOP RESET FUNCTION BLOCK NAME OF FUNCTION BLOCK CONTROL LOOP TO BE RESET DEFAULT TEMPERATURE RESET POINTS AND LIMITS POINT LOW LIMIT HIGH LIMIT

Function-block programming
A second approach taken to improve success with applications software involves refining the preprogrammed approach to give it some additional flexibility in a form that is structured specifically for typical HVAC applications. The manufacturers that adopted this approach decided to break down the factory programmed applications into small program blocks that can be linked together and have parameters assigned by the designer or operator. By assembling these preprogrammed blocks in various combinations and providing some flexibility in assigning points and parameters, manufacturers believe they can satisfy most typical DDC applications while maintaining a simple and easy-touse program format. Fig. 2 is a sample function-block program. This function block provides space temperature reset of the supply air temperature set point of a simple fan system. Some in the industry call this fill in the blanks programming because only a limited number of fields in each program block need to be entered. The primary advantage for function-block programming is its simplicity in standard HVAC applications. Indeed, if the control sequence happens to call for the ex-

RATE OF RESET LOW TEMP: <-ii----) HIGH TEMP: c-?----> LEGEND:

SASP SUPPLY AIR TEMPERATURE SETPOINT (NAME OF THIS FUNCTION BLOCK) AH1 SATLP ANALOG CONTROL LOOP TO BE RESET (A SEPARATE FUNCTION BLOCK) ST1 -ST5 SPACE TEMPERATURES IN AREAS SUPPUED BY AIR SYSTEM c-) AREAS OF FUNCTION BLOCK THAT CAN BE CHANGED BY THE OPERATOR

2 Function-block program for supply air reset.

blocks, the programmer must employ custom blocks or employ blocks for purposes other than those for which they were developed. This usually results in more complicated programs and reduced HVAC system performance. As a result of the relative advantages and disadvantages of the two programming approaches, line programming-based systems are usually far more effective in high-performance building applications that require more in-depth control strategies. Systems with functionblock programming are generally limited to applications employing simple or traditional pneumatic strategies. To see the differences in the two approaches, consider the programs in Figs. 1 and 2. The line program in Fig. 1 and the reset-function block in Fig. 2 are both intended to reset the supply air temperature set

program. By contrast, the function block cannot implement a number of the factors employed in Fig. 1. Function-block programs usually do permit linking blocks together for additional factors in calculations. However, linking causes the calculation to be broken into a number of small relationships that do not appear together on a single screen and are therefore difficult for the operator to follow.

DDC language trends

Because our firm focuses its efforts on high-performance building-control applications, we have favored DDC systems employing line programming. Line programs offer greater power and expanded functions that are necessary in our high-performance DDC projects. With our experience, we have long understood the problems and shortcomings of line-programming
n DECEMRER l!#X~

~(E41l~~i/~lllN~;/~fli~ONI~I1IONINti

73

Programming languages

languages in certain applications. Over the years, we have worked with users and manufacturers to improve the ease with which line programs can be applied to building control. We promote the concept of output-oriented programs wherein every calculation and command that directly affects an output is installed in one (and only

provided updates that simplify the operation of these languages by consolidating and simplifying their rules of application. These improvements are making line programs simpler to apply without compromising the power and flexibility that are inherent in their architecture. Meanwhile, to improve their

3 Typical graphic program expression.

one) section of the program. With output-oriented programming, the operator can quickly trace any control path and adjust the program easily when a point or mechanical system is not operating as desired. In 1988, our firm released a guideline for line-based program languages called the operators control language (OCL). The OCL guide was intended as a functional specification for features our firm and our clients desired in line-program languages. A discussion of the merits of good operators control languages appeared in the September 1990 issue of HPAC. O C L Spells Freedom, by Ken Sinclair, concluded that advanced line program-based DDC systems offer greater flexibility and are easier to use than other approaches. Our firms experience agrees with Sinclairs conclusion. Building operators are easily trained in the operation of most line programs and are able to use many of the advanced features to make their maintenance and troubleshooting duties easier to perform. This is possible because most line program-based systems have recently
74
DECEhlf3EH 1!2!Kl

range of applications, function block-based systems continue to add to their libraries of blocks. But large libraries reduce their simplicity, the primary advantage of function-block programs. These trends are eroding the few advantages that function-block programs can offer when compared to the more advanced line programs.

New approaches
The inevitable demand for systems that operate as Capt. Kirk expected continues to drive manufacturers to develop innovative approaches to building-control applications programming. Several DDC system manufacturers have committed themselves to releasing DDC systems that employ a new approach to the applications programming language-graphics programming. This approach utilizes the powerful graphics capabilities of modern PCs to permit the programmer to sketch out a flow chart for the control sequence desired. A program in the PC is then employed to translate this sketch into a program (usually written in a high-level line language), which is

down-loaded to the selected DDC stand-alone panel. An example of a screen containing a graphics program is shown in Fig. 3. Note that the entire figure would be built by the operator with system points, variables, and a library of mathematical and logical functions. If the operator desired an additional space temperature for the calculation in Fig. 3, he could easily add it by choosing the appropriate system point and sketching a connection to the calculation block. The idea behind graphics programming is to find a way to provide the power and flexibility of line programming with the simplicity of function-block programming. Essentially, the programmer can develop custom function blocks to meet the needs of any particular project application. The idea is enticing, but there are some potential problems with the concept, including: 0 Program execution uncertainty--Once a programmer has developed a graphic, a translator program has to be employed to convert the picture into an executable program to be down-loaded to the appropriate stand-alone panel. Experienced programmers have long realized that there are some special considerations required when writing programs to ensure that they execute as expected. For example, assume the following simple sequence of operation is desired to start supply and return fans. If the weekly schedule is on, start the return fan, and after a 30-sec delay, start the supply fan. Fig. 4a shows how a line program can be written to execute that logic sequence. If the data point RETURN_FAN is turned on only after the entire block is executed, the program will execute properly. However, if RETURN_FAN is turned on as soon as that line is executed, it is clear that SUPPLY_ FAN may be started an instant after RETURN_FAN is started. A better way to write the program is

n H~,~Tl~~;;llllNC/All~CONI~ITIONlS~~

I
CONTROL SEQUENCE: If the weekly schedule Is on, start the return fan, and after a 30 second delay start the supply fan

FIGURE 4a: UNCERTAlN EXECUTION ORDER IF WEEKLY_SCHEDULE = ON THEN BEGIN DOEVERY. SO SEC S T A R T RETURN_FAN ,_/ I F RETURN_FAN O N T H E N START SUPPLY_FAN ENDDO END.

FIGURE 4b: RELIABLE EXECUTlON ORDER IF WEEKLY-SCHEDULE = ON THEN BEGIN DOEVERY 30 SEC IF RETURN_FAN ON THEN START SUPPLY-FAN START RETURN_FAN ENDDO END

L E G E N D
WEEKLY_SCHEDUlE DOEVERY 30 SEC
RETURN-FAN

SUPPLY-FAN

WEEKLY SCHEDULE (SET UP ELSE WHERE) DO LOOP THAT IS EXECUTED ONCE EVERY 30 SECONDS DlGITAL OUTPUT THAT STARTS THE RETURN FAN DlGITAL OUTPUT THAT OPERATES THE SUPPLY FAN

NOTE: The underline Character Is used to join words to make single terms for point names or variables. This follows normal programmingconvention.

4 Line programs to execute simple tan start sequence.

shown in Fig. 4b. Although the order of items in the program is reversed from what we might expect, it is clear that SUPPLY-FAN will never be turned on less than 30 sec after RETURN-FAN. While programmers can expect translators to offer some degree of protection against such translation errors, they can never be entirely certain that the translator is not responsible for operational errors. What is a programmer to do if a program does not appear to be executing as pictured in the graphic? The programmer will inevitably be required to inspect and troubleshoot the translated program if the error cannot be found by reviewing the graphic screens. The line program that is developed by the translator is likely to be very difficult to review because such programs do not have the form and logical flow that programmers typically provide in their line programs. Nor are such programs likely to

have comments or formatting devices that make them easy to read. Graphic programming may seem to be much more straightforward than line-based software, but any debugging effort can become much more complicated, especially if the operator desires to write more complex control algorithms that make the fullest possible use of the energy and comfort-enhancing capabilities of DDC. l Display limitations- Another problem with graphic representation of control programs is they are bulky to display. Note that the averaging calculation of Fig. 3 requires only a single line to represent in the line program of Fig. 1. When several pages or more of graphics are required to represent a control sequence, the sequence can become very difficult to review because the operator cannot look at the whole program at once. Recently, a client of ours experimented with a prototype of a new

graphic program-based DDC system. He translated a line DDC program used to control his buildings air systems to see if the graphic representation offered any advantages. To his surprise, he found the graphic program required eight screens of graphic representation and seven pages of constants and gains to duplicate a line program that occupied (with comments) only one and one-half pages. The resulting program was far more difficult for him to review than the original line program. l Operator interface costMost line-based programming languages can be operated directly or over phone lines with a simple PC and low-cost software. However, the complex hardware and software required to create, test, translate, and compile graphics programs can add substantially to the cost of each such terminal. Many users have developed system-support mechanisms that include offsite access to the system via telephone modem by the engineer or several operators (at night). These multiple-terminal operational schemes can be much more costly to implement because simple PCs may not be capable of the performance needed to accomplish graphic programming. Furthermore, the extensive proprietary software required for graphics programming can become costly, and it is possible a copy will have to be purchased for every computer that may be used. Such costs could substantially impact the flexible DDC operating schemes employed by many users. Graphic programming is another serious attempt to provide DDC system operators, whose primary training and knowledge are in mechanical systems, with the ability to write and adjust high-performance custom DDC applications programs. Today, DDC system users and manufacturers alike understand the need to implement DDC systems that are both functional and easy to use. This represents an important change in operations

Programming languages

philosophy from a few years ago when most manufacturers (and many users) believed DDC system operators should not be permitted direct access to control programs.

knowledgeable operator of an older ticularly adept at understanding system. His ideas and concerns how effective new digital techsuggested to me that however bril- nologies can be until we have some liantly this operator had employed experience working with them. This his old system to provide comfort makes it difficult to look very far in and energy efficiency, his mode of the future with clarity. However, we thinking was now limited by the can look at the issues that need to be Looking to the future operational capabilities of that sys- resolved to continue to improve the While Capt. Kirk might not be impressed with the improvements tem. It is likely he will be able to success of DDC programming lanutilize fully the capabilities of the guages, and this may provide some in DDC programming capabilities new DDC system only after he has answers for the most likely next imthat have been made in the last few developed an understanding of provements. years, DDC system operators them through experience once the should be. Comparing todays DDC programming features with those Combined line and graphics new system is installed. Our firm, and many of our This operators problem is probonly a few years old makes one realize the enormous strides that the ably universal to the building de- clients, continue to believe that sign industry-and to other indus- line-based programming languages industry has made. Recently, I distries that utilize digital technol- provide the best method to develop cussed program specifics for a reogies as well. None of us is par- DDC programs that execute effecplacement DDC system with a very tive energy and comfort-enhancing control strategies. However, we believe it is possible that combining FAN ON/OFF CONTROL certain features of both line-proDOEVERY 1 MIN gramming and graphics-programI F OCCUPlED_MODE = ON OR COOLING_PURGE = O N O R ming techniques may produce a WARMUP-MODE = ON THEN START SUPPLY-FAN ELSE STOP SUPPLY-FAN format for control programming ENDDO that has advantages over the allprogramming techniques generally HEATING VALVE CONTROL available today. IF WARMUP-MODE = ON OR HEATING_REQD = ON THEN BEGIN Earlier in this article, it was illusPID_HTG:SETPOINT = SUPPLY_SETPOlNT_CALC trated that line-based programs HEAT-VALVE = PlD_HTG can be the most effective way to END represent many kinds of matheELSE HEAT-VALVE = 0 matical and logical expressions beCOOLING VALVE CONTROL cause they can make such expressions clearly and concisely. IF MECH_CLG = ON THEN BEGIN However, a problem with line proPID_CLG:SETPOINT = SUPPLY_SETPOINT_CALC grams is the lack of clarity in repreCOOL-VALVE = PID_CLG senting major logic sequences. Fig. END 5 shows the system-level logic that ELSE COOL-VALVE = 0 might be employed to control a MIXED AIR DAMPER CONTROL simple fan system. In Fig. 5, the calculations and IF SUPPLY FAN = OFF THEN MIXED-DAMPER = 0 ELSE lower-level logic can be considered IF MECH_CLG = ON THEN IF ENTHALPY_RA = ON THEN MIXED-DAMPER = 5 to have been made elsewhere and ELSE MIXED DAMPER = 100 ELSE are represented by their resulting IF COOLING_PURGE = ON THEN MIXED-DAMPER = 100 ELSE BEGIN IF MINIMUM_OA < SUPPLY_SETPOINT_CALC THEN variables. Note that the logic conPID_MAD:SETPOINT = MINIMUM_OA ELSE BEGIN trolling the supply fan is very readPID_MAD:SETPOINT = SUPPLY_SETPOlNT_CALC able in this format. However, the MIXED-DAMPER = PID_MAD logic for the mixed air dampers is END not so simple and therefore someEND what difficult to follow even though NOTE: OCCUPIED-MODE, COOLlNG_PURGE, WARMUP_MODE. HEATING_REOD, MECH_CLG, SUPPit is concise. LY_SETPOINY_CALC, ENTHALPY _RA, MINIMUM_OA are all variables representing logical decisions or calFunction-block and graphics culations that are not shown in the program. The program represents the logic flow of a simple fan system that is represented graphically In figure 6. The underline character Is used lo join words to programming as they exist today make single terms for point names or variables. This follows normal programming convention. are also weak in representing system-level logic because they are not concise. In these programs, logic 5 Line program showing control logic for Simple fan System.

Fan system start/stop

Heating valve

5%

Mixed air damper

/ Occ;w;fim$de t Line program module-accessible for viewing or editing

by placing the cursor on the module and clicking the mouse

program modules whose programs are written in line program +5+~/@ Graphic and accessible the same as line-program modules 6 System logic for simple fan system.

Off = Real-time logical or calculation result from line program module -Lines of calculated values -Logic flow line-real-time FALSE indication -Logic flow line-real-time TRUE indication

paths are provided, but each . . component represents a very small portion of the overall program. Therefore, the overall system logic can be represented only after one has pieced together a number of individual blocks. The DDC industry has not yet found an effective means to represent system-level logic. This is a shortcoming of every programming format in wide use today. With the features now generally available in PCs, it may soon be possible to combine line-program blocks with graphic representations of system logic to provide a format for improved DDC control representations. Fig. 6 shows the logic for the simple fan-control program of Fig. 5 in graphic form.

Note that the graphic overview is effective in representing the system logic. The circumstances under which the dampers and heating or cooling valves are operated can be reasonably deduced from the diagram. The labeled rectangular blocks are line-program modules. Imagine that this graphic representation can display real-time result(s) for each line-program module with the current lines of control logic highlighted in special colors for true or false. Assume further that the contents of any of the lineprogram modules can be called up for review or editing simply by clicking on the chosen module. With such a programming technique, the operator could quickly isolate and troubleshoot the exact

areas of logic in effect when problems develop. Such a programming scheme as represented in Fig. 6 may provide advantages over both line and graphics programming while mitigating many of their disadvantages.

Future languages
The programming concepts shown above may be a natural continuation for recent improvements in DDC system programming languages, but a wide variety of other options are possible as well. Whatever the next steps in programming languages may be, they will be successful only if they work toward solving the following current problems: l Concise representation of ef-

Programming languages

fective DDC control strategiesThe key to success with DDC controls is not to emulate traditional pneumatic controls but to use the power and flexibility of DDC systems to provide new, in-depth modes of control that result in enhanced comfort and energy-efficient operation. Such modes of control can be supported only if they are provided in programs that can be understood and diagnosed by system operators. It is not enough to break the program into such small pieces that the overall concept is difficult to determine because the operator has trouble assembling all the pieces together at once. Any language must include representation(s) that show very clearly both the overall concept and the calculation/logic pieces that constitute that concept. l Real-time indications of program calculations and logic

paths-One of the most powerful tools available to a DDC system operator is the ability to watch programs as they execute and see the results as they are calculated. This programming tool is typically available only with interpretive languages. However it is accomplished, languages must be developed that enhance the operators ability to view real-time calculations and logic while the DDC system is operating. This feature allows the operator to check programs easily when their operation is suspect. l Few and simple rules to govern the language-The more rules that govern how a programming language can be applied, the more difficulties the operator has trying to support the programs. Early applications program languages had many rules governing everything from the use of integers and floating point numbers to the use of

math in conditional statements. Manufacturers have done a good job issuing revisions that have simplified language rules for many existing DDC languages. More needs to be done, however, and new languages, whether graphic- or linebased, should be as free of restrictive rules as possible. l Low cost- Improvements in programming languages cannot be permitted to reverse the trend toward lower-cost DDC systems. Manufacturers should consider the enormous market potential for their products when they have combined sufficient function and ease of operation in a package that competes with pneumatics on first cost. Full DDC systems are usually 10 to 30 times larger (in terms of system points) than the DDC systems that go into many buildings today. Retrofit opportunities are even greater. Energy costs are now high

enough that most building owners can find a very attractive rate of return in an investment of between $1 and $2 per sq ft for complete HVAC and lighting control retrofit. If full DDC systems that provide the comfort enhancement and energy reductions of advanced control strategies can be implemented at these costs, the industry will experience an enormous growth in volume over the next few years that will help pay for some of the development requirements.

Summary
The DDC industry has made substantial steps over the last few years to improve the power and flexibility of the control languages supplied with their systems. This is a primary reason DDC systems are better accepted today than ever before. The types of control languages commonly available for program-

ming DDC systems today include line-based programming languages, function-block programming languages, and now graphics programming. Each of these approaches has certain advantages when compared to the others for specific applications, but it is clear further improvements are still needed, particularly improved representations of system logic. When considering programming language improvements, manufacturers should work toward approaches that permit the more indepth strategies possible with DDC control to be represented clearly and concisely and provide methods whereby real-time logic paths and calculation results can be displayed and reviewed as the program is operating. As discussed in last months article, the acceptance of full DDC systems has become a reality by users who wish to have their buildings

perform more effectively than they can with traditional controls. However, to ensure that this higher level of performance can be installed and will be maintained, better performing and more easily supported applications languages need to be developed. As the power, flexibility, and ease of implementing DDC applications programs grow, the demand for DDC products will grow also. fi
The Hartman Co. intends to update its operators control language (OCL) guide to DDC system programming function in 1991. We request comments from all in the industry who are experienced in high-performance DDC systems on the subject of DDC control languages and related items. We will be happy to make a copy of the current OCL guide available to such individuals at no cost. Please write to: OCL Guide, The Hartman Co., 1016 North 36th St., Seattle, WA 98103.

TERMINAL INPUT AND OUTPUT DEVICES

A discussion of the selection and specification of input and output devices as well as some new, inexpensive approaches to l/O device configuration

By THOMAS HARTMAN, PE,

The Hartman Co., Seattle, Wash. n the two previous articles of this series, I have discussed current trends in DDC sysII tem architecture toward what many in the industry call full DDC systems. Full DDC systems are those that incorporate terminal control as well as system control. To develop successful high-performance configurations of full DDC systems requires special attention to some of the features of the DDC system. The previous articles focused on evolving capabilities of terminal controllers and programming languages. They suggested ideas for improvements in those areas that may be beneficial for this current trend toward full DDC systems. The purpose of these articles is to encourage a dialog within the industry aimed at defining features useful in high-performance applications of full DDC systems. The

view of many who are successful with DDC systems is that the industry is flying blind. Manufacturers are developing products based on criteria that may be flawed or lack relevance. Many A/E firms continue to employ DDC specifications and procurement procedures that display a profound lack of knowledge about the features that are most crucial to a successful DDC system. Recent research conducted by trade associations on DDC seems to have reached at least as many wrong conclusions as correct ones about what ingredients are necessary for successful DDC implementation. .i the DDC revolution continues, our industry needs to attract and make better use of the ideas from those associated with successful high performance DDC systems. This final article will focus on input and output devices employed for terminal control. We will start by considering how the evolution to full DDC is placing enormous importance on how the designer

chooses terminal input/output devices. Then, we will look at some of the newer hardware that may reduce the cost of full DDC systems without compromising their quality. Finally, we will explore some approaches DDC system manufacturers could incorporate in their products and interface schemes to provide more functional and economical terminal control. Terminal control devices DDC system inputs and outputs (commonly referred to as points) have increased enormously in significance as terminal control opportunities have encouraged the use of full DDC system configurations. DDC systems that control only fans, chillers, and central plant equipment often require 200 points or less to control a large size building effectively. Full DDC configurations that incorporate lighting and temperature control for every space may require 5000 or more points. This 25-fold increase in points has changed the rules by which DDC systems are
n JANUARY 1991

HEATING/PIPING~AIRCONDIT~ONING

91 I1

I/O devices

configured to be cost effective and perform successfully. The role of input/output device cost has changed from a minor item to the leading factor in determining DDC system cost effectiveness. A rule of thumb is that the installed cost of a full DDC system must be kept under $200 per point to have a chance of being cost effective when compared to other approaches from a purely economic standpoint. This objective can be achieved only if the system designer pays close attention to the selection of input and output devices employed for terminal control. A strong emphasis needs to be placed on the terminal control input/output devices because these often represent more than 90 percent of the total input/output devices in a full DDC system. The designers task is to select components that provide desired sensing or control characteristics at the lowest possible cost. To do this, the designer must first determine what characteristics are realistic. For example, precision RTD temperature sensors are still commonly specified for space temperature sensors. Such devices can cost up to $100 or more apiece. But the limiting factor in space temperature sensing accuracy is almost never the sensing device. The end-to-end accuracy of space temperature sensing for DDC systems involves a number of considerations, the most important of which is usually the space itself. Consider that typical occupied spaces in buildings are supplied with air that is 20 F warmer or cooler than the space. As this air is circulated to mix with the space air to add or reject heat, temperature gradients within spaces are often significant. If several independent instruments are placed within a few inches of a space temperature sensor, their readings will frequently differ by 0.2 to 0.5 F. Stable thermistor-based space temperature sensors cost only a fraction of precision RTD sensors yet provide

stable operation and precision that is far better than the 0.2 to 0.5 F precision that can be realistically achieved in space sensing applications. In selecting input/output devices, designers should remind themselves that if their full DDC systems are configured such that they cannot be economically justified over pneumatic alternates, their clients and the building occupants are the big losers. Compare the precision and stability of pneumatic thermostats and humidistats to those of the most economical electronic devices. If DDC systems cannot be configured to meet budgetary and/or cost benefit requirements, the lower cost alternative is usually pneumatic controls. But pneumatic devices provide control that is several orders of magnitude less precise and also require constant calibration to maintain even that low level of precision. Such an understanding of the role and importance of cost considerations is required for a designer to make reasonable choices for terminal input/output devices. Terminal output devices such as damper actuators and reheat valves have often been considered to be perfect applications for pneumatic actuators. However, when the applications are scrutinized more closely, pneumatic devices are often not a good choice at all for terminal control applications. Furthermore, pneumatic devices do not interface well with electronic control systems. As a result, a number of new economical actuation devices have been growing in popularity over the last few years.

another. The increasing demand for electric control devices that can be interfaced directly with DDC systems appears to be improving the situation. Electric actuators are falling in price and are becoming more standardized on the 0 to 10 v DC control signal to modulate their position. An electric modulating actuator of the type employed to control a VAV box damper is shown in Fig. 1. This type of actuator is now becoming available for $100 to $200. Electric actuators have several features that make them desirable for the control of small dampers. Generally, a unit modulates by comparing the modulating voltage from the DDC system with a voltage that depends on the position of the actuator. Unlike pneumatic actuators whose force is related to the output and position, most electric actuators provide constant force (or torque) at all outputs and stop only when the position corresponding to the DDC voltage is reached. This eliminates the uneven movement that is often associated with pneumatic actuation. Some electric actuators also maintain constant torque whether running or stalled. This feature is particularly useful in applications that require torque to be applied at the end stops to hold dampers fully open or fully closed.

Two-motor actuators
A more economical electric zone damper actuator employs two small constant-speed clock motors connected together. These, as do the previously described actuators, rotate the shaft collar or clamp through a gear drive. The two-motor type actuator is interfaced with the DDC system through two digital output points. The two motors run in opposite directions. One motor drives the damper open, the other drives the damper closed. This device is commonly called a floating control actuator. An example of a two-motor actuator is shown in Fig. 2. This type of

Electric actuators
Modulating electric actuators have been manufactured for many years, but they have not been widely employed in North American HVAC applications. Until recently, these devices were generally comparatively expensive and not noted for durability. Electric controls of different manufacturers were often incompatible with one

1 Electric actuator 0 to 10 v DC modulating signal, 24 v (AC or D C ) power.

actuator typically costs under $50. The speed of rotation of each motor is constant and generally geared to be slow-l to 3 min to rotate fully each direction. In the past, reliability problems with the two-motor actuator some-

times resulted when the control system provided continuous power to the controlling motor at the fully open or fully closed position. The resulting locked rotor condition appears to have led to abnormal failures of the motor or gear train. Re-

liability problems also may have resulted from the mechanical switching mechanism that operates the motors. Switches or relays in the control circuit are subject to a great deal of cycling, and unreliable contact closure may have also caused some problems in the past. Although many of the two-motor actuators are built much sturdier than they were a few years ago, proper operation by DDC controllers have also contributed to reduced reliability problems. Because DDC systems today typically drive their outputs with solid state triacs, there are no mechanical components involved in running or switching the motors. To eliminate locked rotor problems and to provide improved control, several DDC manufacturers have provided special interfaces for the two-motor actuator so that it looks like any other analog point to the operator. The interface automatically calculates the position of the actuator by keeping track of the time each motor has operated. When the actuator reaches one extreme or the other, the position is automatically reset, and the power to the motor is shut off. This approach works well and usually permits two-motor actuators to operate by standard PID controllers just like other analog outputs. Because each motor is shut off at extreme positions, locked rotor problems are eliminated. The two-motor controller provides economical modulating control that is adequate for most terminal control applications_ These devices are typically configured with low-speed gear trains because hunting problems can result when the devices are controlled by simple floating-point controllers. The low speeds may not be adequate for opening or closing in certain circumstances. However, with output controllers that permit PID control of the two-motor actuators, much higher speeds can be controlled with ease, and some two-motor actuators are now becoming available that operate at greater speeds.
# JANUARY 1991

HEATINO/PIPINO/AIRC~NO~T~~N~NO

93

I/O devices

Air velocity sensors

One of the most noticeable advantages of DDC terminal VAV control is the ability to provide air balancing of terminal boxes with the DDC system. DDC air balancing is much easier than traditional methods. The maximum and minimum air flows for each box are set by the DDC system, and each VAV box damper operates to provide a specific air flow within that range based on space conditions. Duct pressure compensation is automatic. To achieve all these features requires an air flow sensor at each VAV box, connected to the DDC system. At present there are two approaches employed by the HVAC industry to measure air flow at terminal boxes. Both actually measure air velocity, which is mul-

2 Two-motor electric actuator, 24 v AC.

tiplied by a constant (the constant includes factors for duct area and flow profile compensation) to obtain air flow. One approach employed to determine air velocity at terminal units is to measure the velocity pressure of the air through the use of a pitot tube arrangement. Some of these velocity pressure devices actually measure the average velocity pressure with a series of upstream orifices to help compensate for the irregular velocity profiles

that occur in many ducts. The velocity pressure air flow measurement technique has one serious limitation. At low air velocities, the velocity pressure of air is very small. Much literature recommends the velocity pressure technique be limited to air velocities above 300 fpm. Tests conducted by our firm found reasonable accuracy somewhat below that figure under certain conditions, but the bottom limit of accurate flow measurement is still above the lowest velocities encountered at VAV terminal boxes. Another problem with velocity pressure type sensors is that the differential pressure/electric transducer required to convert the velocity pressure into an electric signal and the sensing equipment can be costly. Some DDC manufacturers include the pressure/electric transducer with their box controller, which seems to reduce transducer cost dramatically. Controllers incorporating the transducers are usually competitively priced with those not including transducers. A second approach to measuring air flow is to determine the air velocity by measuring its capacity to provide cooling. Methods that employ this approach are sometimes referred to as hot wire or heated thermistor techniques. There are a number of variations in operation from manufacturer to manufacturer and depending on the exact devices incorporated in the design. Fig. 3 shows one type of heated thermistor air velocity sensor. Essentially, one device is heated while some mechanism determines the cooling effect of air passing over it by measuring the actual temperature of the device (or the current required to maintain the device at a constant temperature). The cooling effect approach has one big advantage over the velocity pressure technique: it can accurately measure air velocities as low as 50 fpm. The most economical of the cooing effect velocity sensors employs several thermistors. Exact oper-

ation varies, but a popular technique is to employ one thermistor to measure the ambient temperature of the air stream. The second thermistor (slightly downstream of the first) has a voltage applied to it, and the current required to maintain the voltage is measured. Because the resistance of the thermistors varies with temperature, the temperature of the heated thermistor can be calculated. The combination of ambient temperature, heated thermistor temperature, and energy to the heated thermistor can be used to calculate the air velocity in smoothly flowing air. This sequence may seem complicated, but these comparisons are usually made in a small IC chip. The cost of thermistor-based cooling effect air velocity sensors is about $25 or less, and some of them automatically provide leads that can be connected to another input of the DDC system to provide the air temperature as well. Most do not provide a linear output. Far more expensive devices are required for that. But most DDC systems offered today have the ability to linearize such signals with calibration tables that are included in their input point databases. A disadvantage of the cooling effect air flow measurement technique is that it provides a reading of air flow at only one point in the air stream. To read more points for average air flow (which is easily accomplished with the velocity pressure technique) requires an array of a number of individual sensors. Thus, a sensor designed to measure several points can become expensive. Our firm has also conducted performance tests of air flow measuring devices. The tests concluded that by following some basic rules for sensor location, we can employ a single-point velocity sensor to measure air flow in the duct sizes typically employed for VAV terminal units with adequate precision. In addition to sensor location, these tests found that how the velocity-to-flow factor is determined

is important in achieving accurate measurements. Because they generally offer the lowest cost air flow measurement option, thermistor-based air flow measurement has a jump on other techniques. However, thermistorbased velocity sensors have pitfalls that designers should consider when selecting air flow devices. The low-cost thermistors employed in HVAC applications are generally stable and drift-free in applications up to about 300 F. Above this temperature many are subject to drift. Because the heated thermistor in an air flow sensor usually operates at about 300 F, some in the industry have expressed concern that periodic recalibration of air flow sensors may be required. I have seen no conclusive evidence on the subject of drift for heated thermistor sensors. However, most of the installations we are familiar with are only several years old. Meanwhile, manufacturers are very short on facts that might prove drift will not be a problem. One solution to the drift problem is to employ sensors that use stable RTD sensors in hot wire configurations. Another is to select higher quality thermistors that are not subject to drift at elevated temperatures. Glass-encapsulated thermistors are generally very stable at higher temperatures. A recent entry into the air velocity sensor market utilizes glass-encapsulated thermistors that are still priced at about $20 apiece in quantity.

sensor

3 The photo on the left shows a heated thermistor-type velocity sensor and

integrated circuit chip. The upper diagram shows dimensions in millimeters. The diagram on the right shows connection to IC chip.

Other sensing devices

Temperature sensors, air flow sensors, relays, and damper actuators currently represent the vast majority of terminal end devices typically associated with DDC terminal units. A number of other devices will become widely used as high-performance DDC systems gain in popularity. Occupancy sensing is becoming standard fare for high-performance DDC systems as lighting control is integrated with HVAC terminal control strategies. The simplest

type of occupancy sensing device is a button on the space temperature sensor that occupants push when they arrive at the zone. When the button is pushed, the DDC system turns on the lights and establishes an HVAC occupancy mode for the zone. Typically, the zone remains in the occupied mode for the remainder of the day if the button has been pushed during normal working hours, or for a predetermined time if it is activated during off hours. Many manufacturers now incorporate such buttons in versions of their temperature sensors. The temperature sensor input is also used by some to transmit the occupancy signal to the DDC system. The result is that this type of oc-

cupancy device can be very inexpensive to provide-adding only a few dollars per zone to the cost of the system. A more effective, but also more costly, occupancy sensing device is an infrared or ultrasonic sensor. Such sensors-usually mounted on the ceiling or high on the wall-detect movement and can thus be employed to establish occupancy automatically as people arrive at each zone. This type of occupancy sensor costs about $50 to $100. The added cost over pushbuttons can often be justified because the motion type sensor offers greater energy savings and less inconvenience to building occupants. It is a nuisance for zones to lapse out of occupancy while still ocrorrlrfllllY~orr pug. I(X)
n JANUARY lew 95

HEATING/PIPING/AIRCONDITIONING

I/O devices

cupied, so most pushbutton oc- valves are actuated by thermostatic cupancy strategies employ long pe- elements similar to the bimetal deriods of occupancy each time the vice that regulates the water tembutton is pushed. If someone comes perature in automobiles. But ininto an office to pick up some pa- stead of water temperature, these pers, the zone may remain in the actuators are energized by a small occupied mode for several hours or resistance wire that is heated by all day. But an occupancy sensor just a few watts of power. The can automatically return the zone power requirement is small enough to the unoccupied mode after just a that the valves can be powered directly from the triacs that drive few minutes. Other devices that are growing in digital output points in most DDC popularity are humidity sensors terminal controllers. While these and various kinds of air quality valves are generally used as two-posensors. Such sensors are not re- sition valves, they can be moduquired in every zone. But sensors lated by pulse-width modulation of that provide VOC (volatile organic the digital output points to which compounds) and CO, readings in they are connected. These valves typical and critical areas can pro- cost less than many pneumatic acvide useful information to the tuated reheat valves and have a building control system. The trend good record for reliability. Similarly, electric reheat coils is clearly toward considering air quality control as important as can be modulated by pulse-width temperature control. Most sensors modulation of a digital output sigthat measure the quality of the nal to solid-state power relays. Fig. b u i l d i n g e n v i r o n m e n t a r e b e - 4 shows a typical connection for coming very economical as new in- such modulation devices connected expensive technologies are devel- to a single DDC digital output. To make effective use of such oped to accomplish the meamodulation techniques, DDC prodsurements. ucts require the capacity for operators to configure digital output Whats needed next The quality of typical input/out- points to look like analog points. put devices for DDC terminal con- This is already done by manufactrol is improving while costs are turers who configure two-motor falling. These are good trends, but electric actuators to look like single further improvements are possible, analog outputs. Unfortunately, and cooperation with DDC system these configurations are not now manufacturers is required to see very flexible. Rather, they are typithem through. There remain a cally incorporated into dedicated number of possible improvements controller outputs that are generin terminal input/output devices ally unusable for any other purthat can increase the cost effec- pose. As was pointed out in the article tiveness of DDC systems in the on terminal controllers, developing short term. products to be more flexible should be the goal of every DDC system Modulating digital outputs A variety of electrical actuation manufacturer today. Flexibility in approaches are offered for reheat points configuration is extremely valve operation. Those that involve helpful to operators. With a configmodulation are often more ex- uration option that permits a syspensive than pneumatic alternates. tem to be easily configured with However, several manufacturers digital outputs operating as analog points, the devices shown in Fig. 4 offer small electric two- and threecould be operated with standard way valves that employ a very inexpensive method of actuation that PID loop control to provide effeccould be easily modulated by a tive modulating terminal control DDC terminal controller. These despite the fact that they are digi-

tal output points. DDC system suppliers and manufacturers appear at times to be poorly informed about interface options for their systems. This is understandable since they generally do not manufacture many of the input/output devices employed by their systems. However, if a greater effort were made to seek out effective and economical I/O device options for their DDC systems, controls companies would find a substantially greater demand for their terminal control products. Those manufacturers who are willing to spend the effort to provide effective and flexible interfaces to operate special I/O devices in a manner that is consistent with more standard devices are the ones

DDC system

Thermo-electric actuated valve

r - i

Valve control current

Modulating valve on digital output

DDC system
0

-ISolid state relay

Digital output (Triac)

1 f==

Electric resistance duct heater

IJ Control current Power current Modulating electric duct heater

4 Typical connection for a modulating valve on digital output and modulating electric duct heater connected to a single DDC digital output.

that will dominate the full DDC market in the years ahead.

Terminal DDC input/output devices are becoming more effective and lower in cost. Unfortunately, many DDC systems presently do not fully exploit low cost terminal control concepts. As a result, many DDC systems cannot provide costeffective full DDC for high-performance applications. Such DDC systems are losing out in a very important new DDC market. In these articles, I have suggested three areas that need the attention of manufacturers, engineers, and users to exploit the growing full DDC market fully. These are DDC terminal controllers, programming languages, and terminal control input/output devices. The views I have expressed have been developed from our firms experience and include ideas we have gleaned from our more successful clients. There are many other viewpoints and ideas about the best way to configure full DDC systems for high performance applications. All who have successfully implemented cost effective DDC systems should be heard. Important upcoming forums to discuss high-performance DDC systems will be conducted at AEEs WORLDCON90 this spring and ASHRAEs annual meeting this summer. Its time those who are associated with successful high-performance DDC systems get together and share ideas so manufacturers and designers can get a better picture of what users need to continue their successful trends into the era of full DDC systems. The primary purpose of this article series was to stimulate much needed discussion within the industry about full DDC system function. Everyone stands to benefit from such an exchange as the HVAC controls industry attempts to take the two giant steps to full DDC system configuration and high-performance DDC system operation. Q

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..~..........................

HIGH-TECH SYSTEMS

PROCURlNG HIGH-TECH MECHANICAL SYSTEMS

Alternative purchasing procedures for engineers and owners who are considering procuring advanced technology components for new or existing building mechanical sys terns
By THOMAS HARTMAN, PE, The Hartman Co., Seattle, Wash. ow do engineers ensure that the advanced technology building systems I3 and components incorporated in modern buildings are purchased effectively? Our firm designs high performance DDC systems for building controls. We have for many years worked with owners, engineers, and control vendors to develop more effective procurement techniques with some significant successes-the cost per point of systems employing improved procurement techniques is well below the industry average, and the achievements of these systems are far above the industry norm. Recently, however, I realized how unpolished the process still is in many building construction projects. In a discussion of requests for proposal procurement for advanced technology mechanical systems, I was explaining to a project engineer that typical bidding processes were not effective because equals simply do not exist among most high-tech building systems. The faraway look in the project engineers eyes let me know it was time to stop talking, so I asked his views. He stated, We name names. When we look at a project, we degineer. But my experience is that such techniques often result in control or other high-tech systems that are not the best value and often do not meet the functional expectations of the client. Furthermore, the problems with procurement concepts such as these are increasing dramatically as the use and level of technology involved increase. Fig. 1 shows the increasing role of high-tech systems in our firms projects. We believe the use of advanced technologies in building mechanical systems will be skyrocketing by the turn of the

termine which of the local contractors can be trusted to deliver on the size and level of complexity involved and limit the bid to those firms. It simplifies the process for us. In another recent discussion, a design engineer unabashedly disclosed that he viewed his role as a specifier of the advanced technology portion of the project and implied that the actual design was best accomplished by the supplier who won the bid. Indeed, these approaches do simplify the process for the design en-

1 Historic and estimated trends for high-tech percent of total mechanical budget for The Hartman Co.
HEAT~NG~P~P~NG~A~RCOND~T~ON~NG n NovEbwmI99i 37

High-tech systems

century. My criticisms of procuring advanced technology systems with the traditional specification-bidding process fit two basic categories. First, purchasing equipment for which exact equals do not exist is not genuinely competitive and often results in poor value for the client. The cost penalty is potentially enormous. High-tech HVAC control systems that are procured through more competitive means usually meet the applications requirements more effectively and may cost half as much as similar installations (or even less) purchased by the traditional bidding process. My second basic criticism of purchasing high-tech systems by bidding is that the resulting specifications are generally not clear and require substantial additional design efforts by the successful contractor. By not accepting responsibility for the complete application design of an advanced technology system, the engineer tends to lose touch with these increasingly important advanced technology systems. This can be seen with HVAC control systems, which are now as removed from pneumatic controls as office computers are from the typewriters and desktop calculators they have replaced. Still, a large number of DDC specifications describe function in pneumatic terms. An engineer who loses touch with certain technologies involved in a design hurts the client because such a designer seldom utilizes that technology to its full potential. Additional problems stem from the ambiguity that results from the designers lack of understanding of the technology. Many high-tech system specifications actually discourage contractors from responding with low-cost or optimally configured systems. In this article, I will suggest ideas for alternative procurement procedures to engineers and owners who are contemplating advanced technology components for new or existing building mechanical sys38
NOVEMBER~~~~

tems. The techniques I will discuss are not new or the only possible options. They are techniques our firm and others have used very successfully for a number of years. As buildings continue to employ advanced technology systems, the techniques engineers and owners develop to purchase, install, and operate their advanced technology components will increasingly influence the level of success these systems attain. I recommend that engineers and owners consider establishing a sep-

Step 1 - Whats needed

The first step in conducting an effective procurement process for advanced technology systems is determining what features are needed and desired for the system to operate successfully as part of the buildings mechanical system. l Function-Building owners, including speculative developers, almost always want more than just the lowest cost product. They desire systems that will provide suitable levels of performance at the most reasonable cost. The engineer

Ron Anderson (left) of The Hartman Co. fields some suggested changes in the DDC system for a terminal regulated air volume (TRAV) mechanical system with Bob Cuti and Terry Egnor of Microgrid, the projects management firm. Plan a discussion session to explain options and alternatives so that everyone involved in the procurement is working toward obtaining the best value for the high-tech system.

arate category at the start of each building construction project for systems and equipment that utilize advanced technologies and, because of a lack of equal products, require a special evaluation process for selection. The owner and engineer should then work together to ensure that these systems and components are designed to meet the functional requirements of the project with as high a degree of certainty as possible and that they are purchased as competitively as possible. The focus of my experience with advanced technology systems is controls, but the following threestep process is suitable for a number of other advanced technology systems and components that may be integrated into HVAC system designs.

must determine a part of what constitutes a suitable level of performance by determining what is necessary to meet the rigors of the overall mechanical design. However, the owners concerns should be an important factor in determining what is suitable as well. Design engineers are generally not well acquainted with issues of operability or maintainability-especially as they relate to the rapidly advancing technologies that are incorporated into DDC systems and other energy management devices. Modern mechanical designs are becoming increasingly complex because they must meet increasingly complex demands for space and energy efficiency, comfort, air quality, and fire and life safety. An engineer may spend months developing

n HEATING~PIPING~AIRCONDITIONING

and refining the overall operating strategy for an HVAC system, and he must ensure that the level of performance of each component of the system is adequate to meet the overall system performance requirement. On the other hand, owners and operators usually consider operational issues of primary importance and generally have some strong opinions about what qualities are necessary to achieve the desired performance. The best way to develop basic performance level standards is through discussions between the engineer and owners representatives. In these discussions, the engineer can explain his design from an operations standpoint and solicit the owners views concerning operating and maintenance features of the system. The engineer should act as a resource for the owner to answer questions or concerns with up-todate information about the advanced technological system being considered. In this way, the basic standard of acceptable system performance can be developed for the design. Our firms approach is to prepare draft specifications for these meetings and explain the provisions and why each has been included. As realistic changes or improvements are agreed upon, we simply adjust the specification to incorporate those changes. We have found that the process works well because it prepares the owner for participating in the process of selecting a system from those that will eventually be proposed. l Maintenance and expansion- Most advanced technology energy-related systems include a substantial amount of equipment that has been designed and constructed by only one manufacturer. To repair, expand, or replace this equipment generally entails noncompetitive sole-source procurement. While this is not a major portion of the business of most manufacturers, some local representatives cannot resist the chance

A large number of high-tech products are now entering mechanical system designs. This influx demands improvements in procurement procedures to obtain the best value for the client. Here, Ron Anderson of The Hartman Co. inspects the variable frequency drives installed as part of a recent TRAV retrofit.

to take advantage of the owner when such sole-source purchasing is required. By making unit pricing and extended warranty offers a required part of the procurement process, the engineer and owner can eliminate these potential problems. They have an opportunity to discuss options for supporting, maintaining, or perhaps expanding the advanced technology system before the system is purchased and ensuring that the ongoing needs of the owner are addressed in the procurement process. It should be noted that if it is decided that unit pricing or extended warranty offers are desired for more than several years past the date of completion, potential vendors should be given wide latitude to tie their pricing to price indexes such as the consumers price index, interest rate, or other industry pricing indices. If maintenance of the system is expected to be substantial, or if the owner intends substantial expansion of the system, then the evaluation of the warranty offer and/or unit pricing schedule may be an important part of the evaluation process. Our clients have found the extended warranty offer requirement to be very valuable in comparing

competing systems. Because the offer is made in a competitive process, each extended warranty offer usually represents the contractors best estimate of the true annual costs of maintaining the system. Whether or not the owner intends to contract with the vendor to maintain the system after the oneyear construction warranty expires, the vendors estimate of annual warranty cost is very telling about its confidence in the integrity of the equipment to be supplied. A requirement that the extended warranty agreement remain open until the expiration of the construction warranty period permits the owner a year of operating experience before deciding whether or not to contract emergency repair services or provide them internally.

Step 2 - Procurement
Once the decision to procure a projects advanced technology components competitively has been determined, the owner and engineer must work together to ensure that a competitive selection process is developed to meet the owners procurement rules. If the owner represents a private organization, there are likely rules and guidelines within the organization that may limit the flexibility of the
n Novmmm 1991

HEATINGIPIPINGIAIRCONDITIONING

39

High-tech systems

procurement process. If the owner is a government agency, then laws may limit the procurement. The best rule to follow is to make no assumptions regarding which rules will actually apply. I sometimes think I would own the U.S. Mint if I had a nickel for every client who started the process by telling me his organization has to use the traditional bidding process in its projects. To date, I have not had one client that in the end actually had to do so. What has transpired between initial rebuff and the successful procurement process provides enough material to write a book of humor and several mystery tales. The goal of the procurement process should be to break the advanced technology system(s) out of the basic construction bid documents and conduct a separate procurement, generally with the plan to reattach it to the construction contract once that procurement

be. Once the procurement is completed, the actual subcontract amount is substituted for the allowance, and the general contractor assumes responsibility for managing the entire construction effort. Fig. 2 illustrates how the cash allowance process permits the separation of advanced technology system procurement from the general contract. There are other ways to accomplish the advanced technology installation, but if the system or component is an integral portion of the overall building operation, a cash allowance is recommended because it does not compromise the general contractors control or responsibility for the construction effort. To ensure that an effective procurement process is established for high-tech systems, I suggest the following guidelines be followed when developing the process: l Make the case-There is an

2 Using the cash allowance feature to conduct separate selection process for high-tech systems.

process is complete. The cash allowance feature of construction contracts is an excellent vehicle for accomplishing a separate procurement. The use of cash allowances is widespread in construction contracts today. The presence of the allowance alerts the general contractor to the fact that management resources will be required to coordinate this element of the construction effort, but the general contractor does not decide who the subcontractor will
40
NovEMBERI~~I

important reason you are likely to hear negative responses if you ask about procurement flexibility in large bureaucracies. People like to do things the easiest way possible. Conducting construction procurement with a traditional bidding process is usually the easiest way. However, if the issue is presented candidly, I have found the problems associated with garnering support within the bureaucracys purchasing establishment are generally greatly reduced.

To make the case, the engineer should be prepared to discuss the advantages of a proposal process realistically along with the problems of a strictly bidding procurement. Having as many of the differences expressed in bottom line dollar figures is the best way I know to get the attention (and sympathies) of purchasing people. In making the case, provide a specific procurement plan that you believe will meet the owners organizational rules. The most powerful support you can garner is patterning your plan after a similar process that has been recently employed by the organization with a successful outcome. For this, some research and investigating is necessary. When working with government projects, I have had success contacting the offices of representatives that chair committees involved with energy to find laws that permit (and sometimes encourage) nontraditional procurement procedures. State level energy departments are good resources when working with a state agency as a client. Whatever resources are available, use them to sell the idea of nontraditional procurement rather than ask about it. l The evaluation processOne of the reasons organizations are leery of nontraditional procurement processes is the fear of conflict-of-interest situations. Include a well thought-out evaluation process to allay this fear. Form an evaluation committee that includes the building operating personnel and at least one person from the department in charge of purchasing for the project. This is a good way to eliminate worry that someone may put something over on an organization. In addition, a description of the range of features that will be under scrutiny in the evaluation process and the financial considerations that accompany each feature is a valuable tool in making the case for nontraditional procurement. The more it is possible to put the evalu-

n HEATINGIPIPINGIAIRCONDlTlONlNG

ation process into a formula, the easier it will be to make the process fly. Of course, it is important to understand that many of the items that may distinguish one proposal from another cannot be listed before the proposals are received. Still, that does not prevent the engineer from establishing broad categories of possible differences and making an assignment of relative value to each category.

l A realistic process-Every client organization has a personality complete with preconceptions and prejudices that are usually the result of its past experience-good and bad. The interpretation of these experiences has often led organizations to a particular way of doing things that can conflict with the engineers vision of how to develop the procurement process most effectively. The engineer must be sensitive to the organizations view of the procurement process and help in the development of a procurement plan that will accommodate the organizations way of doing business. Only if harmony would compromise a satisfactory outcome should the engineer challenge established procedures, and if such challenges are necessary, they should be minimized so as not to cause the whole process to be discarded. The cornerstone of successful procurement of advanced technology systems is breaking those components out of the traditional bidding process. Depending on the technological system to be purchased and the organizational constraints and views by which the owners procurement procedures are governed, a successful procurement may be as simple as a prequalification of acceptable system configurations and components. There is no need to rock the organizations boat unduly if a very simple procurement process that is compatible with the organizations general business attitudes will suffice.

Step 3 - Develop specs

The introduction of high-tech systems into mechanical designs increases the integration of mechanical and other building design disciplines. Here, a new lighting interface panel has been mounted below the lighting circuit breaker panel to provide lighting control through the DDC system. To be effective, the procurement of these high-tech systems must satisfy the requirements of each design discipline.

When the process by which the procurement will take place has been determined, the engineer must translate the design of the advanced technology system into a specification document that meets the requirements of the chosen process. The following basic rules should apply to the specification development process: l Design the system only once-Earlier, I discussed the need

for the engineer to be as specific as possible in selecting the advanced technology systems. This requires the engineer to be knowledgeable about the technology to be employed and provide a complete design of the application (with alternates if necessary). If the design is not complete, the contractor will have to provide a design as a part of his contract. The contractors design will require full knowledge of the way the system must integrate into the other mechanical system components, but that information may never be made completely available to the contractor. Therefore, the second (contractors) design almost always has defects that usually dont come to light until startup, at which time they can cause enormous headaches for the owner and operators. To eliminate this problem, the engineer should assume complete responsibility for each advanced technology system that is a part of the overall mechanical design. If the engineer does not have the expertise to provide the design, then an independent outside source should be found to help with the design, the specifications, and the procurement. It is important to remember that the design of the overall mechanical system is often the result of many months of information exchanged among the engineer, owner, and others in the design team. It is not reasonable to expect the system subcontractor to pick up all this information and assume responsibility for the design of the system. 0 Competitive flexibility-It is important for the engineer to provide a complete design so that the potential contractors know exactly what will be expected of them. Also, specifications should not be so specific as to favor one contractors equipment over another without valid reasons. For example, DDC system panels have varying point capacities among the various manufacturers. When specifying for a competitive
41

H EATING/ PIPING/ AIR C ONDITIONING H NOVEMBER 1991

High-tech systems

I.

I. System design
Does design meet functional requirements? Does the design ensure the necessary system function will be achieved? e Can the owner or user operate the system?
n

DDC procurement, our firm tries to detail each point and its interface and establish the level of control, communication, and operator interface required. But we do not specify a point connection list for each panel. If we were to do so, worthy systems might be eliminated simply because they incorporate lower point densities at each panel. Or systems with high panel point capacities might be placed at a competitive disadvantage by requiring a low point density. Unless the point layout is important to system operation, it is left up to each contractor to match his equipment to the points list most cost effectively. Then the configuration becomes an item of the evaluation process. We try to develop specifications so that as many products as possible can be considered. Obviously, the more versatile the procurement process, the more flexible the specifications can be. But flexibility should not be confused with ambiguity. No matter how flexible the specifications are, they must very clearly describe what is required of the system or component.

Are Drovisians!

for otieratina: f6atureti i

Uncertainties
An owner once told me that in discussions with his engineer on a project, he questioned the engineer about the lack of a points list for the DDC system. The engineer replied, Oh, we never include points lists in our specifications. This way the vendor must supply all points that are necessary to make the system work. We never have to worry about leaving any out. That revealing statement should cause engineers and owners alike to pause and rethink the interests and duties in construction projects. We all know that there are some uncertainties in most building system designs. Those that are most aggressive in achieving high levels of energy efficiency, comfort, and air quality probably encompass somewhat larger uncertainties because many of them utilize technologies or strategies that have not been
V

3 High-tech system procurement checklist. widely employed before. Presumably owners who encourage or permit higher performance designs have some understanding of the uncertainties involved. However, the high-tech contractor is the least prepared to evaluate and accept responsibility for a projects uncertainties. The contractor has not been privy to all the discussions and decisions throughout the design process and very likely does not understand the complete mechanical system design. When the specifications are released, each contractor is (or should be) wholly engrossed in a good faith effort to apply his product in the most suitable configuration possible. Putting the uncertainties of the project on the contractors back results in a costly redundant engineering effort. At times it seems that the use of ambiguous procurement procedures is aimed at enticing contractors to underestimate the scope of work and obtain lower prices. More frequently, however, I find ambiguous specifications are the product of unknowledgeable designers. Whatever the reason for a specification that does not clearly define the scope of work for an advanced technology system or component, it is a risky procurement strategy because the project and the owner are almost always the real losers. The staffs of most contractors have been around long enough to make provisions for risk. Their bids or proposals respond to ambiguity with ambiguity. I have heard vendors state that if the engineer requires them to provide Part A,

42

HEATING/PIPING/AIR CONDITIONING

which they believe was not required by the specifications, they intend to respond by downscaling or omitting Parts B, C, or D, whose exact requirements are obscure by the ambiguous nature of the specifications. What those conducting the process must consider is that so long as the requirements of the advanced technology system are realistic, then the more complete the specifications are in describing level of performance and other requirements of the system to be purchased, the more likely suppliers will provide competitive proposals. The specifications must also provide sufficient latitude so that several potential suppliers can offer cost-effective proposals. The engineer must stay up to date with each technology of the design and possess sufficient knowledge of the range of products available to ensure that the balance between desired features and opportunity for multiple vendors is maintained. 86
NOVEMBER 1991 w R

While a great many engineers still use simple bidding procedures to purchase controls and other advanced technology building systems, the bidding process can result in costlier systems that fail to operate up to the owners expectations. Placing a projects high-tech systems and components in a special category and procuring them separately from the general contract is an approach that has proved useful to the owners and engineers who use it regularly. Fig. 3 illustrates the three basic steps that our firm has found to be effective for successful high-tech system procurement. The questions are used as a checklist to ensure that the resulting system meets the demands of the project and will he procured effectively. Seeing that these questions are answered positively and attention is provided to the special procurement process will result in improved performance and cost of each projects high-tech systems. 11

HEATING/PIP AI CONDITIONING

.,........................................................

NEW GENERATION DDC

PROMISING CONTROL INNOVATIONS WITH NEW GENERATION DDC

The new generation of DDC systems now available fulfills past promises of enormous improvements in building comfort and operating efficiency
By THOMAS HARTMAN, PE,

The Hartman Co., Seattle, Wash. romises dont usually mean much in an election year. But for our in0 dustry, this may be the year that some long-standing promises of the newest direct digital control (DDC) systems finally begin to be realized. After years of talk about an impending revolution in the way buildings will operate, the new generation of DDC systems that was to have sparked the revolution has arrived. But the accompanying control revolution has been noticeably absent. Instead, the new DDC systems are being applied to operate in the same manner as systems of the past. Now, developments are starting to paint a very encouraging picture of realistic opportunities possible with these new, more functional DDC systems. Experience has now been gained with some exciting new control concepts and approaches. It is time to consider these new approaches as candidates for current projects. The following is my short list of design innovations made possible by the recent advances in control and related technologies D

that we should all be considering. l Keep buildings runningWith proper control algorithms, keeping the air moving in buildings at low flow rates during unoccupied hours (using variable frequency drives) can, in many buildings, save far more energy than the small extra fan energy it costs. Reducing infiltration and distributing warm core air to the perimeter reduces heating requirements during cold weather.

Purging the building with cool night air saves cooling energy in warm weather. In addition to energy savings, keeping air moving without a large night setback has shown remarkable improvement in the comfort and quality of the building environment. l Integrate lighting with HVAC control-Integrating the lighting control into the DDC system permits the establishment of spaceby-space occupancy conditions so

80

Inlet vane

5: .gl 60

$ a 70 t 3 60

+i 40 5 30 20
10

Variable speed drive TRAV

l .
l

.._
l *

-4

-*

0.

Obh100

--._

1 Fan power vs. air flow for a centrifugal fan with three methods of VAV control. See text for definition of TRAV.
HEATING/PIPING HEATING/PIPIN /AIR /AIR GCONDITIONING CONDITIONIN GM- N OVEMBER 1992

55

New generation

DDC
at significantly higher heat transfer efficiencies. In similar fashion, carefully designed heating and cooling equipment can also be made to operate at higher efficiencies during low load conditions. The key to achieving energy reductions by implementing 24-hr HVAC operation is to develop an efficient building envelope system and high part-load efficiencies for all the HVAC transfer and energy conversion systems. Fig. 1 shows that approximately 2 percent of the design energy is required to operate a VAV fan at 30 percent of design air flow when controlled by a variable frequency drive (VFD) under a scheme that incorporates terminal-regulated air volume (TRAV).* Supply fan output is governed by real-time terminal box air flow requirements rather than the need to meet a duct static pressure set point. Continuous air system operation can be used to reduce infiltration, purge the building with cool outside air during anticipated warm weather, or transfer the heat in the building core to the perimeter areas in cold weather. Continuous operation strategies also permit down-sizing heating and cooling equipment. The continuous operation strategy has an enormously positive impact on comfort and air quality in buildings during normal work hours and an even greater one during the off hours that building occupants increasingly wish to work. In the past, building controls have required the building operator or mechanic to establish one occupied hour schedule for the buildings HVAC system and a second for lighting control because these control functions were usually provided by separate systems. Lighting control has traditionally been accomplished with
*See TRAV-A New HVAC Concept, by T. Hartman, HPAC, July 1989, pp. 69- 73.

that temperature control and outside air ventilation can be supplied only to the zones that are actually occupied. First costs for integrated HVAC and lighting control are often less than those of designs incorporating separate control systems. l Install a space temperature sensor in every office - When a terminal box supplies multiple offices or areas, using multiple space temperature sensors for box control can provide an enormous improvement in comfort over the single sensor concept. Simple averaging logic combined with occupancy signals can increase comfort in each office and improve the energy performance of the mechanical system. l Eliminate VAV box maximum and minimum air flowsComfort, air quality, and energy efficiency can all be improved by better air flow control strategies at terminal boxes. Controlling the air flow at each terminal box based on actual needs for ventilation air and cooling is a more effective control technique than strategies that simply regulate zone air flow between preset minimum and maximum box air flows. l Use the DDC operators console as an electronic drawing file-As-built CAD drawings in color are far easier to follow than blueprints. Installing a CAD system in the DDC host computer as a reference, and for updating tenant changes of mechanical drawings, offers an effective management tool to operate the building more effectively and keep the building record documents current. Each of these innovations represents a substantial break with traditional practices. My experience with initial implementations of each of these strategies is very positive. All appear to have great promise when incorporated as part of a careful controls design. What follows is a more detailed discussion of each.
56 N OVEMBER

Traditional thinking concerning energy conservation in buildings has focused on keeping the HVAC system off when the building is unoccupied. This widely accepted strategy compromises the comfort and air quality for those who occupy buildings beyond regular business hours. In todays commercial and institutional buildings, the concept of building shutdown is rapidly losing its attractiveness because of the occupants demands for more flexibility, better comfort, and improved air quality. This trend is well known to building operators. A trend that is not immediately apparent is that building shutdown is also becoming obsolete as a result of more efficient building envelope technologies. In older buildings with poorly insulated envelopes, turning HVAC systems off was about all operators could do to reduce energy use. But the development of more effective building envelopes is changing the pattern of energy use in buildings. Todays buildings have a much higher ratio of building mass to envelope loss than older buildings. This development has made the thermal inertia of buildings a significant component in mechanical system sizing and building operation. At the same time, the higher resistance to heat flow of modern envelopes has decreased the impact of conduction heat flow on the overall building energy use. For many buildings, it has been shown to be more economical to maintain indoor comfort conditions during low or unoccupied hours than to shut the system completely down. Maintaining more constant space temperature conditions reduces peak building startup energy requirements. Overall fan energy can be reduced with 24-hr operation because fans operate at high speeds for fewer hours. Operating a fan for 24 hr permits energy exchange within the building or with other systems

Continuous fan operation

Lighting and HVAC control

PIPING / AIR CONDITIONING

2 Program for averaging multiple space temperature sensors.

simple time sweeps. When the DDC system is extended throughout the building to each terminal VAV box, lighting control can easily be added to the system at low additional cost. Lighting and HVAC control integration offers excellent opportunities through synergism to improve the performance and efficiencies of both systems. The expanded logic capabilities of new technology DDC systems can offer lighting control schemes that are more effective in accommodating occupant needs, yet result in fewer lighting operating hours than other strategies. Furthermore, monitoring occu-

pancy on a zone-by-zone basis provides an opportunity for improved comfort and HVAC energy savings by allowing the system to focus heating, cooling, and ventilation on those areas of the building that are occupied. The most obvious cost benefit of combining lighting and HVAC control is that the lights indicate the occupancy status of each zone-no other indication is required. Savings benefits stem from the fact that modern buildings are rarely 100 percent occupied. Establishing office-by-office occupancy conditions can limit comfort conditioning to the areas

that are actually occupied. Distributed DDC control of lights can also be more effective than traditional lighting control strategies. Most lighting control schemes use centralized lighting panels with lighting sweeps that turn off large blocks of lights at specific times. Under DDC operation, the lighting zones are typically much smaller (a single enclosed office is one zone), and a control relay is located at each zone. Lighting sweeps during nonbusiness hours are absent. Instead, each zones lights operate individually. Pushbuttons or occupancy sensors, or both, can be employed to determine building occupancy on a zone-by-zone basis. Occupancy sensor technology has advanced rapidly in the last few years. The performance of these devices has improved substantially while the manufacturing costs have continued to decrease. The two major detection methods, infrared and ultrasonic, are both effective in detecting occupancy at the work place, and each has certain advantages in particular applications. The combination of advanced DDC system logic capabilities and occupancy sensor improvements means that fewer sensors may be adequate to monitor open office applications. However, it is very important that occupancy sensing designs be conservatively developed because small difficulties can result in very negative perceptions by the occupants. The most economical occupancy sensing method is a low-voltage pushbutton similar to those employed in low-voltage lighting systems. The pushbutton may be integral with each space temperature sensor, located separately, or a combination of both. When the occupant arrives and pushes the button, the DDC system is alerted to the occupied condition. The lights in the area are turned on, and the HVAC terminal unit(s) is switched to the occu-

New generation DDC

pied mode, assuring outside ventilation air is delivered to the zone. When the zone is occupied, the space temperature is controlled within tighter limits. The logic of the DDC system determines how long the space will assume occupancy, depending on the time of day and whether it is a weekday, weekend, or holiday. Generally,

3 VAV terminal control strategy employing proportional relationship between temperature and air flow.

the conclusion of occupancy is preceded by a brief flash of the lights that reminds the occupant it is necessary to press the button to continue the occupied state. A press of the button at that time resets the occupancy timer to the preprogrammed override time. Pressing the occupancy button any time before the end of occupancy with the lights on alerts the system that the occupant is leaving. The lights shut off, and the HVAC system reverts to the unoccupied control state until the pushbutton is pressed again. For some years, the industry has accepted the limitation of using only one temperature device for each terminal box no matter how many separate offices or spaces that box served. This may have been an acceptable practice in the past, but it is now out of date. The comfort of all areas served by a single terminal box can be improved by installing an inexpensive electronic temperature sensor in each area served and using the math and logic ca58

pabilities of the DDC system to control the box. I recommend the control logic employ a weighted-average technique. The weighted-average approach controls the box according to the average space temperature of the occupied offices. However, if the temperature of any office or area served by the box rises or falls beyond the heating or cooling set points, the weighting of those offices in the averaging calculation is increased. Fig. 2 illustrates a simple weighted-average program for a VAV box that serves three offices. The resulting weighted-average temperature (called BOX_TEMP in Fig. 2) is used to establish the air flow set point for the box. This simple but effective strategy can significantly improve the comfort of a multiple office arrangement. It is not costly because space temperature sensors are among the least costly devices available for DDC systems.

Until recently, nearly all VAV box control strategies employed a proportional relationship between space temperature and air flow, limited by a box minimum air flow and a box maximum air flow. This relationship is shown in Fig. 3. VAV systems have been correctly criticized for their failure to guarantee minimum outside ventilation air. If the box is at the minimum flow, the outside air available for ventilation depends on the percent of outside air in the supply air, which is a variable with systems that incorporate airside economizers. The amount of outside air required for ventilation may also be variable. On the other hand, the box maximum design air flow (which is usually entered as the box maximum) is not properly used in most applications. If a space is well above the space temperature set point, there is rarely any compelling reason to limit the air flow

VAV box maximum/minimum

Multiple space sensors

4 Improved VAV terminal control strategy possible with DDC control.

NOVEMBER 1992 - HEATING / PIPING / AIR CONDITIONING

to this calculated value if more air is available. With new DDC technologies, adding separate set points for heating and cooling along with a weather-calculated air flow for the range between the two set points is now possible and provides benefits in comfort, energy performance, and air quality. I recommend the VAV box control scheme shown in Fig. 4 be used to replace the old pneumatic control concept of Fig. 3. In Fig. 4, there is no fixed box minimum or maximum air flow. The minimum air flow at space temperatures below the heating set point is what at the time provides the correct amount of outside air ventilation to the zone(s) served by box based on current occupancy conditions. The actual air flow is calculated from the current occupancy conditions (number of people in the zones served by the box) and the percent of outside air in the supply air stream. Between the heating and cooling space temperature set points, the air flow set point for the box is based on the space temperature and the projected outside weather conditions. In warm weather, the air flow set point is high during economizer operation to maintain the space temperature at the lower end of the range in anticipation of warmer weather ahead. In cold weather, the air flow set point remains low while the space temperature is between the heating and cooling set points to allow the space temperature to rise in anticipation of heating requirements. Above the cooling set point, the box air flow set point rises rapidly to the design cooling air flow, where it remains level for an approximate 0.3 F space temperature increase. At about 0.5 F above the cooling set point, the air flow set point is raised to ensure the damper is fully open to provide all possible cooling and prevent further temperature increases.

This newly developed box control strategy has been very effective. It provides required minimum ventilation at all times during occupancy and maintains more constant comfort conditions. One of the widely recognized commissioning issues for buildings today is the need to get good and other documentation database management tools into the hands of the operations staff. The term database is used here to include as-built drawings, building system information, controls operation strategies, and documentation for the thousands of DDC points that make up the DDC system. Current building database management techniques already include the use of graphic menus and text file access that permits operators to access control sequence programs and descriptions of operating intent from the DDC system operators console screen. The concept of access to documentation through the DDC system operators console can now be extended to include the mechanical and control drawings as well. In this concept, the buildings mechanical, electrical, and control CAD drawing files are included on the DDC system operators console disk. A file management program that permits the operator to page through the drawings much the same as if the drawings were stacked on the operators desk is installed. In addition, I recommend a support concept that gives the operations staff the responsibility of maintaining up-to-date as-built drawing and database materials. I also recommend the installation of the CAD editor program in one of the DDC system operators consoles so that the inevitable ongoing changes in partitions, temperature sensors, diffusers, lights, and (most important) heat loads can be noted on drawings and listings. The operator is responsible for up-

dating these changes and seeing that the required balancing changes are made to maintain comfortable and efficient system operation. The new generation of DDC systems now available holds the promise of enormous improvements in building comfort and operating efficiency. But like political promises, an overabundance of past pronouncements about DDC systems has already dulled the receptivity of our industry to such promises. Fortunately, improving the comfort and efficiency standards of a buildings operation is a lot easier than resolving national problems. Concerned building owners and operators can themselves pursue these opportunities without depending on manufacturers promises. The industry has started to learn that effective building operation has more to do with how well mechanical, electrical, and control systems are all integrated into a sound design concept than with specifics of the systems themselves. Well-trained building operators together with an experienced and knowledgeable engineering group form the team that can make these new, ambitious opportunities come true. Successful building owners and operators already know that the optimum strategy for developing, constructing, and operating buildings requires a motivated team with strong leadership. Those are precisely the management tools required to implement the kinds of innovations listed above. We can make buildings that are easier to operate and operate them more effectively. To do so requires hard work, a solid team-oriented project organization, and a strong commitment to technical excellence. But the result will be worth the effort, and we dont have to depend on promises to make it all happen. 0
1992

Getting started

CAD and the DDC console

HEATING /PIPING /PIPING/AIR CONDITIONING~NOVEMBER

59

..L.. .

..*......................

INDIVIDUAL TERMINAL CONTROL

NEW ZONE CONTROLS HELP ACHIEVE TOTAL ENVIRONMENTAL QUALITY

Increasing demands for total environmental quality by building occupants are changing HVAC design criteria from uniform to individually adjustable zone conditions
By THOMAS HARTMAN, PE, The Hartman Co., Marysville, Wash.
cool room in cold weather may improve the occupants sense of comfort. At the same time, the heat generated by the lights is helpful to the heating effort. Similarly, decreasing the lighting level under warm conditions provides help for the sense of comfort and the cooling effort of the space. The clients engineer took some exception to this approach even though it was simply a supplemental step that was not intended to affect traditional temperature control strategies directly. We did note, however, that it could evolve into a strategy that permitted a slightly wider range between heating and cooling set points. As I listened to the discussion, I realized that the client feared this approach would reduce the building opera. tors ability to illusr trate to occupants that the building is comfortable because the simple fixed criterion for comfort-space temperature-would quite possibly no longer be the absolute determination of comfort. I thought about a scene I have experienced many times in which a building operator
HEATING/PIPING/AIR CONDITIONING - NOVEMBER 1993 43

veloping regarding lighting control. Our project manager was explaining to the client that dimming lighting 1 ballasts allowed us to include lighting control as part of our comfort control strategy. In addition to compensating for changes in outdoor light, he explained that our lighting control strategy includes further adjustment of lighting levels depending on room space temperatures and weather. Increasing the lighting levels in a

Individual terminal control

or engineer argues with an occupant that he or she must be comfortable because the space temperature is right on set point! I was reminded that despite the limited understanding our building design community has for the elements of human comfort, we continue to try to express comfort in simplified (and unrealistic) terms so we can get on with what many believe to be the real job of HVAC engineers-designing/operating/maintaining chillers, air handlers, and boilers. Those who continue to view our industry in these traditional terms need to understand that an enormous change is taking place in the minds of individuals who own and occupy the buildings we design and construct. This change is being propelled by the isolated but highly visible unhealthy buildings we all have heard about. Building owners and occupants are coming to believe that the industrys emphasis on very simple comfort and air quality criteria is insufficient to meet the needs of building occupants. Occupants no longer believe that being shown that their space is within acceptable temperature limits and the outside air dampers at some minimum position are acceptable assurances of a comfortable and healthy building. Furthermore, owners understand that such simplified building design criteria do not adequately address their concerns over liability. Fortunately, as the individuals who operate and occupy buildings are beginning to hold our industry more accountable for keeping them comfortable and healthy, an astounding array of economical new comfort control techniques are finding their way to the marketplace. What is needed now is an understanding of how these rapidly evolving changes are affecting the basic criteria for building design and how to apply equipment now becoming avail44
G

able to satisfy these changes most effectively and most economically. For years, building design engineers have assumed a simple set of criteria to establish comfort; a single and uniform set point for interior space temperature for comfort and a minimum outside air damper position or opening for air quality. In particularly humid or cold climates, the criteria often included limits on humidity as well. To develop an HVAC system to meet these simple criteria, designers assumed maximum and minimum outdoor temperature and humidity conditions and maximum solar conditions. Thats about all there was to sizing the components of an HVAC system. The industry has long been aware that space temperature sensing by itself is not an effective means for accurately assessing comfort. Even if it were, typical air-based HVAC systems operate with large air temperature gradients within each space that make it very difficult to assess overall thermal comfort conditions from a point temperature within the space. Compounding the problem has been the well-known fact that in typical office environments, comfort is substantially affected by radiant heat exchange between building occupants and surrounding surfaces as well as the amount of air movement. The practical aspects of ensuring air quality have also been largely ignored even as the potential liability for doing so has skyrocketed over the last few years. In low-rise constant volume sys-

New challenges

terns, a prebalanced damper position or inlet is a reasonable method of ensuring a certain amount of outside air enters the building, but this approach has been extended to high-rise buildings and VAV systems even though it is very clear that it cannot be so applied. Furthermore, the integrity of outside air distribution to individual zones in buildings has been severely undermined in many VAV designs. Now, building designers are learning that a number of other environmental factors also need to be considered when developing comfort and air quality criteria. The term total environmental quality (TEQ) has been coined to develop comfort and air quality standards based on the larger number of factors known to contribute to an occupants sense of well being. My list of TEQ items designers should be considering in their designs is provided in Table 1. It is reasonably well known that these (and other less definitive or controllable factors) interact to establish comfort and air quality for building occupants. Exactly how all these factors interact is not well understood. Furthermore, the occupants sense of well being is a very subjective determination that depends not only on clothing, building materials, and levels of exertion but also on a number of cultural and individual criteria that, at least for the present, are far too nebulous to define with precision.

The large number of factors that affect TEQ, together with our growing knowledge about the TABLE 1 - Total environmental quality variations among individuals, cre(TEQ) factors. ates a complex web of information Space temperature that is not likely to be adequately Wall, window, and ceiling surface temperatures understood anytime soon enough Space humidity for even the most sophisticated Air movement control system to anticipate what Outside air ventilation rate (dilution) is required to provide TEQ to each Light level building occupant.

Meeting the challenge

systems can address this concern deserve some discussion. Simply stated, in a high-performance control environment, building occupants express their desired changes to the building control system rather than directly overriding its programmed function. This permits the building control system to determine the most appropriate (and efficient) means in which to act upon each request. How successful a building design team is in meeting the new challenges of building total environmental quality depends largely on whether this challenge is viewed as an opportunity or a burden. For building owners who do view these changes as an opportunity, the benefits include more satisfied (and perhaps more) tenants, less time required by building operations staff to respond to complaints, and significantly lower energy usage. Providing individual terminal control has a number of pitfalls designers need to avoid, and if the designer is not careful, some existing terminal control products may lead the design in the wrong direction. The following are several key concepts that must be applied in developing ITC systems. l The ITC system must have the capability to reset occupant adjustments automatically to a neutral or near neutral state from time to time. A number of problems with permitting adjustment by occupants can be traced to a temporary adjustment that is forgotten by the occupant. For example, imagine that a portion of the cooling plant has failed in a building on a warm afternoon. If an ITC system is present, it is likely that many occupants will call for cooler space conditions because of the failure. However, if the failure is corrected that evening, and no automatic reset of the adjustments is made, the occupants will likely ex-

Applying ITC principles

2 Combined VAV box controller and box damper actuator.

The natural conclusion from these facts is clear: The standard for establishing building comfort must change from a set of uniform building conditions to a range of conditions within which each zone can be individually adjusted to whatever is desired by the occupant(s). In other words, it is not the building owner, the design engineer, or regulations or standards that can define comfort and a sense of well being. It is the building occupant. Fig. 1 illustrates the ITC concept. In this figure, a building occupant is able to communicate desires for changes in lighting, temperature, and outdoor air ventilation levels from his or her workstation. Equipment and devices are now becoming available that afford this type of operator interaction with the HVAC system. To satisfy building occupants, designers must begin to incorporate individual terminal control (ITC) features in their mechanical system designs, This unavoidable conclusion is a reversal of the direction the industry has generally been traveling for the last several decades. The occupant has too often been seen as

the culprit in comfort, air quality, and energy use problems. Removing operable windows and installing locking thermostat covers have been very popular in this time, and the advent of digital controls at the terminal box has often eliminated any capacity for occupant adjustment altogether. In some instances, there have been valid reasons for disabling direct occupant control, but the prospect of true individual control with a high-performance DDC system provides a formidable challenge to the assumptions that have guided such decisions in the past. For example, building operators are frustrated when the adjustment of a thermostat in one office leaves those in adjacent offices (connected to the same terminal unit) uncomfortable. True individual control eliminates this concern by offering each and every office or work area the capacity for an individually regulated environment. Another concern of building operators is the possibility of excessive energy use or the inability to exert control over energy use with ITC schemes. The mechanisms by which high-performance control

Individual terminal control

Space temperature and vices alone. In addition to perience discomfort the product cost savings, the following day because impact on the total promany will have forgotten ject cost is often dramatitheir adjustment the day cally improved because of before. the substantial installal The ITC system tion cost savings from should not provide atten360 deg fewer units to install. tion-diverting informaoccupancy sensor Pendant mounted temperature sensor Figs. 2 through 4 illustion to the occupant. trate the trend toward inSince the advent of ditegrated DDC devices. rect digital controls, I 7 Fig. 2 is a damper mohave too often heard artor upon which the manuguments between building technicians and occu- 3 Combined space temperature and occupancy sensors. facturer has mounted a VAV box controller modpants about the accuracy ule. In addition to the of the space temperature Space temperature, dedicated air velocity insensors employed by the occupancy, and put and the damper mosystem. Rather than enadjustment parameters to/from DDC system tor modulating output, courage the occupant to the module includes an set specific space temper-7 additional three inputs ature set points and proand three outputs, each vide the means for him or of which can be proher to check the systems grammed as either a digiability to meet that tartal or analog type. The get, our experience is that controller also includes more effective ITC results Light display indicating occupancy sensor fully programmable confrom permitting the operadjustment mode trol language, PID conator to reduce less quantitroller blocks, and schedfied adjustment requests. Pendant mounted remote control ule and display features. Warmer/cooler, brighter temperature sensor sensor/decoder This motor/controller de/darker, and ventilate/revice is now available and circulate are probably more effective ITC crite- 4 Comfort control module combining space temperature will soon be released by ria than requesting spe- and occupancy sensing with infrared wireless remote re- several additional manufacturers. Where pricing c i f i c s e t p o i n t s f o r ceptor/decoder and lighted display. sheets have been made temperature, light, and ventilation levels from the occu- must also be a prompt perceptible available, the price of this unit is less than the price of VAV conresponse. pant. troller modules alone several l When an individual makes an years ago. adjustment, some acknowledgFig. 3 is an occupancy sensor Putting the elements described ment of the action must be proabove into a simple and effective that has been integrated with a vided immediately. When an occupant adjustment ITC design is becoming easier and pendant type temperature sensor. is made, some signal must ac- less expensive with the advent of This unit has recently become knowledge the adjustment. Some new products appearing on the available and provides two importype of display to show that the market today. The Fig. 1 illustra- tant zone inputs to a DDC system system has received the adjust- tion of an individual adjusting with a very simple and low-cost lighting, temperature, or outside installation, The method of occument request is essential. air ventilation levels from his or pancy sensing for this particular l Each adjustment must lead to a perceptible change in the direc- her workstation is rapidly becom- unit is passive infrared, though ing a reality. The key to these ultrasonic may soon be available. tion of the intended adjustment. Exactly how the system will re- rapid development efforts ap- The cost of the unit is about the act to each adjustment request pears to be the economics of inte- same as the cost of an occupancy will depend on the circumstances gration. A number of manufactur- sensor alone. Fig. 4 is an enhanced version of at the time of the request. How- ers are combining several devices ever, it is imperative that in addi- or components into single units the Fig. 3 device with several addition to an acknowledgment, there that cost no more than single decontinued on page 92 46
NOVEMBER NovEMBER~~~~~HEATING/PIPING/AIR 1993 - HEATING/PIPING/AIR CO N D I T I O N I IN NG

Individual terminal control

continued continuedfrompage46 from page 46

tional features. This device is now under development by at least one manufacturer and will soon be available. In addition to the temperature and occupancy sensor, this device also has an infrared sensor that receives and decodes wireless remote transmitters of the same type used to adjust television and VCR units, For this product, the hand-held remote unit permits the occupant to request warmer or cooler space conditions, brighter or dimmer lighting, and more or less outside air ventilation. The integration of this unit with standard VAV components to effect economical individual terminal control is shown in Fig. 5. Notice that all components in Fig. 5 are ceiling mounted, which substantially reduces the installation costs and increases the ease of adjustment to meet tenant

space changes. Our firms estimate is that the cost for the Fig. 5 configuration will be substantially less than the present cost of a DDC box control installation using standard VAV control techniques. However, this does not mean that ITC will be less expensive to implement than current DDC VAV control because ITC generally requires substantially more zones than conventional VAV systems. Under an ITC concept, each office and open area (forming a cluster of occupants) requires a separate VAV terminal unit. In typical office buildings, this represents an increase in VAV box count of between two and three times. Even though the per unit cost of each VAV zone can be reduced with ITC concepts, the total cost in all but entirely open office configurations will be greater.

However, the benefits of this approach include increased occupant productivity. A recent study at the West Bend Mutual Insurance Co. showed that an individual control concept improved worker productivity at levels of between 2 and 6 percent. These economies alone justify the cost for incorporating individual control into VAV systems. We have discussed the transformation of typical VAV system configuration to ITC, but it is important to realize that as old HVAC design concepts based on uniform fixed building environments evolve over the next few years to the individually controllable building environment, the ITC concept will exert a strong influence on the remainder of the HVAC system. New and more effi-

New ITC concepts

DDC zone controller

5 Individual terminal control (ITC) employing standard VAV and a comfort control module.

nents rather than a single isolated factor. The increasing demands for total environmental quality by building occupants are transforming HVAC design criteria from providing uniform building conditions to offering individually adjustable zone conditions. DDC components becoming available are integrating previously separate functions and offering capabilities suitable for individualized control at low cost. Designers should begin to develop ITC configurations for their projects as these products are becoming available. n

Summary and conclusion

Comfort control module

cient mechanisms for delivering air to individual zones are being developed, and traditional VAV concepts of zone temperature control are likely to be replaced with new strategies that include a connection between air temperature,

surface temperature, humidity, ventilation air, air movement, and light intensity. When an occupant requests a change, the control system will be able to make complete and prompt response by adjusting the matrix of compo-

..........................................................
OPEN/STANDARD PROTOCOLS

PRACTICAL CONSIDERATIONS FOR PROTOCOL STANDARDS

A realistic look at how open/standard protocols will meet designers and operators needs for improved interoperability among HVAC equipment and control systems of different manufacture
By THOMAS HARTMAN, PE, The Hartman Co., Marysville, Wash.
scar Wilde, a cynical observer of modern mankinds follies, once observed When the gods wish to punish us, they answer our prayers. I fear this insightful commentary is soon to be visited upon the building construction industry. For some years, I have heard my colleagues carp about the lack of progress toward a standard protocol for DDC systems. Well, soon such standards will exist, and for many it may well become more of a punishment than a blessing! Those with hands-on DDC systems experience must shudder at the thought of having multiple DDC systems on a single network. While a standard in communication may lead to a uniform mechanism of displaying data and issuing commands to individual system points, operators will still have to cope with the differences among the various systems in changing programs, displaying trends, and point database operations, not to mention dealing with the idiosyncrasies of each different system. Furthermore, knowledgeable operators must surely wonder to whom they will turn if the network performance does not meet expectations. What if controllers are not receiving necessary information from others of different manufacture, or what if it takes too long for operator displays to be updated? On the other hand, designers and operators are beginning to understand the need for improved communication capabilities among components of different manufacturers. Virtually every piece of mechanical and electrical equipment used in buildings today comes factory assembled with some type of microprocessor conEntering condenser water temperature (AI) T

trol module. Requiring each manufacturer to be compatible with a particular control system to integrate the control of all components is becoming increasingly expensive and slowing the pace of technological advancement that is so important to our industry today.

Importance of controls
Some in the industry question if all this emphasis on controls is really necessary. The answer should be obvious. For years, studies
Leaving condenser water temperature (AI) T

Return chilled water temperature (AI) T

Condenser Evaporator R R R

Supply chilled water temperature (AI) T

PS DDC interface module

CS

Run status (DI) Alarm status (AI) Head pressure (AI)

Start/stop (DO) Chiller power (AI) Vane position (AO) Point type legend DI = digital input DO = digital output AI = analog input AO = analog output

Transducer interface legend T = temperature sensor R = relay PS = pressure sensor CS = current sensor

1 Typical DDC interface to chiller.

HEATING / PIPING / AIR CONDITIONING I AUGUST 1994

45

Open/standard protocols

have shown large numbers of office workers are dissatisfied with the workplace comfort and air quality. The underlying problems only rarely involve a lack of capacity of HVAC components to provide necessary heating, cooling, or ventilation. Rather it is almost

Condenser Evaporator DDC interface connect plug

Connect to DDC system trunk

2 DDC interface to trunk compatible chiller.

Architects and interior designers have traditionally worked to design buildings and spaces that are visually attractive and inviting. HVAC designers are now beginning to realize they must design mechanical/electrical systems that are environmentally attractive and inviting. As building owners and managers are learning to exploit their tenants desires for improved comfort and air quality, our industry is having to learn how to design HVAC systems that provide improved individualized terminal control. The major difference between such systems and the systems of today is the high level of integrated controls required to provide such conditions economically.

Pitfalls of protocol standards

entirely the lack of proper controlparticularly zone control that breeds this dissatisfaction. A major roadblock to HVAC designers understanding of the need for improved controls stems from our industrys historic assumption that uniform temperature and ventilation conditions within each building will adequately satisfy all occupants, and that under normal occupied conditions, the internal loads throughout a building are reasonably uniform. Acceptance of this assumption has reduced the requirement for effective controls in the minds of many designers. However, the notions that uniform load conditions actually exist in buildings today and that uniform comfort conditions are acceptable to building occupants need to be reconsidered. Studies have found that the perception of ideal comfort and air quality may vary substantially among individuals. More recent studies are also showing a strong link between worker productivity and ability to control space conditions. Complicating the picture are the increasingly large variations of heating and cooling loads within areas of buildings today.
46

Some designers and building operators believe the development of communication protocol standards will solve our industrys problems with controls by simplifying the process by which control systems are designed and procured. These individuals may still believe a standard protocol will permit designers to treat DDC components as they remember treating pneumatic controls componentsas interchangeable components. This is an unrealistic expectation for DDC-based systems. The utter simplicity of pneumatic controls, along with long-standing competitive forces, required each pneumatic control manufacturer to manufacture pneumatic devices that were functionally identical to other products. DDC controls by contrast are orders of magnitude more complex than pneumatic-based controls, and manufacturers have chosen substantially different development paths for their products. While many in the industry believe communication standards may foster a competitive environment in which certain functional characteristics are encouraged, even an optimistic scenario puts widespread interchangeable DDC

products well beyond the horizon. The introduction of standard communication protocols will not make controls design easier. Rather, it will add significant pitfalls in the path of such design work. Controls designs that rely on substantial intercommunication via standard protocols among DDC SYSTEM products from difCOMPONENT LEVEL ferent manufacturers will require speOperator's console cial consideration. Such designs will very likely encounter startup and DDC proprietary network ongoing operational problems unless the exact nature of the intercommunication requirements Stand-alone panels are well understood by the designer and clearly expressed in Proprietary subnetwork the specifications. The persisting Unit controllers (UC) idea that simply VAV boxes and other terminal control stating that each control component shall be compatible I/O point device with a particular communication protocol standard will ensure the operational integrity of a DDC system network needs to be DDC SYSTEM COMPONENT LEVEL rebutted firmly. A
Key Modem
Operations Local area network

Host computer
DDC proprietary network

Printer
DDC system controllers Air handler controllers VAV zone controllers Dual duct zone controllers Misc. controllers I/O point device

Standalone panels

DDC controllers

AUGUST 1994 I HEATING / PIPING / AIR CONDITIONING

standard communication protocol may ensure that devices coexist on a network but not ensure that they will operate together. DDC controllers that are connected by a standard communication protocol network require the designer to provide a complete description of

virtually all aspects of the connection and operating criteria in the specifications so that each supplier understands not only the communication requirements but also the interoperability requirements of the product to be supplied. Such descriptions are well beyond the level

of specificity typically seen in building control specifications today.

Opportunities
What communication standards can do is aid those designers who are motivated to improve occupants perception of comfort

SAP 1

SAP 2

SAP 3

UC1

UC2

UC3

UC4

UC5

UC6

UC7

UC8

UC9

UC10

Input/output points

Input/output points

Input/output points

3 Typical present day DDC system architecture.

TEM VEL

ions work

work

lers lers lers llers lers

Cooling tower controller

VAV zone

DD zone

VAV zone

DD zone

VAV zone

Dual duct system controller

DD zone

Auxiliary cooling loop controller

Stand-alone panel (general controller)

VAV zone

VAV zone

VAV system controller

DD zone

VAV zone

DD zone

vice

Input/output points

Input/output points

4 Emerging DDC system architecture.

HEATING / PIPING / AIR CONDITIONING I AUGUST 1994

47

Open/standard protocols

DDC SYSTEM COMPONENT LEVEL

Operations

and air quality by economically integrating high-performance controls into HVAC and electrical systems. Knowledgeable designers will have the ability to employ standard protocols to develop more effective HVAC control strategies with economy. To show how this is possible, consider the integration of a standard centrifugal chiller with a building control system that is of different manufacture than the chiller. Traditionally, many such chillers have operated quite independently of the building controls. Such chillers often start and stop based on outside air temperature rather than demand for cooling and usually operate to a fixed preset chilled water temperature rather than what is required to meet current conditions. Experienced designers know that such operation is often counterproductive to both comfort and economy. The chiller plant may operate many more hours than necessary, and the fixed low chilled water temperature set point may make it uneconomical to reset supply air temperatures upward in cool weather for improved comfort. To achieve improvements in comfort and economy, many designers now look to integrate the chiller operation into the overall building control strategy. Fig. 1 illustrates how a typical DDC system instrumentation configuration can accomplish this. This interface includes a total of 10 DDC system points and would likely cost from $3000 to $6000 to accomplish. While the exact number of points and associated cost will vary depending on the interface requirements, the introduction of a standard communication protocol between the two systems will almost always reduce the interfacing cost. Fig. 2 shows how simple it may be to connect such a network. However, simply having the capacity to connect a piece of equipment to a network does not ensure
48

the equipment can be made to operate as desired. The chiller operating program may not permit the unit to be started or stopped over the communication network, or it may not permit the chilled water set point or the demand limit to be adjusted over the network. These features are operational and are not included in communication standards, but they are of critical importance to ensuring the success of the overall HVAC system operation. To be certain a communication network that relies on a standard protocol operates effectively, system designers must very carefully select all operational features that must pass across the network to components of different manufacture and then be certain that these features are provided by the vendor. As time passes and our industry gains more experience with protocol standards, a number of operational features are likely to become universal. But initially, a great deal of diligence on the part of designers will be required to ensure we are not punished as our prayers are finally being answered!

Key Modem

Local area network

Host computer

DDC proprietary network DDC system controllers Air handler controllers VAV zone controllers Dual duct zone controllers Misc. controllers

Printer

Standalone panels

I/O point device

Network gateway

Protocol standard network Standard compliant Devices from other manufacturers

State of standards
Currently, there are two major efforts underway within the industry to introduce communications standards, both of which are likely to have significant impacts over the next few years. The ASHRAE Standard 135P committee has been working diligently now for seven years to develop a communication standard. This standard is focused on HVAC controls and is expected to be released within the next year under the name BACnet. While the BACnet effort is to be applauded as a true grassroots movement, it should be approached with caution because it will likely be released without a definitive compliance testing format. Because compliance testing may initially be voluntary, manufacturers can claim compliance without demonstrating that such

compliance actually exists. In fact, manufacturers have already claimed compliance to BACnet features in their communication protocol. Since the final standard does not yet exist, these claims are obviously exaggerated. Designers should not react negatively to such claims because they may be demonstrating manufacturers good faith effort to support industry standards. But while we welcome such bold behavior by equipment manufacturers, we must also understand that in the short term, a high degree of diligence will be required on the part of designers to interconnect components in the BACnet environment successfully. The second major protocol standard effort in the industry has been developed by Echelon, an independent com-

I/O point device

DDC SYSTEM COMPONENT LEVEL

Operations Local area network

DDC proprietary network DDC system controllers Air handler controllers VAV zone controllers Dual duct zone controllers Misc. controllers I/O point device

Network gateways

Standard compliant Devices from other manufacturers

I/O point device

AUGUST 1994 I HEATING / PIPING / AIR CONDITIONING

TEM VEL

ions

work

work

lers lers lers llers lers

Cooling tower controller

VAV zone

DD zone

VAV zone

DD zone

VAV zone

Dual duct system controller

DD zone

Auxiliary cooling loop controller

Network gateway

VAV VAV zone zone

VAV system controller

DD zone

VAV zone

DD zone

vice

Input/output points

Input/output points

work iant ther rers

Boiler controller

Chiller controller

Motor control center

vice

Input/output points

Input/output points

Input/output points

5 DDC architecture with gateway to standard protocol network.

TEM VEL

ions

work

work

llers llers llers llers llers vice

Cooling tower controller

VAV zone

DD zone

VAV zone

DD zone

VAV zone

Dual duct system controller

DD zone

Auxiliary cooling loop controller

VAV zone

VAV zone

VAV system controller

DD zone

VAV zone

DD zone

Input/output points

Input/output points

ways

Network gateway

Network gateway

Network gateway

liant other rers

Boiler controller

Chiller controller

Motor control center

vice

Input/output points

Input/output points

Input/output points

6 DDC architecture with gateways to standard protocol devices.

HEATING / PIPING / AIR CONDITIONING I AUGUST 1994

49

Open/standard protocols

puter communications firm. This firm hopes to be the supplier of choice for communications chips in a fashion similar to the way Intel supplies microprocessor chips to the computing industry. Echelons product is called LonWorks. The concept behind LonWorks is that computer manufacturers whose products require communication among processors can buy these chips, which are preprogrammed to provide the network services. Such a scheme may be attractive to manufacturers because they could concentrate their efforts on the performance of their products and not on communications issues. Echelon is in the business of selling chips and development packages and is not really concerned with what protocol is employed in its chips, so long as it works. In that sense, Echelon is not really competing with BACnet. In fact, Echelon asked the ASHRAE Standard 135P committee to approve it as one of the environments for operating BACnet, but the committee turned Echelon down. Echelon has already developed its own proprietary protocol and has aggressively entered the DDC system industry marketing its chips to operate with that protocol.

DDC technologies. The key to the success of LonWorks will be cost. DDC manufacturers incur costs both in purchasing each LonWorks chip as well as the expense of developing interfaces to their controllers. The current explosion in distributed processing for DDC systems today means that a typical building of the future may employ hundreds or thousands of separate DDC controllers communicating together. What looks like a small marginal cost for an individual controller may soon become a large cost in an increasingly competitive environment. For LonWorks to be widely employed, it will have to compete with the costs of employing BACnet or proprietary communications networks in such installations.

Implementing standards
While the rapidly approaching competition for the hearts and minds of communication networks may become interesting, HVAC designers will do better if they take a very practical and conservative approach over the next few years. Designers should understand that a far more important development is now underway in the DDC industry. Fig. 3 shows DDC system architecture as it commonly exists today. The basis of many DDC systems today is the stand-alone panel (SAP). Generally, unitary controllers that operate terminal boxes or other unitary devices exist on subnetworks at the SAP level. However, networks of the future are more likely to connect all controllers at a single network level. Such architectures have been shown to improve the consistency of controller operation as well as integration capacity. Such future control networks may more closely resemble that shown in Fig. 4. Additional devices such as communication controllers and/or routers may be employed to direct messages along this busy network. Many manufacturers are now

Protocol wars
HVAC designers are soon to find that BACnet, LonWorks, and a number of open protocols that are being offered by some manufacturers will be competing over the next few years to set the standard for building control communications. For LonWorks, the motive is financial; for BACnet, the motive is more altruistic. If BACnet is found to be a workable standard, and if our industry has the energy and finds the resources to support, improve, and extend the standards, then it may very well flourish. Otherwise, BACnet will likely wither on the vine and be limited to a few esoteric applications noteworthy only for their distance from the beaten path of
50

developing networks similar to that shown in Fig. 4. This network configuration makes every controller more like the stand-alone panel with which we are familiar. As this trend toward further distribution of processing capabilities continues, the demands upon the communication network are increasing rapidly; and fast, efficient communication networks are becoming necessary. It is within this environment that manufacturers are now being asked to develop standard compliant networks. These new demands on throughput brought on by the increasing distribution of processing resources make control manufacturers nervous (and should make designers nervous too!), especially when they are being asked to connect to equipment manufactured by others as well. Designers need to be very careful about their designs that depend on standard networks to connect components of different manufacturers to operate together. To be certain adequate accountability exists when the equipment is installed, the following rules should be followed in developing such designs. N Focus control system specifications on a complete system that will be the responsibility of a single vendor. N Require that the control system vendor supply the gateway to the protocol standard network(s) that will be employed to connect equipment of different manufacture. N Include precise descriptions of information that must be exchanged between the control system and equipment that may be of differing manufacture. Such descriptions must be included in the control specifications and the equipment specifications as well. N Include a procedure for assessing and resolving network problems that may appear. Remember, startup failure is only one possible problem and probably the easiest to solve. Occasional

AUGUST 1994 I HEATING / PIPING / AIR CONDITIONING

malfunctions or sporadic losses of communications may be more common and certainly are more troublesome types of networking problems. Until the industry has more experience with the emerging protocol standards and their use in DDC networks, I recommend that designers develop their designs around gateways. Fig. 5 illustrates a DDC communications gateway connecting a chiller, boiler, and motor control center to a DDC system network. The gateway may or may not be an actual physical device, but it is a point of responsibility transfer. The DDC system vendor is responsible for maintaining the DDC system network and gateway. The chiller, boiler, and motor control center manufacturers have the responsibility of providing and accepting information as selected in a format compatible with the stan-

dard network. Initially, I recommend designers consider a separate gateway for each different manufacturers equipment that is intended to be interfaced to the DDC system. Such a network is shown in Fig. 6. This approach limits the finger pointing in the event of an interface problem to no more than two separate entities. It also may provide clues to the culprit. For example, if the chiller and boiler are communicating well with the DDC network but the motor control center is not, then likely the problem is in the motor control center controller. However, if each is experiencing similar faults, then the problem likely resides in the DDC gateway module.

Things to remember
The imminent arrival of communication standards in the building control industry can be

an enormous opportunity or a punishment, depending on how they are employed. Building operators and designers who look to communication standards as a means of simplifying the procurement and operation of building control systems may be very disappointed with the results. Those who rise to the challenge and put forward the extra effort to provide effective controls designs will find that protocol standards provide more economical interface capabilities, resulting in integrated control strategies never before practical. At least for the short term, designers should be very cautious with the application of networks employing standard protocols. Selection should clearly and completely describe all communication and interoperating requirements and establish accountability in case of problems associated with such networks.

HEATING / PIPING / AIR CONDITIONING I AUGUST 1994

51

..........................................................
DDC TECHNOLOGIES

DDC technologies are permitting a new flexibility in the traditional rules concerning the need for linear signals and responses with input/output devices

DIRECT DIGITAL CONTROL FUNDAMENTALS

By THOMAS HARTMAN, PE, Principal, The Hartman Co., Marysville, Wash.
ver since the energy crisis, when digital controls (then called EMCS for energy management and control systems) were unceremoniously ushered into widespread use for HVAC control, the industry has tried to make them look and act like the pneumatic controls they have superseded. Only occasionally are some of the profoundly expanded opportunities available with digital

controls applied effectively. Furthermore, terms like reset schedule and direct acting, relevant only to pneumatic systems, are still commonly employed in what is now the digital controls era. While the process of transition to digital control technologies tolerates this mixed bag, a multitude of new demands are requiring our industry to move ahead and realize the full potential of digital control technologies. Building occupants are demanding more comfortable and higher quality environments. Building owners continue to press for greater economies in construction, opera-

tion, and maintenance. Finally, a variety of pressures are upon us to provide more precise control and documentation that standards for temperature, ventilation, and indoor air quality are being met. In this article, I will discuss how DDC technologies permit a new flexibility in the traditional rules concerning the need for linear signals and responses with input and output devices. When properly applied, this new flexibility can reduce the cost of DDC technologies. Next month, I will show how, by combining these fundamentals with emerging inter-manufacturer controls integration, designers can achieve new horizons in performance and energy efficiency.

36 34 32 30 28 26 24 22 20 18 16 14 12 10 8 6 4 30 40 50 60 70 80 90 100 110 120 Temperature, F

Why linear devices?

When pneumatic controls dominated our industry, building owners paid a high price for modulating loop performance and stability. One of the prices paid was the requirement that input and output devices be linear with respect to the system variable they sensed or controlled. This need for linear response was essential to match the limited control capabilities of pneumatic controllers. A number of rules and conventions were established within our industry that made achieving this linear response requirement easier. Among these were the development of the equal percentage valve, which included
63

1 Resistance curve for thermistor temperature sensor.

Resistance, ohms (Thousands)

HEATING / PIPING / AIR CONDITIONING n FEBRUARY 1995

DDC fundamentals

the seemingly backwards rule of thumb that called for sizing control valves smaller than the pipe size. Similarly, mechanical sensing devices were constructed to provide linear change in control air pressure over their entire sensing range. While these conventions and rules of thumb served the days of pneumatics, they now need to be rethought. Requiring what I call external linearization in digital control designs adds costs in two ways. Linear devices are often more expensive than nonlinear devices that may offer improved levels of performance in DDC applications. Further, linear output conventions, such as designing a high pressure drop through valves or dampers, carry a substantial continuous operating energy penalty. By developing new rules and conventions, the knowledgeable designer can produce designs that have lower first and operating costs and may operate more reliably as well.

tal controls offer improved modulating control capabilities, including proportional/integral/derivative (PID) controllers, these control loops continue to be based on the principle of linear response, at least over certain ranges. However, in most typical applications, digital controls can easily internally linearize both input signals and output control functions.

Internal linearization of inputs

One way to reduce the cost of some DDC configurations is to permit nonlinear input devices and use the DDC system for scaling to achieve the correct reading over the range required for the application. I continue to see DDC specifications that limit the selection of input devices to those that provide a linear signal to the DDC system over a wide range of values. Except in special cases, this is an unnecessary requirement that adds costs and may cause other problems. Consider temperature sensors. Fig. 1 shows a resistance curve for an inexpensive thermistor type temperature sensor that may be employed for room temperature sensing. Thermistors are excellent choices for HVAC applications. They are in-

Linear devices in the DDC era

The need for linear response in modulating control loops has not been eliminated by the introduction of digital controls. While digi12

11 10 9 8 DDC linearized curve 7 Sensor curve 6 5 60

Resistance, ohms (Thousands)

expensive, have excellent accuracy and very low hysteresis, and respond quickly to temperature changes. Furthermore, at temperatures normally involved in HVAC applications, thermistors have excellent long-term stability (some care should be taken in choosing thermistors when temperature may rise above 240 F). Finally, because thermistors are typically high resistance (10,000 ohms is typical), they are not affected by variations in wiring distances. However, some designers continue to exclude thermistors because the input signal is not linear with temperature over wide temperature ranges. Instead, low impedance RTD type sensors are often specified. This type of sensor typically requires an electric circuit at the sensor that linearizes and transmits the signal in a way that it will not be affected by wiring resistance (usually a current loop signal is used). Employing low resistance RTD sensors with additional electronics presents a number of potential problems in DDC applications. First is the matter of accuracy. While the RTD sensors themselves provide excellent accuracy, it is not uncommon to find end-toend accuracies (I use end-to-end as the comparison of the value read by a precision thermometer at the device compared with the actual reading at the DDC system operators terminal) out of tolerance. Calibration of the current loop input may be more difficult than that of a simple resistance type thermistor. Other potential problems with RTDs range from the additional electronics (usually located at the device) that may complicate reliability issues all the way to how the
continued on page 66

65

70 Measured variable

75

80

2 Nonlinear sensor resistance curve. The sensor performance curve is a smooth curve over the sensors operating pressure. The DDC linearized curve is a series of straight lines that closely approximates the sensors performance.

64

FEBRUARY 1995 n HEATING / PIPING / AIR CONDITIONING

DDC fundamentals
Capacity, precent (Valve flow, percent)
continued from page 64

100 90 80 70 60 50 40 30 20 10 0 0 10 20 30 40 50 60 70 80 90 100 Valve travel, precent (Coil design flow, percent) Valve characteristic Coil characteristic Combined valve and coil

sensor and electronics are configured, which on occasion has been found to affect adversely the sensor signal. Table functions that are now readily available with DDC products can be employed to scale thermistors and other nonlinear devices over a wide range of values. Fig. 2 shows how a DDC system can linearize a continuous, nonlinear sensor input curve with a table function. A number of straight line curves are established in the table function to approximate closely the nonlinear function of the device. As long as simple, inexpensive devices can meet the repeatability, hysteresis, and stability requirements for an HVAC application, such devices should not be rejected because their signals are not linear.

3 Design of chilled water valve and coil combinations for proportional control.

100 90 80 70 Capacity, percent 60 50 40 30 20 10 0 0 10 20 30 40 50 60 70 80 90 100 DDC analog output, percent (Valve travel, percent) Combined valve and coil Scaled DDC output to valve

Is linear output required?

Once it is understood that input devices need not be linear, it is not a great leap to recognize that the response from output devices controlled by analog outputs similarly need not be linear. However, the issues here are more complex and more ingrained in the rules of thumb that engineers frequently apply automatically, so some indepth discussion is required. Because of the pneumatic background, valve design manuals commonly stress the need to select coil/valve combinations for which equal increments in valve position will effect equal increments in heat transfer of a typical heating or cooling coil throughout the stroke of the valve actuator. Fig. 3 shows how traditional design practice seeks to linearize the overall performance of valve and cooling coil. Carefully selecting a coil and valve combination can provide nearly linear performance over the entire range of load possibilities. Such selection is done because it is assumed that the valve will be operated by a controller with a fixed proportional gain. Though this design principle is still widely employed, it is no
66

4 Use of scaling to linearize control for DDC control.

longer applicable in many modern HVAC applications. In VAV cooling coil applications, the variations of air flow and air/chilled water temperature characteristics act to change dynamically the heat transfer characteristics of the valve/coil arrangement as these parameters change. This makes it very difficult to select a valve/coil combination that will be linear through the variety of conditions that may accompany its operation. The higher performance of DDC systems permits designers much greater flexibility in the design of modulating controls without establishing

static (and therefore unrealistic) design criteria. Fig. 4 shows a valve and coil combination that does not provide a linear response of valve position to coil capacity. However, modern DDC systems permit scaling tables to be applied to analog outputs as well as the inputs. Output scaling permits an inherently nonlinear device combination to respond in a linear fashion to signals from the DDC system. In this example, the valve and coil combination provides about 70 percent of the design cooling capacity at about 20 percent valve travel. The DDC output to the valve can be adjusted with the scaling table to position the valve at 20 percent
continued on page 68

FEBRUARY 1995 n HEATING / PIPING / AIR CONDITIONING

DDC fundamentals
continued from page 66

100 90 80

travel at a 70 percent output signal from the DDC system. The scaling factor allows standard PID control to operate the valve effectively because of a software linearization of the valve/coil combination. However, the chilled water flow and heat transfer performance assumed for Fig. 4 is valid only for constant load-side flows and inlet temperatures and for constant chilled water supply temperatures. Whether inherent in the system design or for optimization reasons, rarely in real HVAC applications do these other variables remain constant as control loops operate. As previously discussed, the issue of linear output combinations has therefore been only weakly resolved in the past by attempting to linearize components at one set of system conditions. Obtaining good control over wide ranges of system conditions can be resolved far more completely and effectively with the higher performance capabilities of DDC systems. The proportional, integral, and derivative gains can be tied to algorithms that adjust their values as the variables such as load-side flow, temperatures, and chilled water temperature change. Even more impressive is the emergence of self-tuning controllers. These controllers continually re-establish the various gains associated with a control loop to provide continuously precise control without hunting. The benefits of self-tuning are especially important because variables beyond the immediate control loop can have profound and widely varying effects on each control loop. Self-tuning features are becoming widely available with DDC systems and are enormously effective in adjusting control loops to continue stable operation as other system variables change.

70 Capacity, percent 60 50 40 30 20 10 0 0 10 20 30 40 50 60 70 80 90 100 Valve travel, percent Valve A designed for 30 ft pressure drop and 13 F approach Valve B designed for 5 ft pressure drop and 13 F approach Valve and coil design B Valve and coil design A

5 Design of valve and coil combinations for proportional control.

Controllability
As previously discussed, selecting equipment for linear response should not be an overriding con68

sideration for designers in this era of digital controls. However, this does not mean designers can be imprecise in their designs or in the selection of control loop components. The issue of controllability is one that will continue to play a prominent role both in the design of systems and the selection of individual components. Controllability remains largely a sizing issue. If a valve is oversized for given conditions such that the smallest increment possible from the control loop will substantially overshoot the desired control conditions, the loop has become uncontrollable. This is a problem that typically emerges during periods of low load. Fully understanding the issue of controllability and applying DDC capabilities correctly allows designers to solve such problems and at the same time vastly improve the efficiency and performance of these systems. Selecting a control valve with a lower pressure drop will reduce the pumping power required to meet the load conditions. Traditional practice strongly condemns the idea of employing large valves with lower pressure drops because of the nonlinear response and the lack of controllability at low loads. Fig. 5 illustrates the dilemma. The valve/coil combination with Valve A may be selected according to traditional design practice because it is reasonably controllable at low

loads. The vertical axis intercept represents the smallest incremental cooling transfer possible as the valve is cracked open. Note that it is smallonly about 10 percent of the design maximum cooling rate. The coil combination with Valve B has a much lower pressure drop because Valve B is a larger size valve. While valve/coil Combination B would require less pumping power, the Y-axis intercept is much higher than that for Combination A. Traditional design criteria typically declare Valve B unsuitable for the application because it is uncontrollable at lower loads and the valve position/cooling capacity relationship is nonlinear. But when it is integrated with a high-performance control system that can adjust both the chilled water temperature and the loop head pressure, will linearity and controllability of Combination B really be a problem?

System dynamics
To see how this question can be answered, consider the graphs in Figs. 6 and 7. Fig. 6 shows the operation curves for valve/coil Combination B at a number of different approach (chilled water supply less air temperature leaving coil) temperature conditions. It is clear that increasing the chilled water temperature relative to the leaving air temperature markedly improves the concontinued on page 70

FEBRUARY 1995 n HEATING / PIPING / AIR CONDITIONING

DDC fundamentals
continued from page 68

trollability at low loads. Similarly, Fig. 7 illustrates that the decrease in pressure across the valve/coil combination also improves the controllability at low loads. Designers can use these relationships to reduce substantially the problem of controllability. At periods of uniform low loads, the DDC system can reduce the head pressure across a valve and increase the chilled water temperature to improve controllability. If all valves on a common chilled water loop experience similar decreases in load concurrently, as is typical in many HVAC applications, this parameter adjustment is a great help in improving controllability at low loads. It is apparent from the two figures that larger rangeability and low load controllability are achieved by controlling the chilled water temperature for load adjustment. Raising the chilled water temperature provides a bonus of chiller efficiency increases, but chilled water adjustment reduces pumping savings because a higher chilled water temperature increases the water flow necessary to meet loads. Additionally, under certain circumstances dehumidification requirements may limit the permissible chilled water adjustment. Exploiting the integrated control capabilities of DDC systems and controlling chilled water temperature and hydronic loop pressure in coordination with the control valves allows valve/coil Combination B to perform very well in many HVAC applications. Next month we will focus on the level of integration required to make valve/coil Configuration B operate effectively. We will discuss integrating the operation of the various equipment involved in providing comfort, possible now through the industry moves to provide communication bridges among manufacturers. By concentrating on selecting the most costeffective input/output devices and by utilizing the emerging commu70

nications pathways between equipment from various suppliers, we will see that new horizons of performance and energy efficiency can be attained with simple and economical controls configurations.

Summary and conclusion

Designers must exploit the benefits of higher performing DDC

systems to develop an understanding of the fundamentals of interfacing hardware points to DDC systems. In so doing, a more in-depth look into total system operation must be evaluated before solutions are selected. Simply following traditional rules of thumb regarding linear input and output devices is a poor design practice in this digital controls era.

100 90 80 Design capacity, percent 70 60 50 40 30 20 10 0 0 10 20 30 40 50 60 70 80 90 100 Valve travel, percent 4 F approach Valve and coil B performance at various approach temperatures (assuming constant differential pressure and inlet temperature) 7 F approach 10 F approach 13 F approach

6 Heat transfer vs. valve travel for various approaches (leaving air temperature minus entering water temperature).

100 90 80 Design capacity, percent 70 60 50 40 30 20 10 0 0 10 20 30 40 50 60 70 80 90 100 Valve travel, percent Valve and coil B performance at various loop head pressures (assuming constant chilled water and air inlet temperatures) 4 ft head 8 ft head 12 ft head 20 ft head 16 ft head

7 Heat transfer vs. valve travel for various loop head pressures.

FEBRUARY 1995 n HEATING / PIPING / AIR CONDITIONING

..........................................................
DDC TECHNOLOGIES

NEW HORIZONS FOR HVAC CONTROL

How installation and operating costs of HVAC systems can be dramatically reduced by extending the opportunities afforded by controls integration
By THOMAS HARTMAN, PE, Principal, The Hartman Co., Marysville, Wash.
n last months article, we saw how the first costs and operating costs associated with modern DDC systems can be reduced by rethinking our selection and application of input and output devices. This month, I intend to carry the idea of new thinking a step further and show how installation and operating costs of HVAC systems can be even more dramatically reduced by further extending the opportunities afforded by controls integration. For example, to see the benefits of more rigorous design approaches and challenge our acceptance of traditional design practice, lets consider the operation of a variable flow chilled water loop. ing energy, the designer decides to incorporate a variable speed pump on the secondary loop. To decouple the constant chilled water flow requirement of the chiller evaporator from the variable flow through the loads, the designer decides upon a two-pump arrangement with a two-way modulating valve for each load and a variable speed drive to operate the secondary chilled water pump. The primary chilled water pump operates continuously with the chiller. For the secondary loop, traditional design manuals suggest sizing control valves for a full-flow pressure drop of at least 30 percent of the total system drop. This factor is often called valve authority, and rules of thumb dictate that it should be equal to or greater than the pressure drop through the load, which in this case is the cooling coils. As discussed in last months article, the substantial pressure drop through each valve is an effort to attain some measure of linearity between valve action and cooling effect delivered to the load. The high pressure drop also serves to isolate the operation of each valve/coil arrangement from others. Lets start by calculating the total head across the secondary pump at full flow, then calculate the pumping power required. Depending on the physical location and piping lengths, a 20 ft drop across the loads, 30 ft drop across the valves, and perhaps 40 ft head
Chiller

Constant speed primary pump

Variable speed secondary pump

Load 1

Load 2

Load 3

Load 4

1 Primary/secondary chilled water loop.

Chilled water loop

Fig. 1 is a schematic of a chilled water system employing variable flow on the secondary (load) circuit. The loads are cooling coils in a buildings various air systems. In this case, the designer is confronted with a chiller that serves an extensive piping array with multiple chilled water coilsa reasonably common occurrence. Assume all the cooling loads vary fairly uniformly with outdoor conditions and the coils operate at low load conditions much of the time, which is typical of HVAC cooling systems. To reduce pump-

drop across the secondary piping at full flow conditions would be typical for a design using published rules of thumb. Assuming a total full-flow requirement of all the load coils of 1000 gpm, we can calculate theoretical pumping horsepower at full flow as follows: Pump hp = [gpm 8.35 lb per gal pump head]/33,000 ft-lb per min-hp Pump hp = [1000 8.35 (30 + 20 + 40)]/33,000 = 22.8 hp A traditional design would involve one or more differential pressure transducers in the loop
MARCH 1995

HEATING / PIPING / AIR CONDITIONING

75

DDC fundamentals
piping head loss is: 40 0.52 = 10 ft The horsepower requirement at 50 percent flow is: Pump hp = [(1000 0.5) 8.35 (50 + 10)]/33,000 = 7.5 hp The designer would then apply appropriate pump efficiencies and simulate or estimate the amount of time the pump and coils would spend at various conditions and calculate the power requirements as above. If the system operates long hours at low loads, the savings from a variable flow secondary loop will be substantial. ductions to maximize pump power savings, but adjusting the chilled water temperature upward offers savings too. The DDC system must be chosen so that it offers the use of advanced algorithms to adjust the chilled water temperature and secondary pump speed as load conditions change. Furthermore, the DDC system chosen should provide the ability to selftune modulating loops. The question remaining is how the system pressure and water temperature changes will affect the other valve/coil combinations on the cooling circuit. The assumption for this example is that all coils will experience similar loads, as is often the case with HVAC systems. However, the designer must be very careful to ensure this is actually the case. If one of the coils serves an interior computer room that has high constant year-round cooling loads or if coils serve different perimeter zones of a building subject to high solar gains, these unusual zones may have to be specially accommodated either by a booster pump or a separate chilled water circuit as shown in Figs. 3 and 4. In Fig. 3, a small booster pump is added to increase the differential pressure for Load 2, which the designer has determined will not fall as quickly as the others on the loop. In Fig. 4, two entirely separate loops have been configured to permit the separation of loads into groups that will have similar part-load patterns. The configuration in Fig. 4 may be a cost-effective configuration if the load groupings are in different locations and do not require extensive additional piping.

Chiller

Constant speed primary pump Variable speed secondary pump Control VFD

Load 1

Load 2

High-performance design
Load 3 DP Load 4

2 Primary/secondary chilled water loop with differential pressure sensor (DP) and controller.

as shown in Fig. 2 to ensure the pump maintains a 50 ft head across the valve(s) and coil(s). If all the coils spend large amounts of time operating at less than design capacity, the piping head loss would drop with the square of the water flow decrease. The pressure drop through the coil would similarly decrease, but the savings would be excluded because the controls keep a constant pressure across the valve/coilsthe pressure drop across the valves actually increases as the flow decreases. If reasonably even load profiles for all coils are assumed, the pipe head pressure drop will increase with the square of the flow. At 75 percent flow the piping head loss is: 40 0.752 = 22.5 ft Using this new pressure drop figure for the piping, we see that the theoretical horsepower requirement at 75 percent flow is: Pump hp = [(1000 0.75) 8.35 (50 + 22.5 )]/33,000 = 13.8 hp At 50 percent of design flow, the
76
MARCH 1995

The simplified analysis above shows how substantial energy use reductions are possible by the application of variable flow to typical hydronic loops in HVAC systems. However, designers with a good understanding of high-performance DDC controls should ask themselves why they allow their designs to be limited by traditional design methodology. Design teams often lose track of the reasons for the rules of thumb they apply so regularly. When this happens, designers run the risk of applying outdated design techniquesan important reason many designs fail to meet expected levels of performance. As we saw in last months discussion, a control scheme that integrates chilled water temperature control with secondary chilled water loop control can provide effective cooling coil control with considerably reduced control valve pressure drop. In such a design, the DDC system operates the integrated system by increasing the chilled water temperature and decreasing the differential pressure of the loop as the loads on the coils decrease. Lets see how applying the design techniques discussed last month would affect this example.

Performance control program

Thinking back to last months article, consider how the chilled water loop of Fig. 1 might achieve improved energy performance when operated by an integrated control program. The designer may develop a high-performance control program for the loop with several basic elements as follows:
continued on page 80

Further improvements
Obviously, it is wise to keep the chilled water temperature adjustment at levels that permit flow re-

HEATING / PIPING / AIR CONDITIONING

DDC fundamentals
continued from page 76

Chiller Constant speed primary pump

The pump speed will be decreased and the chilled water temperature will be increased to ensure at least one of the control valves serving the loads served by the loop is fully opened at all times during operation. The system will further ensure that all loads are met at all times. Moving from a full-load to part-load situation, the pump speed will first be reduced down to 90 percent of maximum speed while the chilled water temperature remains constant. Thereafter, the chilled water temperature will be increased along with pump speed reductions at a relative reduction ratio that maintains a minimum valve opening of 20 percent. Note that the control program outline does not anticipate the need for a loop differential pressure input.

Evaluating improvements
As discussed last month, a highperformance control system permits the control valves to be selected for lower pressure drops when valve control is integrated

Chiller

Constant speed primary pump

Variable speed secondary pump

Load 1 Booster pump Load 2 Check valve Load 3

with chiller and pump control as above to handle the issues of linearity and controllability. Assume the same piping and coils are selected as in the earlier example, but the control valves are sized for a 5 ft pressure drop at design flow. Now the total head for the secondary loop at full load conditions is the sum of 20 ft drop across the loads, 5 ft drop across the valves, and 40 ft drop across the secondary piping. At the full flow of 1000 gpm, the theoretical fullload pumping horsepower can be calculated as: Pump hp = [gpm 8.35 lb per gal pump head]/33,000 ft-lb per min-hp Pump hp = [1000 8.35 (5 + 20 + 40)]/33,000 = 16.4 hp This compares to the 22.8 hp for the traditional loop design. Although this design requires larger valves to achieve a lower pressure drop at full flow, the pump, motor, and variable frequency drive components are all only about 70 percent as large as required for a traditional design, which means that the overall cost of the mechanical components of the design is likely to be the same or less than the initial design. At lower loads, the pumping energy calculations are more complicated because both the chilled water temperature and the loop pressure will be adjusted by the high-performance DDC system to meet the specific requirements of the various loads. Raising the chilled water temperature raises the chilled water flow required to meet the loads but also increases the efficiency of the cooling plant and results in additional energy use reductions. Reducing the total loop pressure as the load decreases will enhance the energy savings beyond the savings from flow reduction. Using the performance graphs for the valves in last months article, we can assume that at 75 percent average load the system operates at an
3 Primary/secondary chilled water loop with booster pump on one load.

Variable speed secondary Pump 2 Variable speed secondary Pump 1 Load 1

Load 2

Load 3

Load 4

4 Primary/secondary chilled water loop with two secondary loops.

Load 4

11.5 F approach. At this operating point the flow would be: 1000 (0.75 13/11.5) = 848 gpm The piping and coil head losses are, respectively: 40 (848/1000)2 = 28.8 ft 20 (848/1000)2 = 14.4 ft The assumed head loss through the valves depends on the control strategy and the piping arrangement to be employed. Generally, the head loss through the valves can be assumed to decrease at the same ratio as the piping and coil head losses, but this is not always the case. Conservatively, it is assumed for this example that the valve head loss remains at 5 ft. Based on these assumptions, the horsepower requirement at 75 percent flow is: Pump hp = [848 8.35 (5 + 14.4 + 28.8)]/33,000 = 10.3 hp This design clearly offers some additional part-load pumping savings over the original example, but more important is that it also offers savings from a 1.5 F increase in chilled water temperature. A chiller saving of 0.02 KW per ton-F energy reductions per degree increase in chilled water temperature yields an additional savings of more than 9 hp at the chiller for a total saving of more than 15 hp.
continued on page 83

80

MARCH 1995

HEATING / PIPING / AIR CONDITIONING

DDC fundamentals
continued from page 80

Variable speed chiller

At 50 percent of design flow, it can be assumed that the system operates at a 10 F approach. At this operating point the flow would be: 1000 (0.5 13/10) = 650 gpm The piping and coil head losses are, respectively: 40 (650/1000)2 = 16.9 ft 20 (650/1000)2 = 8.5 ft To be conservative in advance of a certain piping layout, we assume the head loss of the valves remains constant at 5 ft. With these assumptions, the horsepower requirement at 50 percent flow is: Pump hp = [650 8.35 (5 + 8.5 + 16.9)]/33,000 = 5.0 hp As before, the part-load operation pumping savings are again overshadowed by the 3 F increase in chilled water temperature. A chiller saving of 0.02 KW per ton-F energy reductions per degree increase in chilled water temperature yields an additional saving of about 7 hp at the chiller for a total saving of approximately 18 hp. Clearly, the use of integrated control and low pressure drop valves results in substantial additional energy savings for this chilled water system.

Variable speed chilled water pump

Load 1

Load 2

Load 3

Load 4

5 Single-pump chilled water loop.

In Fig. 5, the chiller is a variable speed unit that offers a high turndown ratio and high coefficient of performance at low loads. The required rate of flow through the chiller obviously depends on the cooling load. This is a good design fit because the flow through the loads also varies with the loading. However, this does not mean that such a system will work adequately under all load conditions without specific attention to the chiller flow.

quence for a single-pump circuit provides a continuous calculation of cooling load requirements for each load. The cooling loop would be shut down until the sum of the calculated load requirements reaches a specific value that depends in part on the anticipated upcoming conditions. Once enabled, the cooling plant will operate as described above in the high-performance sequence except that the chilled water temperature would be raised any time the evaporator heat exchanger temperature differential reaches the maximum differential for the load conditions. Raising the chilled water temperature will require higher flows to satisfy loads and will automatically return the heat exchanger to optimal performance levels. It is clear that the simpler configuration of Fig. 5 reduces the first cost of the system and also provides some further reduction of operating costs over the Fig. 1 configuration.

Benefits of high performance

The above example is indicative of the opportunities available to those designers with the expertise necessary to apply high-performance DDC to HVAC systems. The expected results for mechanical designs that are carefully integrated with high-performance controls are: Smaller (and usually simpler) equipment sizes with accompanying savings that can reduce the system costs or be invested in higher quality, longer lasting components. Lower total system energy use than what is possible with nonintegrated design and control strategies. Control precision that is superior to that of traditional control approaches.

New horizons for designers

So far, our design thinking has dealt only with the controls aspect of a traditional chilled water loop layout. Now that we understand the possibilities of integrated operation, lets rethink the physical configuration of Fig. 1. The striking feature of Fig. 1 is the necessity of having two pumps. Why? We know that below certain velocities of flow through chiller heat exchangers, the heat transfer capacity drops off rapidly. The purpose of the primary pump is to ensure that condition never troubles our system. However, with an integrated system, we can tell when the water flow is too low for chiller conditions because evaporator and water temperatures can both be continuously monitored. So, instead of Fig. 1, lets consider the simpler piping configuration in Fig. 5.

Single-pump systems
Two basic requirements must be met to ensure effective and efficient operation of the Fig. 5 configuration. First, the system must not be operated unless the cooling requirement is above a minimum threshold load. Second, the water flow through the chiller evaporator heat exchanger must be sufficient to maintain less than a predetermined maximum temperature difference between the refrigerant and chilled water under all conditions. In most HVAC applications, these requirements can be met as long as the DDC system has the capacity to integrate the control of the chiller, pump, and valves and the capacity to operate these elements with high-performance control algorithms. The operating se-

DDC cost considerations

Control system costs can be a factor in the feasibility of a highperformance design, but generalizations should be avoided until the exact control point requirecontinued on page 85

HEATING / PIPING / AIR CONDITIONING

MARCH 1995

83

DDC fundamentals
continued from page 83

ments are fully considered. Numerous economies are possible by integrating the microprocessor control units of various standalone elements. Modern, highperformance DDC control options may cost less than nonintegrated electronic control based alternates. The designer needs to consider that integrated, high-performance control layouts can usually be configured with fewer instrumentation points than traditional control approaches. Communication bridges between components of different manufacture are now becoming available. Furthermore, the intelligence of high-performance DDC systems can increasingly be used to replace instrumentation. Lets consider the chilled water loop. The traditional control of the secondary pump requires one or more differential pressure sensors and a controller to operate the variable frequency drive that sets the pump speed. Also required are control points to operate each valve and operate the chiller. If the chiller and air handlers are packaged units with microprocessor control, the only costs incurred for the high-performance control is the capacity to integrate the operation of this equipment. A highperformance DDC configuration can be simpler because the system uses the existing monitoring and control instrumentation. With microprocessor control at each that can be integrated in its operation, no additional I/O instrumentation is required.

Summary and conclusion

To fully exploit the benefits of high-performance DDC systems, designers must engineer designs before solutions are selected. Rules of thumb have no place in a high-performance design process. The benefits in improved energy performance, and in many instances lower first costs, are compelling reasons for designers to acquire the tools and expertise required to apply high-performance design approaches more widely.

NEW HVAC TECHNOLOGIES

The large variation in occupancy conditions typical for library and museum facilities requires designers to establish a means of providing real-time occupancy data to the control system.

Library and Museum HVAC:

New Technologies/New Opportunities PART 1
How new technologies can dramatically improve performance and economy of part-load operation in continuously operating systems
By THOMAS HARTMAN, PE, The Hartman Co., Marysville, Wash. energy, maintenance, and control performance offered by these new technologies make them very attractive not only for new building construction but as retrofit opportunities as well. the opposite is true. Integrated control and variable-speed technologies, properly applied, can actually reduce energy use when systems are kept running! To see how this can be true, lets first be certain we understand the energy implications of part-load operation. According to fan and centrifugal pump laws: M Fluid flow quantity varies directly with speed of fan or pump. M Static pressure delivered varies as the square of the speed. M Power required varies with the cube of the speed. These laws apply to axial and centrifugal fans as well as centrifugal pumps. They dictate that reducing fan or pump speed reduces flow in proportion to the speed but reduces power required by the cube of the speed reduction. This means that it is theoretically possible to provide 50 percent flow (or capacity) for 13 percent (0.53) of the full capacity power, an efficiency
April 1996

owhere are new HVAC technologies more dramatically showcased for their practical application than in buildings such as museums and libraries. These buildings require continuous operation and precise control under widely varying loads for visitors and contents alike, and providing these improvements with economy is the ideal calling for recently introduced HVAC technologies. Today, the knowledgeable designer can develop facilities with performance features far better than ever before while working within limited first and operating cost budget constraints. In Part 1 of this article, I will discuss how new technologies in such buildings can significantly improve part-load operation. The

Continuous HVAC operation

One of the most important but least recognized characteristics of HVAC design is the long hours most buildings operate at low-load conditions. Low-load operation is further increased in archival buildings where uninterrupted control of conditions, essential to the preservation of the materials they contain, requires some form of continuous HVAC system operation. Our industrys failure to capitalize fully on the benefits of new technologies has led operators of such facilities to believe that energy costs will be significantly higher because of the long hours of operation. It has been demonstrated in certain cases that

HPAC Heating/Piping/AirConditioning

57

New HVAC technologies

increase of almost 400 percent. Theoretically, power for 25 percent of the flow requires about 2 percent of full capacity power, an efficiency increase of over 1000 percent! The importance of these high
4000 3500 Chiller plant operating hours

mum efficiencies (usually near peak load) to meet the various intermediate-load conditions. The result could be a complicated mechanical room and even more complicated operating sequence that is costly to install and difficult to support.

Part-load operation

If we consider all the energy consuming compo3000 nents employed to provide cooling in typical buildings, 2500 we find that all of them, the fans, pumps, and centrifu2000 gal chillers, are subject to the basic fan and pump 1500 laws. While we know that delivery efficiencies, the variance of system loads 1000 and outdoor conditions, and other factors make it unre500 alistic to achieve theoretical part-load operating effi0 Seattle Los Angeles Helena New York ciencies, we should expect Annual chiller operating hours to develop cooling systems that can operate much 1 24-hr building chiller operation. more efficiently at partpart-load operating efficiencies is load conditions than at relatively illustrated in Figs. 1 and 2, which infrequent peak conditions. depict simulated chiller load proDesigners often believe the solufiles for a continuously operating tion to saving cooling plant energy building in four U.S. cities. Note at part-load conditions is simply to that despite the wide variations in employ a variable-flow cooling dischiller plant operating hours as de- tribution system with variablepicted in Fig. 1, the chiller plant speed drives operating distribution load profiles for cities as shown in Fig. 2 are very similar and most 60 certainly skewed toward low-load conditions. Note also that these Seattle Los Angeles 50 profiles are based on a perfectly Helena New York sized chiller plant. In practice, designers tend to oversize chiller 40 plant elements, resulting in profiles even more skewed toward part loads. 30 A designer working with traditional technologies might develop 20 an indirect evaporative cooling circuit to maximize chiller plant energy efficiency at all the various 10 load conditions. This configuration uses tower water directly for cooling during low-load, low outside 0 <20 20 to 40 40 to 60 60 to 80 >80 ambient conditions. The design Percent of design maximum load might also employ multiple chillers of various sizes to permit combina- 2 24-hr building chiller operation based on percentage of operating hours at tions of chillers operating at maxi- various loads.
58

pumps. Unfortunately, present designs of these systems fail to optimize the energy reduction potential of such systems. For example, consider the chilled water distribution circuit in Fig. 3. In this widely employed design, the variablespeed distribution pump is controlled to maintain a fixed differential pressure set point. The location of the differential pressure sensor is usually at the end of the piping run. A controller operates each of the load control valves. Even though such systems are likely to be connected together in a DDC system network, the operation of each controller is almost always entirely independent of all others as shown in the figure. In Fig. 3, each individual control valve is typically sized for a substantial portion of the total system pressure drop at full load (usually 30 to 40 percent). The differential pressure set point is the sum of the full-flow valve and load pressure drop, which is in the range of 75 percent of the total full-load pump head, the remainder being the fullflow distribution piping pressure drop. The system operating curve for a system of this design is shown as the bold line in Fig. 4. Fig. 4 assumes the design full-flow head is 100 ft at 1000 gpm, and the differ-

HPAC Heating/Piping/AirConditioning

April 1996

Percent of total annual operating hours

120 110 100 90 80 Head, ft 70 60 50 40 30 20 10 0 1150 rpm pressure curve 900 rpm pressure curve 600 rpm pressure curve 1750 rpm pressure curve
System curve

40

ential set point for the distribution pump controller is 75 ft. The Y axis intercept is the differential head set point because even at zero flow the pump must operate to maintain that pressure. Lets see how this design performs at various load conditions. At full-load the power curve shows that the pump will draw about 33 hp. As flow is reduced to 50 percent, the pump speed is reduced to slightly more than 1450 rpm, and the pump draws about 14 hp. As flow drops further to about one-third, the pump speed remains approximately constant, and the power drops to approximately 11 hp. Further reductions see little further reduction in power. Although a reduction from 33 to 11 hp may seem to be substantial,
Chilled water return to chiller plant Bypass Variable-speed distribution pump C Load 1 C Chilled water supply from chiller plant

24

1450 rpm

power cu

rve
16

1150 rpm e power curv

900 rpm power curve 600 rpm power curve

100

200

300

400

500

600 T00 Flow, gpm

800

900

0 1000 1100 1200

4 Operating curve for system shown in Fig. 3 assumes design full-flow head of 100 ft at 1000 gpm with the differential set point for the distribution pump at 75 ft.

120 110 100 90 80 Head, ft 70 60 1450 rpm pressure curve

1750 rp wer m po curve

40 1750 rpm pressure curve 32 Brake horsepower

24

rv

40 30 20 10 0

900 rpm pressure curve 600 rpm pressure curve

1150 rpm power curve

8 900 rpm power curve 0 1000 1100 1200

Load 2

600 rpm power curve

0 100 200 300 400 500 600 700 Flow, gpm 800 900

C Load 3 C Load 4 Differential pressure sensor and controller DP

5 Operating curve for a system design that permits the pressure differential across the valves and loads to fall as flow requirements are reduced, producing dramatic increases in pumping efficiency.

3 Schematic of typical variable-flow chilled water distribution system.

it is actually a stark display of the failure of this design. The operating efficiency (flow per hp) of the system remains almost constant despite the part-load conditions; one-third of the design flow requires exactly one-third of the design power. At lower flow requirements, the efficiency begins to decrease further.

Better part-load efficiency

The challenge for designers today is to design systems that can

better exploit the benefits of new variable-speed and control technologies. Because HVAC systems, especially for continuously operating public buildings such as museums and libraries, spend long hours at part-load conditions and because new technologies enable such enormous improvements in part-load efficiency, the industrys current focus on full-load performance criteria is missing the mark entirely. Instead, designers must come to view full-load operation as primarily a sizing issue. It is essential that each design adequately address full-load condi-

tions. But lets be clear that at present, operating costs under these infrequent conditions are only a very minor portion of annual operating costs. The failure of the widely employed variable-flow distribution schematic of Fig. 3 is that it does not attempt to mitigate problems associated with the second power relationship between pump head and speed. Because of pressure set point constraints, the pump in this example cannot operate at less than 1450 rpm. This severely limits power reduction as flow requirements are reduced. Now imagine a
April 1996

Sy

50

st

1150 rpm pressure curve

1450 rpm po

em

wer curve

cu

16

HPAC Heating/Piping/AirConditioning

Brake horsepower

1450 rpm pressure curve

17

mp 50 rp

ower

curve

32

59

New HVAC technologies

design that permits the pressure differential across the valves and loads to fall as flow requirements are reduced. Such a system curve may look more like the one in Fig. 5. Note in Fig. 5 that as the flow requirement falls to one-third, the pump power drops to less than 2 hp, a nearly 600 percent increase in the pumping efficiency compared to full-flow conditions. The question posed by Fig. 5 is how to accomplish such part-load efficiencies. Some answers are shown in Fig. 6. In Fig. 6, the control of the variable-speed distribution pump is accomplished by transmitting the conditions at each of the loads served to the pump controller rather than employing a differential pressure sensor as in the Fig. 3 schematic. Also, the load control valves would typically be line sized without a substantial pressure drop.
Chilled water return to chiller plant Bypass

fied, the distribution pump gradually speeds up. Such a simple, network-based control scheme is very effective in substantially improving part-load efficiencies of a chilled water distribution system if the load profiles for the loads served (Loads 1 to 4 in this example) are similar. Note that the control configuration in Fig. 6 is actually simpler than that of Fig. 3 since it employs no differential pressure sensor. Note also that the basic design philosophy employed in Fig. 6 can be extended to other components of a building cooling system, particularly the supply air fans and air distribution system. Operating an HVAC system suitably configured with these components continuously at low loads with corresponding high delivery efficiencies may use less energy than shutting them off and forcing them to start each day at high loads (to bring the Chilled water building back under consupply from chiller plant trol) and correspondingly To other low operating efficiency.
DDC controllers DDC DDC

The wide open areas typical in libraries and museums require special attention to ensure that suitable conditions can be maintained throughout the wide variations of indoor and outdoor loads and conditions they will experience.

Because HVAC design emphasis has not traditionally been placed on Load 1 part-load operating characteristics, the fan, hyDDC dronic, and cooling sysLoad 2 tems in a great many buildings operate at lower DDC efficiencies at part-load Load 3 conditions than at full DDC load. For buildings that Load 4 must operate continuously, this is an especially Integrated DDC enticing design opportucontrollers and network nity because the opportu6 Schematic of variable-flow chilled water nities for energy reduction distribution system designed to maximize are enormous. It is not too part-load energy savings. difficult to see how implementing or upgrading As the cooling load requirements HVAC systems that operate at at each of the zones fall, the corre- high part-load efficiencies can ofsponding valve begins to close. If fer vastly improved performance none of the valves are fully open and reduced energy costs. and the loads are satisfied, the A final note on part-load design distribution pump slowly reduces emphasis regards assessing the its speed. If one or more valves is level of occupancy to provide adefully open and its load is not satis- quate ventilation air. For build60

Variable-speed distribution pump

Part-load design emphasis

ings that may experience substantial variations in occupants or visitors, and museums and libraries fit this category, it is an absolute requirement that building occupancy levels be determined continuously from real-time information. Virtually all museums have some mechanism for tracking visitors; often it is a turnstile count. Nowadays such data can be directed with relative ease to the HVAC control system. Because the occupant density in libraries is usually low, occupancy sensors can be located throughout and connected to the HVAC control system to operate the lights as well as HVAC terminal units. Information from the occupancy sensors may then be suitable to assess occupancy of the building for a determination of the outside air requirement at all times. Whatever mechanism is employed, it is essential both for providing good environmental control and economy that the HVAC system design incorporate a means to assess real-time occupancy conditions accurately. This information is used by the control system to establish and maintain ventilation air flow set points as required throughout the facility. Next month I will continue my discussion on library and museum facilities focusing on humidity, dehumidification, and building HPAC pressure control.

HPAC Heating/Piping/AirConditioning

April 1996

LIBRARY/MUSEUM HVAC

Library and Museum HVAC:

New Technologies/ New Opportunities

Part 2
By THOMAS HARTMAN, PE, The Hartman Co., Marysville, Wash.

Museums often include exhibit areas that require special temperature and humidity conditions. Designers must take care to ensure sufficient isolation is provided for such special areas.

ast month, I discussed how new technologies are especially suited for museums and libraries, which require close control of indoor environments year-round, both for contents and occupants. The focus was on the dramatic improvements in the performance and economy of partload operation that these technologies permit. This month I will focus on year-round humidity control and building pressure control and their role in controlling the indoor environment of these and other institutional buildings.

Humidity control
The HVAC industry has been deluged with a great deal of misinformation about humidification in the last few years. Several years ago the indoor air quality consultant on a new office building on which we were working recommended against a humidifica-

tion system on the basis of IAQ concerns! Certainly very high humidity and water penetrating duct insulation have been shown to be substantial contributors in many buildings that have experienced IAQ problems, but low humidity is a serious IAQ issue as well. The potentially serious problems that can lead to high humidity and water carryover are not difficult to resolve by the disciplined designer whereas the idea of omitting humidity control altogether is no longer a reasonable response. In museums and libraries that contain valuable artifacts, the question of humidity control is generally not one that can be easily ducked. Constant changes in humidity have been shown to contribute to the deterioration of paper, wood fiber, and other waterpervious materials. Maintaining constant humidity as well as temperature is important to preserving the valuable contents of these facilities. The question of how constant the humidity must be re-

mains one that must be determined on a building-by-building basis. Designers should remember that stability and diffusion characteristics of humidity, which is based on vapor pressure, are very different from air temperature, which is based on molecular activity. Still air can be a thermal insulator but is far less effective in inhibiting the transmission of water vapor. It is much more difficult to maintain differing levels of specific humidity within a building than it is to maintain space temperature differences. When curators desire different levels of humidity in different areas, I recommend trying first to compromise on a level that would work throughout the facility and next seeing if the same specific humidity with a variation in temperature to alter the rela-

New technologies can dramatically improve performance and economy of library and museum HVAC systems, which must operate continuously under widely fluctuating loads for occupant comfort and contents protection
HPAC Heating/Piping/AirConditioning

May 1996

63

Library/museum HVAC

tive humidity would provide the desired effects.

Humidification in cold climates

Humidification of buildings in cold climates requires special attention to the building envelope. To protect the building envelope, I recommend operating the building at a static pressure that is slightly negative with respect to the outside anytime the outside air temperature is below freezing and the specific humidity inside is higher than outside. (How to control building static pressure is discussed in greater detail in the next section.) This reduces the potential for building envelope degradation from water vapor migrating into walls and freezing. It should be noted that the high rate of vapor diffusion in air makes it impossible to prevent the migration of mois60 55 50 set point, percent 45 40 35 30 25 20 Jan

specifications are in order, and a rigorous design and testing procedure to ensure they are met must be accepted. Still, the idea of operating buildings at 50 percent relative humidity in subzero weather is not at all an attractive one. The size of the humidification plant required is very large. Puddles of water condensed from the air might form regularly in entrance lobbies. This and the potential formation of ice just outside the building pose hazards. To mitigate such problems, our firm has successfully employed an annual humidity set point schedule for museums and libraries located in cold climates. Such a schedule is shown in Fig. 1. Although humidity is not absolutely constant, there is only one gradual cycle each year. For most books or artifacts, this annual hu-

commonly employed with mechanical amplification to provide humidity control. Calibration and reliability were serious concerns. Today, sensitive membranes provide accuracy and long-term stability not possible just a few years ago without incurring enormous expense. Unless the facility demands unusually rigorous humidity control and/or verification, I recommend installing good commercial quality relative humidity sensors that offer readings well within several percent RH and are very reasonably priced. Installing a number of these inexpensive sensors in various spaces and the return air duct of each air handling system allows the building staff to target those that may be out of calibration (by comparing readings during moderate weather and low building load). At the same time, this configuration ensures humidity control is being provided uniformly throughout the building. Because the relationship between relative humidity and specific humidity is temperature dependent, I also recommend locating each RH sensor within, or immediately adjacent to, a space temperature sensor so that temperature compensation can be performed accurately.

RH

Dehumidifying strategies
For many North American climates, warm weather humidity control can be accomplished by adjusting primary supply air temperature or cold deck temperature. I recommend that designers consider a VAV system with the capacity to provide supply air temperature a few degrees less than 55 F, depending on the climate. In climates that are not prone to warm humid weather, the operation of a VAV system to control humidity as well as temperature entails adjusting the supply air temperature downward below its normal set point any time building humidity reaches the high limit. As the supply temperature is reduced, the air flow required for the cooling load is
continued on page 67

Feb

Mar

Apr

May

Aug Jun Jul Day of year

Sep

Oct

Nov

Dec

Jan

1 Set point schedule for cold climate space humidity.

ture into the walls altogether just by maintaining a slight pressure difference. Therefore, strong attention must be paid to achieving an occupancy-side vapor barrier as absolutely impenetrable as possible. This requires the coordinated efforts of the architect and the contractor. Very tight infiltration
64

midity cycle is not seen as a threat to their longevity.

Controlling humidity
Technologies for the measurement and control of humidity in buildings have improved enormously in the last few years. Fifteen years ago, human hair was

HPAC Heating/Piping/AirConditioning

May 1996

Library/museum HVAC
continued from page 64

How major types of humidification systems compare

Wetted media evaporative humidifierADVANTAGES: N Self-regulating, proper sizing results in limited control requirements. N Control failure cannot result in water discharge to the ductwork. N In dry or moderate climates, building energy use can be reduced. N Extensive water treatment is not required because impurities do not readily enter air stream. N System is inexpensive and easy to control with no concern of oversaturating air stream. Wetted media evaporative humidifierDISADVANTAGES: N Requires an external source of heat to evaporate water in cold weather. N Requires more room in duct, must be accessible, and medium requires cleaning and periodic replacement. Water atomization injection humidifierADVANTAGES: N System is low in cost and easy to control N In dry or moderate climates, building energy use can be reduced. Water atomization injection humidifierDISADVANTAGES: N Control failures can cause water deposition in ductwork. N Water treatment is required to prevent deposits on artifacts and/or IAQ problems. N In cold weather, external source of heat to evaporate water is required. Steam injection humidifierADVANTAGES: N In cold weather, system provides its own heat of vaporization. Steam injection humidifierDISADVANTAGES: N Control failures can cause water deposition in ductwork. N Extensive water treatment is required to prevent deposits on artifacts and/or IAQ problems. N Water treatment and steam boiler are expensive to buy, operate, and maintain. point and that bypassed through nonoperating coil(s) provides the warmer, drier air required for such a load. If the designer takes care to locate coils and mixing plenum such that the highest humidity air stream (usually outside air) is most directed at the lead coil, and configuration ensures condensate will not be evaporated back into the bypassed air, this approach can be every bit as effective as a reheat scheme in warm weather. And it is certainly more economical. (See also the discussion under Coils in Ken Gills article IAQ and Air Handling Unit Design in the January 1996 issue of HPAC.) ests invariably conclude there is no reason to provide some inherent preference of one means of humidification over another. Each method has advantages and disadvantages when compared to others. The knowledgeable designer can quickly sort out salient issues regarding the choice of a humidification system for each project depending on the characteristics of the application, climate, and, certainly not to be forgotten, the preference of those who will be operating and maintaining the equipment. Lets remember that a humidification system is an IAQ friend, not a foe, as long as it is designed and maintained properly. The advent of vastly improved humidity sensors and advanced controls has made this fact more easily proven than ever before. The accompanying sidebar provides a brief comparison of potential advantages and disadvantages of the three major types of humidification systems for buildings. In deMay 1996

also reduced, providing a higher ratio of latent-to-sensible cooling and reducing the relative humidity in the building. For many North American climates, this approach works very well, and the control is simple and stable. In climates that experience mild but humid weather, the building can at times require almost entirely latent cooling. Supplying the low temperature air to provide the required latent cooling in very small amounts because sensible cooling is not required does not provide satisfactory circulation, and it can lead to serious consequences from localized condensation as supply air is introduced at low velocities below the dew point of the space. Before automatically considering a system with summer reheat in these applications, I suggest that designers consider a method first described to me by fellow engineer and frequent HPAC author Ken Gill. This approach employs splitting the cooling coil into two or more separately controlled sections. So long as building humidity is at or below set point, the sections operate in unison. However, as space humidity rises above set point, the coil control reverts to a staging sequence wherein the cooling valve for one stage is fully opened before the next stage valve begins opening. The resulting mix of air that has been cooled well below its dew

Humidifying alternates
Museums located in harsh climates require coordination between the architect and engineer to ensure the envelope will perform as required. This is the Royal Tyrrell Museum/Alberta Community Development in Drumheller, Alberta, Canada.

Providing humidity during the cold months of the year can be a much more substantial design challenge. Objectivity has been notably absent from recent guides and proposed standards. However, discussions by the technical experts representing various inter-

HPAC Heating/Piping/AirConditioning

67

Library/museum HVAC

veloping the humidification design, the engineer should keep the following in mind: L Provide a separate, independent control loop to disable humidification operation and alert staff if ever any failure could cause moisture to be deposited on the duct

walls from oversaturation of the air or excessive carryover. L Make certain the locations of all humidification hardware include suitable access for easy inspection and maintenance. L If a steam or water injection system of any kind is to be em-

ployed, the water must be treated to be pure. The buildup of residue on artifacts or odors from injection-type humidity systems is always a concern. The beauty of humidification systems today is how well they can be controlled with modern sensors that are accurate and reliable, and modern control systems can regulate the operation of such systems in accordance with multiple variables.

Building pressure control

I urge the HVAC designers of museums and libraries to consider implementing variable volume HVAC systems for these facilities. The HVAC systems are typically sized to meet large peak occupant loads. With systems designed to operate at improved efficiencies during part-load conditions, the initial investment for the larger capacity can be earned back through reduced operating costs. However, whatever system is finally chosen, if the air volume does vary with load, the problem of how to control building pressure looms large. In cold weather climates, a positive building pressure can force water vapor from the interior into the structure where it can freeze and cause structural problems. In humid climates, a negative building pressure can draw in high-humidity outside air that can contain three times or more the moisture of the interior air and overload the dehumidification system. Fortunately, technological advances have a solution to these problems. In the early days of VAV, fan flow tracking was commonly employed to provide a balance in outside air and exhaust air. More recently, methods that involve measuring the actual air entering the building and that exiting the building have been proposed. However, technology is now available to measure building pressure directly. This is a much preferred method from my point of view, and it has been employed with success.
continued on page 72

68

HPAC Heating/Piping/AirConditioning

May 1996

CALCULATE AVERAGE BUILDING STATIC PRESSURE NOTE: CERTAIN SENSORS MAY CARRY MORE WEIGHT THAN OTHERS

Library/museum HVAC
continued from page 68

DOEVERY 1 MINUTE STATIC-AVE = (STATIC-1 + STATIC-2 + STATIC-3 . . . STATIC-N)/N CALCULATE SUPPLY/RETURN FAN SPEED OFFSET TO BE APPLIED BY ALL FANS CALC1 = (STATIC-AVE STATIC-SP) * 100 IF CALC1 > 3 THEN CALC1 = 3 IF CALC1 < 3 THEN CALC1 = 3 SUP/RET-OFFSET = SUP/RET-OFFSET + CALC1 IF SUP/RET-OFFSET > 10 THEN SUP/RET-OFFSET = 10 IF SUP/RET-OFFSET < 10 THEN SUP/RET-OFFSET = 10 CONTROL ALL RETURN FANS ACCORDING TO SUPPLY FAN SPEED AND OFFSET RET-FAN1 = SUP-FAN1 + SUP/RET-OFFSET RET-FAN2 = SUP-FAN2 + SUP/RET-OFFSET RET-FAN3 = SUP-FAN3 + SUP/RET-OFFSET RET-FAN4 = SUP-FAN4 + SUP/RET-OFFSET RET-FAN5 = SUP-FAN5 + SUP/RET-OFFSET ... 2 Algorithm used to operate the return air fan in medium-rise buildings.

Measuring building pressure

Pressure sensors are now available that measure ranges of +0.1 to 0.1 in. WG and less with 2 percent or better accuracy over that range. Such devices are entirely suitable for measuring building static pressure directly to provide return fan control. Some of these instruments employ a self-zero feature that periodically connects the two ports together and resets itself to ensure the integrity of the zero reading. This is a most attractive feature because it is a near-zero set point that is desired in such buildings. My recommendation is to consider a building as a single entity no matter how many supply and return/exhaust fans it contains. (However, buildings with kitchens containing large exhaust hoods and makeup that are something other than constant volume may constitute an exception to this rule.) Multiple space static pres-

sure sensors are essential with at least one on every floor and at least one per 20,000 sq ft on a single floor. A common outside air reference is employed for all the sensors. Standard 3/ 8 in. pneumatic tubing can be employed for refer-

ence piping if the runs are short. For longer runs (100 ft or greater), the size should be increased correspondingly. I recommend the space reference be at ceiling level. Having the reference tube exposed apcontinued on page 103

72

HPAC Heating/Piping/AirConditioning

May 1996

Library/museum HVAC
continued from page 72

proximately 1 in. below the ceiling tile away from doors, supply diffusers, and return grilles has worked well. Be certain it is well labeled. The outside reference must connect to at least two separate points of outside air reference on at least two opposite sides of the building. The piping connecting outside references must be larger than that to the sensors. Each outlet to outside air must be protected from direct wind. Even with such precautions, I find the building static pressure readings are subject to continuous variation as wind blows, people move through the building, doors open and close, etc. However, stable building pressure can be maintained as long as sensed building pressures are not permitted to make drastic changes in return fan speed. Fig. 2 shows the algorithm our firm has employed with success to operate the return air fan in medium-rise buildings. Note that the intermediate calculation (CALC1) is limited in magnitude but can gradually increase the magnitude of the offset. Our firm typically limits the return fan speed offset to no more than 20 percent of the supply fan speed. However, the exact magnitude and range of the offset are determined empirically based on the operational performance of the two fans and actual operation of the building.

Summary and conclusion

Thanks to new technologies available to the HVAC industry, providing constant environmental conditions in libraries and museums to protect valuable books, artifacts, and art is a challenge with increasingly effective and economical solutions. When designers configure systems employing these technologies properly, a suitable environment can be maintained for first and operating costs that are only a slight premium over what is required to maintain comfort conditions in typical commercial buildings. HPAC

BUILDINGS AS HABITATS

Trends Toward More

User-Friendly
Building Environments
By THOMAS HARTMAN, PE Principal, The Hartman Co., Marysville, Wash.

ne very cold winter morning, I was touring an office building with its operating engineer when a tenant approached to ask if it would be possible to raise the temperature in her office for a short time each morning during this cold weather. The building engineer patiently explained how costly it would be, and the tenant departed disappointed but accepting of the explanation. I was reminded that years ago, when my parents remodeled our home, my mother had been convinced that her new all-electric kitchen would be so expensive to operate that she stopped much of her baking for a time. When we finally did an analysis to determine the costs, we were all surprised at how little baking actually cost. This tenants request was in truth much less costly than suggested by the engineer. I mused to myself that an hours extra warmth was certainly far less than the premium she had paid for the latte she carried with her. So why, I wondered, hasnt our industry like the coffee industrybegun catering to individual desires?

HVAC designers are beginning to realize that building occupants are demanding more from their environment

has begun to raise occupants concerns about the building environment in which they work. Building engineers are learning to take occupant complaints seriously as they are increasingly challenged with probing questions aimed at ensuring that the buildings environment will not contribute to long-term health problems.

Sophisticated occupants
HVAC designers are beginning to realize that building occupants

have become much more demanding and sophisticated. In recent years, a lot of industry discussion has revolved around the sophisticated building owner, but owners are just trying to keep up with their tenants demands. Building owners are faced with demanding environmental requirements from prospective tenants. And when a tenants employee has a complaint about comfort or air quality, the building owner often finds him- or her-self in a face-to-face discussion with a whole troupe of experts retained by the tenant to represent just such issues. It is not difficult to understand why tenants are so concerned about comfort and air quality issues. Employers are interested in creating an accommodating workplace to hold valued employees, build or improve company morale, and improve worker performance. Employers have heard of the disruptive horrors that air quality complaints cause. Meanwhile, building occupants are not waiting for their employers to act. Publicity over the last few years

Renewed focus on comfort

This renewed focus on building environments forces engineers to also face the nagging issue of occupant comfort. Studies are beginning to show a strong link between comfort and worker performance and an even stronger link between individual control of building environments and worker performance. Several studies have shown us that workplace performance does improve when comfort is improved or when occupants perceive control over their environment. Further, it is suggested by a recent survey that sick leave may be dramatically reduced when occupants are afforded control over their workplace environments. The building industry has for too long adhered to the dated concept that a building environment acceptable to at least 80 percent of the occupants is adequate. The industry must evolve from this rule of thumb because it is well known that a single-space thermal condition leaves a lot of building occuFebruary 1997

HPAC Heating/Piping/AirConditioning

63

USER-FRIENDLY BUILDINGS
pants unsatisfied. There is a misconception among many designers and building operators about human comfort. A widely held view by occupants of commercial buildings is that they demand very precise space temperature conditions that are best provided by a sophisticated centralized control system. Those who view building operation in these terms often applaud the trend away from local thermostat control as a positive development that will, in the long run, lead to fewer comfort complaints.
High Unlikely to complain Likely to complain

Low 70 71 72 73 74 75 76 77 Space temperature

1 Typical occupant comfort curve.

Individual comfort criteria

Though I am sympathetic to the plight of building operators, I also know that people have difficulty distinguishing any thermal difference at all when space temperatures are changed by up to 1 F. So the problem of satisfying more building occupants is not solved by tighter space temperature control but rather by accepting and accommodating the variations in thermal environments required by the variety of building occupants in each building. Let us consider the interaction of individual comfort requirements in populations to see why many building operators have been sidetracked by the tight control solution and discover a more effective solution to raising comfort levels in buildings. Generally, it is accepted that when holding other contributing factors constant, an individuals response to space temperature conditions is represented by a curve similar to that in Fig. 1. On the X-axis is space temperature. The Y-axis records the individuals perception of comfort at each space temperature. A boundary separates levels of perceived comfort for which the individual is unlikely to complain from the levels at which the individual is likely to complain. The range for which an individual is unlikely to complain about comfort is about 4 to 6 F. The individual whose comfort
64

curve is represented in Fig. 1 is unlikely to complain as long as the temperature range remains between 72 and 76 F. The center point of this individual comfort curve is 74 F. Generally, it is assumed that differences such as clothing, physical size, conditioning, metabolism, gender, culture, and perhaps a number of other factors do not significantly change the shape of the curve in Fig. 1 but may alter where on the curve the space temperature spectrum falls. So, for example, if the individual upon whom Fig. 1 is based were to put on a sweater or jacket, we would expect the curve to move to the left so it is centered on a lower space temperature. If the individual were to change into light summer clothing, we would expect the curve to move to the right. For a building inhabited by a variety of occupants with differences in clothing and other factors, one would expect a variety of individual comfort curves centered around different space temperatures. From these, it would be easy to calculate the number of expected complaints if one knew the center point of each individuals comfort curve. Fig. 2 shows what percentage of occupants would be likely to complain at various space temperatures if there were a normal distribution of individual temperature curves for building occupants. In Fig. 2, the line plot shows the number of people whose individual comfort

curves are centered on the various space temperatures, and the bar graph shows the percent of occupants who would be likely to complain when the building is operated at that space temperature. Surprisingly, the curve in Fig. 2 does not correlate well with the experience of many building operators 78 because it does not predict the large increase of comfort complaints originating from very small changes in building space temperature. If, however, it is assumed that the occupants of buildings generally fall into two groups, one that likes the space warmer than the current set point and one that likes the space cooler than that set point, the resulting satisfaction curve takes on the appearance of Fig. 3. Note that in Fig. 3, nearly two-thirds of the building occupants individual comfort curves are assumed to be centered on two space temperatures that, rather than being next to one another, are about 4 F apart (72 and 76 F). The temperature/complaint graph in Fig. 3 looks much more like what one would expect from the experience of building engineers. Comfort complaints are low as long as building space temperature is maintained within an extremely narrow temperature range. If the building temperature varies only slightly, the number of occupants likely to complain rises rapidly. So despite each individuals tolerance for a rather wide variation of space temperatures, a group of people can require a very narrow space temperature range.

Occupant satisfaction rating

Satisfying all occupants

Many explanations have been presented for a polarized distribution of comfort curves in buildings. I suggest that it is the uniform conditions themselves that our industry struggles to maintain in buildings that tends to po-

HPAC Heating/Piping/AirConditioning

February 1997

USER-FRIENDLY BUILDINGS
larize individual groupings. By the economy of the controls tech- front edge of the desk. The unit understanding the dynamic na- nologies upon which they are drew air from the floor and disture of comfort curves ( e.g., they based improve. In my view, indi- charged the condenser cooling air change with activity and/or vidual environmental control is in- vertically from the rear of the desk. stress), one does not need to make deed a concept whose time has For heating, it employed a small a great leap in faith to see that come. radiant panel on the modesty subjecting these dynamic individpanel. The occupant could adjust ual curves to a static environment Development of individual control cooling air flow and set the radiant can act to exaggerate their comOur firm first started working panel to any power setting. Maxifort curves away from that static with individual control concepts mum power drawn off the unit was set point condition. nearly two decades ago. At that a mere 175 watts, and tests deterWhatever the true distribution time, we designed and tested a mined that the unit could provide a of comfort curves in buildings, the small air conditioning unit to be perceived thermal sensation of 3 industry has begun to realize mounted in the modesty panel of to 5 F from the ambient space temwere nearing the end of the era of office desks with outlets on the continued on page 67 uniform conditions in buildings for many rea100% sons. First among them is that despite improvements in control technologies, oc80% cupants do not appear to complain less about com60% fort than they did decades ago! Another problem with 40% uniform building conditions is that the work many 20% in the industry have done with dynamic control has 0% shown that operating 70 71 72 73 74 75 76 77 78 buildings at constant therSpace temperature, F mal conditions is not the Percent of occupants likely to complain at this temperature most energy-efficient mode Percent of occupants comfort curves centered at this temperature of operation. And finally, polarization as represented in Fig. 3 requires a space 2 Predicted comfort complaint chart assuming a normal distribution of individual temperature that provides comfort curves within the building. only marginal satisfaction for most of those who do not 100% complain. Rather than pursue tech80% nologies that are aimed at maintaining uniform fixed 60% space thermal conditions in buildings, it is far more prudent now to seize the 40% opportunity to develop technologies and HVAC de20% signs that offer individual environmental control for building occupants. Indi0% vidual control concepts pro70 71 72 73 74 75 76 77 78 vide the opportunity of opSpace temperature, F erating buildings more Percent of occupants likely to complain at this temperature comfortably with lower enPercent of occupants comfort curves centered at this temperature ergy use. The first cost of individual control system 3 Predicted comfort complaint chart assuming a polarized distribution of individual concepts is improving as comfort curves within the building.
February 1997

HPAC Heating/Piping/AirConditioning

65

USER-FRIENDLY BUILDINGS
continued from page 65

perature. We were pleased with the result, but the office furniture manufacturer with whom we were working lost interest in the concept, and it was never integrated into their office furniture. By the mid 1980s, workstationbased comfort control systems that connected directly to the buildings air distribution system were becoming available. Today, the workstation-based individual control system still offers the best capacity for satisfying individual environmental requirements, especially in open office areas. But this type of system also has some drawbacks. The requirement that the workstation be connected to the air system may limit the space designers flexibility. Such systems may also negatively affect the esthetics of the space and make even minor rearrangements costly. Furthermore, adding HVAC components to and around workstations adds clutter to space the occupants would often like to use to store their clutter! In the last few years, a number of new approaches to individual control have been advanced by manufacturers and design engineers. Workstation-focused individual control products available today range from systems that provide a totally stand-alone HVAC system for each workstation to those that simply employ fans in the workstations and permit the occupants to direct more or less ambient air over and around them. Other approaches, such as underfloor air distribution systems that afford occupants the ability to adjust the flow of air around their workstation, have also been successfully employed.

4 Personal diffuser installed in an office. This style provides air diffusion by injecting a stream of air into the room. Inset photo is a closeup of the diffuser. Photos courtesy of Zero Complaints.

Personal diffuser technologies

About the same time the workstation-based individual control concepts were being introduced, work began on other paths toward meeting the growing demand for more individual control within

buildings. Most air-based HVAC systems employ a single terminal unit for every two to four offices. A single controlling thermostat is placed in just one of the offices, and the rest operate with essentially no control at all. This loose zone control is based on the premise that offices sharing a common building exposure experience nearly identical heat load characteristics. However, improvements in building envelope technologies over the last several decades together with variations of internal heat loads now make the heat balance equations of offices widely variable even when they share the same perimeter exposure. To deal with heat load variations in different spaces served by a single terminal unit, engineers began developing designs that could change the balance of air flow from the terminal unit among the individual offices or areas it served based on the thermal conditions of each area. For example, if one office was at its thermal set point, a damper at the diffuser supplying air to that room (or area) would close, directing most of the terminal units supply air to the other offices or areas served by that terminal unit. This concept has evolved to what is today called personal

diffuser technologies. Originally, personal diffusers were connected to standard VAV boxes and employed to redistribute air among the areas served by the box. However, the technology has evolved to one where the VAV box now regulates the downstream static pressure and switches from heating to cooling, depending on commands from the personal diffusers it serves. Typically, when configured with low pressure rooftop air supply systems, personal diffusers require no intermediate control boxes or devices at all. As personal diffuser technologies have matured, a number of approaches have been taken to improve control capabilities. One important improvement is the use of wireless infrared remote controls that allow occupants to make temperature adjustments with ease. Fig. 4 shows a personal diffuser in an office application. Here, two personal diffusers are employed in a master/subordinate configuration. An in-depth explanation of this system concept appears in the March 1995 issue of HPAC. A more typical personal diffuser, which usually fits in a 2 by 2 ft ceiling
February 1997

HPAC Heating/Piping/AirConditioning

67

USER-FRIENDLY BUILDINGS
grid, is shown in Fig. 5. the cost premium for individual attractiveness of individual enviMost personal diffuser designs control will keep it out of main- ronmental control. Now, the emerdo not offer complete individual stream designs for the foreseeable gence of deregulated electric utilicontrol because personal dif- future. Those involved in the indi- ties, some of which are anxious to fusers are typically connected to vidual control market are experi- invest in high-quality building an air supply system serving sev- encing an expanding demand for comfort systems, may radically eral offices or spaces and cannot such systems, but the market shift the way engineers view simultaneously provide heating share for these products is still HVAC system design. to one diffuser and cooling to an- quite small. Few anywhere in the other. But personal diffuser tech- industry will go on record with Buildings in the next century nology does go a long way toward predictions for the role of individThe term human habitat is used improving building environments ual control. to describe the places where we from the days of one thermostat Since our firms experiences spend our time because they are for every three or four offices. with an individually controllable the places where we feel most comEach personal diffuser employs desk-based comfort system, I have fortable. Office buildings have hisan integrated space temperature been convinced that a confluence torically not been seen as consensor to provide local tempera- of trends would some day make in- formable environments. Instead, ture control. Furthermore, the dividual control-based HVAC/ people flock to them simply bepersonal diffuser technology can lighting the obvious choice for cause their jobs require it. Now, be easily added to any type of commercial buildings. Emerging modern technology offers us standard overhead air dischoices. We can now be nearly as connected tribution system common from our homes as at our in office buildings today. offices. This trend to forAnd it is an economical sake the office environtechnology, often adding ment and, despite the only modestly, if at all, to problems of isolation and the overall cost of a tradidisruptions, have employtional HVAC system. In ees work primarily at some ways, the personal home is growing. If office diffuser is a true crossbuildings are to retain roads product. Although their economic viability its name suggests individand flourish into the next ual control, it really cannot century, we engineers provide distinct thermal must sharpen our focus environments, especially on comfort. To do so, we in open office environmust reach beyond a dements. What the technolsign methodology that ogy can do is provide a more precise, uniform sees comfort as a simple space temperature matter of maintaining throughout the building. 5 Technician installing a lay-in style of personal dif- heat balances. Instead, The question is, will this fuser. This unit mixes supply air with room air by main- we must implement detechnology be enough to taining a fixed outlet velocity at reduced supply air signs that make buildings more environmentally atsatisfy building occupants, flows. Photo courtesy of Warren Technology. tractive. Individual comor will they be captivated by the possibilities it suggests digital control and network tech- fort control is an obvious vehicle for nologies have continuously re- turning unfriendly offices into deand want more? HPAC duced controls costs for individual sirable worker habitats. Industry developments control concepts while at the same There is by no means a consen- time, building occupants have Background information on technolosus on the future of individual en- been increasing their demands for gies discussed in this article are availvironmental control in buildings. better comfort and control over able at www.hartmanco.com/engr. Any questions or comments about this As indicated earlier in this article, their work environments. Also, article may be addressed to Mr. Hartsome still believe it is more useful building owners and managers man at tomh@hartmanco.com or to apply advanced control tech- have been looking for amenities to Compuserve 104067,3463. Thomas nologies to maintain rigidly uni- attract potential tenants who are Hartman is a member of HPAC s form thermal conditions in build- increasingly working at their Board of Consulting and Contributing ings. Others are concerned that homes. These trends increase the Editors.
68

HPAC Heating/Piping/AirConditioning

February 1997

ADVANCED CONTROLS CONCEPT

Packaging DDC Networks with

Variable Speed Drives

By designating complex network-based control sequences as packaged products with responsibility for their performance vested to manufacturers, it may be possible to simplify designs and increase the utilization of VFDs and other advanced control technologies in the field
By THOMAS HARTMAN, PE, Principal, The Hartman Co., Marysville, Wash.
Network to other controllers and operator workstation C Chilled water supply Chilled water return VFD DP C T C DDC controller T Temperature sensor VFD Variable speed drive DP Differential pressure sensor DDC point connection DDC communications network T C T C T C

1 Typical variable flow chilled water distribution system.

wo of the greatest recent technological breakthroughs in our industry have been variable frequency drives (VFDs) for AC motors and direct digital control (DDC) networks. But if there were a law against under utilizing either of these technologies, I fear all HVAC engineers would be in jail today; VFDs and DDC networks are the least effectively applied technological tools in our industry. Difficult as it may seem, many designers are still questioning whether these two technologies are even useful in projects where their proper application could reduce energy
80

use and improve overall system performance by an order of magnitude or more. Variable speed drives and network-based DDC controls can be applied to nearly all of the major HVAC componentsfrom fans to pumps to chillers. In fact, an increasing number of these are being installed with VFD speed regulation and DDC control. But, make no mistake about it, even though variable speed drives and DDC controls are becoming widely employed, the technology is overwhelmingly underutilized. To show why and offer suggestions for improving the situation, this article uses the example of a chilled water distribution pumping system in which a variable flow chilled water pump serves multiple loads with modulating, two-way control valves (Fig. 1).

Here, a variable speed pump is operated to maintain a differential pressure setpoint for a distribution system that serves a number of loads. Each load has a modulating valve operated by a DDC controller. Fig. 1 also shows the DDC network that connects all controllers together such that the value and status of all connected points are accessible from an operator workstation. Note that the network is used to transmit the differential pressure value to the controller operating the pump. So, we have a variable speed pump and networked controls operating the system. Whats wrong with this picture?

Pump and fan laws

Fan and pump laws dictate that a centrifugal pump can sup-

HPAC Heating/Piping/AirConditioning

November 1998

140

ply 50 percent of design flow at 50 percent speed and require only 12 12 percent (0.5 cu) of the full flow power, but only if the pressure at which the fluid is supplied is permitted to fall to 25 percent (0.5 sq) of the full flow pressure. If the supply pressure does not fall with the square of the flow requirements, then the pump speed cannot be reduced as the law permits. This results in a reduction in pump efficiency. These relationships are shown in Fig. 2. A zone of highest pumping efficiency throughout the speed range of the pump is highlighted in blue. If the system curve approximates that of a simple circulating system (e.g., the head pressure requirements fall with the square of the flow), the system curve is approximated by Curve A. Such a system curve ensures that the pump operates at the highest pumping efficiency through all flows and that the power required falls with the cube of the flow requirements. The system curve for a typical variable flow distribution system, as represented in Fig. 1, is shown by Curve B. In Curve B, the pumping efficiency quickly falls as flow requirements are reduced.

Design maximum pump head capacity, percent

120

1750 rpm pump speed Curve "B" for systems operating with fixed differential pressure control 1450 rpm pump speed High efficiency area

100

80

60 1150 rpm pump speed 40 Curve "A" for systems operating according to a "natural" system curve

900 rpm pump speed 600 rpm pump speed

20

10

20

30

40

50

60

70

80

90

100 110 120 130 140

Design maximun flow, percent

2 Pump/system curves for pressure and network pump controls.

Regarding chillers, manufacturers try to remedy this loss by offering screw compressors or other positive displacement-type compressors that do not lose efficiency under such conditions. Multiple units are often employed to reduce part-load inefficiencies. But these approaches do not attack the second penalty, which is the extra power requirement due to the high head requirements (that are unnecessary) at low flows.

Centrifugal fans, pumps, and compressors are simple devices that are very efficient at their sweet spots (the high-efficiency zone in Fig. 2). It is reasonable to find ways to accommodate limitations of these devices with system concepts permitting the pressure to decrease as flow decreases to operate at the highest possible efficiencies throughout all flow conditions. The value of such an approach
continued on page 83

Part load operation

There are actually two costly energy penalties for systems that follow Curve B. Loss of pump efficiency is one of them. The second is the high pump-head pressure required at decreased loads. Because the majority of variable flow pumping systems are operated to maintain a fixed differential pressure between the supply and return distribution lines, the pumping pressure is not permitted to fall significantly as the load decreases, and a substantial energy penalty is the result. This combination of reduced pump efficiency and a higher head pressure limits opportunities for reducing pump power at part-load conditions.
25 Cooling load profile tropical climates Cooling load profile temperate climates Operating hours at load, percent 20

15

10

15

25 35 45 55 65 75 Design maximum cooling load, percent

85

95

3 Cooling load profile for comfort conditioning in two major climates.

November 1998

HPAC Heating/Piping/AirConditioning

81

ADVANCED CONTROLS CONCEPT

continued from page 81

is illustrated in Fig. 3, which shows the expected load profiles for chilled water plants employed for comfort conditioning. Fig. 3 also shows the percent of total operating hours the system spends at various load capacities for two basic climates. Assuming the chilled water flow is proportional to the load, Fig. 3 approximates the flow requirements for a distribution system to the loads served. Note that the overwhelming majority of pumping hours is spent at reduced flows in both climate types. Systems operating according to Curve A in Fig. 2 can achieve enormous energy savings over Curve B systems at these flow conditions.

ever, the pump operates to maintain a constant pressure across the valve and coil, so the pressure drop across the valve increases. This increase in pressure across the valve at low flows reduces controllability and is one reason why control instability is more likely to occur at low load conditions.

New rules are needed

The old valve-sizing rule of thumb served an important purpose when linear pneumatic controllers were prevalent, but it has always had significant limitations, and it is not an energyefficient approach in the era of variable speed pumping. Knowledgeable designers recognize immediately the two questions that have to be answered to move beyond this reliable, but less efficient, approach. The first question is What is the most effective design direction to take? and the second is How can I be sure it will be implemented smoothly without wreaking havoc on my projects? The answer to the first question is to employ the DDC network rather than differential pressure to operate the pump speed. Such an approach would permit the pump to slow, so long as all the loads are satisfied, without regard to a differential pressure. Such control is now possible because DDC technologies are maturing and developing more sophisticated software controllers. Control manufacturers have had sufficient experience with software loop control to add features like variable gains, anti-

Change is needed
Rules for sizing valves and certain pump control features predate the availability of variable speed drives. To achieve good control, it is still recommended that control valves be sized so that 25 to 50 percent of the full flow system pressure drop occurs across the valve. Rules of thumb in hydronic systems vary, but it is common to see systems configured so that at full flow the control valve pressure drop is one-third of the total system head; the pressure drop through each load is one-third of the total head; and the remaining one-third consists of piping and all other pressure drops. This is the criterion employed for Curve B in Fig. 2 wherein the differential pressure setpoint is about two-thirds of the total full flow system head. While these rules did provide successful control for many applications back in the days of simple pneumatic controls, the solution is energy intensive and also has negative side effects. For example, consider the Fig. 1 system operating at low loads. Because the flow through the coil is low, the pressure drop across the coil approaches zero. How-

MMM

While the industry employs both variable speed drives and networked DDC technologies with increasing regularity, the full energy savings and performance capacities of these remarkable developments are rarely achieved.

windup, and self-tuning capabilities. Perhaps the most important feature that is evolving with DDC systems is their ability to network points among various controllers. It may be time to consider putting these features to work. The two primary reasons the rule of thumb concerning valve selection seeks a relatively high pressure drop across control valves are 1) to ensure linear control response, and 2) to ensure that the operation of each valve is undisturbed by the action of adjacent valves. If the full flow pressure drop across the valve is low, then linear response is lost. One valve suddenly opening or closing as its load changes state could change conditions at an adjacent load. It is not difficult to imagine that instabilities could occur as two or more adjacent valves each correct themselves in response to changes of the other. This type of instability would have been enormously troublesome in the days of pneumatic controllers, but networked DDC systems can accommodate such changing conditions smoothly and easily. Imagine that the communications network in Fig. 1 is used to control the variable speed pump and valves instead of the differential pressure sensor. Each valve could be modulated not only according to local load changes but also in response to system changes as communicated over the network. The pump could be operated to meet flow requirements as efficiently as possible. A system schematic that provides such control is shown in Fig. 1

November 1998

HPAC Heating/Piping/AirConditioning

82

ADVANCED CONTROLS CONCEPT

just remove the differential pressure sensor. Properly pro grammed, such a configuration can permit the chilled water distribution pump to operate in the high-efficiency zone of Fig. 2 at nearly all conditions. The control valves for each load can be sized for a significantly lower pressure drop, thus reducing the total system head at all loads and reducing the size of the pump. In simulations and field tests, such chilled water distribution system configurations have been shown to cut the required annual pumping power by one-half to two-thirds below that of a traditional system. This is exciting. These designs provide savings in both first costs and annual energy costs, resulting in energy savings in the range of $1 to more than $3 annually per gpm of chilled water pump capacity. design approach they know can be implemented without troublesome startup problems. stantial. We in the industry should consider working together to package specific networked technologies so they can be employed efficiently and effectively as applications arise. To get the benefit of all voices in our industry, I urge interested readers to visit our Website frequently over the next year to get a better understanding of and provide comments to a developing vision as to how networking technologies can be packaged most effectively. I also urge readers to write to HPAC to voice your opinions on this important topic.

Productization
To try to develop a solution to this dilemma, imagine that the pump applications technology outlined in this article was a product that designers could specify, such as air handlers or VAV boxes. In these products, it is the product supplier and not the designer who assumes ultimate responsibility for the performance of the product. That performance is based on sizing and other general information provided by the designer in the construction documents. Software products are widely used today but not as elements in design documents for the building construction industry. However, by designating certain more complex network-based sequences as products whose basic features are generally understood by a wide segment of our industry, it may be possible to develop an implementation path that is similar to that of other products. This process of productization is a promising model for successfully implementing new network-based application controls. Manufacturers of related HVAC components and controls manufacturers are both logical choices for delivering networkbased software applications. My company is working with interested firms in these two categories to try to define and modularize reasonable networking technology packages such as the example of low power pumping outlined in this article. Other network application technologies that can be modularized in similar fashion are chiller plant control, VAV system control, and comfort controls operated by individual tenants. The improvement in performance and reduction in energy available by implementing such networked controls are sub-

Summary and conclusion

While the industry employs both variable speed drives and networked DDC technologies with increasing regularity, the full energy savings and performance capacities of these remarkable developments are rarely achieved. The present design and implementation process that places complete responsibility for the implementation of such technologies on the designer is effectively blocking such advancements in our industry. A promising prospect that could change this situation is the development of DDC network products. To undertake this challenge, the industry needs to hear all interested voices to determine if and how such packaged products can be developed and implemented in systems to achieve higher performance. HPAC Mr. Hartman is a member of HPACs Board of Consulting and Contributing Editors. Additional information on technologies discussed in this article is available at www.hartmanco.com. Any questions or comments about the article may be addressed to Mr. Hartman at tomh@hartmanco.com. See pg. 7 for contact information.
Circle 504 on reader service card if this article was useful; circle 505 if it was not.

A change worth pursuing

Such an advance in pumping control technology is an exciting prospect, but the second question still begs an answer. How can the designer be sure the new ap proach will be implemented effectively and without hassles? It is not too difficult to envision the basic changes in controls and operation required to capture these substantial pumping energy savings; however, designers cringe at the prospect of trying to describe networked solutions in a sequence of operations. Most designers will dismiss entirely the chances that a controls contractor will provide a trouble-free implementation of a network-based sequence no matter how well it is described. The unhappy truth seems to be that the process by which control sequences are implemented is an enormous part of the problem in raising the level of DDC technology applied in the HVAC industry today. Most designers realize that the valve sizing rules result in a premium for pump-power costs, but they are reluctant to change a
84

HPAC Heating/Piping/AirConditioning

November 1998

THE HARTMAN LOOP

ALL-VARIABLE SPEED CHILLER PLANT DESIGN AND OPERATING TECHNOLOGIES

FREQUENTLY ASKED QUESTIONS

Last Updated September 7, 2001

The Hartman Company

www.hartmanco.com 755 County Road 247 Georgetown, Texas 78628 Phone: 254-793-0120 FAX : 254-793-0121 E-mail: thartman@hartmanco.com

The Hartman Company

September 7, 2001

HARTMAN LOOP CHILLER PLANTS

FREQUENTLY ASKED QUESTIONS

Table of Contents
INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 GENERAL QUESTIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 LOOP CHILLER PLANT ENERGY USE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 LOOP PLANT DESIGN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 CONSTRUCTION COST ISSUES FOR LOOP PLANTS . . . . . . . . . . . . . . . 6 LOOP RETROFITS OF EXISTING PLANTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 LOOP PLANT OPERATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 LOOP EQUIPMENT REQUIREMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 SUPPORT FOR LOOP DESIGN & OPERATION . . . . . . . . . . . . . . . . . . . . . . 12

The Hartman Company

September 7, 2001

HARTMAN LOOP CHILLER PLANTS

FREQUENTLY ASKED QUESTIONS

INTRODUCTION
Hartman LOOP all-variable speed chiller plant technologies offer revolutionary reductions in chiller plant energy use with only modest alterations in plant equipment configurations. This document is intended as a supplement to other technical information with answers to practical questions that arise when owners, operators and designers consider employing LOOP technologies in specific applications. If additional technical information is desired, please visit The Hartman Company website at http://www.hartmanco.com or contact Hartman by e-mail for the information you desire. Hartman LOOP chiller plant design and operating technologies are straightforward to apply, but certain aspects of these technologies such as chiller loading and sequencing control are quite different from traditional chiller plant operating concepts. Owners, operators and designers contemplating the use of these exciting cost reducing technologies often have questions about whether their application is well suited to LOOP technologies. As a part of developing LOOP technologies, Hartman has initiated testing of LOOP features, and has discussed aspects of LOOP operations with a wide range of industry experts that include owners, designers, and equipment manufacturers. LOOP technologies have been evaluated for many climate regions and facility types. Hartman has vigorously pursued all issues that have been raised in these tests and discussions such that an owner can be confident a LOOP plant implemented anywhere in the world will achieve the level of performance projected by the performance calculations without unanticipated costs or additional work. It is our firm's intent to keep LOOP technologies completely open to all in the industry. This FAQ document is intended to provide information to answer questions being asked when LOOP technology is first considered for a project. Detailed design and operating information is also available. To recover our considerable development and support costs, LOOP technologies are patented and a low cost site license is required for each application. We have tried to make the licensing process as simple and easy to use as possible. We are always eager to hear our users' comments about improving the information and licensing procedures. If you have any ideas, I would like to hear from you! I hope this information helps to answer your questions about the LOOP chiller plant technologies, and I trust you will not hesitate to contact us if you have additional questions or comments. Let us know if you have further questions or need more information so you can better evaluate the cost and energy impact of LOOP technologies on your chiller plant. Tom Hartman

The Hartman Company

September 7, 2001

Page 1

HARTMAN LOOP CHILLER PLANTS

FREQUENTLY ASKED QUESTIONS

GENERAL QUESTIONS
1. Why is this called LOOP technology?

ANSWER: During the development process Hartman named these LOOP technologies because they employ integrated closed LOOP controls for the entire plant whereby the operation of all chillers, pumps and towers is coordinated in order to optimize total plant efficiency under all conditions. Most equipment in conventional plants operates in stand alone fashion, responding only to certain temperature setpoints rather than operating in coordination with related equipment. LOOP is also used to describe the chilled water distribution system in this technology which involves a fully determinant, single circuit chilled water loop instead of the common but less efficient indeterminate primary/secondary systems.

2.

What is new about LOOP technologies?

ANSWER: The most significant improvement of LOOP all-variable speed chiller plant and distribution systems over conventional plants is the significantly reduced energy use in comfort conditioning applications. Under LOOP operation, the entire chiller plant annual energy use usually averages about 0.6 kW/ton or less for most comfort cooling applications. This represents an annual energy reduction of 25% to 50% (depending on climate and application) below the most highly optimized conventional configurations of components of the same operating efficiencies. In LOOP chiller plants, all components are variable speed and chiller sequencing endeavors to keep chillers and towers operating at lower loads and flows rather than shedding them to keep the on-line equipment at high loading as in conventional plant operating strategies. To achieve these higher levels of performance, an entirely new approach to operating the equipment in chiller plants has been developed. Aside from the use of variable speed drives for all pumps, chillers and tower fans, LOOP plant configurations are very similar to conventional chiller plants. It is how the equipment is sequenced and operated with simple, straightforward and stable network based controls that is really new.

3.

In what applications are LOOP technologies most effective?

ANSWER: LOOP technologies have been developed specifically for chiller plants that serve comfort conditioning loads. Industrial process loads may be suitable for LOOP technologies if the process loads are variable because LOOP technologies reduce energy use only at part load conditions. Warm, dry climates usually offer the best savings opportunities, but LOOP technologies offer huge savings in comfort conditioning applications all over the world. The energy savings calculator at The Hartman Company website can estimate energy savings for a chiller plant that is employed for comfort conditioning in any of more than 200 different climates

The Hartman Company

September 7, 2001

Page 2

HARTMAN LOOP CHILLER PLANTS

FREQUENTLY ASKED QUESTIONS

worldwide. To estimate the potential reduction for an industrial process cooling plant, contact The Hartman Company with the plants estimated load profile information.

The Hartman Company

September 7, 2001

Page 3

HARTMAN LOOP CHILLER PLANTS

FREQUENTLY ASKED QUESTIONS

LOOP CHILLER PLANT ENERGY USE

1. How are the savings estimates calculated that are employed in the web site Energy Savings Calculator and how accurate are they?

ANSWER: A combination of hourly simulation and spreadsheet calculations are employed for this calculator and designers can see the comparison results for any application they wish to consider using the Savings Calculator at the Hartman website. This on-line program has been thoroughly tested and provides a good first cut estimate; savings are likely within 10% to 30% of the true value. Care has been taken to ensure the savings estimates in the savings calculator are not inflated. More accurate estimates can be easily developed when additional information about the load being served is available. The most useful additional information is the actual cooling load profile (e.g. the percent of time the plant spends operating at various loads) and/or total chiller plant operating hours for a typical year. At present, this additional information cannot be input directly into the on-line calculator, but if you have this information, you can e-mail it to us using the contact us page, and we will quickly send you back a corrected savings calculation that is adjusted by this additional information.

2.

What efficiency equipment is used to calculate energy savings and what would be the effect on savings if chillers of different efficiencies or towers of different approaches were employed?

ANSWER: Savings comparisons between LOOP plants and optimized conventional plants are calculated by simulating chillers with nominal full load ARI rated efficiency of 0.62 kW/ton and towers with 8 to 10oF approach. However, the percentage savings between LOOP and conventional plants is independent of chiller efficiency or tower approach temperatures. Thus, as efficiencies of chillers and related equipment are improved, LOOP configuration efficiencies will also improve offering approximately the same percentage of energy savings over the conventionally configured and controlled chiller plant.

3.

Is water side economizing used in the LOOP system to achieve reduced energy use?

ANSWER: No, the savings estimates do not consider water side economizing but a direct tower cooling cycle could be incorporated into a LOOP chiller plant just as it can into conventional plants. However, the energy saving comparisons for LOOP plants are not based on varying equipment configurations or equipment efficiencies, but on straight-across comparisons employing identical equipment efficiencies, approach temperatures, and weather and load data. The only change made to compare LOOP performance with conventional plants is that the equipment in the LOOP plant is operated by variable speed drives and employs network based LOOP control technologies for sequencing and equipment speed control.

The Hartman Company

September 7, 2001

Page 4

HARTMAN LOOP CHILLER PLANTS

FREQUENTLY ASKED QUESTIONS

LOOP PLANT DESIGN

1. How does a LOOP chiller plant layout differ from a conventional chiller plant?

ANSWER: On the chilled water generating side, there is usually little or no difference in layout between a conventional chiller plant and LOOP chiller plant. Designers may decide to employ common headers or dedicated chilled and condenser water pumps with either system, although LOOP control considerations may influence which approach to choose. Though LOOP plant layouts are the same or very similar to conventional plants, for LOOP configurations employing multiple chillers, it is recommended that all chillers be the same in size and have the same operating characteristics, and it is helpful, though not necessary, to have the same number of towers or tower cells as chillers.

2.

How does LOOP chilled water distribution differ from standard Primary/Secondary systems?

ANSWER: LOOP chilled water distribution technologies employ a single circuit chilled water distribution system based on Low Power Pumping technologies that are a subset of the LOOP chiller plant technologies. In these recommended distribution systems, there are no decoupler lines. For single building and small distribution systems Hartman recommends that the same set of pumps that pump the chillers also pump the distribution system and the loads. In large systems, primary pumps pump the chillers and distribution system (maintaining a neutral pressure in the distribution system), and booster pumps (that are connected in series with the primary pumps without decoupling lines) pump each load, major aggregate of loads, or building. Note that instead of a primary/secondary pumping system, LOOP plants employ primary only or primary/booster arrangements with the booster pumps in series with the primary pumps. A bypass valve may be installed at the end of each main to ensure a minimum flow is maintained at all times. Operation of this simple configuration is optimized with network based control sequences. Enormous efficiency improvements come from network optimization of the pumping, load flow control, and the elimination of direct mixing of supply and return chilled water.

3.

How can LOOP technologies be applied; must the chiller plant and distribution system both employ LOOP technologies for proper operation?

ANSWER: No, LOOP chiller plant technologies are modular. The two major parts are the LOOP chilled water generation technology and the LOOP chilled water distribution technology. LOOP technologies can be applied to each independent of the other. It is possible to further modularize LOOP technologies such that only the heat rejection circuits employ LOOP technologies. This is being done cost-effectively for existing chilled water plants that employ constant speed chillers and cannot justify the expense of changing them at this time. However, energy reduction opportunities are substantially increased by implementing a complete LOOP network based system to the entire chilled water plant and distribution systems.

The Hartman Company

September 7, 2001

Page 5

HARTMAN LOOP CHILLER PLANTS

FREQUENTLY ASKED QUESTIONS

CONSTRUCTION COST ISSUES FOR LOOP PLANTS

1. Are construction costs higher for LOOP plants?

ANSWER: No, not necessarily. It is sometimes assumed that a cost premium equal to the cost of the variable speed drives less the cost of the across-the-line starters for the chillers, pumps and fans will be necessary. However, in many applications constructing a LOOP all-variable speed chiller plant in place of a constant speed plant of the same size and nominal efficiency costs about the same. Below about 80% loading, a LOOP all-variable speed chiller plant configuration incorporating variable speed chillers that are somewhat less efficient at full load, but cost the same as more efficient constant speed chiller of the same capacity will begin to operate more efficiently than a conventional plant with the more efficient constant speed chillers. This means that when chiller plants are sized with a 20% or greater margin of excess capacity, the operating efficiency of a LOOP all-variable speed chiller plant incorporating equipment of about the same cost will operate more efficiently even at peak load conditions than a conventional constant speed plant which loses efficiency when the equipment is oversized. So, anytime a chiller plant is oversized for failure or standby protection the nominal efficiency of the chiller plant should be based on the actual peak load served by the plant rather than the total capacity of the plant. Doing so reduces the nominal full load efficiency requirements of variable speed plant components and therefore lowers their cost. This cost reduction, along with further reductions from effective network control connections offsets the extra cost for the variable speed drives and allows all-variable speed chiller plants to provide substantial annual energy savings while costing about the same to implement as an optimized constant speed alternative of the same capacity.

2.

Are the network controls required for LOOP plants more expensive or complicated to operate than standard controls?

ANSWER: No. these controls are usually the same DDC controls that are employed in conventional plants. Most modern DDC systems have the capacity for network control, but it is seldom employed. See the section on LOOP equipment requirements for more information on LOOP control system requirements.

The Hartman Company

September 7, 2001

Page 6

HARTMAN LOOP CHILLER PLANTS

FREQUENTLY ASKED QUESTIONS

LOOP RETROFITS OF EXISTING PLANTS

1. Is it cost-effective to retrofit an existing constant speed chiller plant to LOOP technologies, and if so, how is that accomplished?

ANSWER: Many existing chilled water plants can be cost-effectively retrofitted to LOOP plants. Chiller plants that are located in warm climates and do not employ effective tower optimization strategies are the very best candidates as they will provide the greatest annual energy savings. Because LOOP chiller plant technologies are modular, it is possible to upgrade a plant in stages, or to limit the upgrade to only certain elements of LOOP technologies. For example, applying LOOP technologies to the heat rejection circuits (condenser pumps and tower fans) is almost always a cost effective measure and therefore a good first step in an upgrade program. A heat rejection LOOP upgrade can be done without any configuration changes to the existing constant speed chillers. For chiller plants that serve large distribution systems, it is usually cost effective to convert the distribution system from primary/secondary to primary/booster and upgrade to LOOP low power pumping technologies. Whether or not it is beneficial to convert the existing constant speed chillers to variable speed chillers and apply LOOP operation technologies depends on the age, efficiency and configuration of the current chiller plant. The Hartman Company will gladly provide preliminary guidance on the costs and energy reductions associated with various upgrade options. Simply e-mail or mail the basic plant equipment schedule, configuration schematic, and load profile or description of loads served. You will receive a good first cut estimate of what the various upgrade options will cost and save.

2.

Can any existing constant speed centrifugal chiller be upgraded to variable speed for LOOP operation?

ANSWER: Like all existing motors, those that drive centrifugal chillers can be retrofitted with variable speed drives. Like other motors, if the compressor motor is in good condition, many years of excellent operation can be expected. However, there are some limits that may make variable speed conversion uneconomical. Reasonably priced variable speed drives are generally not available for high voltage motors (over 600 volts), and for motor sizes on chillers rated at over 2000 tons. Furthermore, some constant speed chillers cannot be significantly slowed due to minimum refrigerant pressure requirements or certain mechanical limitation. Finally, chillers more than ten years old may employ refrigerants that are being phased out, and may operate at much lower efficiencies than modern chillers. Chiller plants with older, less efficient chillers may not be good candidates for variable speed upgrades. Because of low efficiency and lack of useful service life, it may be better to replace older chillers outright with new variable speed chillers, or focus immediate upgrade efforts on the heat rejection and distribution circuits. New variable speed chillers can then be retrofitted when the existing chillers reach the end of their useful life, or when scheduled as part of a refrigerant phase-out program.

The Hartman Company

September 7, 2001

Page 7

HARTMAN LOOP CHILLER PLANTS

FREQUENTLY ASKED QUESTIONS

LOOP PLANT OPERATION

1. How are the chillers staged in a LOOP chiller plant?

ANSWER: Hartman has developed an entirely new method of chiller sequencing called the "Natural Curve" method of sequencing. The "Natural Curve" is a term coined to describe the most efficient operating load point of a chiller at various head (condenser and evaporator temperature) conditions. Typically, the Natural Curve for a constant speed chiller is at or very near full load at all head conditions, but for a variable speed chiller the most efficient operating point is at much lower loads and varies with the head conditions. Thus, a curve can be developed for variable speed chillers that plots their most efficient operating point at various head conditions, and this curve is called the "Natural Curve" of the variable speed chiller. In this easily applied Natural Curve sequencing method, chillers (and towers) are staged in LOOP chiller plants such that chillers operate at all times closest to their Natural Curve. Typically, LOOP plant chiller shedding occurs at much lower loads than in conventionally operated plants. As the load falls from full load, all equipment is operated at reduced speed until the Natural Curve algorithm calculates that the plant can operate more closely to the remaining chillers' Natural Curves if a chiller is shed. The same is true when a chiller is staged on. The exact points of this staged operation depend on the characteristics of the variable speed chillers and towers employed, as well as the current characteristics of the load served and the chiller head conditions.

2.

Is control of chiller capacity included with speed control in LOOP operation?

ANSWER: Yes. Currently, a DDC controller is configured and programmed to operate chillers in a LOOP plant just as DDC controllers typically operate chillers in a conventional plant. Thus, the operation of variable speed chillers in LOOP chiller plants is very similar to the operation of conventional plants. A LOOP DDC controller controls both chiller sequencing (on/off control) as well as the amount of capacity (demand limit) of each chiller. All factory built variable speed chillers include internal logic that is intended to continuously optimize vane and speed control to meet current conditions and variable speed chillers can also be configured to accept demand commands from the LOOP plant controller. Thus it is not difficult to establish LOOP plant operations with any variable speed chillers. In plants that are retrofit from constant speed chillers, the speed/vane control logic may be externally applied in some circumstances.

3.

How are the condenser water pumps and tower fans controlled in a LOOP chiller plant?

ANSWER: In LOOP chiller plants, the condenser water pumps and tower fans are variable speed as are all other components. LOOP control of these components is accomplished with very simple algorithms that tie the pump and fan operation directly to the power input to the chiller(s). Some special care must be taken in the choice of the towers for LOOP operation such that each tower is able to handle a range of flows and still achieve full coverage of its fill and provide efficient

The Hartman Company

September 7, 2001

Page 8

HARTMAN LOOP CHILLER PLANTS

FREQUENTLY ASKED QUESTIONS

air/water surface exposure. It is also important that the fixed head requirement of the tower be considered when selecting the condenser pump(s). There are, however, many tower and condenser pump configurations that easily provide the variations in flow required for efficient LOOP operation. While condenser pump speed and tower fan speed are both adjusted in accordance with chiller loading, the control also employs limits on this preset relationship to ensure that maximum efficiency for the plant as a whole is achieved, and that certain temperature and flow operating limits are not exceeded.

4.

How realistic are water flow, temperature values and power requirements that are employed in sample LOOP energy savings models?

ANSWER: The values for outdoor dry and wet bulb temperatures come for actual weather data for the specific location chosen. Operating flows and temperatures are all within manufacturers limits for the chiller or tower to which they are applied. Pump flows and power requirements are calculated from standard manufacturers pump performance curves. Chiller and tower operating temperatures and power requirements at each load condition come directly from the ARI and CTI performance data for the equipment involved. This data has been determined from testing at those specific flows, loads, and temperatures. Therefore the values and power requirements are considered to be very realistic.

5.

If chillers run longer at lower loads in LOOP plants, does that mean chiller maintenance costs will rise?

ANSWER: No, but this is a very important point. This question was first raised when the LOOP technology was being developed. We have discussed maintenance issues with chiller manufacturers and others who agree that chiller maintenance based on component wear will likely be reduced from the application of LOOP operation. The reason for reduced maintenance despite the longer operating hours of each chiller are 1) fewer starts, 2) softer starts, and 3) lower average loading on each machine. While there is not yet sufficient data to show conclusively that maintenance costs are reduced in a LOOP plant, there is strong agreement among those expert in variable speed operations, including chiller manufacturers, that maintenance for wear and tear certainly does not increase in LOOP plants. Currently, much periodic chiller maintenance is triggered by runtime hours. It is generally agreed that for a LOOP plant these should be adjusted or replaced with new PM guides that recognize the reduced wear per operating hour that LOOP plants achieve. Also, the newer studies that show mechanical failures are not generally reduced by periodic maintenance based only on run time, but only when maintenance is triggered by vibration, power or other operating anomalies.

6.

Are there any new operating or maintenance issues that should be considered with a LOOP chiller plant?

ANSWER: Yes. Because LOOP plants operate at reduced condenser water flows at low loads, there is the possibility of a greater rate of condenser tube fouling with a LOOP plant than with a

The Hartman Company

September 7, 2001

Page 9

HARTMAN LOOP CHILLER PLANTS

FREQUENTLY ASKED QUESTIONS

conventional plant in some configurations. Because the flow is variable, it is difficult to precisely plan the frequency of required tube cleaning of a LOOP plant. However, if specific tube cleaning intervals are essential, there is a great deal of flexibility in LOOP plant design and operations that should be considered by the designer or plant operations manager. To reduce the frequency of tube cleaning, the designer may decide to employ a three pass condenser bundle. This design can raise the flow rate in the chiller and eliminate any potential problem altogether. Once the plant is operating, the minimum condenser flow can be adjusted by the plant operators at any time to establish any tube cleaning intervals that are required. These steps may have a small effect on construction or operating costs for a LOOP plant, but they can keep the tube cleaning at present levels, or even reduce the frequency of cleaning . Furthermore, there are now several different approaches to automatic tube cleaning that can be implemented to ensure that chillers operate at all times with the highest possible condenser heat transfer.

7.

How does the owner know if LOOP technologies are performing as projected?

ANSWER: The Hartman LOOP Design Guide outlines a simple and low cost means of integrating real-time chiller plant efficiency monitoring into the plant controls. This added feature costs very little, but it is of enormous assistance in operating the plant and managing maintenance activities. The energy performance instrumentation provides a continuous readout of the current total chiller plant operation effectiveness in kW/ton and also accumulates data that can be compared with previous periods during which the plant operated under similar load and weather conditions. This information helps operations staff and management know very quickly when the plant operations stray from projected and historic energy use patterns, and it helps provide direction for getting the operation back on track.

The Hartman Company

September 7, 2001

Page 10

HARTMAN LOOP CHILLER PLANTS

FREQUENTLY ASKED QUESTIONS

LOOP EQUIPMENT REQUIREMENTS

1. What type of chiller and cooling tower are required by LOOP plants, and are these products readily available?

ANSWER: LOOP technologies are specifically developed for centrifugal chillers. Hartman has worked with the major chiller and cooling tower manufacturers during the development of the LOOP chiller plant technologies. Optimum LOOP plant configurations require that the centrifugal chiller be variable speed; all major chiller manufacturers make such variable speed chillers suitable for LOOP operation. LOOP plants also require that the cooling tower be a low head type with gravity or rotating sprinkler hot water distribution. Towers must also be constructed such that the tower works effectively with a condenser water flow turndown ratio of approximately 2.5:1. Many US and international manufacturers of cooling towers make such towers that are suitable for LOOP plants.

2.

Are manufacturers concerned about applying their equipment in LOOP plants, and does LOOP operation have any effect on equipment warranty?

ANSWER: No. LOOP designs never exceed the operating limits for the equipment selected. There is no effect on warranty, and manufacturers are generally pleased to have their equipment chosen for this ultra-efficient application. While in some locations the local manufacturers representatives for chillers and cooling towers may not fully understand LOOP technologies or envision widespread applications, the major manufacturers are supportive of LOOP configurations for their equipment.

3.

What type of control system is required by LOOP plants to provide network control?

ANSWER: A LOOP chiller plant can be implemented with any of a number of direct digital control (DDC) systems to operate the chiller plant and/or distribution equipment. The fundamental requirements are that: The DDC system has a functional and flexible programming language that uses floating point math and allows multiple layers of custom mathematical calculations and logic statements. You can find more about this requirement by viewing our Operators' Control Language (OCL) guide to DDC manufacturers on our web site. The DDC system has extensive and automatic network management features such that each controller can employ point & variable data from any other controller in its control programs. The DDC system employs standard or gateway protocol features such that it can be connected to communicate with chiller and variable speed drive equipment. Not all, but many of the existing DDC systems on the market today incorporate these features such that they are suitable for LOOP chiller plant control.
The Hartman Company September 7, 2001 Page 11

HARTMAN LOOP CHILLER PLANTS

FREQUENTLY ASKED QUESTIONS

SUPPORT FOR LOOP DESIGN & OPERATION

1. What is the history and present status of LOOP technologies?

ANSWER: Hartman first began investigating the benefits of all-variable speed chiller plants in 1992. At that time Hartman encouraged manufacturers and industry organizations to develop this promising technology. However, none elected to do so. Because even at that early date, the impact on plant energy use appeared to be very beneficial, Hartman determined to make the investment to develop the technologies internally. Along with the development, Hartman was encouraged by industry members to develop a mechanism for ongoing support since concerns arose that plant operators would not be able to tap their normal sources of operations information to keep all-variable speed plants operating at peak efficiency levels. Since 1996 when development began in a large scale, Hartman has invested nearly a million dollars in the development of LOOP technologies and plans to continue to invest at this rate for ongoing development and support for LOOP plants that are now being implemented. LOOP technologies are protected by three patents (US Patent No. 5,946,926, US Patent No. 6,185,946, and US Patent No. 6,257,007) and one other pending. In early 2000, the LOOP technologies were released to the public for use in new chiller plant designs as well as retrofits of existing plants. Projects incorporating LOOP technologies are now under design and construction.

2.

What is The Hartman Companys role in LOOP projects?

ANSWER: Hartman continues to be an advanced technology engineering firm and does not represent any manufacturer or sell any equipment. Nor does Hartman specifically recommend any one manufacturer for chillers, towers, pumps or DDC systems. While Hartman can act as the design engineer when desired by the owner, the firms main focus is on providing implementation support of LOOP technologies to design and contracting teams and ongoing operations support as required to the plants operations staff.

3.

How are LOOP technologies implemented and what is the cost?

ANSWER: Although LOOP technologies are protected by patents, Hartman endeavors to make them all open technologies subject to industry discussion and testing as any other new technology. There is no need to employ black box controls or other secretive devices or products in a LOOP chiller plant. LOOP technologies are implemented with standard mechanical systems and controls through standard plan/spec or design-build contract procedures. To recover development costs, Hartman issues a "Site License" for each installation. The fee for each site license is a one-time payment of $5.00 per installed ton of the chiller plant employing LOOP technologies. The site license entitles the licensee to employ all Hartman LOOP technologies at the site.

The Hartman Company

September 7, 2001

Page 12

HARTMAN LOOP CHILLER PLANTS

FREQUENTLY ASKED QUESTIONS

For additional support, a Support License provides the site license; an updated Hartman LOOP Design Guide, and a two year Ongoing Operations Support Agreement. The Hartman LOOP Design Guide includes suggested plant configurations, equipment specifications, LOOP control sequences and other important design considerations and guides. The Ongoing Support Agreement provides support for plant operations staff with on-line information, Q&A resources, and membership in the Internet based LOOP users group. The ongoing support services also include direct contact channels to LOOP operations specialists for assistance in troubleshooting for an initial two year period which can be easily extended to the entire life of the plant if desired. The cost for the Support License option is a one-time payment of $10.00 per installed ton. However, experienced designers know that in order for a newly developed technology to be implemented effectively, the design team should have access to special expertise in that technology throughout the design and implementation process. In LOOP projects, this can be accomplished with a LOOP Engineering Agreement which includes all the features of the site and support Licenses plus direct engineering support from The Hartman Company to the design and construction team. In Engineering Agreements, Hartman acts as a LOOP technology specialist to the design and construction team(s) and provides one or more of the following as deemed necessary by the design and construction team: A. Engineering and peer review services to the Owner, designer or contractor to assist in developing the configuration of chillers, towers, pumps, piping and controls that is most suitable for the plant needs and ensures optimal LOOP plant operation. B. Specifications for critical elements of the plant that include the chillers, towers, and DDC system, along with procurement services in order to ensure that the purchase of major equipment employs value analysis principles in a pre-purchase or similar process that fits the needs of the Owners organization. C. Construction services to review and support the DDC control contractors software and hardware design, and assistance to the startup, test and balance, or commissioning agencies in order to ensure that LOOP hardware and control sequences are implemented correctly and that optimum operational efficiency of the plant is achieved. The cost of an Engineering Agreement varies depending on the amount and areas of support that are identified as necessary. However, an Engineering Agreement option which includes site license, design guide and ongoing support in addition to this direct engineering support will cost approximately $20.00 per installed ton. Such an agreement is adequate to fully support a design team new to LOOP technologies but otherwise experienced in chiller plant design. The annual energy reduction achieved by employing LOOP technologies usually makes the payback for this one-time fee less than a year.

4.

How do I start evaluating LOOP technologies to see if they fit my application?

The Hartman Company

September 7, 2001

Page 13

HARTMAN LOOP CHILLER PLANTS

FREQUENTLY ASKED QUESTIONS

To start the process of implementing LOOP technologies into a chiller plant project, the first step is to see if LOOP technologies energy savings can meet the financial threshold required for your specific application. The Savings Calculator on The Hartman Company web site http://www.hartmanco.com/innovate/savecalc/index.htm - be employed for this purpose by giving a quick first cut estimate of the savings that can be achieved by implementing the LOOP technologies in a specific application. The savings calculator also provides a range of first cost differentials for implementing a LOOP plant in place of a conventional plant for new, upgrade, or retrofit applications. These broad ranges can be refined by your design team or with our help. If the decision is made to employ LOOP technologies, the project engineer or owner should contact Hartman to purchase a Site License, Support License, or an Engineering Agreement depending on the level of support required. With the license or agreement in place, any change in plant capacity requirement that may be made during the design process can be easily accommodated by a simple adjustment to the license or agreement. Such flexibility for plant size adjustments is built into all licenses and agreements.

The Hartman Company

September 7, 2001

Page 14

USING GATEWAY TECHNOLOGIES

Practice eliminates need for duplicating sensing and control points
By THOMAS HARTMAN, PE, The Hartman Co., Marysville, Wash.
TYPICAL CHILLER-PACKAGE POINTS Chilled-water supply temperature Chilled-water return temperature Chilled-water flow status Entering condenser-water temperature Leaving condenser-water temperature Condenser-water flow status Chiller power (amperes) Evaporator refrigerant pressure Evaporator refrigerant temperature Condenser refrigerant pressure Condenser refrigerant temperature Compressor discharge refrigerant temperature Oil pressure Oil temperature Chiller status Alarm status Chiller start/stop command Chilled-water set-point control Chiller maximum-demand control TYPICAL DDC POINTS Chilled-water supply temperature Chilled-water return temperature Entering condenser-water temperature Leaving condenser-water temperature Chiller power (amperes)

Chiller-Plant Control

n the building-construction industry, a design that reduces operating costs is appealing, but only when its cost fits the construction budget. This truism was clear in our minds as we began implementing a new all-variable-speed-chiller-plant design that provides significant reductions in annual cooling-energy costs. Adopting an approach employed for many new technologies, we aimed for both lower installation and lower operating costs. However, because the variable-frequency drives (VFD) of chillers, pumps, and tower fans in an all-variable-speedchiller plant add to equipment costs, we began looking for ways to reduce the total capital cost of this new plant. We found at least part of the answer in another emerging industry technology control gateways, which are eliminating the need for duplicating sensing and control points in an increasing number of HVAC applications. REDUNDANT CONTROLS AND INSTRUMENTATION Generally, commercial-building construction projects involve a single manufacturer and contractor for controls. The job of this contractor is to make all HVAC equipment operate according to a sequence of operations spelled out in the project specifications. In doing this, the controls contractor applies the control system he represents to all elements of the project. The contractor supplies controllers and actuators, modulating valves, dampers, and other equipment to the mechanical contractor(s) and

Chiller status Alarm status Chiller start/stop command Chilled-water set-point control Chiller maximum-demand control

TABLE 1. Points included in chiller controllers and chiller-plant DDC controls.

equipment manufacturers for factory and, ultimately, field installation. As the project nears completion, the controls contractor goes through the building, adding equipment and connecting the devices previously supplied to the others. Though still common, this practice of single-sourcing controls is not particularly efficient because it can lead to construction delays and/or cost increases attributed to delivery and other coordination problems and can result in substandard operation or reliability when components do not fit together properly and fail to perform as expected. In recent years, this approach also has resulted in an increasing amount of control redundancy, including duplication of controls instrumentation. This is particularly true in the case of chillers, which now are supplied

from the factory with a complete direct-digital-control (DDC) and monitoring package. The redundancy in chiller-plant control systems is illustrated in Table 1, which shows the monitoring and control points typically included in a chiller package and those points that generally are field-installed separately in the DDC controls installation. A TREND TOWARD GATEWAYS Before the mid-1990s, many manufacturers of digital HVAC controls and equipment utilizing digital control were secretive to the point that the communications details of their products often were not made available to purchasers. Today, the trend is for manufacturers to open communications architecture and encourage (or at least not discourage)

Heating/Piping/AirConditioning January 2000

HPAC
ENGINEERING

81

CONTROL GATEWAYS

others to communicate with their systems through published interface information using communication gateways. Many manufacturers of HVAC controls and equipment have an in-house interface standard that can be employed by others for connection to and communication with their equipment, while others are embracing one or more of the emerging industry communication protocol standards. Currently, the primary method of connecting different systems involves gateway-type connections, but in the future, systems may employ industry-standard architecture throughout or have gateways embedded in all controllers so that connection to other systems can be made throughout their network architecture. The result of this trend is that controls and equipment from different manufacturers usually can be connected

and interoperated to some degree. This development is being embraced because: Manufacturers of equipment such as air handlers, cooling towers, and VAV boxes, who typically either have shipped components without controls or had to factory-install controls and devices supplied by others, realize that designing their equipment around their own controls and devices enables them to better control manufacturing costs, streamline production, and capture more value. Design engineers realize that it may simplify and speed the construction process and help them satisfy their clients desire not to be tied to a single controlsystem vendor. Building owners and managers believe that it will facilitate integration with other building and computing sys-

tems and improve the enterprise-level performance of their facility-management system with features that reach far beyond the simple HVAC monitoring and control that is the limit of most building-control systems. GATEWAYS FOR CHILLERS AND VARIABLE-FREQUENCY DRIVES As the trend toward gateways developed, the first two types of equipment to become available with gateway-type communication interfaces were chillers and VFDs. For this reason, an all-variable-speed-chiller plant is an excellent starting point for employing this new approach. A comparison of a standard chillerplant DDC connection and a DDC interface with a gateway-type connection can be made by considering figures 1 and 2. Figure 1 shows how a conventional chiller plant that employs constant-speed chillers, pumps, and fans typically would be instrumented using discrete devices for each DDC I/O point. The power- and flow-metering equipment is included to provide chilled-water production and plant-efficiency information to the owner and operations staff. The DDC interface shown in Figure 1 requires 45 separate points. An all-variable-speed plant would be expected to have the same or even a larger number of discrete points, because each VFD must be operated both digitally (start/stop) and via analog control (motor speed). However, by employing a gateway approach as diagrammed in Figure 2, the number of discretely connected points required in an all-variablespeed-plant design is reduced to nine. All other input and output points are connected through the networks. This large reduction in discrete hardware connections is made possible by several networking and interconnecting features of VFDs. Note that the VFD operating the chiller compressors in Figure 1 are connected to the chiller controller, but in the connection scheme of Figure 2, each also is connected to the chiller-plant network. Depending on the chiller connection to the VFD, this is useful because it makes information accessible via a net-

What Is a Gateway?

any digital systems used in building operation today employ rules for communicating among their units that are unique to their system. That is why even when two different systems employ the same RS485 or Ethernet local-areanetwork architecture, it usually still is not possible for them to exchange information. Only when both systems also employ the same communication protocol, such as BACnet or LonWorks, can intercommunication be counted upon. For systems that do not employ the same network architecture and communication rules, a gateway (sometimes called protocol interface) is required. A communication gateway is a device that connects two networks and converts dissimilar protocols. Some gateways employed in large wide-area-network and Internet applications are complex and expensive, but those required for direct-digital-control (DDC) systems usually are quite simple. Typical communication gateways consist of three elements: G A communication link that connects the two (or more) systems. G One or more translators (or drivers) that understand the communication rules of each system being connected so that data streams can be converted into useful information for the system or systems that require the interconnection. G A simple operating system and section of memory (or buffer) where required intercommunication data is mapped so that relevant data is captured and stored until it can be communicated to the other system(s). Today, most DDC field panels are built with gateways. Gateways enable simple variable-air-volume terminal-control networks to communicate on a higher-speed field-panel communications network. Many such panels also include configurable gateways. Depending on the requirement for intercommunication, such control panels can be shipped from the factory with the communications link and drivers required for intercommunication or these features can be field-installed and programmed. Thus, the cost of gateway integration of systems has fallen dramatically in recent years and can be expected to approach zero for common systems in the near future. Many in the industry see gateways as a temporary fix as the move toward standard communication protocols takes place. However, whether the entire building-construction industry ever will embrace a single communication standard remains an open question.

82

HPAC
ENGINEERING

January 2000 Heating/Piping/AirConditioning

CT1 SS DO Relay CT1 STS DI Current switch CT1 KW AI kw transducer CT1 ALRM DI Dry contact CT1 HTR DO Relay CT1 STEMP AI
Wtr temp sensor

DO

kw

kw

CT2 SS Relay DI CT2 STS Current switch AI CT2 KW kw transducer DI CT2 ALRM Dry contact DO CT2 HTR Relay AI CT2 STEMP
Wtr temp sensor

CDWP1 SS DO Relay CDWP1STS DI Current switch CDWP1 KW AI kw transducer

C
kw kw

AI CH1 ECDWT
Wtr temp sensor

DO CDWP2 SS Relay DI CDWP2STS Current switch AI CDWP2 KW kw transducer

AI CH1ECDWT
Wtr temp sensor

AI CH2ECDWT Wtr temp sensor AI CH2 ECDWT

Wtr temp sensor

DO

CH1 SS Relay DI CH1 Dry contact AO CH1CWTSP 4-20ma signal

kw

COND

COND

COND

EVAP

AO CH1DMDSP 4-20ma signal Al CH1 KW kw transducer

AI CH1 SCWT
Wtr temp sensor

AI CH2 SCWT
Wtr temp sensor

EVAP

COMP
kw

Chiller 1

AI CH1 RCWT
Wtr temp sensor

AI CH2 RCWT
Wtr temp sensor

Chiller 2

DO CH2 SS Relay DI CH2 Dry contact AO CH2CWTSP 4-20ma signal AO CH2DMDSP 4-20ma signal AI CH2 KW kw transducer

PCWP1 SS DO Relay PCWP1STS DI Current switch PCWP1 KW AI kw transducer

C
kw kw

DO PCWP2 SS Relay DI PCWP2STS Current switch AI PCWP2 KW kw transducer Point type DI Digital input DO Digital output AI Analog input AO Analog output

AI CWFLOW Flow meter AI SCWT AI RCWT

Point name

Wtr temp sensor

Wtr temp sensor

DO CT2 HTR Relay Chilled-water supply and return to and from the distribution system I/O device DDC Ppoint Legend

FIGURE 1. Conventional constant-speed-chiller-plant DDC interface.

work connection that otherwise would not be accessible through the chiller controller. While such a dual-network connection may be unusual, it will not compromise effective chiller operation because all operating commands to the VFD still would be coordinated through the chiller controls. The network connection to the VFD may be used to report only the power ( KW ) draw of the chiller. With pump and fan VFD, the network connection is used to turn the VFD on and off, set its speed, determine its status, and read the instanta-

neous KW use of the motor controlled by the VFD. The points that are communicated to the chiller-plant controller by the network are shown in Table 2. We estimate that the reduction in devices and field work attributed to these network connections will cut controls costs by 25 to 30 percent, which is sufficient to pay the premium for VFD for all fans and pumps in the plant. With this network-based control scheme and other economies, it is possible to implement all-variablespeed-chiller plants that cost the same as or less than and that significantly

outperform conventional high-efficiency constant-speed-chiller plants. PITFALLS OF GATEWAY CONNECTIONS However, before implementing a controls design that employs gateways, a designer should be aware of the common pitfalls of gateway connections and take steps to avoid them. The three most common problems with gateway connections we have seen are: Information is not accessible across the network as desired. A commandable point such as a digital output

Heating/Piping/AirConditioning January 2000

HPAC
ENGINEERING

83

CONTROL GATEWAYS

(DO) or analog output (AO) may be unable to receive a command across the network because operating capabilities across gateways usually are limited by equipment manufacturersunless, that is, requirements are specified clearly. Timing and precision issues across the network hamper smooth operation. The network interface may be so slow that information exchanges take minutes to perform or a gateway may have such a large change-of-value

(COV) limit that an analog input (AI) must change significantly before the change will be reported across the network. Both of these limitations may make cross-network control unstable or erratic if timing and COV reportinglimit requirements are not made clear in the specifications. Point value or status becomes unstable under certain conditions. If a point value exceeds the range that is expected by the gateway or if commu-

nication across the network is lost, the receiving controller may not retain a default value or status. The designer must be certain that these possibilities have been carefully considered and dealt with effectively to ensure the operational integrity of a network-based control system. These potential problems and others that can crop up easily make it imperative that a designer use great care when employing gateway data-transfer de-

VFD CT1 ALRM DI Dry contact CT1 HTR DO Relay CT1 STEMP AI
Wtr temp sensor

VFD DI CT2 ALRM Dry contact DO CT2 HTR Relay AI CT2 STEMP
Wtr temp sensor

VFD Network connection

VFD Network connection

Network connection COND Network connection VFD COND EVAP COND EVAP COND

Network connection Network connection VFD

Chiller 1 Network connection VFD

Chiller 2
Network connection VFD

AI CWFLOW Flow meter AI SCWT AI RCWT

Point type DI Digital input DO Digital output AI Analog input AO Analog output AI

Point name

Wtr temp sensor

Wtr temp sensor

RCWT

Wtr temp sensor

I/O device DDC Point Legend

Chilled-water supply and return to and from the distribution system

FIGURE 2. All-variable-speed-chiller-plant DDC interface using network connections to chillers and variable-speed drives.

84

HPAC
ENGINEERING

January 2000 Heating/Piping/AirConditioning

C H I L L E R - P L A N T - N E T W O R K
Description Point name System Bldg System Point Chiller plant networked points schedule Chiller No. 1 control panel Chilled-water supply temp CP CH1 CWST Chilled-water return temp CP CH1 CWRT Chilled-water flow status CP CH1 CWFLO Entering condenser-water temp CP CH1 ECDWT Leaving condenser-water temp CP CH1 LCDWT Condenser-water flow status CP CH1 CDWFLO Evaporator refrigerant pressure CP CH1 EVAPP Evaporator refrigerant temp CP CH1 EVAPT Condenser refrigerant pressure CP CH1 CDNP Condenser refrigerant temp CP CH1 CDNT Compressor-discharge temp CP CH1 DISCHT Inlet vane position signal CP CH1 VANE Oil pressure CP CH1 OILP Oil temperature CP CH1 OILT Chiller status CP CH1 Alarm status CP CH1 ALRM Chiller start/stop command CP CH1 SSC Chilled-water set-point control CP CH1 CWSPC Chiller-demand control CP CH1 DMDC Chiller No. 1 variable-frequency drive VFD compressor speed signal CP CH1D RPM VFD status CP CH1D VFD alarm CP CH1D ALRM VFD power CP CH1D KW Chiller No. 1 CW pump variable-frequency drive VFD pump speed command signal CP CWP1 RPMC VFD status CP CWP1 VFD alarm CP CWP1 ALRM VFD power CP CWP1 KW Chiller No. 1 CDW pump variable-frequency drive VFD pump speed command signal CP CDWP1 RPMC VFD status CP CDWP1 VFD alarm CP CDWP1 ALRM VFD power CP CDWP1 KW Chiller No. 1 tower-fan variable-frequency drive VFD fan speed command signal CP CT1F RPMC VFD status CP CT1F VFD alarm CP CT1F ALRM VFD power CP CT1F KW Chiller No. 2 control panel Chilled-water supply temp CP CH2 CWST Chilled-water return temp CP CH2 CWRT Chilled-water flow status CP CH2 CWFLO Entering condenser-water temp CP CH2 ECDWT Leaving condenser-water temp CP CH2 LCDWT Condenser-water flow status CP CH2 CDWFLO Evaporator refrigerant pressure CP CH2 EVAPP Evaporator refrigerant temp CP CH2 EVAPT Condenser refrigerant pressure CP CH2 CDNP Condenser refrigerant temp CP CH2 CDNT Compressor discharge temp CP CH2 DISCHT Inlet vane position signal CP CH2 VANE Oil pressure CP CH2 OILP Oil temperature CP CH2 OILT Chiller status CP CH2 Alarm status CP CH2 ALRM Chiller start/stop command CP CH2 SSC Chilled-water set-point control CP CH2 CWSPC Chiller demand control CP CH2 DMDC Chiller No. 2 variable-frequency drive VFD compressor speed signal CP CH2D RPM VFD status CP CH2D VFD alarm CP CH2D ALRM VFD power CP CH2D KW Chiller No. 2 CW pump variable-frequency drive VFD pump speed command signal CP CWP2 RPMC VFD status CP CWP2 VFD alarm CP CWP2 ALRM VFD power CP CWP2 KW Chiller No. 2 CDW pump variable-frequency drive VFD pump-speed command signal CP CDWP2 RPMC VFD status CP CDWP2 VFD alarm CP CDWP2 ALRM VFD power CP CDWP2 KW Chiller No. 2 tower-fan variable-frequency drive VFD fan speed command signal CP CT2F RPMC VFD status CP CT2F VFD alarm CP CT2F ALRM VFD power CP CT2F KW Type Range Comments

P O I N T S

L I S T

AI AI DI AI AI DI AI AI AI AI AI AI AI AI DI DI DO AO AO AI DI DI AI AO DI DI AI AO DI DI AI AO DI DI AI AI AI DI AI AI DI AI AI AI AI AI AI AI AI DI DI DO AO AO AI DI DI AI AO DI DI AI AO DI DI AI AO DI DI AI

20F - 70F 20F - 70F Yes/No 50F - 120F 50F - 120F Yes/No See comment. 20F - 70F See comment. 20F - 70F 50F - 120F 0% - 100% 0-50 psi 50F - 160F On/Off Alarm/Off Start/Stop 38F - 54F 10% - 100% 0% - 100% On/Off Alarm/Off See comment. 0% - 100% On/Off Alarm/Off See comment. 0% - 100% On/Off Alarm/Off See comment. 0% - 100% On/Off Alarm/Off See comment. 20F - 70F 20F - 70F Yes/No 50F - 120F 50F - 120F Yes/No See comment. 20F -70F See comment. 20F -70F 50F - 120F 0% - 100% 0-50 psi 50F - 160F On/Off Alarm/Off Start/Stop 38F - 54F 10% -100% 0% - 100% On/Off Alarm/Off See comment. 0% - 100% On/Off Alarm/Off See comment. 0% - 100% On/Off Alarm/Off See comment. 0% - 100% On/Off Alarm/Off See comment.

Connect via specified interface between chiller controls and chiller-plant controller. Range is minimum.

Scale point as required depending on operating pressure of refrigerant employed in chiller. Scale point as required depending on operating pressure of refrigerant employed in chiller.

Connect via specified interface between VFD network and chiller-plant controller.

Connect to provide 0- to 100-percent outside-air damper control. Actuator by AHU mfg. Scale point as required depending on operating power range of unit. Connect via specified interface between VFD network and chiller-plant controller.

Scale point as required depending on operating power range of unit. Connect via specified interface between VFD network and chiller-plant controller.

Scale point as required depending on operating power range of unit. Connect via specified interface between VFD network and chiller-plant controller.

Scale point as required depending on operating power range of unit. Connect via specified interface between chiller controls and chiller-plant controller. Range is minimum.

Scale point as required depending on operating pressure of refrigerant employed in chiller. Scale point as required depending on operating pressure of refrigerant employed in chiller.

Connect via specified interface between VFD network and chiller-plant controller.

Scale point as required depending on operating power range of unit. Connect via specified interface between VFD network and chiller-plant controller.

Scale point as required depending on operating power range of unit. Connect via specified interface between VFD network and chiller-plant controller.

Scale point as required depending on operating power range of unit. Connect via specified interface between VFD network and chiller-plant controller.

Scale point as required depending on operating power range of unit.

TABLE 2: Networked all-variable-speed-chiller-plant DDC points.

86

HPAC
ENGINEERING

January 2000 Heating/Piping/AirConditioning

CONTROL GATEWAYS

sign schemes. While there is a developing consensus about how gateway interfaces should perform, individual opinions vary widely at this early stage, with some manufacturers new to the idea of cross-network operations. This can lead to misunderstandings about the role of each gateway and result in unsatisfactory performance of systems that rely on gateway communication. To minimize such problems, I recommend that special consideration be given to specifying each gateway connection or interface. SPECIFYING GATEWAYS While much has been written about how to specify interfaces between systems, much more needs to be developed. My rule for myself is never specify features I do not fully understand or that are too difficult to evaluate to be enforced. To ensure an effective interface, our firm believes one should: Base the gateway or interface specification on a well-known industry standard. We recommend never basing an interface specification on a single companys open standard. Instead, we specify at least one industrystandard communication-interface option. However, after all of the equipment and systems have been selected, it may be useful to entertain proposals for an inter-system interface employing a non-standard open interface. Such a proposal could be accepted if it results in reduced costs, it comes with a guarantee that the operational capabilities will not be compromised, and the owner agrees that such change will not limit the ability to implement desired upgrades. Describe exactly what data need to be transferred across the interface and how that data is to be employed by each system involved. For example, if a chiller must be controlled on and off and have its maximum power setting set across a gateway interface, it is necessary to call for the chiller start/stop and demand-control points to be networked across the gateway. It also is necessary to explain how and where in the network the commands are to be made. Finally, it is useful to detail how the information will be employed on each side of the interface,

how often data must be sent and/or received, and any other information that helps ensure that the system operates effectively. Assign a single source of responsibility and authority in making each interface work. From chiller and VFD interfaces to DDC controls, we recommend that the DDC-controls vendor be responsible for each interface and be given the authority to determine if the interfaces provided by the chiller or VFD vendor meet the requirements of the specified interface standard and other specification items. If the controls vendor concludes that an interface does not meet specified requirements, then the provider of that interface must make the changes mandated by the controls vendor or clearly show where the controls vendor erred. This approach greatly reduces the finger pointing that can overwhelm a designer when a multi-vendor network is brought on-line. By basing gateway connections on industry-standard communications, fully specifying the data that must be transferred and how it is to be transferred, and assigning a single point of authority and responsibility for the interconnections, confusion and finger pointing during installation and startup is minimized. This approach is a good model for designers approaching a future in which multi-vendor controls networks likely will become standard practice. GATEWAYS: CHANGING HVAC CONTROLS FOREVER As the trend toward gateways and other interface standards is accelerated by the increasing demands and expectations of major industry players, I believe that the way our industry works will change forever. Controls contractors increasingly will work to fill a serious void in construction services and increase their sales by recasting themselves as system integrators. As part of this changing role, they will lose their loyalty to the control-system manufacturer to which they historically were bound and start looking to expand their product line and their knowledge of other systems so that they will be able to offer the most at-

tractive combination of hardware products and capture a larger piece of the growing equipment and data-integration market. Manufacturers of control products will begin to find the construction market for their products shrinking, as more and more controls are factory-installed under OEM agreements. Contractors will find the construction process to be simpler and startup easierbut only if designers have done their job well. The design stage, therefore, will become more important and require designers to develop digitalcommunication expertise that is not widely available in the industry today. But if designers are successful in developing this expertise and employing it in well-structured controls specifications, then putting together an HVAC system will, from the standpoint of controls, become more like putting together an office data-communications network. It will be easier and less costly than controls implementation is today and result in a better-performing system. SUMMARY AND CONCLUSION Employing gateways to connect systems effectively requires substantial adjustments to the design and specification of affected systems. However, the results are worth the effort, as construction costs and startup problems are reduced through the elimination of point redundancy. Because manufacturers of chillers and VFD are leaders in the trend toward gateway connection, employing gateways to connect their equipment to DDC systems offers a good opportunity for designers to begin employing this important new technology and prepare for a future in which multivendor control systems will be common. Additional information on technologies discussed in this article is available on the Web at www.hartmanco.com . Questions and comments about the article can be addressed to the author at HPAC tomh@hartmanco.com.
Circle 502 on reader service card if this article was useful; circle 503 if it was not.

Heating/Piping/AirConditioning Heating/Piping/AirConditioning January Month2000 Year

HPAC
ENGINEERING

87 6

Hartman
Company
%fax: 254-793-0121

The

Consulting Engineers & High Performance Building Designs /755 County Road 247, Georgetown, Texas 78628 (ph: 254-793-0120 7E-mail: tomh@hartmanco.com

A HARTMAN LOOP EXAMPLE AN ENERGY COMPARISON OF A LOOP CHILLER PLANT WITH CONVENTIONAL CHILLER PLANTS
by THE HARTMAN COMPANY

NOTE: This document shows a specific example of how a LOOP chiller plant reduces energy use compared to conventional constant speed or variable speed chiller plant configurations. In order to present an example with verifiable numbers, this example uses and identifies specific pieces of equipment. However, any one of a number of different equipment configurations could be employed to achieve nearly identical results. This document is not intended to indicate endorsement or support of one manufacturers equipment over another. The purpose of this document is to evaluate chiller plant performance for a typical application at a specific operating point to show how LOOP technologies can reduce energy use compared to conventional chiller plants. This example is for a chiller plant that serves a load that peaks at 1800 tons of cooling. The load is a commercial office building and the chiller plant consists of three 600 ton York centrifugal chillers, each with a nominal 0.6 kW/ton efficiency. Each chiller is connected to a Marley NCB2A1 Tower 600 ton cross flow cooling tower. For simplicity, a constant 42oF chilled water supply temperature is assumed at all loads (though actual LOOP projects employ variable chilled water temperature when possible for additional energy savings). The plant is located in Detroit, and has a load profile shown in the figure below, which is typical for such a facility in the Detroit climate. Detroit is chosen for this specific example because the Detroit load profile is typical of cooling load profiles for chiller plants throughout much of the US. A design tower approach temperature of 8oF establishes an 85oF entering condenser water temperature at the design condition or 77oF wet bulb.

Page 1 of 5

Hartman
Company
%fax: 254-793-0121

The

Consulting Engineers & High Performance Building Designs /755 County Road 247, Georgetown, Texas 78628 (ph: 254-793-0120 7E-mail: tomh@hartmanco.com

Chiller Plant Load Profile for Detroit

Cooling Load Profile 20% Average Wet Bulb Temperature 80

75 Percent of Operations Time at Load 15% 70 Average Wet Bulb Temperature

10%

65

60 5% 55

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

50

Chiller Plant Loading

Using this chiller plant load profile, this example considers a specific part load operating point of 1/3 total plant capacity. Note from the chart above that this chiller plant spends most of its time operating at and around this point. We will compare operating energy consumption at this point for: 1. 2. 3. a constant speed centrifugal chiller plant a conventional variable speed centrifugal chiller plant a LOOP all-variable speed centrifugal chiller plant

First, consider a constant speed plant. At 1/3 plant capacity, one chiller, one tower, and one condenser pump are operated. Assume a 3 gpm/ton condenser water flow, and a constant speed tower fan with an optimization strategy aimed at minimizing condenser water temperature. The Marley tower head requirement is 12 ft, the chiller condenser head requirement is 16 ft, and our assumed piping loss is 32 ft. for a total pump head of 60 ft. A B&G 1531, 6BC pump yields an operating power requirement of 34 hp at these conditions. The Marley NCB2A1 Tower achieves an 8oF approach for an 85oF entering water temperature at design conditions. This tower employs a 25 hp fan. The Marley performance data shows that at a 600 ton load and the 56oF Wet Bulb temperature shown on the chart above for 1/3 plant capacity, the tower will deliver 69.9oF leaving Tower Water temperature with the fan in constant operation. From York constant speed chiller

Page 2 of 5

Hartman
Company
%fax: 254-793-0121

The

Consulting Engineers & High Performance Building Designs /755 County Road 247, Georgetown, Texas 78628 (ph: 254-793-0120 7E-mail: tomh@hartmanco.com

power curves, the chiller power requirement is 84% of the power at design entering condenser water conditions. So, power requirements for the entire plant are as follows: Chiller - constant speed 600 tons x 0.6 kW/ton x .84 Demand = 302 kW Condenser pump 34 hp x 0.746 kW/hp / .92 efficiency = 28 kW Tower Fan 25 hp x 0.746 kW/hp / .92 efficiency = 20 kW -----------------------------------------------------------------------------------------------------------------------------------TOTAL 350 kW Now assume a plant with variable speed chillers of the same nominal efficiency. A conventional variable speed plant still operates only a single chiller at this operating point. Making allowance for the VFD losses, the York power curves show that the power requirements from the York variable speed chiller curves reduces the power from 84% to 75%. Thus the power requirements for this operating point with a variable speed chiller are: Chiller - variable speed 600 tons x 0.6 kW/ton x .75 Demand = 270 kW Condenser pump 34 hp x 0.746 kW/hp / .92 efficiency = 28 kW Tower Fan 25 hp x 0.746 kW/hp / .92 efficiency = 20 kW -----------------------------------------------------------------------------------------------------------------------------------TOTAL 318 kW Now lets look at an all-variable speed LOOP chiller plant in operation. In a LOOP chiller plant, all chillers, pumps and tower fans are variable speed. The operations calculator in the DDC controller that operates the plant calculates that two chillers is optimum for this operating point, and sequences the operation as follows: Chiller Capacity (each for two chillers ) = 50% (300 tons) Effective entering condenser water temperature = 64oF Because each tower is only loaded to 300 tons, this optimized lower leaving tower water temperature is obtained with the following operating parameters for each cooling tower: Condenser Pump Flow (Gpm) = 1440 gpm Condenser Pump Head (feet) = 42 ft. Condenser Pump speed (RPM) = 1470 rpm Condenser Pump motor hp = 19.2 hp Tower Fan speed (RPM) = 1440 rpm Tower Fan motor hp = 13 hp Actual tower leaving temperature = 62.9oF Note that the 64oF effective entering condenser water temperature is based on 1800 gpm flow. To adjust for heat transfer changes due to flow reduction and the decrease in log mean Page 3 of 5

Hartman
Company
%fax: 254-793-0121

The

Consulting Engineers & High Performance Building Designs /755 County Road 247, Georgetown, Texas 78628 (ph: 254-793-0120 7E-mail: tomh@hartmanco.com

temperature in the bundle, the calculator shows a true tower leaving temperature requirement of 62.9oF is equivalent to the 64oF entering water temperature at full flow. Using these operating parameters, the power requirements are obtained from York, Marley and B&G performance curves and data and adjusted for variable speed drive losses. The total power consumed is as follows: Chillers - variable speed 300 tons x 0.6 kW/ton x .51 Demand x 2 = 184 kW Condenser pumps (variable speed) 19.2 hp x 0.746 kW/hp / .88 efficiency x 2 = 33 kW Tower Fans (variable speed) 13.0 hp x 0.746 kW/hp / .88 efficiency x 2 = 22 kW ----------------------------------------------------------------------------------------------------------------------------------------------------------TOTAL 239 kW The constant speed plant uses 46% more power and the conventionally operated variable speed plant approach uses 33% more energy than the all-variable speed LOOP chiller plant at this operating point. By making the same calculation for each of the intervals shown in the above chart, we find that in this application on an annual basis, the LOOP plant offers more than double the energy reduction of a conventional variable speed chiller plant compared to a constant speed plant. These results are shown in the chart below:
Comparison of Hartman LOOP vs Traditional Chiller System Performance
At various load conditions in Detroit application 1.00 Performance Map of Conventional VS Chiller Plant Performance Map of Hartman LOOP with constant supply CW temperature 25% Performance Map of Conventional CS Chiller Plant Load Profile of Chiller Plant

0.75

15% 0.50 10%

0.25 5%

0.00 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

0%

Plant Capacity Requirement

Page 4 of 5

Percent Of Time Plant Operates At Load

Total Chiller Plant Energy Use (KW/ton)

20%

Hartman
Company
%fax: 254-793-0121

The

Consulting Engineers & High Performance Building Designs /755 County Road 247, Georgetown, Texas 78628 (ph: 254-793-0120 7E-mail: tomh@hartmanco.com

This improved energy savings coupled with a simpler plant configuration and other features make LOOP chiller plants cost about the same, or in some cases less than conventional plants to construct. Also, in many climates, LOOP technologies makes changing out chillers that employ phased out refrigerants a very attractive investment. While it may appear that the patented control sequences that constitute LOOP chiller plant technologies result in a more complex plant, they do not. In fact, the sequence of operations for a LOOP plant is actually simpler than a conventional optimized chiller plant because LOOP technologies include a new simple set of parameters that provide much more straightforward control of all plant equipment. Please review this example to see if the concepts employed hold true for the plant you are considering constructing or upgrading. If you have any questions or comments please call, write or e-mail us, and we will be happy to respond promptly with whatever information is required to complete your analysis of this exciting new technology for your facility!

March, 2000

Page 5 of 5

THE COMFORT INDUSTRY:

a 21st-Century Opportunity
A market worth tens of billions of dollars awaits those courageous and creative enough to give building occupants what they want most
By THOMAS HARTMAN, PE Principal, The Hartman Co. Marysville, Wash.

n designing HVAC systems for commercial buildings, many engineers believe that comfort and ventilation will be suitable as long as a minimum outside-air flow is maintained and heat flows are balanced in each zone. What they fail to consider is that building zones often encompass large areas, multiple offices, and a number of occupants, whose perception of comfort and environmental quality is influenced by thermal variances, radiant energy, the movement and distribution of air, and a host of other factors. Consider Figure 1, which shows the elements of a typical HVAC system in a commercial building. When I present this to industry professionals, most can cite rules of thumb for sizing and selecting chillers, boilers, air systems, and piping. Few, on the other hand, can recall basic facts about the influence of humidity, air movement, and radiant energy on occupant comfort. What this shows is that while the purpose of an HVAC system is to keep building occupants comfortable, environmental quality for individuals seldom is a significant factor in design. Usually, occupants have no direct control

A member of HPAC Engineerings Editorial Advisory Board, Thomas Hartman, PE, is principal of The Hartman Co., a Marysville, Wash.-based firm that employs new technologies in offering enhanced, individually adjustable, and accountable comfort systems for commercial buildings. More information about these technologies can be obtained at the firms Website (www.hartmanco.com) or by contacting Mr. Hartman at tomh@hartmanco.com.
Background: Stone/Terry Vine; hand with remote control: Photodisc/Nancy R. Cohen. Composited by: HPAC Engineering.

HPAC Engineering January 2001

41

T H E

C O M F O R T

I N D U S T R Y

Building-control network

Tower
DDC CDWR CWS DDC

C C

Fan
DDC box

FIGURE 1. Elements of a typical HVAC system.

CWR DDC DDC

Cond.
DDC

HWS HWR

CDWS

Boiler
DDC

DDC

Chiller

over their thermal environment. Indeed, most do not even have a temperature sensor or thermostat located near them. This article discusses the emerging comfort industry, the forces driving it, and ways designers, manufacturers, and system integrators can benefit from it.
THE ENEMY IS US

What we discovered was that comfort ranks first in importance to both occupants and building owners, which means the HVAC industry cannot reasonably blame developers and building owners for the failure to implement more comfortable systems. Increasingly, it looks as if, We have seen the enemy, and it is us. While buildings may be more comfortable than they were decades ago, it is clear that the industry is

focused industries in which innovation in product design is discouraged. Manufacturers know that if they develop a product with features that significantly set it apart, a substantial portion of the market will be closed to that product because designers are trained to specify only those components that have equals so that any one of several (usually at least three) products can be applied. This is because the number of products required

For decades, surveys have shown that building occupants biggest complaint about their workplace is lack of thermal comfort. A recent Building Owners and Managers Association (BOMA) survey reaffirmed this, showing that lack of occupant control also is of utmost concern. Yet despite these long-standing complaints, low-cost sensing and processing capabilities and other technology enhancements have not been widely employed in products and rarely are sought for HVAC-system designs. In an effort to understand why occupant desires for improvement have not been heeded as enabling technologies have advanced, my firm co-sponsored and conducted a survey of attendees at the 1997 BOMA Convention and Office Building Show in Minneapolis. We were interested in seeing if building owners and operators were impeding the implementation of better comfort systems. The results of the survey are shown in Figure 2.
42

The HVAC industry is one of the few technologyfocused industries in which innovation in product design is discouraged.
not meeting building occupants rising expectations for comfort and control.
INSTITUTIONAL IMPEDIMENTS

So, if technology has made possible more comfortable buildings, and both occupants and building owners want them, why are they not being designed and built? The answer is that the building-construction industry has institutional impediments to the type of change required to make systems more comfortable. Such impediments are: Innovation is discouraged. The HVAC industry is one of the few technology-

for a typical HVAC system is large, and current construction procedures do not provide designers with the capacity to negotiate and enforce pricing. Pressure on first costs. This is not new to anyone in the building-construction industry. Many designers have seen comfort-enhancing features such as improved controls, smaller zones, and better glazing minimized and even eliminated as final budgets were developed. This happens with surprising frequency and discourages the industry from focusing more on comfort, even when doing so requires only a very small cost premium.

January 2001 HPAC Engineering

T H E

C O M F O R T

I N D U S T R Y

60 50 Superior comfort Flexible and easily configured Efficient and cost-effective servicing Temperature and occupancy sensing Increased worker productivity Detailed monitoring of energy usage Automatic self-diagnostics

PERCENT OF ATTENDEES

40 30 20 10 0

Dont know

None

Little

Some

Substantial

Extreme

FIGURE 2. Perceived value of improved zone-control features.

Operations-support issues. The relationships between designers and between manufacturers and building-operations staffs never have been strong. All too often, we hear, The operator will reduce any system to his/her level of understanding. True of all systems in all industries, this should challenge designers and manufacturers to put more emphasis on operations-support issues. But in the HVAC industry, it often is used as an excuse to not work harder to achieve more effective systems, particularly at the zone (comfort) level. Lack of integration. Because the HVAC industry continues to consider controls and hardware independently of one another, the functionality of zone products is compromised. Comfortenhancing features such as integrated occupancy sensing and individual occupant control could be implemented in a zone system easily and inexpensively; however, terminal hardware is crude, and terminal-controller program capacity is aimed at hardware adaptability. Thus, the cost of each terminal unit with separate controls and hardware is high while function is low. This lack of controls integration is one of the most substantial impediments to improved occupant comfort. Lack of vision. Perhaps the greatest impediment of all is the profound lack of vision on the part of the HVAC industry as a whole. Although engineers, manufacturers, contractors, developers,
44

and building operators seem to understand the issue of comfort, only a few see value in working to provide better comfort solutions. Many are dismissive of criticisms from occupants, whom they consider to be complainers.
NEW IMPLEMENTATION PATHS

Removing these impediments would require a restructuring of the building-

While buildings may be more comfortable than they were decades ago, it is clear that the industry is not meeting building occupants rising expectations for comfort and control.
construction process, which is unlikely in the short term. However, with both building owners and tenants beginning to view occupants as profit centers, change is comingalthough it likely will occur very differently than most in the industry are anticipating.

Consider that individual-comfort devices that can be employed with underfloor or ceiling air-supply systems or incorporated into work stations are being developed. Consider also that recently developed technologies that allow occupants seated adjacent to one another in open office areas to perceive a difference in thermal comfort of 2-3 F can be integrated into low-cost terminal comfort devices and that these devices can provide individualized comfort billing, just as the phone company provides monthly bills for each office phone. Because these technologies show such great promise, it is reasonable to envision that truly effective personal-comfort devices soon can be connected to standard HVAC systems. Consider for a moment that such products are available. Because they employ specially developed, integrated features designed for particular applications, the market does not offer equals. Cost depends on a devices application focus, which varies from, say, a large individual office to a densely populated open office area. That such products could be implemented in a building intended for multiple tenants through a change in the building-construction process is not likely for several reasons. First, because the products do not have equals, the specifier would lose the ability to control the cost of the products. Second, because building owners often have little idea

January 2001 HPAC Engineering

about how space in a new building will be utilized, the types and number of each type of products would not be known at the time the building was built. It is becoming increasingly clear that a buildings tenantsnot its owner should determine the types of comfort systems to employ in offices. This is important because if a mechanism to market end-use comfort systems directly to tenants were developed, many of the institutional impediments listed above could be avoided. Consider that employers normally spend $8,000 to $12,000 per employee on workspace fixtures and equipment to enable each employee to perform his or her tasks effectively. In this context, the cost of a personal-comfort device does not seem large, especially when one considers that, according to the aforementioned BOMA survey, comfort is regarded as not only the most important aspect of a workspace, but the area most in need of improvement. Marketing directly to tenants would leave value issues to be weighed by employers, who would benefit from the investment in improved comfort for their employees, and eliminate the need for equals, as tenants would appreciate differentiating features and have the ability to negotiate for products best-suited for their particular work environment.

A NEW PARADIGM

Marketing end-use terminal comfort components more or less independently of an HVAC system would not be difficult because, as shown in Figure 1, HVAC systems typically are quite disconnected from the distribution system at each occupants workspace. Furthermore, building projects increasingly have moved the purchase and installation of terminal distribution equipment to the tenant build-outs. Often, a credit is applied, and the tenant is required to adhere to certain building standards regarding the selection and layout of equipment. The ease of implementing individual terminal control is further reinforced by Figure 3, in which an individually controllable terminal comfort device has replaced a standard diffuser at the work station. Such a personal-comfort device could be ceiling-mounted as shown, work-station-based, or part of an underfloor distribution system. Functional individually controllable terminal comfort devices are becoming available for nearly every type of delivery system. Figure 3 shows that the most important network connection for an end-use device is not to the building-control network, but to a network that allows occupants to request comfort adjustments from their personal computers and

enables the exchange of data necessary to provide a record of activity for end-use accounting and/or billing purposes. With the communication standards available today, it would not be difficult to connect the two networks and achieve certain optimization functions. It also would not be difficult to make individual-comfort products completely independent of an existing HVAC system so that such a comfort-system network could be implemented in an older building lacking a modern building-system network. To be cost-effective, an integrated, individually controllable terminal comfort product must enhance employee performance by about onequarter of 1 percent or reduce sick leave by about a half-day per year per employee. Studies generally show both performance gains of 5-15 percent and sick-leave reductions of at least several days.
THE POTENTIAL OF THE COMFORT INDUSTRY

As noted earlier, a recent BOMA survey found that lack of comfort and lack of individual control are building occupants two biggest complaints about their workplaces. Integrated technologies that permit microclimates at each work station through the use of radiant thermal

Building-control network Office c LAN or Internet n Optional connection networks

DDC

DDC CDWR CWS

DDC

C C

Fan

CWR HWS HWR CDWS

DDC DDC

Individually controlled terminal unit

Cond.
DDC

Boiler
DDC

DDC

FIGURE 3. Building- and comfort-control networks.

Chiller

HPAC Engineering January 2001

45

T H E

C O M F O R T

I N D U S T R Y

exchange and local air-movement patterns have been developed and are ready for implementation into more functional individual control products. Such products will provide exactly what occupants desire: an immediate and noticeable (but not excessive) response to requests for thermal changes. Our firm estimates that terminal units with individual comfort controls could be applied for some 20 percent of the approximately 60 billion sq ft of commercial floor space in the United States without modifications being made to the remainder of the building system. This amounts to approximately 50 million workspaces and represents an immediate potential market of more than $50 billion. We estimate that as the focus of HVAC design shifts to comfort, this element of the building-construction industry will attain a size comparable to that of the entire HVAC industry today.

A LOOK TO THE FUTURE

SUMMARY

The future of the HVAC industry will involve not an evolutionary change in the way buildings are designed and constructed, but the development of a parallel comfort industry that is network-based and looks more like the PC and peripherals industry than it does the building-construction industry. This new industry will market to tenants, and products will be installed by office-furnishing installers, not building trades. Those who stand to benefit most from this new industry are manufacturers who build products that integrate state-of-the-art controls with terminal hardware in a single package that is designed to capture the economies of high-volume production. Designers and system integrators will be able to capture value by applying these new products effectively and integrating them into functional control and accounting networks.

A lack of focus on occupant-comfort issues in the HVAC industry has created an enormous gap between terminal systems that are being installed in buildings today and those that could be installed to enhance comfort and provide the individual-control features building occupants want. This gap has created an enormous opportunity that is unlikely to be captured by working within current building-construction processes and procedures. Those who wish to participate in capturing this value must be willing to work to develop alternative implementation paths that include direct or indirect sales to building tenants and the development of comfort systems and networks that can operate relatively independently of building HVAC systems. Enabling procedures and technologies have been developed. The market is ready. Are we?

www.ventprod.com

46

January 2001 HPAC Engineering

C O N T R O L

F R E A K S

PID Control: May It Rest in Peace

Technology is obsolete in the age of digital networks
By THOMAS HARTMAN, PE Principal The Hartman Co. Marysville, Wash.

ID (proportional + integral + derivative) control has been the foundation of building control for several generations, but with the emergence of digital networks, the drawbacks of PID control are becoming more apparent, especially as our industry becomes more concerned with energy efficiency. PID control has persisted as long as it has largely because of a lack of competition (i.e., an alternative that can be universally applied when modulating control is required). That is changing, however, with the recent development of methods of control that take advantage of network integration to improve performance and efficiency. The time has come, then, to bid farewell to PID and focus on new, more-efficient means of control.

DRAWBACKS OF PID CONTROL

PID controls most serious drawback is inefficient operation. Application rules dictate that for effective PID operation, valves and dampers must be undersized so that a substantial pressure drop occurs across them under all operating conditions. When applied to chilled- or heating-water distribution systems, this recommended valve pressure drop can as much as double the pumping power required. This loss is magnified because PID control moves va r i a b l e - s p e e d pumps and fans away from their natural A member of HPAC Engineerings Editorial Advisory Board, Thomas Hartman, PE, is an internationally recognized expert on the use of advanced building controls and c o n t rol network s . Comments and questions can be addressed to him at t o m h @ hartmanco.com.

curves (the curve of the pumps or fans highest efficiency at various speeds), thus reducing operating efficiency while adding substantially higher operatingpressure requirements. As much as threefourths of annual power consumption in some distribution systems can be attributed solely to PID-control losses. The problems of PID control do not end with inefficient operation: Because PID loops operate in isolation from each other, they cannot assure that all loads will be satisfied at any given time. PID control can be costly to implement and support. The goal for any distribution system employing PID control is to have the pressure differential across supply and return headers at each load be constant as flow in the system changes. To accomplish this, designers often ove r s i ze distribution mains or employ reverse return-piping configurations. These pressure-leveling techniques add costs. PID control is a headache for operations staffs because to operate effectively, it requires frequent valve- and damperposition readjustment (1- to 5-sec reposition intervals often are recommended). This nearly continuous repositioning shortens actuator life, adds to maintenance costs, and makes control stability an ongoing issue.
INTERDEPENDENT NETWORK CONTROL

All of these problems can be mitigated or eliminated by replacing independent PID control with interdependent network control. Historically, PID control loops have operated independently of other equipment to maintain a stable pressure or temperature condition that is employed for control by the next system down the line. Network-enabled interdependent c o n t rol combines these subsystems and operates them as a single system. In addition to automatically optimizing operation, effective network control avoids the costly and energy-wasting decoupling required for local PID control. The integrated network control of

HVAC systems is not unlike the control approach employed in the informationtechnology (IT) world to manage Internet services in an office network. Think of the loads in a distribution system as clients and the pump or fan as a server interconnected through a network. The direct-digital-control system acts as a re s o u rce manager, processing client requests and directing the server accordingly. Each load is satisfied while resource expense is minimized. There are several important advantages of applying IT resource-management strategies for HVAC systems. Among them are lower costs, simpler configurations, more-efficient and e f f e c t i ve operation, reduced maintenance, and greater long-term stability and reliability. In implementing these strategies as a replacement for PID c o n t rol in a distribution system, the pressure/temperature-setpoint control of subsystems is replaced with energyoptimized control based on the Equal Marginal Performance Principle (see the authors feature article Ultra-Efficient Cooling With Demand-Based Control, December 2001). Designers understand that the purpose of an HVAC system is not to maintain distribution pressures or temperatures, but to keep all of the people in a building comfortable as efficiently as possible. A network control system designed to connect every occupied zone d i rectly to comfort - re s o u rce systems can do a far better job of maintaining c o m f o rtable conditions in all zones with less energy than can a system in which these components are isolated and operated independently. Achieving such a system, however, requires a significant change in operating strategies. In August, Thomas Hartman, PE, will outline some demand-based-control operat ing strategies and provide some tips for setting up effective network controls for typical distribution systems. For previous Control Freaks columns, visit www.hpac.com.
9

HPAC Engineering July 2002

C O N T R O L

F R E A K S

Out With the Old, In With the New

Replacing PID control with demand-based control
By THOMAS HARTMAN, PE Principal The Hartman Co. Marysville, Wash.

ast month, I discussed cost- and e n e r g y - related drawbacks of continuing to employ PID (proportional + integral + derivative) control in HVAC systems and introduced the concept of interdependent network control, which can replace PID control to improve the performance and efficiency of many systems. Interdependent control employs resource-management techniques similar to those used in information-technology networks. This month, I will provide a brief overview of a demand-based-control application to show how temperature- and pressuresetpoint control can be eliminated while the performance and efficiency of an a l l - variable-speed HVAC system are improved.

ALL-VARIABLE-SPEED HVAC SYSTEMS

Applying variable-speed AC drives to HVAC components with built-in extra capacity saves energy at all expected load conditions. For this energy saving to be maximized, pressure differentials must change as the load conditions change. This is why temperature- and pressuresetpoint reset commonly is used as an optimization technique in conventional control systems.
DEMAND-BASED CONTROL

matically optimizes overall system operation, thus eliminating the setpoint-optimization step. Demand-based control also reduces construction, energy, and operation-and-maintenance costs by simplifying system configurations, reducing pressure losses at all load conditions, and eliminating much of the wear and tear on modulating components. Imagine designing the controls for a HVAC system in which all cooling-tower fans, condenser-water pumps, chillers, chilled-water distribution pumps, and supply fans employ electric variablespeed operation. Because of mechanicaland electrical-efficiency considerations, each of these components operates most efficiently within a capacity span that varies for each system element and may depend on more than a single operating parameter. Demand-based control recognizes that the operation of each piece of equipment influences the operation of eve ry other piece of equipment in a system. In a demand-based-control system, network controls coordinate the operation of all equipment as a single system to meet all of the cooling loads effectively and efficiently.
Demand-based control at the zone level.

cooling demand by adjusting the speed of the fan(s) and/or the flow of chilled water through the cooling coil. Supplyair temperature and pre s s u re are not directly controlled, acting only as limiting values to ensure equipment operates within manufacturer and system design parameters. This approach offers a great advantage in meeting zone loads. If one or more zones is not obtaining adequate cooling effect, the system determines whether it would be more efficient to increase airflow or reduce air temperature. If the system chooses to reduce air t e m p e r a t u re, it will slightly incre a s e chilled-water flow to the air-handlingunit (AHU) cooling coil. VAV zones that already are satisfied will automatically reduce their airflow, resulting in both an increased air volume and lower-temperature air being immediately available to the zones that require additional cooling. Such control substantially improves system re s p o n s i veness and operating efficiency. The readjustment takes place about once a minute.
CHILLED-WATER-DISTRIBUTION DESIGN

Demand-based control is a networkenabled method of control that modulates equipment to meet current loading conditions and, at the same time, autoA member of HPAC Engineerings Editorial Ad v i s o ry Board , Thomas Ha rt m a n , PE, is an internationally recognized expert on the use of advanced building controls and control networks. Contact him at t o m h @ hartmanco.com.

Variable-air-volume (VAV) boxes in a demand-based-control system employ cooling-effect operation rather than just airflow control. Cooling effect is a combination of airflow and temperature effect and is calculated approximately once a minute in each VAV zone. The calculation considers airflow, supply-air temperature, and zone temperature. The cooling effect is compared to the deviation from the space-temperature setpoint in each zone, and a damper-position adjustment is made to increase or decrease the cooling effect as required in each zone. A self-balancing calculation also may be included so that the coolingeffect parameters for each zone are automatically adjusted. Each zone communicates certain information to the fan system serving it.
Demand-based control at the fan-system level. With demand-based control, a fan

system can respond to changes in zone-

Adjusting chilled-water valves once a minute simplifies control and makes control stability much less of a problem than it is with PID control. The lessstringent incremental requirements of demand-based control afford designers an opportunity to improve efficiency. Valves can be sized for lower pressure losses, as linear response to actuation is not required. Furthermore, with network-management techniques, there is no requirement that each valve operate independently of the others. Demandbased-control valve-selection rules call for a full flow pressure drop of 1 to 2 ft of head in worst-case configurations. The rules also require that some method of automatic coil-overflow protection be provided. This feature may be incorporated into the va l ve itself or the AHU controls. Simple ball or butterfly modulating valves work well in demandbased-control applications. Modulating valve actuators should be slow-acting. Although 120- to 300-sec actuators are
9

HPAC Engineering August 2002

C O N T R O L

F R E A K S

ideal, faster-acting actuators will work fine as long as their total movement during each adjustment period is limited. With demand-based control, there is no need to have identical pressure differentials at each load, so reverse return configurations are not necessary. Pipes should be sized to keep velocities and p re s s u re losses within recommended limits, with some system analysis p e rformed to select the distribution pump(s). Slightly oversizing the pumping capacity in a variable-speed system carries no energy penalty; however, to maintain good electrical efficiency at lower flows, pump motors should be sized as small as possible. Operating the pump motor into its service factor at peak conditions is what the factor is designed for. There is no concern about motor overload because the variablefrequency drive can automatically limit the power to its rated maximum.
DISTRIBUTION-PUMP CONTROL

no direct temperature- or pre s s u re setpoint control at the fan or pumping system. This frees the system to respond to changes in cooling requirements with simple network-management techniques that automatically optimize the overall system efficiency. When demand-based control is ap-

plied to fan systems and chillers, as well as to chilled-water distribution systems, it has the ability to drastically improve the c o m f o rt of buildings and cut total HVAC electricity use in half. For previous Control Freaks columns, visit www.hpac.com.

There are several methods of applying demand-based control to chilled-waterdistribution pumping systems. The one I p refer is the orifice-area method, by which pump speed is regulated according to the percentage of total valve orifice area opened. For ball and butterfly valves, fractional valve opening can be calculated by raising the fractional actuator position to a power of about 2.7. In single-pump systems, multiplying this fractional opening by the valves full flow as a percentage of total-system maximum flow and summing the value for all valves is used to directly set pump speed as a percentage of the maximum speed when the pump is sized to meet full-flow requirements. The advantage of this technique is that it allows the pump to operate at its highest efficiency at all flows. Experience shows that making this valvearea calculation and using it to set pump speed once every minute works very well regardless of the size of the pump and distribution system. When multiple pumps are employed, it is not difficult to determine the optimum orifice area at which a pump should be added or subtracted and to incorporate these switching points into the pump control.
ELIMINATION OF SETPOINTS

With demand-based control, there is

Circle 309 HPAC Engineering August 2002 11

ULTRA-EFFICIENT

COOLING
with Demand-Based Control
The value and methods of applying direct-coupled network control to building-energy-system design
magine operating a chilled-water plant with- decouple) one system element from another. out controlling the chilled-water temperature Figure 1 shows basic control modules for a typical and operating a variable-flow chilled-water H VAC system. The cooling towers, chillers, distribution network without a differential-pressure distribution pumps, and supply fans are controlled setpoint. Why would you operate a independently with temperature or cooling system this way? The rea- By THOMAS HARTMAN, PE pressure setpoints that ensure the sons may surprise you. First, operatThe Hartman Co. surrounding equipment also can ing plants without directly controlMarysville, Wash. operate independently over a wide ling the temperature or pressure of range of loading re q u i rements. their outputs can be much more efficient. Second, Although a network-capable direct-digital-control temperature and pressure control is considerably (DDC) system may be employed for control, the more complex and less stable than are control network typically is used only to collect informastrategies that combine more direct control and tion for operations. In many systems, additional optimization. isolation is provided with primary - s e c o n d a ry Through a new approach to HVAC control pumping, bypasses, decoupling lines, and valves or called demand-based control, building systems dampers that have large pressure drops. are operated using the network capacity of modern building control systems. Combining variableDirect-coupled control can speed-drive equipment with network-enabled demand-based-control technologies can make reduce the size and simpler, smaller, and lower-cost building energy systems operate as much as 30- to 50-percent more complexity of both controls efficiently than conventional system configurations with the same basic HVAC-component efficiencies can. Also, demand-based control enhances the and system components. comfort of buildings and provides a platform for individual control and other valuable occupant amenities. This focus on independent equipment operation wastes energy. Normally, the chillers in Figure 1 HOW CONVENTIONAL CONTROL WASTES ENERGY would be operated at a fixed chilled-water temperaConventional HVAC controls employ pressure- ture. At low-load conditions, the chiller compressor or temperature-setpoint control to isolate (or would operate at higher-than-needed head (and reduced efficiency) to provide colder-than-required chilled water, which the distribution pump would A member of H PAC En g i n e e r i n gs Editorial distribute at a higher-than-necessary pressure. To Ad v i s o ry Board, Thomas Ha rtman, PE, is an ensure stable and independent chilled-water coil i n t e rnationally re c o g n i zed expert on the use of operation under these conditions, the valves would advanced building controls and control networks. be selected for high pressure drops. All of these Comments and questions can be addressed to him at equipment-isolation measures would reduce overall tomh@hartmanco.com. system efficiency.

HPAC Engineering December 2001

29

E M A N D -

A S E D

C O N T R O

when the operation of all components is coordinated to meet actual needs in the spaces served. Demand-based control is intended to fill the vacuum in controls technologies created by the development of variablef requency drives (VFDs) for HVAC equipment. Prior to the introduction of VFDs, coordinating the operation of HVAC equipment mattered little because equipment efficiency remained almost constant over ranges in loading. Now, with VFD modulation, the efficiency of HVAC components can change dramatically as load conditions vary. Although coordinated operation is essential to maximizing overall energy efficiency today, most conventional control schemes operate VFD equipment as they do mechanically modulated equipment. Demand-based control can solve this problem; however, applying it effectively requires new thinking about h ow VFD equipment operates most efficiently.
HOW DEMAND-BASED CONTROL WORKS

FIGURE 1. A conventional HVAC system in which each piece of equipment is operated with an independent control loop that maintains a temperature or pressure setpoint regardless of the end-use requirements.

If current HVAC-control practices were applied to automobile operation, drivers would be taught to control the accelerator to maintain a fixed engine rpm, operate the clutch to provide a specific torque, and use the brake to control the vehicles speed. Consider what this would do to a cars average miles per gallon! Traditional optimization approaches cannot overcome the built-in inefficiencies of conventional HVAC control. Instead, they typically create one or more additional layers of software that reset the various setpoints to marginally improve operating efficiency as conditions change. These optimization approaches add complexity to the controls and are limited in their ability to reduce energy consumption. Direct-coupled control, on the other hand, can reduce the size and complexity of both controls and system components. However, a basic change in how
30

the system is operated must be made first.

DEMAND-BASED CONTROL: A NETWORK-CONTROL SOLUTION

Demand-based control is a method of applying direct-coupled network control. It is based on the idea that a building HVAC system is a single system the energy efficiency and comfort p e rformance of which are optimize d

With conventional control strategies, little if any information concerning upstream or downstream loading/operating conditions is employed to adjust the operation of equipment. Ty p i c a l l y, HVAC components operate to maintain a single temperature or pressure setpoint. If a number of spaces in a building were to begin to overheat, most central systems would not self-adjust to provide more cooling and, once the spaces were satisfied, readjust to meet the reduced load with greater efficiency. Accomplishing this would require a network-control scheme that communicates with the loads being served. Network-enabled demand-based control is very cost-effective because, in most

If current HVAC-control practices were applied to automobile operation, drivers would control the accelerator to maintain a fixed engine rpm, operate the clutch to provide a specific torque, and use the brake to control speed.

December 2001 HPAC Engineering

E M A N D -

B A S E D

C O N T R O

With variable-speed equipment and network-control capabilities, the long-standing dictum that equipment must be decoupled to operate effectively has been reversed.
cases, it does not re q u i re additional equipment or controls. In fact, when properly applied, demand-based control requires less equipment and often employs a simpler configuration than the c o n ventional system and control it replaces. Although demand-based control ties the operation of all equipment to end-use re q u i rementsactual space requirements in single-building HVAC applicationsthis does not mean that chillers and cooling towers operate directly from space-temperature sensors. R a t h e r, as in automobile operation, demand-based control optimizes an HVAC configuration by directly coupling the operation of all components so they operate as a single system in meeting the actual needs of spaces. This single-system approach generally is not considered because HVAC designers a re too used to ensuring that their designs isolate the operation of HVAC components to give an alternative a thought. For example, while low delta T is a serious problem in many chiller plants, designers continue to employ decoupling or bypass lines that permit d i rect mixing of supply and re t u r n chilled water. With variable-speed equipment and network-control capabilities, the longstanding dictum that equipment must be decoupled to operate effectively has been reversed. Direct coupling leads to simpler, more-efficient operation. It is intuitive that coordinating the operation of a chiller plant and chilled-water distribution network is re q u i red to a c h i e ve the highest overall coolingsystem efficiency. The question is how does one control these together so they can be operated most efficiently. The answer is to consider all equipment involved in cooling a building as a single system instead of a series of systems. When cooling needs to be adjusted in response to space conditions, demand-based control coordinates the operation of all elements to provide cooling where it is needed according to predefined efficiency relationships.

We know that coordinating the operation of two identical pumps in a direct-coupled parallel or series circuit based on power or speed is the most straightforward method of optimizing the pumps overall efficiency. But what if the pumps (or other type of equipment) are not identical? In a similar fashion, one or more power-based relationships could be developed using the Equal Marginal Performance Principle (see sidebar), which would optimize the pumps operation under all loading conditions. A circuit consisting of cooling towers, chillers, pumps, and conditioning fans with variable-speed drives could be optimized in this manner. Like

Equal Marginal Performance

ormulated more than a decade ago, when variable-speed drives were first applied to pumps and fans, the Equal Marginal Performance Principle (EMPP) states that the operation of a system comprised of multiple modulating components (in series or parallel) is optimized when the marginal system output divided by the marginal system input is the same for all components. To better visualize how the EMPP is applied to an entire cooling system, consider the schematic below.

Imagine that the knobs below each piece of equipment in this schematic adjust the capacity of that element by changing its speed or by some other means. Assume that instrumentation for measuring the system output and system input is provided. How would you optimize this system at its current point of operation? According to the EMPP, it could be done by making small adjustments to the capacity of each element and noting the changes in system output and input. Then, system efficiency could be improved by: (1) reducing the capacity setting of elements that show relatively small marginal-system capacity change per unit input change and (2) increasing the capacity setting of elements that show larger changes in capacity per unit power change so that the total system output remains at its current point of operation. These processes of testing each element and resetting the system would be repeated until all elements had exactly the same marginal output per unit change in power input. At that point, the system would be optimized.

HPAC Engineering December 2001

31

E M A N D -

B A S E D

C O N T R O

comfortable building and a higher level of indoor environmental quality.

CASE STUDY

FIGURE 2. A comparison of the annual energy use with conventional optimized HVAC systems and equipment of the same basic configuration and efficiencies except operated with demand-based control of a new mid-rise office building in Denver.

dissimilar pumps, these components could be directly coupled and controlled using demand-based control to operate the circuit as a single system and provide the cooling capacity re q u i red. Such c o o rdinated control is the simplest method of obtaining the highest overall operating efficiency. Thus, with demand-based control, equipment is c o o rdinated to operate according to p ower (kilowatt) setpoints, which is simpler, more stable, and much more efficient than the use of temperature or pressure setpoints is. Furthermore, because demand-based control is network-based, enhancing system performance by making system adjustments that focus on meeting exceptional spaceconditioning requirements that occur from time to time is very easy.
ENERGY-REDUCTION IMPLICATIONS OF DEMAND-BASED CONTROL

designed to use less than 50,000 Btu per sq ft annually. Note that despite the initial low-energy design, the networkbased control further reduced the electric-energy use of the chiller plant and HVAC distribution system by nearly 50 percent. At the same time, the networked control resulted in a more

Bellevue Corporate Plaza is a multitenant mid-rise office building in Bellevue, Wash. It is an all-electric building that employs va r i a b l e - a i r volume (VAV) air systems with fanpowered perimeter boxes and electric reheat. The chiller plant consists of two air-cooled centrifugal chillers and a constant-flow primary-only distribution system. In 1993, an integrated lighting and terminal-re g u l a t e d - a i r - vo l u m e (TRAV) controls retrofit was ordered. T R AV was the first demand-basedc o n t rol strategy developed, and its application reduced total building energy use from about 80,000 Btu per sq ft annually to about 50,000 Btu per sq ft. Seven years later, it was time to upgrade the chiller plant. The plant employed a phased-out refrigerant, and the chillers were near the end of their useful life, as evidenced by a dramatic increase in maintenance and numerous failure s .

By directly coupling HVAC components into a single system with network communication and acting on end-use requirements in setting system capacity, demand-based-control strategies can improve building comfort, environmental quality, and energy perf o r m a n c e . While the assessment of comfort and environmental quality usually is subjective, the energy savings attributable to demand-based control can be accurately determined through hourly simulation. The simulation results for the electric portion of the HVAC system in a new mid-rise office building in Denver are shown in Figure 2. The building was
32

FIGURE 3. The new Bellevue Corporate Plaza cooling system. The Loop demand-basedcontrol network is shown schematically. Its purpose is to connect equipment to the network for coordinated operation rather than to maintain temperature or pressure setpoints.

December 2001 HPAC Engineering

E M A N D -

B A S E D

C O N T R O

Fu rt h e r m o re, the plant lacked the capacity to meet growing tenant internal heat loads at peak conditions. After exploring several replacement options, the building owner, Hallwood Commercial Real Estate, adopted an upgrade approach that employed the n ew Loop demand-based-contro l -

sequence package. The Hartman Co. helped implement Loop ultra-efficient demand-basedcontrol sequences and bring the new system on line. The Loop control sequences were installed in the buildings existing DDC system without the need for additional control equipment, save

some input/output devices. The resulting cooling system provided additional capacity with a lower design chilledwater temperature, allowing the existing cooling coils to be used. The Loop demand-based-contro l technologies significantly improved the chiller plants operating efficiency by using the control network to coordinate cooling supply with the actual cooling demand. All balancing va l ves we re removed, and the three-way modulatingflow control valves on the cooling coils

Direct coupling leads to simpler, more-efficient operation.

we re changed to two-way line-size d modulating valves. Variable-speed drives were added to the pumps. The distribution system was reconfigured to variable primary flow without a bypass valve. The chiller-plant components are now controlled at optimal levels of relative p ower use rather than to maintain a specific chilled-water temperature or distribution-pressure setpoint. Whenever possible, when cooling is called for, the chiller plant, distribution pumps, and fans are optimally operated together. Under normal operations, there is no direct control of chilled-water temperature or distribution pressure. At low-load conditions, chilled-water temperature floats upward, and pump-head requirements for the distribution system fall, resulting in very stable operation and low power requirements. Even at full loading, the chilled-water temperature often remains above the design minimum, and pumping head is only half of what is was before the retrofit. The operation of the chillers, distribution pumps, and main supply/return fans is coordinated for the highest overall cooling and distribution efficiency at all times. This automatic network optimization is especially useful during periods of demand limiting. In conventional systems, demand limiting may be applied only to chillers. The chilled-waterdistribution pumps and air-distribution fans speed up during demand limiting
34 Circle 000 December 2001 HPAC Engineering

E M A N D -

B A S E D

C O N T R O

because their loads are not met when the chillers reduce their loading. This results in greater power use by pumps and fans, which reduces the demandlimiting function of the chillers. At Bellevue Corporate Plaza, the Loop demand-based-control sequences coordinate the operation of all equipment and provide more - e f f e c t i ve demand limiting with less impact on occupants because system efficiency increases during demand limiting so that the cooling effect remains high. Despite a larger cooling system and lower design operating temperatures, the buildings summer peak electrical demand has been reduced by about 20 percent. And as a result of extending demand-based control to the cooling system, the buildings daily energy use

strategies are such that this change in building-control technology should be considered as part of national energy policies. When demand-based c o n t rol is installed throughout most typical HVAC systems, one can expect a 30- to 50-percent annual electricalenergy reduction. Equipment configura-

tions require only slight modifications to be operated in accordance with demandbased control in place of conventional o p t i m i zed control. Fu rt h e r m o re, a system specifically configured for demand-based control can be less expensive to design, install, and maintain than a conventional system.

When demand-based control is installed throughout most typical HVAC systems, one can expect a 30- to 50percent annual electrical-energy reduction.
during cooling-system operation has been reduced by about 22 percent. The building is using 50-centsper-square-foot less energy annually than it would be if the demand-based-control program, which began with the TRAV retrofit, had not been undertaken. In addition, the buildings occupants report being much more comfortable.
SUMMARY AND CONCLUSION

Energy efficiency and comfort can be i m p roved substantially by applying network-enabled demand-based-control strategies to operate building HVAC equipment. The enormous electricenergy reductions possible from the application of demand-based-control
Circle 000 HPAC Engineering December 2001 35

Presenting Intelligent Iterative Control: PID Replacement for Setpoint Control

New approach makes better use of digital controls
Editors note: This is the first part in a three-part series.
By THOMAS HARTMAN, PE Principal The Hartman Co. Marysville, Wash.
AO FC1CCV V AI FC1DAT T
1

n a December 2001 article, a new network control approach that can be used in place of PID control for the operation of much of the equipment that constitutes a buildings HVAC system was introduced. Called demandbased control, it avoids the extra energy requirements, control-stability problems, and frequent repositioning of PID control. Furthermore, it is simpler to apply because it provides stable, optimized control in a single step, while PID control requires two separate steps: (1) setting up a control loop to maintain a setpoint and (2) continuously adjusting the setpoint to optimize operation. Using the equal-marginal-performance principle, the 2001 article showed that demand-based control enables stable, optimized control of building-comfort-system components in a single step and does not require any of the energywasting design rules that PID control does. That is why energy costs with demand-based control are 30 to 50 percent lower than they are with PID control.

DO FC1SS

DI FC1

FIGURE 1. A simple cooling-only fan coil.

However, demand-based control typically uses direct power relationships, rather than temperature and pressure setpoints, to control equipment. In some instances, temperature- or pressuresetpoint control is required to achieve effective HVAC-system operation. Some consider this to be a reason to continue using PID control. But with the availability of digital network controls, there is a far more efficient and effective alternative.
REPLACING PID CONTROL WITH INTELLIGENT ITERATIVE CONTROL

A member of HPAC Engineerings Editorial Advisory Board, Thomas Hartman, PE, is an internationally recognized expert on the use of advanced building controls and control networks. Contact him at tomh@ hartmanco.com.

When temperature- or pressuresetpoint control is necessary, designers should consider new approaches that make better use of modern networked digital controls before relying on PID control. As indicated earlier, the use of PID control usually wastes energy by requiring 25 percent to more than 90 percent of the fluid total dynamic head to be dissipated through the control valve under various operating conditions. This

makes the power requirements of the system far greater than should be necessary to distribute air or water to the loads served. Furthermore, a number of other factors in modern HVAC systems, including the application of optimization setpoint reset, cause PID control loops to be non-proportional much of the time. This requires modulation readjustments at short intervals, which reduce equipment life and add to maintenance costs. For applications requiring temperature- or pressure-setpoint control, a new approach called intelligent iterative control can avoid these PID pitfalls, providing more efficient and stable systems with longer-lasting devices requiring less maintenance. Intelligent iterative control is a method of control based on the iterative problem-solving techniques used by computers since the beginning of digital processing. To show how an intelligent-iterative-control procedure can be developed to operate more efficiently and effectively than PID control in a temperature-setpoint-control appli13

HPAC Engineering September 2003

cation, consider the simple cooling-only fan coil depicted in Figure 1. The fan coil in Figure 1 serves a space the temperature of which is monitored with a space-temperature sensor. To ensure even comfort, control of the fan coil involves maintenance of a discharge-airtemperature setpoint in response to space-temperature conditions. The control we are interested in for this discussion is that of the cooling coil in maintaining a specific fan-coil discharge-airtemperature setpoint. If conventional PID control were to be used, the designer would face several problems. First, conventions dictate that the cooling-coil control valve be sized for a substantial pressure drop at full flow (the ASHRAE 2000 HVAC Systems and Equipment Handbook recommends the pressure drop across the valve at full flow be 25 to 50 percent of the total pressure drop of the chilled-water distribution

network). This radical undersizing of control valves adds dramatically to pump size and cost and, just as troublesome, long-term pumping-energy cost. Also, to maintain stable PID control, the valve may have to be repositioned every several seconds, reducing its life and adding to maintenance costs. These problems can be avoided by using an iterative-control technique that marshals the computing and networking capacity of modern DDC systems more effectively. Instead of PID control, imagine the incorporation of line-sized and full-ported control valves on all of the fan coils in an effort to reduce pump size and the pumping-energy requirements of the chilled-water distribution system. We then could adjust pump speed using demand-based control and the valve-orifice-area method outlined in my August 2002 column.2 That approach would reduce the pumping power from fixed

pressure-setpoint control. The question then becomes how can line-sized valves be operated to provide stable and precise control to maintain the fan-coil discharge-air-temperature setpoint. Next month, Part 2 in this series will discuss the implementation of an intelligent-iterative-control algorithm.
REFERENCES

1) Hartman, T. (2001, December). Ultra-efficient cooling with demandbased control. HPAC Engineering, pp. 29-32, 34, 35. 2) Hartman, T. (2002, August). Out with the old, in with the new. HPAC Engineering, pp. 9, 11. For previous Control Freaks columns or to visit the Networked Controls microsite, go to www.hpac.com. Send comments and suggestions to controlfreaks@penton.com.

Circle 209
14 September 2003 HPAC Engineering

Presenting Intelligent Iterative Control: PID Replacement for Setpoint Control

Implementing an intelligent-iterative-control algorithm
Editors note: This is the second part in a three-part series.
By THOMAS HARTMAN, PE Principal The Hartman Co. Marysville, Wash.
AO FC1CCV V AI FC1DAT T C

hough widely used to solve complex problems, iterative logic is well-suited to providing modulation control. Indeed, the iterative process can be easily applied to control applications. For the fan-coil-valve-control application in Figure 1, assume the unit is operating, but the discharge-air temperature is deviating from the setpoint value. The iterative process of control involves: (1) estimating the change in valve position required to achieve the desired change in discharge-air temperature, (2) making the change, (3) waiting to see how close to the desired temperature the result is, and (4) adjusting the estimation logic (if necessary) and repeating the process. The process becomes intelligent when the estimate is enhanced by logic and real-time information derived from other elements of the fan coil and/or related systems. With an iterative-control scheme, control for the cooling valve in Figure 1 (FC1CCV) involves estimating and implementing a capacity-adjustment

DO FC1SS

DI FC1

FIGURE 1. A simple cooling-only fan coil.

A member of HPAC Engineerings Editorial Advisory Board, Thomas Hartman, PE, is an internationally recognized expert on the use of advanced building controls and control networks. Contact him at tomh@ hartmanco.com.

algorithm intended to bring the loop to its current setpoint. Because this is an iterative process, the algorithm employed to make the estimate does not need to be precise; however, it should be as accurate as possible and contain the primary data points needed to affect the desired change in valve position, along with a factor that correlates the relative importance of each. Among the substantial benefits of applying intelligent iterative control are that the intervals between valve repositionings can be much longer than those typically employed with PID control, and there is no need for interval times to be fixed. This has the potential to greatly reduce the frequency of required valve repositioning and, thus, extend valve and actuator life. From an energy-performance perspective, a significant benefit of replacing PID control with intelligent iterative control is system head pressure can be reduced dramatically at all load conditions because valves can be line-sized and

full-ported. With intelligent iterative control, there is no requirement for a linear response between valve actuation and the cooling effect it produces. To achieve the linear relationship required for PID control, valves must be undersized with significant pressure drops, while controllability requires only that the relationship between actuation and coil cooling output be a continuous curve and that the slope be positive at all times. As shown in Figure 2, these conditions are not difficult to meet with line-sized valves. A linear relationship as shown by the red line in Figure 2 can be established when the valve is undersized and consumes 25 to 50 percent of the total system head-pressure drop. Though not offering a linear relationship between actuation and coil cooling capacity, the line-sized valve in Figure 2 does meet the requirements for controllability. The iterative logic used to effectively control a line-sized valve such as the one in Figure 2 is based on information coming from the fan coil and across the
9

HPAC Engineering October 2003

Do every 30 sec :
100

A = C2 1.0 + C1 PumpRPM 0.3

80 Valve actuation, percent

(FC1DAT FC1DATSP) (0.2 + FC1CCV )

If absolute value B > 15% 0.15 , then FC1CCVnew = A + FC1CCV and B = 0; otherwise, B = B + A

( (

)))

()

( )

60

40

20

where: A and B = intermediate variables used in this algorithm only This iterative process compensates for the fact that changes in pump speed may cause changes in pressure across the valve and, thus, change the valve movement necessary to attain the same change in cooling effect at different pump speeds. This valve-control algorithm is simpler than many of the PID algorithms now in use. With some experience, intelligentiterative-control algorithms are relatively easy to sketch out during design and can be easily inserted into sequences of operation. Initial values for C1, C2, and C3 usually can be estimated based on coil characteristics and the location of the fan coil in the distribution circuit. If a fan-coil supply fan is variable-speed, and a variable chilled-water-supply temperature is incorporated, additional factors can be implemented easily. Sometimes, it is helpful to incorporate into iterative algorithms self-learning, whereby constants are automatically set and adjusted to reduce startup-time requirements and the frequency of repositioning. Lastly, in critical systems, it sometimes is useful to adjust valve position for system-dynamics features, in addition to marginal coolingload requirements. Such adjustments may be made independently of valveposition changes attributed to changes in load requirements. All of these controls are surprisingly easy to set up, as long as the digital controller has flexible controlprogramming capabilities. Next month, the third and final part in this series will discuss limitations in designing with line-sized chilled-water valves and benefits of intelligent iterative control.

0 0 10 20 30 40 50 60 70 80 Cooling-coil capacity, percent of design maximum 90 100

FIGURE 2. Actuation vs. coil capacity for undersized valve used in PID control (red line) and line-sized valve in intelligent-iterative-control application (green line).

network from other system elements. During the design process, it is useful for designers to learn what factors are involved for each modulating control loop. A reasonably effective approximation of the incremental cooling effect required to meet a change in load is not difficult to approximate. When a line-sized valve is employed, and the chilled-water temperature is fixed for the application shown in Figure 1, the iterative-control algorithm often can be:

(C2 (1.0 + C1 (PumpRPM 0.3)))

FC1CCVnew = FC1CCV +

(FC1DAT FC1DATSP) (C3 + FC1CCV )

where: FC1CCVnew = the estimated new valve position required to meet the load FC1CCV = the current valve position C2 = a constant that depends on the relative capacity of the cooling coil and the chilled-water lines that connect it. Initially, it can be set at 0.05. From there, it generally is adjusted between 0.02 and 0.10, with the lower values ensuring less potential for hunting and the higher ones ensuring faster response C1 = a constant between 0.0 and about 20 that is adjusted depending on the proximity of each valve to the pump (C1 should be 0.0 for the load farthest
10

from the pump) PumpRPM = the speed of the distribution pump (100-percent speed = 1.0) FC1DAT = the current fan-coil discharge-air temperature in degrees Fahrenheit FC1DATSP = the current fan-coil discharge-air-temperature setpoint in degrees Fahrenheit C3 = a constant developed to compensate for the slope of coil capacity vs. the actuation curve in Figure 2 as a function of valve position. The value of C3, which must be greater than zero, generally ranges between 0.1 and 0.5 Note that this algorithm calculates valve position as a fractional value (0.0 to 1.0) and uses pump speed as a fractional value. Some control systems provide these percentages as whole numbers (0 to 100) or other values. Constant and variable values may need to be scaled to be compatible with the control system employed. As discussed earlier, one of the purposes of this type of control is the minimization of valve repositioning. For the fan-coil application shown in Figure 1, the minimum valve-repositioning interval typically is 30 sec. Following is the final form of a useful control algorithm for the fan coil in Figure 1:

October 2003 HPAC Engineering

Presenting Intelligent Iterative Control: PID Replacement for Setpoint Control

Limits of designing with line-sized chilled-water valves
Editors note: This is the third part in a three-part series.
By THOMAS HARTMAN, PE Principal The Hartman Co. Georgetown, Texas
Chiller C Load 1-n C Load 1-2 C Load 1-1 Chiller C Load 2-1 Chiller C Load 2-2 C Load 2-n C Load 4-1 C Load 4-2 C Load 4-n C Load 6-1 C Load 6-2 C Load 6-n C Load 3-n C Load 3-2 C Load 3-1 C Load 5-n C Load 5-2 C Load 5-1

lthough designing chilledwater distribution systems with line-sized control valves reduces pump-head requirements and, thus, reduces both first costs and energy costs, there are restrictions and limitations when such a design path is taken. One of the most critical of these concerns the pressure differential across valves. I recommend that systems be designed so that the pressure across a control valve does not exceed double the design pressure drop through the load it serves. This means that if a cooling coil is rated for a pressure drop of 10 ft at full flow, the distribution system should be designed so that the valve serving the coil has a differential that remains below about 20 ft. As an example, consider the variable-primary-flow distribution system shown in Figure 3. In this centralized pumping scheme, chilled water is pumped through the chillers and into the distribution network. If the distribution system is large, then loads near the chillers (such as loads

FIGURE 3. A variable-primary-flow distribution system.

1-1 and 2-1) may be subjected to large supply-header and very low returnheader pressures at high-flow conditions. These high pressures may be necessary
C Load 1-n C Load 1-2 Chiller C Load 1-1

to transport water to the farthest reaches of the system. Thus, if iterative control with line-sized valves were employed, the loads close to the chillers could become
C Load 3-n C Load 3-2 C Load 3-1 C Load 5-n C Load 5-2 C Load 5-1

Chiller

A member of HPAC Engineerings Editorial Advisory Board, Thomas Hartman, PE, is an internationally recognized expert on the use of advanced building controls and control networks. Contact him at tomh@ hartmanco.com.

Chiller

C Load 2-1 C Load 2-2 C Load 2-n

C Load 4-1 C Load 4-2 C Load 4-n

C Load 6-1 C Load 6-2 C Load 6-n

FIGURE 4. Alternative configuration using distributed pumps.

HPAC Engineering November 2003 9

uncontrollable at high-flow conditions. To avoid this potential condition, an alternative configuration using distributed pumps (Figure 4) should be considered. In this pumping scheme, pumps are

in Figure 4 is more energy-efficient, in addition to being more controllable at all flows. Readers may notice that the variableprimary-flow configurations in figures 3

BENEFITS OF INTELLIGENT ITERATIVE CONTROL

When coupled with demand-based control in plant and distribution systems, this control is far more effective, with the opportunity to ensure that all loads are satisfied all of the time.
placed at the distribution mains, rather than centralized at the plant. In other configurations, it may be advantageous to employ a primary-booster pumping scheme. Because the pressures across the valves in Figure 4 are lower (on average) than those in Figure 3, the configuration and 4 do not employ a bypass valve for low-flow conditions. When intelligent iterative control is employed, a bypass valve usually is unnecessary, as the intelligent, network-based valve control can ensure a minimum level of flow through the loads at all times a chiller is operating.

The benefits of replacing PID control with the type of iterative-control scheme described in this series of columns have been shown to include a reduction in distribution-pumping-energy requirements to less than one-half of conventional distribution-energy requirements. Furthermore, average valve-repositioning frequency can be reduced from one-tenth to a hundredth of what is required conventionally. In addition, much greater control stability under nearly all operating conditions usually is achievable. Finally, when coupled with demand-based control in plant and distribution systems, this control is far more effective, with the opportunity to ensure that all loads are satisfied all of the time. For previous Control Freaks columns, visit www.hpac.com.

Circle 161
10 November 2003 HPAC Engineering

DIRECT NETWORK

CONNECTION
of Variable-Speed Drives
Achieving the highest levels of system performance and efficiency through reliance on control networks
n recent years, there has been a trend toward system when operated with variable speed. using variable speed to operate motors for An obstacle on the road to variable speed has equipment and systems that provide heating, been the cost associated with variable-frequency ventilation, and air conditioning for buildings. drives, which are the most popular and efficient Variable speed has been shown to method of applying variable speed improve the operating efficiency By THOMAS HARTMAN, PE to AC motors. In recent years, howThe Hartman Co. ever, that cost has fallen dramatiof equipment such as fans, chillers, Marysville, Wash. cally. For the most popular sizes of and pumps, the capacity of which motors for HVAC fan and pump must be modulated for proper system operation. Furthermore, designers now applications, variable-frequency-drive cost is understand that motors powering devices tradition- approaching $50 per horsepower. When all costs ally operated at constant speed, such as fan coils and are considered, variable speed is now much less fan-powered terminal units, can benefit enor- costly than two-speed-control alternatives. In mously in terms of both performance and efficiency addition, the soft start from variable speed adds from the application of variable speed. It is becom- enormous life to belts and other drive components. One of the largely untapped cost-saving features ing clear that virtually every motor involved in a building HVAC system today has the capacity of variable-speed drives is built-in intelligence. to improve the efficiency and effectiveness of the The internal control logic of variable-speed drives is microprocessor-driven. Most drives have capacity to contribute to building High-speed building network control systems with networking services and processing capabilities. It is possible to envision a system in which VFD VFD DDC DDC DDC much of the control sequencing takes panel panel panel VFD VFD VFD place in intelligent end devices conVFD VFD VFD nected to the system, such as variablespeed drives, rather than in dedicated AHU AHU VFD VFD control panels of the building control system, where it takes place now. RF RF Unfortunately, control-sequence capabilities from separate components are not easily integrated into building FIGURE 1. Preferred direct network connection of variable-speed drives.

A member of HPAC Engineerings Editorial Advisory Board and a regular contributor to the magazines Control Freaks column, Thomas Hartman, PE, is an internationally recognized expert on the use of advanced building controls and control networks. Comments and questions can be addressed to him at tomh@ hartmanco.com.
22 March 2003 HPAC Engineering

D I R E C T

N E T W O R K

C O N N E C T I O N

High-speed building network

VFD gateway Low-speed VFD network DDC panel DDC panel

VFD

VFD gateway Low-speed VFD network

VFD VFD

VFD VFD VFD VFD

DDC panel
VFD

AHU

VFD

VFD

AHU

RF

RF

when it was developed a decade or more ago, network speed and reliability have improved substantially since that time. Also, the industry has realized that the most effective control of any variablespeed device almost always requires the use of multiple factors in making control decisions. Furthermore, it now is understood that very frequent speed adjustments to large fans or pumps can waste energy. For these reasons, relying on building control networks to operate HVAC equipment is not just acceptable, it is absolutely necessary if the highest possible levels of system performance and efficiency are desired.
DIRECT-NETWORK-CONNECTION CONFIGURATION CONSIDERATIONS

FIGURE 2. Alternate network connection of variable-speed drives.

control systems because each building control system employs proprietary programming functions that are not compatible beyond the product. The use of coor-

(DDC), when, from reliability and timeof-response standpoints, it was recommended that all input and output points of control for each system be imple-

One of the largely untapped cost-saving features of variable-speed drives is built-in intelligence.
dinated, distributed processing without some level of programming compatibility complicates the operation of such a configuration. However, one area in which standards have greatly progressed and now permit a certain level of integration and resulting economy is the incorporation of variable-speed drives directly into the control network, rather than through discrete point connections, as has been the norm.
HISTORICAL IMPEDIMENTS TO DIRECT NETWORK CONNECTION

mented in a single panel. That is why control designs still are made with remote static-pressure sensors wired all the way back to the panel in which the fan or pump they are controlling is connected. Although this approach had value

Once the decision to consider the use of direct network connections for variable-speed drives is made, the operator needs to determine the nature of the integration of the drives into the building control system. Leading the list of considerations is the type of connection to be made. Presently, the most popular means of connecting variable-speed drives to building control networks are BACnet, LonWorks, and Modbus interfaces. The benefit of a BACnet or Lonworks connection is that many systems employ one of these communication protocols as their network backbone. Thus, it may be possible to connect

High-speed building network

While the direct network connection VFD VFD of variable-speed drives and other equipVFD VFD VFD VFD ment no longer is rare, the reliance on the VFD network for control, as well as monitoring function, still appears to be. The reason AHU VFD for this is that design engineers continue to be somewhat hesitant to rely on the RF control network to execute direct control actions. This hesitance goes back to the early days of direct digital control FIGURE 3. Indirect network connection of variable-speed drives.
26 March 2003 HPAC Engineering

DDC panel and gateway

Low-speed VFD network

DDC panel and gateway

Low-speed VFD network

DDC panel and gateway

VFD

VFD

AHU

RF

D I R E C T

N E T W O R K

variable-speed drives directly to a building control network with relative ease. Such a connection is shown in Figure 1. Figure 1 illustrates the direct connection of variable-speed drives to a building network as nodes, just as the building DDC panels are connected. The DDC panels connect to equipment and devices associated with the chillers, towers, and air handlers. The variable-speed drives are connected directly to the high-speed building network. The chillers may be connected via either direct network connection or the DDC panel, as shown. The chillers variable-speed drives usually are part of that overall chiller interface. Although the means of connection shown in Figure 1 is preferred because of its simplicity, many manufacturers of variablespeed drives continue to employ older communications and have developed gateways so that their drives can be connected to the more-advanced building control networks. So while these drives may advertise BACnet or LonWorks compatibility, they actually employ a gateway to provide that compatibility,

Virtually every motor involved in a building HVAC system today has the capacity to improve the efficiency and effectiveness of the system when operated with variable speed.
and it may not be economically attractive to buy a gateway for each drive. This results in many direct-connect networks that need to be configured as shown in Figure 2. In Figure 2, the connections are functionally the same as in Figure 1, but the variable-speed drives are connected to their own network(s) and then connected to the high-speed building network with special gateways. While the configuration shown in Figure 2 does result in connection of the variable-speed drives to the high-speed building network, it is more complicated. Also, because the variable-speed-drive network may employ older communications that are slow by todays standards, it can result in some communication bottlenecks, which can lead to data-exchange delays (the issue of timing will be discussed later). In short, while the configuration in Figure 2 is acceptable and widely employed, the configuration in Figure 1 is preferred. Another connection option is shown in Figure 3. This configuration permits the variable-speed drives to be integrated with the building control system, but without a direct connection to the building network. This configuration usually is employed when the high-speed building network is proprietary and the interface, therefore, must be provided by the supplier
Circle 174
HPAC Engineering March 2003 27

D I R E C T

N E T W O R K

C O N N E C T I O N

of the DDC system. For such connections, gateways often are provided as an integral component of the control panels, as shown in Figure 3. The most common interface technique is to connect the variable-speed drives on their own network and integrate them into the control system via one or more DDC panels that have gateway features. In some systems, it is preferred that each variable-speed drive be connected to the panel in which the related systems are connected. This type of connection is shown in Figure 3. The reason for this is that many of the proprietary building networks in use today are legacy networks that do not employ the most modern network-management technologies and, therefore, can be subject to certain network-traffic limitations.
DIRECT-NETWORK-CONNECTION OPERATING CONSIDERATIONS

Response Time and VFDs

There has been a great deal of discus-

ome designers avoid the direct network connection of variable-speed drives (VSDs) because they are concerned about the time it takes a VSD to receive and execute a command after a signal is issued from DDC logic. Although modern network capabilities are lightning fast, at this early point in building-system networking, many equipment manufacturers employ crude network interfaces that result in signaltransfer delays, which can add up to several seconds in certain applications. While response time for direct network connection eventually will take care of itself, the question is whether such delays are detrimental to proper system operation. The answer comes in part from looking closely at the system that each variablefrequency-drive (VFD) operates. For the most part, each system involves motors, pumps or fans, and fluid circuits, all of which contain substantial inertia. Adjusting the speed of these systems at short intervals not only is ineffectual, it also usually is detrimental to the energy performance of the system. It is like driving a car by constantly pressing down and letting up on the accelerator. Most of us were taught in drivers education that this is an inefficient method of operating a car, and for similar reasons, it also is true for fans and pumps. I recommend that control for VFD-operated motors be developed so that a speed adjustment is required at intervals no shorter than 30 seconds under normal operating conditions. This constraint can be implemented easily, and when done properly, it almost always leads to more stable and more efficient operation of the system. Furthermore, with this larger interval, response-time issues associated with the direct network connection of VFDs disappear.

Circle 167
28 March 2003 HPAC Engineering

D I R E C T

N E T W O R K

C O N N E C T I O N

sion about the advisability of the direct network connection of variable-speed drives in certain critical applications. It has been argued that certain motors operated with variable-speed drives require speed adjustments once every second or less and that network connections in many systems may delay such signals. Depending on the network, it is possible to see a delay in information transmission and reception of up to a second or more under certain circumstances. Therefore, when a network connection is employed to turn on lights from a manual switch, occupancy sensor, or similar control, the maximum end-to-end signal transmission/reception time should be evaluated carefully. However, implementing control for a variable-speed drive that subjects a motor to speed-adjustment intervals as short as 1 second is not recommended because it is not energy-efficient. Consider that pumps and fans have substantial inertia, especially when the fluid involved also is taken into account. When speed adjustments are made so frequently, energy can be wasted by the continuous change in inertial energy. With modern control techniques,1 speed adjustment can be much less frequent without compromising control stability. The direct network connection of variable-speed drives sometimes is avoided because the designer desires to have the controlled variable (usually a pressure or temperature sensor) and controlled device (the variable-speed drive)

RS485 Connections

S485 is a method of transmitting and receiving serial digital data over twisted-pair wiring. Many manufacturers of equipment with internal digital controls, including variable-frequency drives, offer an RS485 connection option or port. As a result, the question, Can Manufacturer A connect to the RS485 port of the equipment supplied by Manufacturer B if Manufacturer A supports RS485 communications? often is asked. The answer is, That depends. The RS485 protocol establishes the lower hardware layers of communications, but does not ensure communication will occur. Imagine you want to communicate with your long-lost cousin in France. You decide to contact him by telephone, knowing that if you dial the right number, a connection will be made, and your cousins phone will ring. However, if you do not speak French, and he does not speak English, you may not have a meaningful conversation. The situation is similar when only an RS485 connection is specified or offered. It is very difficult to ensure communications because RS485 is a very popular Institute of Electrical and Electronics Engineers standard link for proprietary communication protocols. Many protocols operate on RS485 connections that cannot be understood by others. For example, a great many variable-air-volume terminal networks employ RS485 communications. Until BACnet and Lon, they all were proprietarynone could communicate with anything but its own kind, even though all were RS485. In the HVAC industry, the closest thing to an open RS485 protocol may be Modbus. Modbus is a higher-level protocol implemented with software that allows communication over the RS485 physical link. It is the language spoken over the twisted pair. But beware. Even specifying the protocol may not end the potential for communication problems. RS485 links operate at a variety of speeds (baud rates). Most ports offer a range of speed settings, but some do not. Therefore, care must be taken to ensure that a common protocol and communication speed can be established when RS485 communication is employed. ADVANTAGES OF DIRECT NETWORK CONNECTIONS

this additional information is used to adjust the setpoint of the controlled variable; in others, it is used to construct a virtual controlled variable or provide other means of control. However, in many (if not most) cases, much of the information employed for effective

The idea that effective and efficient control can be achieved in modern buildings without the use of the network simply is out of date.
connected directly to the DDC panel containing the control algorithm or relationship. To employ variable speed effectively, however, the control is required (by codes, in many instances) to be composed of more information than a single controlled variable. In some cases, variable-speed control is not available directly from a single DDC panel; therefore, network communication is required for effective control. The idea that effective and efficient control can be achieved in modern buildings without the use of the network simply is out of date.

Before embarking on the direct network connection of variable-speed drives, it is imperative that designers develop a thorough understanding of all relevant issues. Some earlier articles focus on such issues.2,3 The benefits of the network connection of variable-speed drives usually far outweigh potential problems. Photo A shows an all-variable-speed chiller plant. The direct network connection of the numerous variable-speed drives in this plant reduced control costs and simplified the control system. Furthermore, the operator can access remotely or from the plant workstation all of the variable-speed-drive parameters available from the individual drive panels. On this project, the direct network connection of variable-speed drives has resulted in greater economy and simplicity and improved the operability
29

HPAC Engineering March 2003

D I R E C T

N E T W O R K

C O N N E C T I O N

of the central plant.

REFERENCES

1) Hartman, T. (2001, December). Ultra-efficient cooling with demandbased control. HPAC Engineering, pp. 2932, 34, 35. 2) Hartman, T. (2001, August). Effectively integrating controls utilizing gateway connections. HPAC Engineering, pp. 11, 15. 3) Hartman, T. (2000, January). Chiller-plant control using gateway technologies. HPAC Engineering, pp. 8184, 86, 87. Additional information on technologies discussed in this article is available at www.hartmanco.com. For HPAC Engineering feature articles dating back to January 1992, visit www.hpac.com.

PHOTO A. All-variable-speed chiller plant configured with direct network connections for the variable-speed drives (silver and black enclosures).

Circle 187
32 March 2003 HPAC Engineering

Circle 192

FEATURE

Achieving and Verifying Design Intent:

It Takes a
By THOMAS HARTMAN, PE Principal The Hartman Co. Marysville, Wash.

Team
I
t is not a simple matter to successfully implement a large-scale integrated network control system. Many such systems as applied to building control have fallen short of their expected performance and reliability. Howe ve r, the potential benefits of network controls are enormous, including greater building comfort for happier occupants, vastly improved energy e f f i c i e n c y, more effective life s a f e t y, and lower-cost operations. In many building projects, these benefits add significantly to the bottom line and make integrated network control configurations essential to the pro fitability of the building. An important missing ingredient in many of the less-successful network building control systems is teamwork. To achieve a successful integrated network contro l system, a design-and-implementation process that includes effective teamwork throughout is every bit as important as the systems or technologies employed.
FOCUS ON DESIGN INTENT

It is no secret that todays building control systems are often poorly specified with enormous deficiencies in sequences of operation and that most mechanical system specifications lack a clear design intent. Many such specifications reveal that the designer hasnt given much thought to how elements of the va r i o u s building systems should work under certain conditions. For some simple applications, this approach may be acceptable. Control systems subcontractors usually do have extensive practical experience in implementing control for the simpler types of comfort systems incorporated in many buildings, and their internally deve l o p e d control sequences may work satisfactorily. Also, because the control routines for each manufacturers application-specific hardware are different and lack flexibility, it can be very difficult for manufacturers to convert precise, specified sequences into code that fits their

simplest and least-costly hardware. The result is that, for very simple systems, the use of detailed sequences in specifications can result in confusion and duplication of effort. Furthermore, a design intent document for such simple systems, though still important, can be very brief and should be easily developed. At present, simple standalone systems for comfort control, lighting, fire, and life safety may be less costly than an integrated netw o rk-based control system in many small buildings. A simple control system specification and design intent statement that place certain aspects of the control system design on the control system subcontractor can work in these applications. Howe ve r, in larger buildings, building complexes, and certain special-purpose buildings, the economies that can accrue from a well-designed, integrated network control system can be substantial. Facility managers can expect very significant performance improvements, as well as reductions in en-

N ETWORKED CONTROLS

14

July 2002

Integrated Network Controls Require the Right Process and Teamwork, as Well as Effective Technologies

ergy and maintenance costs of 50 cents to more than a $1 per square foot per year that are directly attributable to integrated network control. But achieving these cost reductions and the other benefits of network control requires special attention not only to the technologies applied, but the process of applying them effectively. The key to success is the development very early in the project of an effective design team and a design intent that the entire team, including the project owner/operator, understands and agrees with. Then, further project stages must be organized to ensure that design intent is an active participant in every decision that follows thereafter and that the design team will support it effectively throughout the project.
DEVELOPING AND COMMUNICATING DESIGN INTENT

When designers decide to consider implementing integrated network control aimed at capturing the benefits listed above, it is

imperative to start the process by developing a design-intent document during the schematic design phase with the participation, full understanding, and agreement of the building owner and the entire design team. The design intent must include the role of the integrated network control system. Creation of the design intent is followed in detailed design by an indepth specification that effectively communicates design intent to potential contractors. These steps are important and often not properly executed. The specification must reflect the design intent with realistic technologies and provide specific achievable operational sequences and goals. However, producing an effective design-intent document and well-organized and detailed specifications requires good teamwork among the design disciplines and the owner. This team must remain throughout the project because it is teamwork in applying technology effectivelyas much as the technology itselfthat determines h ow successful an integrated building control network will be. Figure 1 shows a flow chart that, if followed prudently, can lead to successful integrated network controls. The first two steps are those of developing and communicating the design intent for the network c o n t rol system. As discussed above, this process is often poorly performed in designs for building projects. Failing to develop a universally agreed-upon design intent and communicating it effectively in the specifications can be disastrous to the success of integrated network control technologies. And the lack of teamwork in developing an effective vision for the operation of the building systems is often the root cause of this failure. To protect against this early f l a w, design engineers need to

communicate effectively to the project owner and design team the potential risks and benefits of an integrated network control system. If it is decided that such a system should be considered for the project, the design team must make a strong commitment to interdisciplinary teamwork and owner involvement, and a designer experienced in network contro l technologies appropriate for the project should be selected to join the team. This designer should assume responsibility for network issues, which include developing preliminary costs and benefits so that the cost-effectiveness of network options can be discussed. Such a designer may be part of the mechanical design team (in rare cases, the electrical design team), a separate consulting engineer working directly with the mechanical and electrical designers, or an engineer from a systems integrator or control system subcontractor (this may limit technology options that are considered, but can work well in design-build projects in which the contractor already has selected the control system subcontractor). It is up to the project owner and the initial design team to select the members of the group who will develop the schematic design, which will include a design-intent document and preliminary plans for the integrated network control system. That group must then become the core of the team that will see the implementation of these technologies through. It is imperative that the designer experienced in network technology join the group during the schematic design stage. Integrated network control technologies have subtle but very important influences on equipment and configurations. If the detailed design of equipment or specific configurations has begun before the start of the integrated netNETWORKED CONTROLS
July 2002

15

work design, the potential benefits of the network may be limited by a degree of incompatibility bet ween the equipment and netw o rk, an incompatibility that could easily be avoided by concurrent development. Working closely with the other engineers, the architect, and the owner, the designer experienced in network control technologies should be a key player in developing the design-intent document and have responsibility for integrating this narrative into the control specifications. The network technology designer should also have primary responsibility for developing the control sequences, equipment interfacing, and system operation details sufficient to communicate the design intent to the specific systems and equipment that are to be connected to the integrated network. The network technology designer must include in the specifications the methods and manner in which the contractors interact with the core team to oversee the implementation and performance verification of the integrated network control system.
HIGHLIGHTING AND PROTECTING DESIGN INTENT DURING PROCUREMENT

Using teamwork to develop a sound design intent and effective specifications is only a start on the road to a successful network control project. The next major challenge is effectively communicating all aspects of the design intent and implementation process to the technology subcontractors, who are so used to seeing traditional specifications that they cannot easily understand approaches that a re substantially different. To counter this potential derailing of the process in the transition from design to construction stages, one or more pre p roposal or pre b i d conferences (and after the con-

tract has been a w a rded, pre c o ns t ruction conferences) are necessary to explain to potential contractors what is expected of them, both from the standpoint of technology and in terms of the p rocess that will be employed to see that the technology is implemented properly. While all technology items for a building construction project should be procured using a proposal-call approach, some owners and architects continue to feel they need to use s t a n d a rd bid processes for all building systems in order to meet certain legal req u i rements. While not the best approach, a bid process, if constructed pro p e r l y, can be made to accommodate the procurement of the technology components in a building construction project that employ s advanced-technology integrated network controls. However, no matter what type of procurement is employed, a meeting with prospective contractors is absolutely essential before the bids or proposals are received. In the pre p roposal or pre b i d conference, the design team highlights the design intent insofar as the technology systems are concerned and also highlights areas of the integrated network control system that may be unusual or challenging. Finally, the process of implementation oversight and review (which are described later) is carefully spelled out, and questions are posed about any and all aspects of the network system(s). To ensure the issues discussed in

this conference are not forgotten, the minutes of the meeting can be included in the contract documents. If, as a result of the meeting, changes to the specifications a re necessary, an addendum is promptly issued after the meeting. Once the bids or pro p o s a l s have been received, they must be evaluated to determine which will be accepted. As noted previously, an advanced-technology system can be effectively procured with a bid process, but it is absolutely essential that the bid be accompanied by some additional information as set forth in the specifications so that the team can make a determination whether or not the lowest bid meets the intent of the specifications. An article I authored titled Procuring High-Tech Mechanical Sy s t e m s in the Nove m b e r 1991 issue of H PAC m a g a z i n e provides an in-depth discussion
Ju ly 2002

Figure 1. This flow chart illustrates a stepby-step method to successfully integrate network controls and those who should participate in each step.

NETWORKED CONTROLS

16

FEATURE

of what should be included in specifications for effective evaluation of integrated network cont rol system proposals and bids and why.
REALIZING DESIGN INTENT THROUGHOUT CONSTRUCTION

The traditional submittal/review process is insufficient for assuring the performance of a netw o rk control system. The submittal/review process was designed for hard w a re and work s well to determine whether sensors, actuators, and some other equipment meet specified requirements. But ensuring a good opportunity for success for an advanced network control system requires a continuous discussion between the control system subcontractor (now more likely to be a systems integrator) and the designer(s) responsible for the network control design. It also requires that the control software as well as the operator-interf a c e software be reviewed by the designer as it is being deve l o p e d . This includes certain softwaremodule and interf a c e - p e rf o rmance testing and ve r i f i c a t i o n well in advance of the complete software implementation. This c o n t i n u o u s submittal re v i ew process, which starts with hardware and carries on to software, has been shown to highlight potential problems so that they can be solved before the network control system and its associated software have been finalized. From an organizational standpoint, one of the greatest problems that occurs during construction is that an adve r s a r i a l relationship between the control system designer and the systems integrator often develops. This is most unfortunate, and this inability to incorporate the systems integrator into the core team is the most notable direct cause of failure in network control projects. It
NETWORKED CONTROLS

is up to the design team to foster and maintain a spirit of teamwork as the contractors are selected. While the designer is required to approve or disapprove submittals, this should be a cooperative process that is greatly enhanced by the design teams understanding of the particular control equipment and technologies that are being applied. If there is no one on the design team with such specific system or equipment know l e d g e , that expertise should be added as the contractor and systems integrator are brought on board.
DEMONSTRATING AND VERIFYING DESIGN INTENT

ally readily solved by working together to determine the best resolution. And resolving them before the trades have left the site is much more economical than bringing people and equipment back after the job is completed.
SUMMARY AND CONCLUSION

Activities of the final step shown in Figure 1 are rarely well executed except in some performance contracts. Demonstrating and verifying that the entire system is actually operating as expected is of crucial importance in any advanced network contro l system. The activities in this step give the designers and contractors an opportunity to discuss with the operating staff operations issues that support or supplement what was discussed in training sessions, helping operators participate better as team members contributing to system perf o rmance. This step involves initial testing, commissioning, and performance verification of the system and development of an ongoing process to maintain system performance over the years of operation. Some designers find little incentive to participate in these activities, but participating actively in this step is very important. The purpose of all the project events up to this point is the satisfactory (or even outstanding) performance of the resulting system. Especially at this stage, teamwork is crucial because the most common problems encountered at startup are those involving multiple disciplines and/or trades. These problems are usu-

The technologies described in the articles of this supplement offer improved building perf o rmance and significantly reduced building owning and operating costs easily amounting from 50 cents to $1per square foot per year. But successfully implementing these technologies re q u i re s special attention to the processes employed, from initiating the design all the way to verifying its performance. To ensure the success of the processes and, therefore, the technologies to be employed, it takes a special emphasis on teamwork that crosses disciplines and project stages. Without attention to such teamwork, advanced technologies can lose their value and become a liability to the project rather than an asset. However, with a process that incorporates effective teamwork throughout the pro j e c t , advanced-network technologies can help building projects rise far above their competition in profitability and attractiveness to occupants! NC Additional information on technologies and issues discussed in this article is available at www.hartmanco.com. Comments and questions about the article may be add ressed to Mr. Ha rtman at tomh@hartmanco.com. A member of HPAC Engineerings Editorial Advisory Board, Thomas Hartman, PE, is principal of The Hartman Co., which employs new technologies in offering enhanced, individually adjustable, and accountable comfort systems for commercial buildings.

18

July 2002

FEATURE

Whole Building Networks Beyond HVAC

System designers no longer think about digital controls solely in terms of the mechanical portion. We are now considering entire buildings being controlled via digital integrated networks. Indeed, the Integrated Facility Networks (IFNs) of the future will produce a multitude of benefits for building owners, managers, and operators
By THOMAS HARTMAN, PE Principal The Hartman Co. Marysville, Wash.

N
A member of HPAC Engineerings Editorial Advisory Board, Thomas Hartman, PE, is principal of The Hartman Co., which employs new technologies in offering enhanced, individually adjustable, and accountable comfort systems for commercial buildings.
NETWORKED C ONTROLS

etwork-based building control systems offer many advantages including easier and more convenient monitoring, improved energy efficiency, simplified HVAC system maintenance, self-balancing and self-setup features, and more. As building professionals adopt a broader view of building networks, they are beginning to see the full range of benefits achievable through the use of Integrated Facility Networks (IFNs). This new perspective includes many different electronic systems beyond HVAC that are involved in the operation of typical commercial buildings. With the application of network management in other arenas as a model, our industry is realizing how important it is to bring information and c o n t rol capacity together in a straightforward, uniform way to c a p t u re value by stre a m l i n i n g building operations and maintenance (O&M) activities.

EVOLVING ROLE OF BUILDING NETWORKS

When direct digital contro l s (DDCs) were first introduced to this industry around 25 years ago, their primary purpose was to replace pneumatic controls and electric time clocks with a more accurate way to control building HVAC systems. Networks were set up mainly so information could be exchanged between controllers and equipment operators. Digital controls, which have been employed in building environmental control and fire alarm systems for many years, are now universally applied to almost all types of commercial building equipment. Today, digital controls operate security, emergency, standby generation, uninterruptible power, and all kinds of other specialty systems found in modern buildings. They provide not only control capability, but also monitoring and alarm capability. Figure 1 shows a digital control scheme for a typical office build-

ing. Note how each system has its own means of operations interface. The interface is usually either by a dedicated computer work station, or, as in the case of simpler systems, a control panel with perhaps a small digital display and keypad. Modern buildings have many different systems in place, so it can be difficult for operators to keep track of them all. When a computer-based work station is provided, the staff may have the opportunity to consolidate the work stations together in a building operations center. Yet even when this is done, staying on top of operations is not easy and often requires that one or more people be in the operations room at all times to keep track of all the systems. With a greater industry emphasis on open communications and communications standards for effective building networks, it is becoming increasingly possible to integrate all the major building control/monitor systems together so that a single point of operations

36

interface can be created to connect building operators to each digitally controlled and monitored system in the building (Figure 2). While it is exciting to think that a single communication standard could be stipulated so that all equipment could be connected together seamlessly, were not quite there yet. There are many different manufacturers involved, and many of the products found in commercial buildings are also used in very different types of facilities. Today, and very likely well into the future, one or more gateways (shown in Figure 2 as the Building Controls Network Interface) are required to connect the systems together to share necessary information and control features, and to employ a common operator interface. Methods for interfacing equipment in todays buildings vary. Figure 2 shows each DDC system di-

rectly connected to the gateway, which is in turn connected to a building network. Work stations can be installed anywhere in the building or be accessed remotely to

connect the operator to any of the active building systems. Alternative integration schemes that connect each system to a building netw o rk and employ a translating

Figure 1. Digital systems in a typical commercial building.

Figure 2. A typical building system integration network scheme. N ETWORKED CONTROLS

37

FEATURE

HVAC and Lighting Control Integration

An integrated HVAC and lighting system can create cost savings, improve energy efficiency, and enhance occupant comfort. It can also make the building operators job easier. The proper point for integrating the two systems is within the building environmental control system (Figure 1). Lighting and HVAC systems share the same basic operating parameters. Furthermore, they are interrelated in terms of building energy use, due to the effect of lighting on the HVAC system. Coordinating the operation of the two systems at the zone level is an important design and controls issue in any commercial building. Both need to react to occupancy and occupant preferences at the smallest practical zone size. Zones should be the same, or coordinated, for both systems. Successful integration of HVAC and lighting requires good coordination between the mechanical and electrical designers. It also requires effective leadership on the architects part. The mechanical and electrical designers must understand integration issues, and be anxious to secure the benefits of integration for their client. Designers who have successfully integrated HVAC and lighting control should use the procedures as a model for integrating other building systems. Those who have not yet successfully integrated HVAC and lighting controls should consider it a first step in their building system integration efforts.

room need not be staffed continuously to provide 24/7 support of critical building systems. Another benefit of network integration is the need for interoperability among the various building systems to minimize the disruption caused by an abnormal event in any of the systems, and to coordinate the operation of these systems to minimize overall building operation costs.
MULTI-BUILDING MANAGEMENT

gateway separately, usually as part of the operator workstation are possible. In other schemes, all gateway connections are made directly to special DDC controllers on the dedicated control network and the building Internet protocol (IP) network shown in Figure 2 may not be employed at all. Whatever method of interconnection and integration is used to tie the various building control and monitoring systems together, the advantage of such network integration is apparent. Instead of having to physically go to each system to check its operation, everything can be checked at one location. This provides much better assurance that critical building systems will be supported properly. This integration also permits a single dial-in/dial-out modem connection so that the operations
NETWORKED CONTROLS

The type of system integration shown in Figure 2 gives the operations staff located in a stand-alone building or complex improve d oversight capabilities. However, this building-by-building management approach to O&M is rapidly giving way to multi-building approaches where integrated digital networks are an essential component. Consider that nearly all large buildings have chillers, cooling towers, standby generators, and a host of other equipment. The operations staff in a single building cannot afford the time or resources to become experts on every piece of equipment. In most single-building O&M schemes, the support of this complex equipment is provided by a combination of the buildings maintenance staff and outside contractors. These arrangements usually cost more and are not as effective as specifically trained O&M staff whose responsibilities are focused on the equipment and who have d i rect access to it at all times. However, such specially trained technicians, whether they are on staff or with an outside contractor, may be economically justified when a large amount of equipment is involved. To achieve these economies of O&M, some building management and contract maintenance firms have conceived and developed multi-building management concepts where buildings are grouped together to maximize the cost-effectiveness of providing certain services. For example, say a

p ro p e rty management firm has contracts to manage a dozen or so buildings in a metropolitan area. The firm may assign one or more general technicians to each building, or to small groups of buildings located near one another. However, one group of specialty technicians may be responsible for the fire alarm systems, another for the chillers and towers, and another for the standby generators in all the buildings. This type of O&M arrangement, coupled with an integrated facility management system that ties all the buildings together, could greatly reduce O&M costs. The opportunity to reduce maintenance costs through networked management approaches are especially attractive in light of recent research that shows equipment such as chillers, fans, and pumps are most effectively maintained when maintenance is scheduled based on attention to certain operating parameters rather than equipment runtime. An Integrated Fa c i l i t y Network (IFN) connecting the building systems together (Figure 3) provides direct access to the types of operating parameters necessary to provide more effective fault based maintenance programs for any type equipment in the integrated building network. It is also not difficult to envision overlapping networks wherein separate building management firms and specialty O&M contractors cooperate together and with others so that each can get information from the buildings and systems they are responsible for to support them more efficiently. In Figure 3, the Internet provides the inter-building communications backbone. Use of the Internet in place of dial-in/dial-out connections vastly increases the speed of data transmission and usually lowers the costs associated with the inter-building network. The use of the Internet for a communications backbone has some other potential advantages. Although it is not a requirement, it is becoming a significant advantage

38

Figure 3. Integrated facility management system for multiple buildings.

to employ certain Internet standards such as the use of standard Web browsers for the communication and display of information to and from the systems in each building. By adopting widely employed Internet approaches, the IFN will be less dependent on the various control systems vendors whose products are used throughout the buildings in order to be connected to the network.
CHALLENGES OF MULTIBUILDING NETWORKS

The Role of IT Departments in Integrated Facility Networks

Although Information Technology (IT) departments generally, and correctly, view building systems integration as relatively minor applications for their Intra and/or Inter building networks, they should be consulted early and continuously while the Integrated Facility Network (IFN) is being designed. The main reason is so the designers can get information from the IT managers about the corporation or institutions approach to building and supporting the networks involved. The most important questions designers need to answer are: 1. How critical do the IT managers view this network? 2. How is network security provided, and how can security for this application be made compatible with others? 3. How does the IT management intend to support this network and network applications? The answers to these questions will help the design team map out an integration scheme that 1) maximizes the use of internal network resources without jeopardizing this or any other application, and 2) is compatible with development and operations standards and procedures that are used for other network applications. Such compatibility is essential to ensure that the IFN can and will be readily supported along with the other applications on the network.
N ETWORKED CONTROLS

Although the potential opportunities for implementing IFN systems are substantial, a number of issues persist that make implementing such networks a serious challenge for the designer. These include:

39

FEATURE

Benefits of Interoperability Among Building Systems

One of the great things about integrated building systems is the opportunity they present to garner the benefits of interoperability among systems which can result in enhanced control, more flexibility, and impressive cost savings. Designers have long accommodated the need for interoperability between fire alarm and environmental controls systems. However, consider the further benefits of integrating security systems to provide higher levels of security without limiting egress during emergencies. Network-based integration may also permit standby generation to be used as an automatic peak power reduction tool, and the uninterruptible system to shed all but the most critical loads as available backup power is consumed. The opportunities from network integration also include reduced capital costs. Smaller central heating and cooling plants can sometimes be specified when the operation of heat generation equipment in many types of buildings is better coordinated with integrated network control. So, while Integrated Facility Networks (IFNs) are often seen primarily as information gathering tools, designers should never lose track of the potential benefits of incorporating system interoperability as a part of their IFN design.

1. A lack of uniformity in how a point of control or monitoring is represented, adjusted, or overridden at the operations work stations can make it difficult for operations staff to interact with individual systems within an IFN. In some cases, operations staff is required to know and understand a whole array of individual digital control system feat u res even though they are represented on a common platform. This increases the complexity of such a platform and decreases its effectiveness. 2. Understanding the critical components of each building system when thinking in terms of integration. Fire alarm, security, and u n i n t e r ruptible power systems need to stand alone within each building so all critical functions will operate through specifically approved equipment and communication links, regardless of the state or condition of the integration network components. 3. Regulatory and administrative practices for the non-critical functions of many building systems. For example, some fire jurisdictions expect non-critical fire alarm management tasks to be administered entirely fro m within the facility served by the fire alarm system. Remote monitoring and operations of non-

alarm functions that may include resetting trouble messages or adjusting detector sensitivity are new concepts for some officials, and can be difficult for them to accept and approve . The fact that fire alarm systems can be better supervised by grouping them together and increasing the attention and supervision over a facility network may take some time for these officials to acknowledge. Patience by the building management and operations team in developing such integration pro g r a m s may be required. 4. Lack of communications standards. A lack of standards is the greatest barrier to widespread implementation of IFNs. Although many digital contro l s manufacturers claim to use open and/or standard communication p rotocols, connecting dive r s e systems effectively to integrated networks continues to take special attention and oversight and is often more expensive than nece s s a ry. The near-term goal for our industry should be to convince all the various manufacturers of building systems that have digital controls to subscribe not just to open communications protocols, but to adopt industry recognized standard communication protocols that allow their

equipment to be tied easily and effectively into larger integrated building management networks.
SUMMARY

Owners stand to achieve significant O&M cost savings through the use of IFNs that connect systems within buildings, and then by aggregating multiple buildings together. However, implementing e f f e c t i ve and economical IFNs that enable effective re m o t e O&M management continues to be more complicated and more costly than necessary. Fortunately, the industry is beginning to understand the opportunities and impediments to effectively implementing IFNs. All parties should join together to encourage the use of industryrecognized communication protocol standards for all building equipment that has digital controls. This will help us move toward more widespread application of IFNs, which will result in greater economies for building owners. NC Thomas Hartman, PE, Fi rst Ri ght s, T h e Ha rt m a n Company,www.hartmanco.com, 9905 39th Dr. NE, t o m h @ hartmanco.com, Marysville, Wash ington 98270, Fe b ru a ry 22, 2001, 360.658.1168.
NETWORKED CONTROLS

43

WAYS TO SAVE

ENERGY
Then and Now
With another energy crisis possibly looming,

HPAC Engineering dusts off and revises

a list of ways to conserve fuel and electricity
n May 1975, HPAC Engineering published a Engineers must evaluate items on these lists to two-page data sheet listing 133 ways to save ensure they are suitable and safe for their buildings. energy in existing buildings. Modifications to the building shell, HVAC and lighting systems, and BEST PRACTICES 1. Participate in voluntary above-code performoperations-and-maintenance practices were suggested. A lot of emphasis was placed on industrial ance programs, such as Building Research Establishment Environmental Assessbuildings, as well as general HVAC considerations. Published By MICHAEL G. IVANOVICH, Editor; ment Methodology (BREEAM) and the U.S. Green Building at the peak of the oil embargo, the THOMAS HARTMAN, PE; and Councils Leadership in Energy list was very well-received. JACK TERRANOVA, PE and Environmental Design Once again, energy conservation is a priority for property ownersa trend that (LEED). 2. Verify that the design conforms to or surpasses is likely to last beyond the current surge in oil, gas, and electricity prices and regional supply uncertain- the states energy-conservation code (if applicable) ties. Responding to reader inquiries for an updated and/or ASHRAE standards. 3. Keep operation-and-maintenance (O&M) list, we revised the 1975 data sheet. Although the list from 1975 has many brute- documentation up-to-date and be certain that force energy-conservation measures that are valid O&M staff are knowledgeable, well-trained, today, there have been a lot of refinements to the art informed, and proactive in their duties. 4. Establish energy-consumption and demand of saving energy. To list them all would be impossible. The point of this list is to inspire progress in goals and track and discuss them with operations achieving energy-efficient systems, not to docu- staff at regular intervals. 5. Work as a team to keep in mind the building ment every way to save a Btu. The following is the revised list. Some old meth- life cycle throughout the design, construction, and ods, such as, Fix broken windows, were dropped, operations phases. 6. Educate occupants on recognizing and reportwhile a new categorybest practiceswas added. Also, the list was broken down into more subcate- ing IAQ and energy problems and reward them for gories to make it easier to use. For comparison, the doing so. 7. Participate in public and peer reviews of 1975 list is contained in the sidebar on pages 51 design for adherence to new codes and standards. and 52. 8. Encourage participation in professional Prior to becoming editor of HPAC Engineering, organizations, provide training in sustainable pracMichael G. Ivanovich was a research scientist working tices, subscribe to quality trade publications, and on environmental projects in a variety of areas, includ- bookmark useful Websites. ing energy efficiency. Thomas Hartman, PE, is the principal of The Hartman Co., a Marysville, Wash.based high-technology engineering and development THE BUILDING firm. Jack Terranova, PE, is division manager for Envelope 9. Design for low infiltration by employing a Colliers ABR Inc. in New York. Both Hartman and Terranova are members of HPAC Engineerings Edito- high-quality envelope and isolating elevator shafts rial Advisory Board. Continued on page 54

50

March 2001 HPAC Engineering

W A Y S

T O

S A V E

E N E R G Y

133 Ways to Save Energy in Existing Buildings

The following list originally was published in the May 1975 issue of HPAC Engineering.
STRUCTURE 1. Add additional insulation to roofs, ceilings, or walls where practical. 2. When reroofing, use light-colored material to reduce solar gain on air-conditioned structures. 3. Ventilate attic spaces. 4. Put solar film on windows to cut cooling loads. 5. Install solar screens on windows to reduce cooling loads. 6. Install double glazing in place of single glazing. 7. Recaulk window and door frames. 8. Reseal curtain walls. 9. Eliminate excessive crackage between double-entry doors. 10. Install weather stripping around windows and doors. 11. Repair broken windows. 12. Keep garage and warehouse doors closed as much as possible. LIGHTING AND POWER 13. Install higher-efficiency lighting systems where possible. 14. Reduce overall illumination levels. 15. Implement a lighting-maintenance program to obtain maximum efficiency from existing systems. 16. Use supplemental lighting for specific tasks instead of increasing the overall illumination for a given area. 17. Utilize natural lighting in perimeter office spaces. 18. Utilize multiple switching for selective lighting levels in offices, conference rooms, etc. 19. Reduce lighting in areas not requiring higher levels (stockrooms, corridors, etc.). 20. When redecorating, use light colors on ceilings and walls to achieve good illumination levels with less lighting. 21. Reduce decorative and advertising lighting. 22. Use timers or photocells to control outdoor lighting. 23. Reduce parking-lot lighting to minimum levels required for safety. 24. Use proper-sized motors. Grossly oversized motors operate at a low power factor. 25. Apply power-factor correction where applicable. 26. Install demand-limiting equipment. CONTROLS 27. Recalibrate all controls. 28. Lock thermostats to prevent resetting by unauthorized personnel. 29. Check room thermostats for proper location. 30. Install individual room control whenever possible. 31. Install temperature-control valves (self-contained) in radiators controlled by hand valves. 32. Install enthalpy controls to optimize use of outdoor air for building cooling. 33. Install building automation system if feasible. HVAC AND MISCELLANEOUS 34. Study system carefully before making changessome changes may increase energy usage. 35. Retest, balance, and adjust systems. 36. Turn off air-conditioning machinery during unoccupied hours. 37. Revise cleanup schedule so lights and system can be turned off earlier. 38. Optimize system-startup times. 39. Shut off outdoor air during unoccupied hours. 40. Reduce outdoor-air quantity. 41. Reduce system air volume. 42. Reduce air-duct leakage. 43. Adjust outdoor-air dampers for tight closure. 44. Replace dampers with higher-quality ones whenever possible. 45. When balancing or rebalancing a system, consider outdoor-air leakage when making minimum-outdoor-air settings. 46. Adjust dampers in mixing boxes and multi-zone units so that they shut off tight to reduce leakage. 47. Avoid use of preheat coils if possible. 48. Raise the mixed-air temperature. 49. Reset hot and cold deck temperatures in direction of reduced heating and cooling. 50. Set reheat schedule as low as can be tolerated. 51. Reset chilled-water and heating-water temperature in accordance with loads. 52. Use lowest radiation temperature possible in perimeter spaces. 53. Do not permit perimeter and interior systems to buck one another. 54. Optimize multiple chiller operation. 55. Run heating and cooling system auxiliaries only when they are required. 56. Whenever possible, only operate return-air fans for heating during unoccupied hours. 57. Install auxiliary air risers to reduce fan horsepower. 58. Convert constant-volume fan system to VAV. 59. Reduce heating in unoccupied areas. 60. Reduce heating in overheated spaces. Do not open the window to cool these places! 61. Shut off exhaust fans during unoccupied cycles. 62. Check exhaust systems to ensure they are exhausting only the amount of air required. 63. Reduce exhaust-air quantities from toilet rooms, laboratories, etc. when feasible. 64. Convert toilet-room exhaust fans to operate only when room is occupied. 65. Reduce supply temperature of domestic hot-water systems. 66. Use condenser water to preheat domestic hot water. 67. Use condenser water for air-conditioning reheat. 68. Retrofit solar collectors to the building to heat domestic and process water.

Continued on page 52

HPAC Engineering March 2001

51

W A Y S

T O

S A V E

E N E R G Y

Continued from page 51

69. Install heat-recovery device to reclaim heat from building, kitchen, and process exhaust. 70. Replace forced-air heaters with infrared heaters. 71. Replace indirect-fired makeup-air units with direct-fired equipment. 72. Insulate piping and ductwork located in unconditioned spaces. 73. Replace worn insulation on boilers, furnaces, pipes, ducts, etc. 74. Reduce hot- and chilled-water flows. 75. Trim pump impellers to match load. 76. Convert three-way valves to two-way operation and install variable-speed pumping. 77. Clean strainer screens in pumping systems. 78. Check vents in hot-water and steam systems for proper performance. 79. Check expansion-tank size. Undersized tanks can cause excessive water loss. 80. Determine whether the boiler plant can be shut down and small boilers and water heaters can be used during the summer. 81. Do not waste condensate; return it to the boiler. 82. When high-pressure steam is available, consider use of steam turbines for pump and fan drives. Turbines can operate as a PRV valve to meet low-pressure-steam needs. 83. Repair all leaks. 84. Use proper water treatment to reduce fouling of transfer surfaces in boilers, chillers, and heat exchangers. 85. Use no more water-treatment chemicals than necessary. 86. Check cooling-tower bleedoff periodically to ensure that water and chemicals are not wasted. 87. Maintain cooling towers, evaporative coolers, and air-cooled condenser for peak efficiency. 88. Periodically inspect and repair faulty equipment. 89. Implement a filter-maintenance program to ensure peak efficiency. 90. Clean and maintain cooking equipment to maintain peak efficiency. COMBUSTION EQUIPMENT 91. Check buildings for negative pressure, which can reduce combustion efficiency. 92. Check flues and chimneys for blockages or improper draft conditions. 93. Clean combustion surfaces. 94. Check and adjust fuel-air ratios. 95. Replace atmospheric burners with power burners. 96. Install pressure controls on furnaces (industrial). 97. Install automatic air-gas combustion controls. 98. Do not overfire equipment. 99. Repair furnace linings frequently. 100. Reduce production-equipment preheat times to minimum required. 101. Reduce production-equipment temperatures to holding temperatures when production stops for relatively long periods. 102. Shut off drying and curing ovens when not in use. Do not start them until just prior to shift. 103. Seal all cracks in furnaces, ovens, etc. 104. Preheat combustion air with waste heat. INDUSTRIAL PLANTS 105. Study plant heating and air-conditioning systems to determine if they are of correct design. Many are not. 106. Reschedule operations whenever possible to second and third shift to get them off the 10-a.m.-to-2-p.m. peak electric-demand period. 107. Plan work so that the whole plant can be shut down on given weekends. 108. Shut off machinery when not in use. 109. Keep covers on tanks and vats closed to reduce evaporation losses. 110. Use push-pull ventilation on open surface tanks; 50 percent or more of the air can be saved. 111. Use immersion heaters whenever possible. 112. Use cold-water detergent in washers whenever possible. 113. Combine operations where possible to reduce the number of washers. 114. Shroud openings of furnaces, ovens, paint booths, and washers so that only the minimum amount of exhaust air will be required. 115. Eliminate stratification of air in the plant during the winter, thereby warming the floor. This can be done easily with fans (high) blowing down ducts terminating close to the floor. 116. Use spot heating or cooling of people when they are located far apart. Each should have control of the air direction and velocity over them. 117. Use evaporative cooling for human cooling whenever practical. 118. Determine whether pressure blowers could replace some compressed-air usage. 119. Do not use compressed air at higher pressures than required. 120. Do not permit compressed air to be used for people cooling. 121. Reduce the quantity of exhaust air; use local, not general, exhaust. 122. Use low-volume, high-velocity exhaust systems whenever possible. 123. Analyze all solid waste to determine whether it can be recycled, burned, or composted. 124. Salvage all oil used in the plant. It either can be reused through refining or burned in the boilers. 125. If exhaust air is contaminated, evaluate air-cleaning devices to determine if the air could be cleaned and recycled. 126. Determine feasibility of utilizing energy in production operations before exhausting it. 127. Analyze interplant truck runs. Consolidate loads and eliminate trips. 128. Shut off interplant truck engines when not in use. 129. Shut off fork-lift engines when not in use. 130. Replace worn-out machinery with modern, efficient equipment. 131. Keep heat and smoke-relief vents closed during the winter. 132. Consider using waste water for roof sprays during the summer to reduce heat load on the plant. 133. Use automatic regulators to control the volume of water used.

52

March 2001 HPAC Engineering

W A Y S

T O

S A V E

E N E R G Y

Continued from page 50

and other ventilated areas. Specify tests to ensure that design criteria are met. 10. Employ the highest practical insulation values for the roof, walls, and glazing. Consider reduced mechanical-system size and improvements in comfort. 11. Use swing/revolving doors at building entrances. According to the American Society of Heating, Refrigerating and Air-Conditioning Engineers, the use of these doors can reduce infiltration by 75 percent compared with standard single and double-leaf doors. 12. Specify Low E, double- or triplepane argon-filled glass with a factoryapplied coating (low shading coefficient). Use interior blinds and/or shades to help further reduce cooling loads during peak hours. 13. Consider planting shade trees and shrubbery to reduce cooling loads. 14. Maintain and improve the envelopes resistance to infiltration through

aggressive building-entrance and envelope-maintenance/upgrade programs.

Mechanical/electrical systemsgeneral

15. Perform an energy study or visit a nearby site with the application that you are considering. 16. Periodically re-evaluate loads to ensure that the HVAC system and components are properly sized and operationally balanced. 17. Incorporate distributing scheduling so that comfort and lighting energy is expended only when building areas are occupied. 18. Maintain good control of outside air to eliminate excessive outside-air flow at extreme temperature conditions while maintaining adequate air quality at all times of occupancy. Use enthalpy control for air economizers.
Controls

19. Design controls for straightforward operation and maintenance with self-diagnostics and other integrated

fault-detection capabilities. 20. Check the location of sensors and control loops and recalibrate them as required to maintain design conditions for comfort and energy efficiency. 21. Keep the control system operating as designed by correcting problems that require manual overrides and other performance-limiting fixes. 22. Upgrade to higher levels of roomor work-station-based control whenever possible. 23. Maintain control software to ensure that programs do not contain errors. Upgrade as required to add control features such as improved scheduling or more operator-adjustment capacity. 24. Upgrade controls or the control system when feasible. The replacement of pneumatic systems with direct digital control can be particularly effective. 25. Consider adding network features to improve alarm and maintenance response and to react to real-time energy-

54

March 2001 HPAC Engineering

W A Y S

T O

S A V E

E N E R G Y

pricing signals. 26. Maintain a sound maintenance program that includes periodic checks for broken, stuck, or loose dampers, linkages, control valves, and other mechanical control devices. 27. Consider upgrading work stations to provide more information to operators. 28. Consider installing power submeters. Also, consider installing temporary data loggers or extending the control system to detect load and scheduling anomalies.
AIR SYSTEMS

take of exhaust air. 39. Design and maintain air outlets and returns to make them free from obstructions.
EXHAUST SYSTEMS

40. Shut off exhaust fans during all unoccupied cycles except economy

purge cycles. 41. Check exhaust systems to ensure that they are exhausting only the amount of air required. If possible, reduce exhaust-air quantities from toilet rooms, laboratories, etc. to minimum acceptable levels. 42. Control room and process exhaust

29. Convert constant-volume fan systems to variable-air-volume (VAV) systems to reduce system air volume and energy use at part-load conditions. 30. Replace mechanical air-volumecontrol VAV systems with variable-frequency-drive (VFD) air-volume control. 31. Place fixed duct static setpoint control of fans with network-based duct static reset or direct zone control of fan volume to reduce energy at part loads. 32. Improve design, installation, and maintenance standards to reduce damper, air-duct, and plenum penetration. Also, improve the design and maintenance of ducts and duct insulation to promote smooth and efficient air flow. 33. Design and maintain a system that can closely control outside air to the actual requirements at all times. 34. Avoid the use of preheat and reheat coils, as well as lockout operation of such coils, when possible. 35. Design and operate systems to raise the mixed-air temperature and to reset hot and cold deck temperatures at part-load conditions. 36. Size filters properly. Do not overfilter. Also, choose the proper filter efficiency for the job. Ensure that filters do not leak around the edges. Install/specify alarms to measure the differential pressure across each filter bank with alarms back to the burner-management system. Implement a filter-maintenance program to ensure that changes are made to maximize efficiency and minimize filter costs. 37. For VFDs and soft-start fans, use sprocketed belt drives for higher efficiencies. Inspect all fan belts periodically and replace them as needed. 38. Relocate or redirect outdoor-air intakes as required to minimize the inHPAC Engineering March 2001 55

W A Y S

T O

S A V E

E N E R G Y

fans to operate only when a room is occupied or equipment is running. 43. Install an energy-recovery device to reclaim energy from building and process exhaust-air systems. 44. Develop operations strategies that keep laboratory fume-hood doors closed and exhaust fans off when they

are not in use.

STEAM AND HYDRONIC SYSTEMS

45. Reset chilled-water and heatingwater temperature in accordance with loads. 46. Optimize cooling-tower operation, and use cooling-tower water for hy-

dronic free cooling when possible. 47. Find and stop steam leaks; institute a trap-, vent-, and strainer-maintenance program; and insulate steam, hotwater, and condensate lines. 48. Replace old chillers and boilers with new, energy-efficient equipment. 49. Design or convert to variable-flow hydronic systems with VFDs to reduce hot- and chilled-water flows at part-load conditions. Use network-based reset of differential-pressure-set-point or directload control of pump speed for lower energy use. 50. Increase pipe diameters to decrease pump head when feasible. 51. Determine whether the boiler plant can be shut down and replaced with smaller boilers and water heaters during the summer. 52. Ensure that all condensate is returned to the boiler plant. 53. Consider replacing PRVs with steam turbines for pump and fan drives. A turbine can operate as a PRV valve to meet low-pressure-steam needs. 54. Use proper water treatment to reduce the fouling of transfer surfaces in boilers, chillers, cooling towers, and heat exchangers. Ensure that only the necessary amount of water-treatment chemicals is used. To cut down on water usage, use improved chemical treatment. 55. Maintain cooling-tower and boiler bleed and blowdown cycles to ensure that water and chemicals are not wasted. Consider sidestream filtration for cooling towers to reduce water and chemical use. 56. Install water-efficient plumbing to reduce hot-water consumption and pumping. 57. Maintain cooling towers, evaporative coolers, and air-cooled condensers for peak efficiency. 58. Use low-flush (1.6-gpm) toilets, occupancy sensors at flushometers, etc. 59. Add a water-side economizer cycle for free cooling instead of operating the chiller.
MISCELLANEOUS

60. Periodically retest, balance, and adjust air and hydronic systems. 61. Commission building systems after retrofits and renovations. 62. Improve access to filters, coils, and other system components to make maintenance easier and more frequent.
56 March 2001 HPAC Engineering

W A Y S

T O

S A V E

E N E R G Y

63. Optimize multiple-chiller operation to stage chillers for maximum efficiency and the lowest electrical demand. 64. Design and operate systems for coordinated control to minimize simultaneous heating and cooling in perimeter and interior systems or within air handlers and unitary equipment. 65. Reduce heating and raise cooling setpoints in unoccupied areas. 66. Notice when personal heaters and fans are being used. Find out why and then solve problems with the building HVAC system.
ALTERNATIVE ENERGY

COMBUSTION EQUIPMENT

67. Design or retrofit solar collectors to heat domestic and process water. 68. Consider photovoltaic roofing materials and curtain wall panels. 69. Consider infrared heating in place of forced-air heaters for areas subject to much outside-air infiltration.

70. Design and periodically check combustion equipment for proper air supply and adequate flue conditions. 71. Periodically check and adjust fuelair ratios on combustion equipment. 72. Consider installing automatic airgas combustion controls. 73. Do not overfire equipment. Carefully plan and facilitate operation to minimize energy usage while meeting process or comfort-load requirements. 74. Replace indirect-fired makeup-air units with direct-fired equipment when feasible. 75. Preheat combustion air with waste heat. 76. Check boilers, furnaces, ovens, etc. for cracks or damage and repair as required. Keep combustion surfaces clean and well-maintained.
INDUSTRIAL PLANTS

77. Update plant heating and air-

conditioning systems as comfort and process requirements change. 78. Consider operations schedules to minimize peak electric-demand periods. 79. Plan work so that the whole plant can be shut down on given weekends. 80. Shut off machinery when not in use, such as during lunchtime. 81. Keep covers on tanks and vats closed to reduce evaporation losses. 82. Use cold-water detergent in washers whenever possible. 83. Combine operations when possible to reduce the number of washers. 84. Shroud openings of furnaces, ovens, paint booths, and washers to minimize the amount of exhaust air required. 85. Eliminate stratification of air in high-bay plants during the winter by heating at the floor level. 86. Use spot heating or cooling for people when they are located far apart. 87. Use evaporative cooling for human cooling when practical.
Continued on page 75

HPAC Engineering March 2001

57

W A Y S

T O

S A V E

E N E R G Y

Continued from page 57

LIGHTING AND POWER

88. Determine whether pressure blowers could replace some compressedair usage. Do not use compressed air at higher pressures than required. 89. Design for reduced exhaust air with the use of local, rather than general, exhaust systems, such as ventilated welding guns, hoods for portable grinding equipment, and local traveling hoods for molten-metal pouring. 90. Consider recycling industrialwaste streams. 91. Consider heat recovery or filtration for recycling industrial exhaust air. 92. Determine the feasibility of utilizing waste heat in production operations before exhausting it. 93. Analyze interplant truck runs. Consolidate loads and eliminate trips. 94. Replace worn-out machinery with modern, efficient equipment. 95. Keep heat- and smoke-relief vents closed during the winter.

96. Design and retrofit high-efficiency lighting systems. 97. Reduce excessive illumination. 98. Consider a group relamping program for bulbs and ballasts. 99. Install daylighting utilizing dimming ballasts and photo sensors. 100. Utilize multiple switching for selective lighting levels in offices, conference rooms, etc. Utilize dimmingballast technology to have the light level match the task. 101. Reduce lighting in areas that do not require high levels of it, such as stockrooms and corridors. 102. When redecorating, use light colors on ceilings and walls to achieve good illumination levels with less lighting. 103. Reduce decorative and advertising lighting. 104. Use timers or photocells to control outdoor lighting.

105. Implement bi-level lighting triggered by motion sensors to lower light levels when no one is around. 106. Reduce parking-lot lighting to the minimum levels required for safety. 107. Use proper-sized high-efficiency motors operated by VFDs when capacity requirements fluctuate. 108. Apply power-factor correction when possible. 109. Stage the startup of loads to keep demand down, and install demand-limiting equipment. 110. Connect lighting and comfortconditioning systems to the building automation system to minimize energy use in unoccupied areas. 111. Replace low-efficiency fixtures with high-efficiency fixtures (HID and HPS). 112. Install motion sensors and dimmable fluorescent fixtures in warehouses for maximum energy savings.

HPAC Engineering March 2001

75

W A Y S

T O

S A V E

E N E R G Y

Continued from page 57

LIGHTING AND POWER

88. Determine whether pressure blowers could replace some compressedair usage. Do not use compressed air at higher pressures than required. 89. Design for reduced exhaust air with the use of local, rather than general, exhaust systems, such as ventilated welding guns, hoods for portable grinding equipment, and local traveling hoods for molten-metal pouring. 90. Consider recycling industrialwaste streams. 91. Consider heat recovery or filtration for recycling industrial exhaust air. 92. Determine the feasibility of utilizing waste heat in production operations before exhausting it. 93. Analyze interplant truck runs. Consolidate loads and eliminate trips. 94. Replace worn-out machinery with modern, efficient equipment. 95. Keep heat- and smoke-relief vents closed during the winter.

96. Design and retrofit high-efficiency lighting systems. 97. Reduce excessive illumination. 98. Consider a group relamping program for bulbs and ballasts. 99. Install daylighting utilizing dimming ballasts and photo sensors. 100. Utilize multiple switching for selective lighting levels in offices, conference rooms, etc. Utilize dimmingballast technology to have the light level match the task. 101. Reduce lighting in areas that do not require high levels of it, such as stockrooms and corridors. 102. When redecorating, use light colors on ceilings and walls to achieve good illumination levels with less lighting. 103. Reduce decorative and advertising lighting. 104. Use timers or photocells to control outdoor lighting.

105. Implement bi-level lighting triggered by motion sensors to lower light levels when no one is around. 106. Reduce parking-lot lighting to the minimum levels required for safety. 107. Use proper-sized high-efficiency motors operated by VFDs when capacity requirements fluctuate. 108. Apply power-factor correction when possible. 109. Stage the startup of loads to keep demand down, and install demand-limiting equipment. 110. Connect lighting and comfortconditioning systems to the building automation system to minimize energy use in unoccupied areas. 111. Replace low-efficiency fixtures with high-efficiency fixtures (HID and HPS). 112. Install motion sensors and dimmable fluorescent fixtures in warehouses for maximum energy savings.

HPAC Engineering March 2001

75

By THOMAS HARTMAN, PE The Hartman Co. Georgetown, Texas

Relational
Network-enabled approaches to improving the efficiency and effectiveness of HVAC operation

NETWORKED CONTROLS

10
October 2005

espite numerous technological adThis article will examine the broad concept vances, digital building controls are of relational control, focusing on its potential woefully underutilized. Why? Because benefits and long-range prospects. the fundamental control methodologies they most often are configured to apply CURRENT CONTROL METHODOLOGIES are little changed from the early 1900s, the time For about as long as building controls have automatic building controls were introduced. been around, the control model shown in Figure 1 As a result, the control applied in many building has been employed in one form or another in comfort systems consists primarily of a large HVAC-system applications to adjust capacity or number of independent, stand-alone control flow by modulating valves, dampers, vanes, motor loops. As building systems become more inte- speed, and other devices and variables. Once a grated and complex, the continued use of this very simple analog device, the controller module control threatens to severely undermine perform- has grown in sophistication over the last several ance, precision, stability, efficiency, and reliability. decades. Today, this PID (proportional, integral, Rising energy costs and a widening gap between derivative) controller is extremely flexible, electricity-generation capacity and electricity incorporating variable-gain and even self-tuning demand are driving interest in the networking capabilities of Feedback signal modern control systems. Taking advantage of these capabilities and creating tremendous synerSensor gies between equipment and operational requirements is a series Controlled Controller of new control methodologies device 1 Control Device output termed relational because Reference signal signal (controlled variable) they optimize the operation of (set point) HVAC-system components in FIGURE 1. Closed-loop control. relation to one another.

l Control
capabilities, which ensure device output remains at or near the reference signal (set point) under widely varying conditions. Unfortunately, for modern HVAC applications, the control architecture is fundamentally flawed, most significantly in terms of: Applications-level control . Figure 1 depicts closed-loop control. Closed loop means the feedback signal links the controller and controlled variable so that the output signal to the controlled device is adjusted automatically to maintain the desired reference signal (set point) as conditions change. In the very simple comfort-conditioning systems employed a century ago, closed-loop control could be provided at the applications level easily. For example, by modulating a steam valve to a simple heating device, with a space-temperature sensor providing feedback to the controller, a space-temperature set point could be maintained under a wide range of load conditions. In todays typical HVAC system, closed-loop control at the applications level is not so easily achieved. For example, in a variable-air-volume (VAV) system, if spaces are not being satisfied, there normally is no feedback to drive down supplyair temperature. This open-loop characteristic significantly limits the performance of todays more complex HVAC systems. Control coordination. Todays VAV-system controls employ two major control loops: one for supply-air temperature and another for duct static pressure. For optimum efficiency and comfort, changes in airflow and temperature in response to changes in a VAV systems cooling requirements need to be coordinated. Most VAV systems, however, lack the control mechanism necessary to coordinate the temperature and flow of supply air in response to changes in load. Likewise, VAV-box data seldom are employed effectively enough for the air system to be adjusted optimally as cooling load changes.
RELATIONAL CONTROL FOR SYSTEMS SERVING MULTIPLE PROCESSES

Consider a chilled-water distribution pumping system. If the system were to be controlled traditionally, a pressure sensor would be incorporated across the supply and return headers at the end of the distribution loop, the idea being that as long as differential pressure between the headers is adequate, every load in the circuit will have sufficient pressure to deliver design flow and, therefore, meet design load conditions. For simple distribution configurations, this strategy works. In many configurations, NETWORKED CONTROLS 11
October 2005

NETWORKED CONTROLS

12
October 2005

however, branches from mains have adjacent loads that can experience high concurrent loading, meaning local areas may experience low differential pressure and flow starvation under certain conditionseven when the design pressure is maintained at the sensor location. At least as problematic is the fact this method of control results in poor pumping efficiency at part-load conditions, which constitute nearly all of a typical pumping systems operating time. Relational-control techniques replace this pressure-based control with more-direct flowbased control. The network collects data on the condition of each load served, which can be condensed easily for muchmore-efficient pumping-system operation at all load conditions. Because each load is checked continuously, proper servicing can be ensured. Such a method of control is shown in Figure 2. Here, pump speed is calculated at timed intervals based on valve-position (and load) data retrieved from the local load controllers (ASC [applicationspecific controller] 1 through ASC 5). This multi-input method of control can be much more stable than conventional PID control because it does not involve continual readjustment based on a set point and error signals. Methods of self-correcting control, which make adjustments at loads, rather than at plants, have been developed.2,3,4 These result in smoother, more stable control. At the applications level, the type of relational control depicted in Figure 2 is closed loop because of the direct link between multiple processes (cooling loads) and a single controlled resource (a chilled-water distribution pump). To a limited extent, this link can be made with conventional PID controllers by using valve-position feedback from all of the loads

Pump-speed controller Chilledwater supply

Load 1

Load 2

Load 3

Load 4

ASC 1

ASC 2

ASC 3

ASC 4

Load 5 Cooling tower Load 5

Chilled-water distribution pump

ASC 5

Chilled-water return

FIGURE 2. Closed-loop relational control.

to reset the distribution-pressure set point. Because the controlled variable (pressure) is non-linear with respect to the controlled output, however, such an approach often becomes

unstable under certain circumstances. Not only must gains change as set point and load conditions change, the accuracy of pressure readings at low-flow conditions often is a problem.
Cooling-tower-fan controller

Chiller

Condenser-pump controller

Chiller controller

Pump-speed controller Chilledwater supply

Load 1

Load 2

Load 3

ASC 1

ASC 2

ASC 3

Load 4

Chilled-water distribution pump

ASC 4

ASC 5

Chilled-water return

FIGURE 3. Closed-loop relational control.

RELATIONAL CONTROL FOR SYSTEMS WITH MULTIPLE COMPONENTS

Many HVAC systems today employ multiple components to provide desired output. Consider the system in Figure 2. While the chilled-water distribution pump appears to be the only element required to provide chilled water, we know that is not the case. Figure 3 is a more realistic diagram of the system. According to the Equal Marginal Performance Principle,5 to optimize operating efficiency, every component in a cooling system must be adjusted simultaneously as load conditions change. Although such integrated, multicompo-

hand, opens our industry to new methods of relational control that offer complete closed-loop and fully coordinated control. These new approaches will significantly improve both the efficiency and effectiveness of HVAC operation while allowing simpler and more stable systems. In this era of rising costs and limited electric-energy resources, relational-control methods have an important role to play in the development of energy-efficient HVAC systems.
REFERENCES

1) Anderson, R., & Hartman, T. (2005). Internal technical communications.

Relational control will significantly improve both the efficiency and effectiveness of HVAC operation while allowing simpler and more stable systems.
nent control is beyond the capacity of conventional control with independent PID controllers, it can be achieved with a flexible, network-based control system. To get an idea of how much more effective a cooling system will be with relational control based on the Equal Marginal Performance Principle, consider that chilled-water systems generally incorporate some degree of decoupling to accommodate stand-alone control of their various components and that low delta-T almost always is an issue. When relational control is applied, however, such decoupling is unnecessary, and chiller-capacity-loss problems from low delta-T vanish. In addition to solving problems that have plagued HVAC systems for decades, relational-control strategies greatly enhance the operational efficiency of systems with multiple components and, when coupled with relationalcontrol approaches for systems serving multiple processes or loads, ensure that each load in a system is met effectively.
CONCLUSION

2) Hartman, T. (2003, September). Presenting intelligent iterative control: PID replacement for setpoint control (pt. 1). HPAC Engineering, pp. 13-14. 3) Hartman, T. (2003, October). Presenting intelligent iterative control: PID replacement for setpoint control (pt. 2). HPAC Engineering, pp. 9-10. 4) Hartman, T. (2003, November). Presenting intelligent iterative control: PID replacement for setpoint control (pt. 3). HPAC Engineering, pp. 9-10. 5) Hartman, T. (2005). Designing efficient systems with the equal marginal performance principle. ASHRAE Journal, 47, 64-70.
ABOUT THE AUTHOR

The use of simple stand-alone, singleprocess feedback control limits both the effectiveness and efficiency of modern HVAC systems. The networking capacity of digital control systems, on the other
Circle 00

Principal of The Hartman Co. (www.hartmanco.com) and a member of HPAC Engineerings Editorial Advisory Board, Thomas Hartman, PE, is an internationally recognized expert in the field of advanced high-performance building-operation strategies. His accomplishments include development of one of the first hourly building energy simulation programs and refinement of an integrated approach to chiller-plant control that reduces commercial-building annual cooling-energy requirements by more than 50 percent. He can be contacted at tomh@hartmanco.com. NC