See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/267498500

On Compressor Station Layout

Conference Paper · January 2003

DOI: 10.1115/GT2003-38019

CITATIONS READS
8 7,402

3 authors, including:

Rainer Kurz Matt Lubomirsky

Solar Turbines Incorporated Caterpillar Inc.
162 PUBLICATIONS   1,295 CITATIONS    16 PUBLICATIONS   67 CITATIONS   

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Rainer Kurz on 22 December 2015.

The user has requested enhancement of the downloaded file.

OGA TOC
Proceedings of ASME Turbo Expo 2003
Power for Land, Sea, and Air
June 16–19, 2003, Atlanta, Georgia, USA

GT2003-38019

ON COMPRESSOR STATION LAYOUT

Rainer Kurz, Sebouh Ohanian and Matt Lubomirsky

Solar Turbines Incorporated
San Diego, CA

ABSTRACT NOMENCLATURE
This paper discusses issues that influence the decision on A = Area
the arrangement of compressors and the type of equipment in a = Elevation
gas pipeline compressor stations. Different concepts such as C = Constant
multiple small units versus single large units are considered,
c = Cost
both regarding their impact on the individual station and the
D = Diameter
overall pipeline. The necessity of standby units is discussed.
Various concepts for drivers (gas turbine, gas motor and electric e = Pipeline Efficiency
motor) and compressors (centrifugal and reciprocating) are fDW = Darcy Weisbach Friction Factor
analyzed. SG = Specific Gravity Relative to Air
H = Head
The importance of considering all possible operating conditions ki = Constants
is stressed. With the wide range of possible operating
L = Length
conditions for the pipeline in mind, the discussion will be
brought into the general context of operational flexibility, N = Speed
availability, reliability, installation issues, remote control, and Ns = Specific Speed
operability of gas turbine driven centrifugal compressors NPV = Net Present Value
compared to other solutions such as electric motor driven P = Power
compressors or gas engine driven reciprocating compressors. p = Pressure
The impact of different concepts on emissions and fuel cost is Q = Volumetric Flow (Capacity)
discussed.
R = Gas Constant
Among the assumptions in this paper are the performance r = Discount Rate
characteristics of the compressor. It will be outlined how these T = Temperature
performance characteristics influence the conclusions. t = Time
V = Flow Velocity
INTRODUCTION W = Mass Flow
The pressure and flow characteristics of pipelines and other x = Distance Coordinate
factors may influence the arrangement of compressors in a Z = Compressibility Factor
station as well as the type of equipment used. Besides the r = Density
question about series or parallel arrangements in a station, the
more general question arises about the number of units, standby h = Efficiency
requirements, type of driver, and type of compressor.

1 Copyright © 2003 by ASME

Subscripts
a = Average
b = Base Condition for Standard State
i = Inner
end = End of Pipeline
begin = Beginning of Pipeline
Acronyms
DLN = Dry Low NOx Combustion System
EMD = Electric Motor Drive
STD = Standard (i.e., Diffusion Flame) Combustion System
OPERATING SCENARIOS
When planning a compressor station or, for a new pipeline,
a number of stations, certain considerations have to be made.
These include:
· Steady-state and transient capabilities and requirements
of the system
· Growth requirements and capability
· Total cost of ownership and delivered cost to shippers
and customers
The first consideration involves the capability to cope with
changes in flow capacity on all time scales (i.e., hourly, daily,
seasonally). The pipeline hydraulics relate pressure losses to the
flow through the pipeline, determine the compressor operating
conditions in terms of head and actual flow, and subsequently
determine the required power from the driver. Contractual
requirements and obligations, such as pressures and volumes at
transfer points, have to be considered.
The second consideration deals with the fact that the nominal
capacity of a pipeline may grow when additional customers
demand a higher supply of natural gas. In fact, many new
pipelines start out with 50% and less capacity and grow to full
capacity over several years, or are sized for easy expansion.
Often, the prediction of the rate of growth shows a significant
degree of uncertainty. The growth scenarios, if foreseeable,
drive a station layout to possibly allow additional power to be
installed at the station level later or additional stations along the
pipeline. The alternative scenario, where the pipeline usage
declines over the years (e.g., because the gas supply from the
field declines), is also a possibility.
We have to distinguish between growth scenarios that increase
pipeline capacity by adding power along the pipeline and
scenarios that add power and loop the pipeline1. The former
scenario will always require an increase in pressure ratio in the
station. It is often necessary to replace single-stage compressors
1
Looping a pipeline means installing an additional pipeline parallel to the
existing one.
with two-stage compressors or install compressors in series to
meet the higher pressure ratio. The latter scenario will usually
increase the flow though the station and will be covered by
installing additional units parallel to the existing ones.
Total cost of ownership reflects the cost to install, operate and
decommission the stations. While the first two considerations
reflect the capability to generate revenue, the latter focuses on
the necessary costs. These costs (ci) may appear at any point in
time during installation, operation and decommissioning of the
station. An easy way to compare cost of ownership is to use a
net present value (NPV) calculation, assuming a fixed discount
rate "r" for "n" time periods:
n ci (1)
NPV = å
i =1 (1 + r ) i
LOST REVENUE
Revenue reduction resulting from equipment downtime is
an important element of the total cost of ownership. To
determine the lost revenue, total annual downtime is multiplied
by the estimated lost revenue per hour (Hsu and Hasselfeld [1]).
Another problem lies in the modeling of risk through the useful
life of the project, as well as the economic value associated with
this risk (Smalley and Mauney [2]). The risk can range from
hardware selection to maintenance practice and control system
set points. The net effect of these risks translates into downtime
or added fuel costs and, hence, added cost or lowered
production output2. In many cases, lost production for a day
creates a loss in the same order of magnitude as the fuel cost for
one driver for a whole year. A few days of otherwise lost
production can "pay" for the cost of a spare gas turbine. The
requirement derived from this is to plan the stations such that
they are tolerant to planned and unplanned outages. This could
mean installing a spare unit or to optimize the installation such
that the failure or downtime of one unit has the smallest
possible impact on the capacity of the overall pipeline. It also
means that the downtime in case of failure or planned outage
has to be minimized. Possible concepts include engine
exchange programs, available spare engines, and preventive
maintenance to name a few.
COMPRESSOR REQUIREMENTS
The study of operating scenarios suggests certain
requirements for the compression system. Beyond the quest for
higher compressor peak efficiencies, the operating requirements
set forth in this study, as well as in other references (Kurz [3])
2
The lost revenue can be considerable. Assuming gas prices of US$3 per
MMBtu and an LHV of 900 Btu/SCF, a pipeline pumping 500 MMSCFD
achieves a revenue of 500 MMSCFD x 900 MMBtu/MMSCF x $3/MMBtu =
US$1,350,000 per day. For comparison, the fuel cost for a typical 7000-hp
driver would be approximately US$1,500,000 per year.
2 Copyright © 2003 by ASME
require a compressor capable of operating over a wide
operating range at high efficiency.
Wide operating range in a centrifugal compressor can be
achieved by a combination of means. Aerodynamic theory
suggests a strong relationship between operating range,
efficiency and impeller backsweep (Cumpsty [4]). However,
there is a practical limit to the amount of backsweep. In
particular, increasing backsweep reduces the capability of an
impeller of given tip speed to make head. With the capability to
use two impellers in a casing, this perceived disadvantage can
be eliminated. The operating range is further increased by the
use of a vaneless diffuser.
The question whether a station should be equipped with
compression units in series or in parallel cannot be answered
universally. While the series approach can have advantages in
case one of the units fails (Ohanian et al. [5]), the decision
process has to take into account issues such as further
expansion, back-up strategies, operational strategy and
aerodynamic performance.
The aerodynamic aspects can be considered with the following:
Q1 / 2
Ns = N
H 3/ 4
Equation (2), and the subsequent discussion use N in min-1, H in
J/kg, and Q in m3/s.
Centrifugal impellers can be described by their specific speed,
where a high specific speed depicts a low head, high flow
impeller. There is a range of specific speeds where centrifugal
impellers tend to exhibit good aerodynamic performance, while
both very low and very high specific speeds penalize the
performance. From data presented in Figure 1, one can
IMPELLER EFFICIENCY (polytr.), %
95
90
85
80
75
70
High Backsweep
65
Moderate Backsweep
60
0 0.1 0.2
SPECIFIC SPEED, Ns
Figure 1. Impeller efficiency versus specific speed for
various well-designed impellers (Rogers [6]).
conclude that specific speeds between 0.09 and 0.14 tend to
yield good efficiency. Mixed flow impellers can extend this
range to higher specific speeds.
It must be noted that the actual running speed (N) is determined
by the power turbine speed of the gas turbine, unless a gearbox
is used. For aerodynamic and mechanical reasons, power
turbine speeds are lower for larger gas turbines than for smaller
ones. For example, a typical 15-MW (20,000-hp) class gas
turbine may have a power turbine speed of about 8500 rpm,
while a typical 3.5-MW (5000-hp) class gas turbine may run at
about 16,000 rpm.
For any given pipeline compressor station, two units in series3
will yield a higher specific speed than two units in parallel.
Thus, once the driver size (and thus the power turbine speed)
and the desired head and flow through the station are known,
one can conceptually decide whether the series or the parallel
approach would lead to better aerodynamic performance (see
example in Appendix 1).
GAS TURBINE REQUIREMENTS
Certain gas turbine operating characteristics need to be
considered4:
(2)
· Gas turbine power and heat rate depend on ambient
temperature and pressure.
· Driven equipment speed has an impact on power turbine
efficiency. The impact of different speeds can be
described by the difference between the operating speed
and the optimum power turbine speed. This optimum
power turbine speed is a function of engine load and
ambient temperature.
· Gas turbine efficiency is reduced at part load (Figure 2).
OPERATIONAL FLEXIBILITY
Operational flexibility under a larger number of different
operating scenarios has to be studied. Figure 3 (Kurz [3]) shows
operating points of a compressor station gathered during six
months of operation. Due to hourly, daily and monthly
variations in demand, the only way to cover the large range of
flow and head requirements efficiently was to install three gas
turbine driven centrifugal compressors. Depending on the
demand, the station either operates one, two or all three of the
compressors. The map shows the operating envelopes of the
3
With modern compressors and stages with a wide operating range, it is
0.3
usually possible to have identical stages for both the low pressure and the
high pressure compressor in a series application. Intercooling is usually not
necessary nor does it typically yield significant savings.
4
The gas turbine used to drive mechanical equipment is usually of the two-
shaft design. The power turbine thus allows variable speeds in the range of at
least 50-to-100% speed.
3 Copyright © 2003 by ASME
 1.2
1 1
REL ETA_TH
0.8
0.6
0.4
0.2
0
50 60 70 80 90
LOAD, %
Figure 2. Typical two-shaft gas turbine relative thermal
efficiency (hth/hth(100%load)) versus load (P/Pmax.) for different
gas turbines. The steeper drop in thermal efficiency with
part load is associated with dry-low-emissions engines.
1 Unit 2 Units 3 Units in Parallel
HEAD
Site Power
at 75°F
0 are required.
0 FLOW
Figure 3. Operating points collected over a six-month
period in a gas compression station, Case A (Kurz [3]).
compressors. Also indicated are the driver power limitations,
based on 75°F (23.9°C) ambient temperature and site
conditions. The gas turbines allow for immediate starting
capability if the need arises. Similarly, transient studies on
pipeline systems (Santos [7]) can reveal the often large range of
operating conditions that needs to be covered by a compressor
station. An example is shown in Figure 4. Lastly, scenarios that
arise from failures of one or more systems have to be
considered (Ohanian and Kurz [8]).
The quest for operational flexibility can be satisfied on various
levels: the compressor and the driver should have a wide
operating range. Using multiple smaller units per station rather
than one large unit can be another way. Here, the arrangement
in series or in parallel will impact the flexibility.
STANDBY UNITS
Because the failure or unavailability of compression units can
cause significant loss in revenue, the installation of standby
1.2
LOAD VARIATION
0.8
0.6
0.4
0.2
100 0
0 96 192 288 384 480
HOURS
Figure 4. Averaged load variation for four stations of an
interstate pipeline during summer and winter scenarios,
Case B.
units must be considered. These standby units can be arranged
such that each compression station has one standby unit, or that
some stations have a standby unit, or that the standby function
is covered by oversizing the drivers for all stations. It must be
noted that the failure of a compression unit does not mean that
the entire pipeline ceases to operate, but rather that the flow
capacity of the pipeline is reduced. Since pipelines have a
significant inherent storage capability ("line pack"), a failure of
one or more units does not have an immediate impact on the
total throughput. Additionally, planned shutdowns due to
maintenance can be planned during times when lower capacities
Standby units are not always mandatory because modern gas
turbine driven compressor sets can achieve an availability of
97% and higher. A station with two operating units and one
standby unit thus has a station availability of 100 (1 - 0.032) =
99.91% (because two units have to fail at the same time in order
to reduce the station throughput to 50%). A station with one
standby unit and one operating unit also yields a 99.91% station
availability. However, while failure of two units in the first case
still leaves the station with 50% capacity, the entire station is
lost if both units fail in the second case. Arguably, installing two
smaller 50% units rather than one larger 100% unit could avoid
the need for installing a standby unit.
A concept that has been discussed from time to time is the
approach to have "power" back up rather than "unit" standby on
a station. The idea is to have an oversized driver, operating at
part load during normal operation rather than an additional
standby unit. In this case, the driver has to be sized such that it,
together with other (equally oversized) units located further
downstream along the pipeline, can pick up the duty of an
unavailable station upstream. The advantage of this concept lies
in the reduction in the number of units. The disadvantage is that
the gas turbine operates at part load and, thus, at lower
efficiency (Figure 2) for most of the time. Even if the larger gas
4 Copyright © 2003 by ASME
turbine has an efficiency advantage over a smaller gas turbine at
full load (which is, however, not necessarily the case), operating
it at part load makes that advantage disappear. Oversized units
also limit the turndown capability of a station significantly.

It has often been assumed that for two-unit stations without a

standby unit, a parallel installation of the two units would yield
the best behavior if one unit fails. However, Ohanian and Kurz
[8] have shown that usually a series arrangement of identical
compressor sets yields a lower deficiency in flow than a parallel
installation. This is due to the fact that pipeline hydraulics
dictate a relationship between the flow through the pipeline and
the necessary pressure ratio at the compressor station. This
follows from the fundamental flow equation for a pipeline:

Tb Tim e 1.0
Q=C× ×e×
Pb
0.0375 × SG × (aend - abegin ) × pa2 ù (3)
0.5
é 2
ê pbegin - pend -
2

Z a × Ta ú
Di ê ú
2.5

ê SG × Ta × L × Z a × f DW ú
ê ú
ë û

For parallel units, the failure of one unit forces the remaining
unit to operate at or near choke, with a very low efficiency.
Identical units in series, upon the failure of one unit, would
initially require the surge valve to open, but the remaining unit
would soon be able to operate at a good efficiency, thus
maintaining a higher flow than in the parallel scenario. Figure 5
shows the operating point of the respective compressor after the
other unit is shut down at T = 1.0. While the remaining unit in
the series arrangement requires to recycle initially, it moves,
starting at about T = 2.0, slowly to efficient operating points.
The remaining parallel unit, on the other hand, moves further
and further into choke. At time T = 10.1 and 7.26, respectively,
either scenario would lead the pressure at the pipeline outlet to
drop below the minimum level (Ohanian and Kurz [8]).
Therefore, the compressor in series arrangement gives the
operator more time for corrective action.
Figure 5. Operating points of a parallel (top) and a series
(bottom) compressor arrangement after one of the two units
fails at Time = 1.0.
system is built up to full capacity over a number of years. The
project can then be started with fewer small units. Additional
units are installed later to meet the increasing demand on
compression. The expenses for portions of the equipment can,
therefore, be delayed. On the other hand, if larger units are
used, all of them have to be purchased at the beginning of the
project. A net present value (NPV) analysis immediately reveals
the advantages of the former approach (Appendix 2). It shows
in particular that even a higher total cost can translate into a
lower NPV, if parts of the cost can be deferred (Table A-3).

Given the fact that the gas stored in the pipeline will help to
maintain the flow to the users, a series installation would often
allow for sufficient time to resolve the problem. Therefore, a
standby unit could be avoided.
ECONOMIC CONSIDERATIONS
Pipeline systems usually are not operated at full capacity from
the very beginning of the project, but rather increase the
demand when new customers for the gas are found. Smaller
units tend to be advantageous in cases where the pipeline
ENVIRONMENTAL CONSIDERATIONS
Any activity has to be judged by its impact on the
environment. This includes the requirement to keep NOx, CO,
and UHC emissions at an acceptable level. For many
operations, meeting certain maximum emission levels is a
requirement for continuous operation. Further, the amount of
CO2 produced as a result of the combustion process has to be
considered. For a given fuel, the only way to reduce the amount
of CO2 produced is to increase the engine efficiency.

5 Copyright © 2003 by ASME

EVALUATION
Figure 3 outlines a typical operating scenario for pipeline
stations, showing a wide variety of operating points (Case A).
Figure 4 shows another scenario for a typical interstate pipeline
(Case B). Here, data for four stations along the pipeline under
summer and winter conditions were averaged. For the purpose
of this evaluation, the load (i.e., the power requirement relative
to the available power at each station and respective ambient
conditions) for each of the two scenarios was mapped into
different load classes, shown in Figures 6 and 7 respectively.
The data in Figures 6 and 7 lend themselves to a study to assess
the effect of different station designs, in particular the number
of units used. We can calculate the fuel usage, as well as the
amount of emissions for Case A and Case B, assuming the
following scenarios5:
0.3
0.25
RELATIVE
USAGE
I. One 100% unit with hth = 35% at full load and a
compressor with hs = 87%
II. Two 50% units with hth = 34% at full load and a
compressor with hs = 86%
III. Two 50% units with hth = 35% at full load and a
compressor with hs = 87%
IV. Three 33% units with hth = 32% at full load and a
compressor with hs = 85%
V. Three 33% units with hth = 35% at full load and a
compressor with hs = 87%
Figure 8 shows the relative fuel usage (and thus also CO2
emissions) for the different scenarios, based on a part-load
efficiency penalty as outlined in Figure 2. Because different gas
turbines exhibit different behavior regarding part-load
efficiency, each series of calculations is performed using a

0.2
0.15
0.1
0.6
0.05 Series1
0 0.55
FUEL USAGE Series2
20 20
30 30
40 40
50 50
60 60
70 70
80 80

to 0
0
10 10

90 o 9
10
to
to
to
to
to
to
to
to

0.5
t
0

CLASS, % power
0.45

Figure 6. Power usage class representation of Case A

0.4
(Figure 3).
1 2 3 4 5
NUMBER OF UNITS
0.6
0.5 a. Case A Scenarios I, III, and V
RELATIVE
USAGE

0.4
0.3 0.6
0.2 0.58 Series1
0.1 0.56 Series2
FUEL USAGE

0 0.54
0.52
0

30

40

50

60

70

80

90

0.5
to

to

to

to

to

to

to

to
30

40

50

60

70

80

90

10

0.48
0

CLASS, % power 0.46

0.44
0.42
Figure 7. Power usage class representation of the data 0.4
(prior to averaging) used for Case B (Figure 4). 1 2 3 4 5
NUMBER OF UNITS

b. Case A Scenarios I, II, and IV

5
Figure 8. Fuel usage for Case A scenarios. Series 2 is for the
Obviously, this calculation can be performed for a real scenario by taking into
steeper drop in part-load gas turbine efficiency than Series
account a large number of different operating points, with the actual engine
and compressor performance for each of these points. 1 (see Figure 2).

6 Copyright © 2003 by ASME

curve that reflects the steepest drop in part-load efficiency and
one that reflects the least steep drop in part-load efficiency in
Figure 2. There is a minimum in fuel usage for the curve that
reflects the steepest drop in part-load efficiency and one that
reflects the least steep drop in part-load efficiency in three units
(which incidentally is the station layout). The more stations
used, the less important the slope for part-load efficiency
becomes. The minimum in fuel usage also implies a minimum
in CO2 production.
Case A (Figure 8) exhibits a clear advantage of multi-unit
stations. Because the smaller units are operated closer to full
load for most of the time, the resulting fuel usage is lower than
for single-unit stations. This holds true for both slopes in part-
load efficiency and even if the smaller units achieve a lower
base efficiency than the larger units. For virtually all cases, a
station with three or four units minimizes the fuel usage.
Additional units yield no additional benefits.
Case B gives a somewhat different picture. Comparing Figures
9a and 9b shows that the conclusion regarding the optimum
0.92
Series1
0.9
FUEL USAGE
number of stations depends highly on the baseline efficiency of
the packages involved. If the smaller units have the same design
efficiency as the larger units, then a three-unit station is
advantageous. If we assume lower efficiencies for the smaller
units than for the larger units, a one or two unit station uses less
fuel.
Having said that, it again needs to be emphasized that a station
outage may cause significantly higher cost due to lost revenue
than the fuel cost for an entire year. Obviously, a standby unit
reduces the exposure significantly. Also, if the station uses
multiple units, then the unavailability of one of these units has a
smaller impact on the amount of gas that can be produced
(admittedly, the chances that one out of four units fails are
higher than the chances that one out of two units fail).
One result seems to be that for EMDs, with a relatively smaller
penalty for part-load operation, the differences in power usage
become very small. This means that the decision will depend
more on the installation and first cost, thus probably favoring
the larger units. This is particularly true if a significant impact
on availability is driven by the reliability of the power supply,
where obviously the number of units has little impact on the
overall availability.

Series2
0.88 The emissions calculations are based on the assumption that
most DLN engines achieve a constant level of NOx emissions,
0.86 on a ppmv or mg/nm3 basis, for a load range from 50 to 100%.
Standard combustion systems achieve reduced NOx levels on a
0.84 ppmv or mg/nm3 basis at part load because the fuel-to-air ratio
is reduced at part load. CO2 emissions, on the other hand, are
0.82
entirely dependent on engine efficiency (assuming the fuel
1 2 3 4 5 composition, especially the carbon-to-hydrogen ratio, remains
NUMBER OF UNITS unchanged).

a. Case B Scenarios I, III, and V Interestingly, multiple units also provide for advantages in the
1.05 absolute amount of emissions. As outlined in Figure 10, the
Series1
1 Series2
FULL LOAD EMISSIONS,

0.92
FUEL USAGE

STD
0.95 0.9
DLN
(lb/hr)/(lb/hr)

0.88
0.9
0.86
0.85
0.84
0.8 0.82
1 2 3 4 5 0.8
NUMBER OF UNITS 1 2 3 4 5
b. Case B Scenarios I, II, and IV NUMBER OF UNITS

Figure 9. Fuel usage for Case B scenarios. Series 2 is for the Figure 10. NOx emissions for Case B comparing standard
steeper drop in part-load gas turbine efficiency than Series and DLN combustion systems (scenarios I, III, V).
1 (see Figure 2).

7 Copyright © 2003 by ASME

lowest absolute NOx emissions for engines with dry-low-NOx
systems occur for a four-unit station. This is due to the usually
lower part-load efficiency of dry-low-NOx engines. Standard
combustion engines are less sensitive, but even here there is a
slight advantage for a three-unit station. It should be noted that
the emissions displayed in Figure 10 are normalized, with the
numbers from the respective combustion systems at full load.
Thus, the absolute amount of NOx emissions in lb/hr for a dry-
low-NOx system is significantly lower than for a standard
combustion system. The other cases follow a similar trend as
the calculations for fuel usage.

EQUIPMENT CONSIDERATIONS
In designing a pipeline, it is important to evaluate the most
appropriate type of compression equipment. In general, three
different concepts can be considered:

1. GT-CC: Centrifugal compressors driven by gas turbines

(Figure 11)
2. EM-CC: Centrifugal compressors driven by variable
speed electric motors (Figure 12)6
3. GM-RECIP: Reciprocating compressors driven by gas
engines (Figure 13)

The decision criteria are often based on life-cycle cost,

flexibility considerations (that usually consider cost as well as
the capacity to generate profit), or (in rare cases) first cost.
Table 1 outlines some general tendencies.

Figure 12. Electric Motor Driven Centrifugal Compressor.

Figure 11. Gas Compression Station with Three Gas

Turbine Driven Centrifugal Compressors.

Figure 13. Gas Compression Station with Numerous Gas

6
Engine Driven Reciprocating Compressors.
We did not consider constant speed electric motor drives because their limited
flexibility does not meet the requirements of typical pipeline operations.

8 Copyright © 2003 by ASME

Table 1. Comparison of Concepts compressors also are at an efficiency disadvantage compared to
centrifugal compressors, especially under the typical conditions
CONCEPT GT-CC EM-CC* GM-RECIP of a pipeline operation with large flow and relatively low head
requirements. Even the manufacturers of reciprocating
Compressor High High Medium
equipment acknowledge the fact that reciprocating machines are
Efficiency
at a disadvantage for pipeline applications (Cierniak et al. [10]).
Fuel Cost Low High** Low to
Medium
CONCLUSIONS
First Cost (incl. Medium Low to Medium to This paper raises issues that should be considered when the
installation) Medium High general arrangements of compressors, compressor stations, and
equipment are discussed. The general relationships and effects
Reliability / High Medium to Medium that play a role in the design and evaluation have been
Availability High*** identified. The generic relationship that the number and
arrangement of compressors in a station may have on the
Maintenance Medium Low High
transient capabilities, the growth capacity, the flexibility, the
Environmental Low to Low Medium to availability and the total cost of ownership have been assessed.
Impact Medium High The key point is that the planning and evaluation of compressor
stations is not possible based on a single operating point.
Rather, the full range of operating and growth scenarios has to
* Variable frequency drive be considered. The paper describes methods for evaluating
** Depending on true cost of electricity these issues.
*** Depending on the reliability of the electricity supply

REFERENCES
The difference between GT-CC and EM-CC is the type of [1] Hsu, L.L., and Hasselfeld, D.E., 1998, “Total Cost of
driver used. While the gas turbine uses small portions of the
pipeline gas as fuel, the electric motor requires a reliable source
of electric power. The cost of electricity is, thus, a key factor in
the life-cycle cost assessment. Unless the true7 cost of [2]
electricity is below about 0.02 US$/kWh, electric motor driven
concepts are at a disadvantage compared to gas turbine drivers
(McKee et al. [9]). It also should be noted that, while the local
environmental impact of an EMD is low, transmission losses
and the source of electricity (which could be a pollution [3]
generating power plant) can increase the overall environmental
impact. If the necessary electricity infrastructure is not yet built, [4]
the disadvantages become even greater. In particular, the lost
opportunity cost during downtime due to an unreliable
electricity supply can be significant. It should be noted that an [5]
electric motor drive as a system can be very reliable, as long as
the electrical power supply is assured. Since the key factor that
determines the availability of an EMD station may be the
reliability of the power supply, multiple units per station will [6]
not increase the reliability significantly.
While many of the older pipeline systems in the USA still use a
[7]
significant number of gas engine driven reciprocating
compressors, virtually all new installations use centrifugal
compressor concepts. The reasons typically include the very
high maintenance cost of reciprocating machines, as well as the [8]
lower availability and reliability, and environmental concerns
(emissions, lube oil, cooling fluids, etc.). Reciprocating
Ownership as a Business Tool in Turbomachinery
Decisions,” Turbomachinery Technology Seminar, San
Diego.
Smalley, A.J., and Mauney, D.A., 1997, “Risk-Based
Maintenance of Turbomachinery,” Proceedings 26th
Turbomachinery Symposium, pp. 177-187, Houston,
Texas.
Kurz, R., 2001, “A Question of Design,” WorldPipelines,
Vol. 1, No. 1, pp. 61-64.
Cumpsty, N.A., 1989, “Compressor Aerodynamics,”
Longman, Essex, UK.
Ohanian, S., Kurz, R., Lubomirsky, M., 2001,
“Compressor Station Layout and Operation
Considerations,” Proceedings of the Beijing International
Pipeline Conference, PRC.
Rogers, C, 1980, “Efficiency of Centrifugal Compressor
Impellers,” Centrifugal Compressors Flow Phenomena
and Performance, AGARD CP 282
Santos, Sidney Pereira dos, 1997, “Transient Analysis - A
Must in Pipeline Design,” Pipeline Simulation Interest
Group.
Ohanian, S., and Kurz, R., 2001, “Series or Parallel
Arrangement in a Two-Unit Compressor Station,”
TransASME JEng GT and Power.

7
The true cost of electricity includes the cost of generation and transmission
(both capital and operating expenses as well as transmission losses).

9 Copyright © 2003 by ASME

[9] McKee, R.J., Durke, R.D., Kuhl, C.A., 1999, “Factors that Table A-1 defines the number of units required by year and unit
Influence the Selection of Electric Motor Drives for size. Table A-2 assumes one standby unit per station, while
Natural Gas Compressors,” The INGAA Foundation. Table A-3 assumes no station standby unit.
[10] Cierniak, S., Schliesser, F., Kartano, J., 1998,
“Gasmotorisch getriebene Kolbenkompressoren im Table A-1. Number of Units Necessary
Vergleich,” Industrie-pumpen und Kompressoren, Heft 1,
Essen. Year/Unit 5,000 hp 10,000 hp 20,000 hp
Size
1 1+1 1+1 1+1
APPENDIX 1: SPECIFIC SPEEDS
Consider a pipeline application with two units in a station. 2 2+1 1+1 1+1
The total station flow is 7 m3/s. The total station head is 50,000
J/kg. The driver required has an optimum power turbine speed 3 3+1 2+1 1+1
of 11,000 rpm.
4 3+1 2+1 1+1
The specific speed for the first unit if the units are in series 5 4+1 2+1 1+1
(the second unit would have to use the same impeller, but at a
different operating point closer to surge), is:

Ns = 11000 x 70.5 / (60 x 250000.75) = 0.244 (A-1) Table A-2. NPV Evaluation (assuming standby unit)

The specific speed for the first unit if the units are in parallel Unit Size 5,000 hp 10,000 hp 20,000 hp
is: Total First 5 x $4MM = 3 x $7.5MM = 2 x $14MM
Cost $20MM $22.5MM = $28MM
Ns = 11000 x 3.50.5 /(60 x 500000.75) = 0.103 (A-2)
Net Present $17.67MM $21.2MM $28MM
Value (10%
In this case, parallel operation would yield an impeller that Discount)
potentially has a better efficiency.

APPENDIX 2: NPV EXAMPLE – POWER DEMAND Table A-3. NPV Evaluation (assuming no standby unit)
BUILD-UP DUE TO INCREASED FLOW DEMAND
Unit Size 5,000 hp 10,000 hp 20,000 hp
Power Demand:
Total First 4 x $4MM = 2 x $7.5MM = 1 x $14MM
Year 1: 5,000 hp Cost $16MM $15MM = $14MM
Year 2: 10,000 hp
Year 3: 13,000 hp Net Present $13.67MM $13.7MM $14MM
Year 4: 15,000 hp Value (10%
Year 5 and thereafter: 20,000 hp Discount)

Installed Cost Assumption for

5,000 hp unit: $4 MM
10,000 hp unit: $7.5 MM
20,000 hp unit: $14 MM8

8
The cost figures here are based on arbitrary monetary units.

10 Copyright © 2003 by ASME

View publication stats