1.industrialdryingheatpumps Minea 2011

See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/289757334
Industrial drying heat pumps
Article · January 2011
CITATIONS READS
8 4,437
1 author:
Vasile Minea
Hydro-Québec
31 PUBLICATIONS 508 CITATIONS
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
Preliminary Assessment of Deep Geothermal Resources in Trois-Rivières Area, View project
Overview of Heat-Pump–Assisted Drying Systems, Part II: Data Provided vs. Results Reported View project
All content following this page was uploaded by Vasile Minea on 25 July 2017.
The user has requested enhancement of the downloaded file.

In: Refrigeration: Theory, Technology and Applications ISBN: 978-1-61668-930-8
Editor: Mikkel E. Larsen, pp. 1-70 © 2011 Nova Science Publishers, Inc.
The exclusive license for this PDF is limited to personal printing only. No part of this digital document
may be reproduced, stored in a retrieval system or transmitted commercially in any form or by any means.
The publisher has taken reasonable care in the preparation of this digital document, but makes no expressed
or implied warranty of any kind and assumes no responsibility for any errors or omissions. No liability is
assumed for incidental or consequential damages in connection with or arising out of information contained
herein. This digital document is sold with the clear understanding that the publisher is not engaged in
rendering legal, medical or any other professional services.
Chapter 1
INDUSTRIAL DRYING HEAT PUMPS
Vasile Minea
Hydro-Quebec Research Institute, Laboratoire des technologies de l’énergie - LTE,
Shawnigan, QC, Canada
1. INTRODUCTION
Most high-value agricultural, food and wood products need drying to minimize spoilage,
preserve quality and reduce transportation costs [1]. Drying is a complex and energy intensive
process where the water is evaporated from a product by supplying heat by convection,
conduction, radiation, microwave, etc. About 85% of industrial driers are the convective type
with hot air as drying medium, and 99% of them involve removal of water [2]. They consume
up to 25% of the national industrial energy in the developed countries.
Driers incorporating dehumidification cycles are called drying heat pumps. They recover
energy by condensing moisture from the drying air. The recovered waste heat is recycled
back to the dryer at higher temperature for reuse in the drying process. This is a method of
improving the energy efficiency and reducing primary energy consumption of drying
processes.
Drying heat pumps based on the vapour compression cycle are largely the most used.
Compared to heat pumps for space heating, they have several advantages, such as:
High annual factors of utilisation;

High energy efficiency due to low temperature lifts;Efficient control of product
moisture content, and air dry and wet-bulb temperatures resulting in better final
quality, especially for heat-sensitive food, agricultural and biomedical products;
Waste heat recovery and heat demand occur simultaneously;
Reduced global (electric and fossil) energy consumption; and
Short pay-back periods.
Among the limitations of drying heat pumps compared to basic hot air convective driers
can be maintioned:
2 Vasile Minea
 Higher initial capital cost;

 Higher maintenance cost due to need to maintain compressor, refrigerant filters and
charging of refrigerant;
 Possible leakage of refrigerant;
 More complex operation;
 Additional floor space required;
 Requirement for competent design engineers and operating technicians.
Although heat pump-assisted drying technology is generally considered as a mature

technology being used during the last 2–3 decades, its industrial application has not been as
widespread as it should have been, particularly in the field of drying agro-food and in wood
industries [3]. Even where the requirement for very short payback periods has been met, heat
pumps have not so far been installed. Among others, some reasons for their neglect were the
following:
 Uncertainty by potential users as to heat pump reliability,

 Lack of good hardware in some types of potential applications,
 Lack of experimental and demonstration installations in different types of industries,
 Lack of required knowledge of chemical engineering and heat pump technology in
target industries, and
 Relative cost of electricity and fossil fuels affecting the commercial viability of
drying heat pumps.
Another problem consists in frequent errors and/or mistakes spread by a large number of
research studies involving drying heat pumps. These errors mainly concern the heat pump –
dryer integration and control, as well as the correlation between the product moisture content
and the heat pump dehumidification rate. Inadequate integration and wrong operation
parameters of the drying heat pump may cause troubles such as high discharge pressures, low
dehumidification efficiency, and even compressor mechanical damages.
Most of these obstacles can be overcome by further R&DD activities aiming at
developing cost-effective and reliable drying heat pumps. Optimal integration of driers and
heat pumps remains a challenging R&D task. Wider dissemination of existing and coming
knowledge may also contribute to avoid future failed applications.
This contribution provides general information on drying heat pumps, and focuses on
laboratory and industrial-scale wood drying applications. A number of design and control
mistakes found in the drying literature are discussed and some solutions are proposed. The
main scope is to avoid undesirable operation parameters, provide safe operation conditions to
the heat pump components, improve the system‘s overall energy performances and accelerate
future successful implementations of industrial drying heat pumps.
Industrial Drying Heat Pumps 3
2. DRYING PROCESS
2.1. Generalities
Drying is a highly nonlinear coupled heat and mass transfer process. It occurs by
vaporization of a liquid by supplying heat to a wet material. Drying with hot air implies
humidification and cooling of the air in a well-insulated (adiabatic) drying chamber. Heat is
transferred by convection from drying air to the product drying surface where heat is needed
for evaporation of moisture. Figure 2.1 shows a schematic representation of energy balance of
a drying system. The material enters the drying chamber at state a with an initial mass
M dry  M wet Ta . After drying, the dried material leaves the drying
and temperature
M dry T
chamber at state b with the final mass and at temperature b . On the other hand, in the

kg /s
drying chamber enters relatively dry air (at state 2) with the flow rate m air ( dry,air ). It
absorbs water from the material during the isenthalpic process 2-3. If the drying air is heated
at constant absolute humidity before entering the drier, the heating thermal power (kW) will
be:
 
Q  mdry, air (h2  h1 ) (2.2)
where h1 and h2 are the air mass enthalpy ( kJ / kg ) entering and leaving the air heater,
respectively.
The maximum rate of water extracted from the material will be:
  
M wet  mdry, air ( 3   2 )  mdry, air ( 3   1 ) (2.3)
where  1 , 2 and  3 are the absolute humidity of the drying air during the process (
kgwater / kgdry, air
). If c is the specific heat of the material, the energy balance of this
simplified drying process will be:

 
 
mdry, air h3   mdry, air h2  M wet cTa   M dry cTa  Tb 
  (2.4)
Heat supplied at the boundaries of the drying product diffuses into the material primarily
by conduction. The heat and mass transfer coefficients which control the heat and moisture
migration rate inside the product are strong functions of the moisture content as well as of the
product temperature. Moisture content represents the weight of water present in the product
4 Vasile Minea
expressed as a percentage of the weight of over-dry product. As the moisture content at the
drying surface of the product drops, liquid moisture migrates from the product bulk to the
drying surface before it is removed by the hot drying air. Local moisture diffusivity controls
liquid moisture migration rate from bulk to the drying surface of the product. The transport of
moisture can occur by one or a combination of several mechanisms as liquid (if the wet
material is at a temperature below the boiling point of water) and vapour diffusion (if the
water vaporizes inside the material), hydrostatic pressure differences (when internal
vaporization rates exceed the rate of vapour transport through the solid to the ambient
medium) or Knudsen diffusion (in the case of freeze drying) [1].
3
Exhaust
air T3
g
Material
J/k
inlet
Absolute humidity, kg/kg

y,
Mdry + Mwet Mdry
alp
Tb
nth
a b 3
D
se
ry
in
Ta
s
Material
g
outet Ma
1 Heating 2
2
T2 Drying air
mdry, air
1
Air heater T1
T1 T2
Dry-bulb temperature, °C
Q
Figure 2.1. Simplified schema of a drying system.
For materials of high diffusivity, moisture migrates rapidly to the drying surface even at
low moisture content of the product. Therefore, the drying rate of such materials is controlled
by external transport rates while the drying rate improves with improvement of the external
conditions. Evaporation of liquid moisture takes place from exposed surface by absorbing the
heat of vaporization Hot air is used both to supply the heat for evaporation and to carry away
the evaporated moisture from the solid. At low moisture content of the product, the effect of
low relative humidity becomes less significant. At low relative humidity of air, partial vapour
pressure of drying air becomes low. This leads to a higher driving potential for mass transfer
resulting in increased moisture evaporation rate.
Liquid moisture evaporates from the drying surface of the product because of the
difference of partial vapour pressures between the air and the surface of the product. Partial
vapour pressure at the drying surface is function of the drying surface temperature and its
water activity which is a function of surface moisture content and temperature. Water activity
(
 w ) depends on relative pressure and is defined as the ratio of the partial pressure (p) of
water over the wet solid to the equilibrium vapour pressure (

p w ) of water at the same
temperature. Thus,
 w , which is also equal to the relative humidity of the ambient humid air,
is defined as:
p
w 
pw (2.1)
As the moisture content and temperature of the drying surface continuously change
during drying process, partial vapour pressure at the drying surface becomes an
uncontrollable parameter. The diffusivity at the exposed drying surface significantly drops
during drying process particularly when the moisture content of the product becomes low.
The moisture content of a wet material in equilibrium with the air of given relative humidity
and temperature represents the equilibrium moisture content. The evaporation rate of moisture
depends on the mass transfer coefficient of the drying air which depends mainly on the types
of flow used such as stagnant, laminar or turbulent. Higher volume flow rate if drying air
increases the mass transfer coefficient, but the size of blower becomes large and power
consumption rate increases. Therefore, maintaining optimum flow of drying air is important
for the economic operation of a dryer. The isotherm obtained by exposing the solid to air of
decreasing humidity is known as the desorption isotherm. As the maximum allowable
temperature of the drying air is limited particularly for heat sensitive materials, partial vapour
pressure of the drying air is usually controlled by condensing moisture of the drying air by
using heat pumps.
2.2. Drying Schedules
Drying schedules are expressed in terms of dry-bulb temperature and either wet-bulb
depression and equilibrium moisture content, or both. Because the relative humidity does not
directly indicate the drying capacity of the kiln atmosphere, this parameter is determined by
use of dry- and wet-bulb temperatures measured by psychrometers. Relative humidity is a
measure of the amount of water vapour in the air, expressed as a percentage of the total
amount contained in saturated air at a given temperature and pressure. Increasing the
temperature of the air without adding more moisture will cause the relative humidity to
decrease.
To reduce the energy consumption per unit of dried product, it is necessary to reduce the
drying time. To reduce the drying time and obtain the desired final product moisture content
and quality, different time-dependent drying schemes, as intermittent, cycling and air reversal
drying are used. With food products as corn, peanuts, maize, wheat, drying and rest periods of
few seconds to few minutes provide shorter effective drying times, thermal energy savings,
higher moisture removal rates and lower product surface temperature.
The intermittent profiles are prescribed by raising the inlet ait temperature for a defined
period and dropping the air temperature back to its original level until the periodic cycle
interval. The intermittency i is defined as the fraction of time during which the inlet air
temperature is raised (
 on ), to the cycle time:
6 Vasile Minea
 on  on
i 
 cycle  on   off
(2.5)
In the case of drying heat pump it is not possible to use too short off periods during
intermittent cycles because the heat pump‘s overall efficiency can be significantly affected.
The transient regimes of heat pumps generally are of at least 10 minutes, before permanent
regime is achieved.
2.3. Control of Driers
Modern drying technologies make it imperative to found solutions allowing optimizing

the processes with respect to complex quality factors of products. The drier control systems
are important for energy efficiency since a poorly controlled process is likely to be wasteful,
in terms of energy, throughput and quality of the product. It is also likely to have a reduced
throughput and to produce inferior quality product. Theoretically, a large number of
parameters can be measured and controlled in a dryer. However, most practical control
systems limit themselves to one or two outputs and one or two inputs. The most important
inputs to a dryer are the heating, air flow and material feed rates. The most common
disturbance inputs are the ambient air and humidity and feedstock moisture content and
composition.
The most commonly used outputs are the moisture content of the dried material,
temperature and humidity of exhaust or re-circulated air, and a measure of the product
quality, as discoloration, etc. The moisture content of the dried material is often difficult to
measure. With certain assumptions, it can be derived from the temperature and humidity of
the exhaust air. Different types of control systems are available. Combinations of line-sensors
and expert systems with feedback response allow immediate quality-related decision to be
made. Sensors are placed in strategic locations to measure real-time quality parameters and
the signals are fed to expert systems using software. Drier control systems are designed to
change the moisture content of the product while minimizing the drying time and energy use,
improving the product quality and reducing the maintenance costs. Since no control system
can act directly on moisture content, drier operation is a function of parameters as product
moisture content, air dry and wet-bulb temperatures, and velocity. In drier control systems,
proportional integral derivative (PID) regulators control the heating, ventilation and
humidification/dehumidification devices. The control loops provide automatic temperature
maintenance at the product stack entry point. A comparator determines the difference
between the actual and the target temperature set by the drying schedule. The most of product
quality degradation is mainly due to the thermal effect of the drying air. It is thus possible to
reduce these quality effects (surface cracking, nutrient degradation) through a proper
feedback system to regulate the air and/or product temperature. To capture the product
surface temperature can be used thermo-vision cameras or hypodermic thermocouples.
2.4. Drying Efficiency
Due to affinity of most materials to water and high latent vaporization heat of water,
drying is highly intensive in energy. The energy consumption for thermal drying with
conventional convective-type dryers is typically high. Contacting efficiency between the
products and the drying medium determines the drying performance. Moreover, when heat
losses exist through the air vented without recycle or energy recovery, overall efficiency
decreases. A good insulation and reduced leakage of the drier minimize heat losses and
increases the overall energy efficiency. The performance of a dryer also depends on its
geographical location, ambient conditions, controls, maintenance, etc. Selection of energy
efficient dryer depends on the physical properties of the product and the ratio of fossil fuel
and electricity costs.
Low energy performance of drying processes results in adverse environmental impact.
Because the oil price has increased in recent years, the drying energy costs have escalated as
well. The forecast rising energy costs may lead to a carbon tax which will add additional
burden on industry. The solution is to develop highly energy-efficient technologies to reduce
the energy consumption and mitigate the environmental impacts.
The energy efficiency of a drying process can be expressed by the following equation [2]:
Eevap

Etot (2.6)
Where:
Eevap
is the energy used for water evaporation, and
Etot - total energy supplied to the dryer.
Because at low humidity and temperatures, energy is proportional to temperature

difference, the energy efficiency may be approximated to thermal efficiency:
Tin  Tout
T 
Tin  Tamb (2.7)
Where:
Tin  Tout is the difference between inlet and outlet temperatures, and
Tin  Tamb - difference between inlet and ambient temperatures.
The maximum thermal efficiency is defined as the ratio of the highest temperature
difference between inlet (

Tin ) and outlet temperature to difference between inlet and ambient
Tamb
( ):
8 Vasile Minea
Tin  Twb
 max 
Tin  Tamb (2.8)
It is obtained when exhaust is saturated at wet-bulb temperature (Twb).
3. INDUSTRIAL DRIERS
Drier selection approach has to include process specifications and economic evaluations.
Are required quantitative information, as the mode of feedstock production, drier throughput
and variability, and product quality parameters as physical, chemical and biochemical. At the
dried inlet, the products can have different forms as solid (wood, ceramics), granular,
particulate, suspension, solutions, sludge, crystalline, liquid, etc. Some materials involve
special requirements as corrosion, toxicity, flammability, fire hazards, color, aroma, etc.
Location of the moisture (near surface or distributed in the product), nature of moisture (free
or strongly bound to solid), mechanisms of moisture transfer, physical size of product,
conditions of drying medium (e.g., temperature, humidity, flow rate of hot air for convective
dryers), pressure in dryer, as well as the operating conditions also influence the selection of
the best drier. Location and average weather, drying kinetics, moist solid sorption isotherms,
drying curves, effect of process variables, changes in operating conditions that may affect the
quality of the product, as well as inlet and outlet moisture contents have to be determined.
Final value of the product, need to automatic control, type and cost of fuel, cost of electricity,
environmental regulations, safety aspects (fire and explosion hazards, toxicity) and available
footprint of drying system and accessories in the facility have also to be known to avoid
errors in selecting the driers [2].
Depending on heat input type, driers can be continuous or intermittent, adiabatic or non-
adiabatic, and operate by convection, conduction, radiation, electromagnetic fields, or
combinations.
The convective dryers are the most common. About 85% of industrial driers are of this
type despite their relatively low thermal efficiency caused by the difficulty in recovering the
latent heat of vaporization contained in the dryer exhaust air in a cost-effective manner. Hot
air is generated by indirect heating or direct firing. Superheated steam can also be used. As
primary energy sources, oil, natural gas or electricity are used. Combustion gases can be used
when the product is not heat-sensitive or affected by the presence of combustion products. In
direct dryers, the drying medium contacts the material to be dried directly and supplies the
heat required for drying by convection. The evaporated moisture is carried out by the same
drying medium. Drying hot air or gas temperatures may range from 50°C to 400°C depending
on the material. Dehumidified air can be needed when drying highly heat-sensitive materials.
Industrial driers can operate at atmospheric pressure or below it (vacuum).
Drying temperatures can be below or above the water boiling temperature, or below the
water freezing temperature. According to the relative flowing between the drying air and the
drying products, driers can be in co- or counter-current, or mixed. Several drying stages can
also be used in more complex industrial processes. Finally, the product residence time inside
the drier can be short (less than, for example, one minute), medium (i.e., between one minute
and one hour) and long (more than one hour). Each type of dryer has specific characteristics
which make it suitable or unsuitable for specific applications. Certain types are inherently
expensive (e.g. freeze dryers) while others are inherently more efficient (e.g., indirect or
conductive dryers). Thus, it is necessary to be aware of the wide variety of dryers available,
as well as their special advantages and limitations.
Solar-assisted heat pump dryers can be used in regions where solar energy is abundant in
order to further improve the energy efficiency of overall drying systems. Instead of using
conventional heating systems to provide auxiliary heating, the solar energy can be stored in
phase-change materials for discharging sensible energy to the drying air. The solar-assisted
heat pump dryer may operate at higher drying temperatures, is environmental-friendly process
and is relatively easy to implement and control. Such a system offers the flexibility of
operating with the heat pump, solar system or with both, and consists of solar collectors,
blowers, phase-change storage tanks, air ducting and valve. However, higher capital costs are
incurred for additional solar panels, blowers, storage tanks, ducting, valves and controls. The
amount of stored solar energy is greatly subjected to the weather conditions. The
implementation of such a system is cost-effective if the average annual sunshine is greater
than 2600 hours per year and if the annual total quantity of radiation is more than 6x106kJ/m2.
The products can be dried in batch or continuous modes, and they can be in stationary,
moving, agitated or dispersed state inside the dryer.
In batch driers, the material to be dried is introduced on trays inside the drying enclosure.
Batch dryers can utilise cross- or through circulated convective and contact heating. In
continuous driers, a series of trays is moved slowly through a heated tunnel. Drying takes
place in a current of warm air. Conveyor driers consist of perforated metal conveyor belts
passing through an oven. Most solid is fed onto the conveyor from an extruder, and heated air
is forced through the bed of solid on the conveyor as it passes through the oven. Conveyor
dryers operate continuously and are particularly suitable for solids requiring gentle physical
handling. Heat loss sources of these dryers are the dryer air leaks and exhausts, and the
typical specific energy consumption rages from 0.945 – 4.32 kWh/kgwater.
Spray dryers consist of a vertical cylindrical or conical chamber into which a liquid or
slurry is spayed using a rotating wheel or nozzle atomiser. Droplets from the atomiser come
into contact with hot air. The dried solid is separated from the exit air in a cyclone. Spray
dryers operate in continuous mode and are suitable for heat-sensitive materials.
Rotary dryers consist of a rotating cylinder inclined a few degrees to the horizontal.
Rotation speed is generally in the range of 5-20 RPM. The cylinder is rotated by small
electrically driven rollers on which it rests. There are various types of indirectly heated rotary
dryers. The most common is the steam tube dryer having a bundle of steam-heated tubes
inside the dryer cylinder itself. Another type of indirectly heated rotary dryer has a heated
jacket around the dryer cylinder. Rotary dryers usually operate in continuous mode. The
material to be dried is fed in at the higher end of the cylinder while air is drawn in by an
induced-draught fan and moves either co-current or counter-current to the solid. The air is
heated either by a heat exchanger or directly by a flame. The specific energy consumption of
spray and rotary dryers varies between 0.8 and 3.24 kWh/kgwater.
Fluidised bed dryers consist of a vessel in which air is blown through a perforated plate
above which lie a bed of solid particles. The air-flow rate is such that the solids become
suspended in the upward flow of air. Fluidised bed dryers can operate in batch or continuous
mode, and are mainly suitable for drying granular solids. In batch mode, the dryer is operated
until a desired level of moisture has been achieved. In continuous mode, fresh material can be
10 Vasile Minea
fed from above and dried material removed from a suitable point on the side. The main heat
loss source is the exhaust air, and the specific energy consumption is of 0.945 – 2.16 kWh/
kgwater. Fluidized bed drying is applied for drying products as granular solids in the food
industry, ceramic, pharmaceutical and agricultural industries. This method has high drying
rates due to excellent gas-particle contact leading to high heat and mass transfer rates. Smaller
flow area is required compared to other drying technologies as rotary, conveyor or tunnel.
The thermal efficiency is higher, and the capital and maintenance costs are lower compared to
rotary dryers. However, the power demand is high due to the need to suspend the entire bed in
gas phase leading to high pressure drops. There is high potential of attrition, and in some
cases, of granulation and agglomeration. The flexibility is low and there is a potential of de-
fluidization if the feed is too wet. The dyer chamber receives wet material and discharge dried
product through the product inlet and outlet ducts.
3.4. Heat Recovery Methods
Dryers are usually one stage of complex manufacturing processes. Energy efficiency
measures, including drying controls, for a drying operation firstly consist in identifying the
parameters which minimize the heating requirement. Because energy efficiency requires
capital investment, it can be considered attractive if the resulting cost savings are sufficiently
high. The most common form of heat recovery option on a dryer involves the use of waste
heat to pre-heat the inlet air to the dryer. This is used in industries as food, ceramics,
chemicals, wood, paper, laundries and textiles. There are cases where heat is recovered
elsewhere in the process and transferred to the drying.
In drying using electric heating, the key energy savings measures may include infra-red,
microwave and dielectric technologies. These technologies can be highly efficient and cost-
effective in certain situations because of their ability to deliver the heat directly to the
moisture to be evaporated, thereby avoiding the losses incurred through heating large
volumes of air. Direct heating involves the use of hot combustion gases directly from a gas or
oil burner, usually diluted with fresh air, as the drying medium. It improves efficiency by
avoiding the losses associated with boilers, steam distribution systems ani heat transfer
equipment.
Heat recovery is one of the most important energy saving measure for any drier. It refers
to any operation which heat from the exhaust air or the product is transferred to the input air
or the dried material. Heat can be recovered from exhaust air either by a heat exchanger or by
recycling part of the exhaust and mixing it with fresh input air. Heat recovery is applied to a
wide range of dryers in a wide range of industries. These include batch, fluidised bed, rotary
and spray dryers in food and drink, and chemical industries, tumble dryers in textiles industry
and paper machine drying in paper industry. Heat recovery can be applied by several
methods: exhaust air recycle, plate heat exchangers, run-around coils, heat wheels, heat pipes,
phase changing materials, special membranes and heat pumps.
Exhaust air recycle, used on both batch and continuous, recovers a part of the enthalpy of
the hot and humid exhaust air. It is important for drying processes operating at high air
velocity and short contact time between the drying medium and the material.
It involves taking a proportion of the exhaust air and mixing it with fresh heated drying
air before it enters the dryer itself. Such a system has the advantage that no heat exchangers
are required and so the capital cost is less. The problem is that in most industrial operations,
the humidity if the exhaust air is well below its equilibrium value in relation to the moisture
content of the material being dried. This means that it has removed less water from that
material than it could have done and that more energy has been used to heat the air than was
necessary. It is not possible to achieve 100% of equilibrium humidity because the rate of
drying is proportional to the difference between the equilibrium and actual humidity: the
smaller the difference, the slower the rate of drying. There are many ways in which this
problem could be handled. The most obvious is to increase the efficiency of mass transfer
between the material to be dried and the drying air so that the air picks up more moisture from
the material. It may be possible to do this by maximizing the area of contact, or by increasing
the length of the dryer. However, increasing the length of the dryer substantially increases the
cost. Another approach could be to reduce the flow rate of air through the dryer. Because this
technology reduces the throughput, it is not usually an option.
Plate heat exchangers consist of an array of parallel thin metal plates which separate the
two streams. They are relatively low cost although not particularly compact.
Run-around coil is used when it is not possible to transfer heat directly between the
exhaust and inlet air streams on a dryer, usually because they are too far apart and ducting
would be too expensive. In such cases, heat can be recovered using a secondary heat transfer
medium. The heat transfer medium is usually a mixture of water and antifreeze. The heat
exchangers are usually tubular recuperators with liquid the liquid on the tube side. The
advantage of the run-around coil is that expensive air ducting is replaced by lower cost small
diameter pipe. Space constraints are therefore not a problem, and the cost can be less.
However, the heat transfer efficiency is less because two heat transfer steps occur in series.
Heat wheels consist of rotating matrices through which hot and cold streams pass
alternately. They are most commonly used in high temperature applications. The main
disadvantage for drying is that they do not cope well with entrained particles.
If the exhaust air of a dryer is cooled below its dew point, the vapour will condense,
releasing its latent heat of vaporisation. Since this latent heat represents the majority of the
heat input to the dryer, its reuse is essential of major improvements in energy efficiency are to
be achieved. However, this heat is usually released at too low temperature for much of it to be
cost-effectively transferred to the input air. Other energy savings measures are mechanical
dewatering, insulation and use of superheated steam as a drying medium.
An alternative method of dehumidification drying is to use dessicant wheels. It is similar
to a heat wheel heat exchanger, but which consists of a powerful dessicant material. One half
of the wheel passes through the moist air, removing water from it, and the other half is heated
by a gas burner, driving off its absorbed moisture. The dehumidified air remains as hot as it
was when it emerged from the dryer and can be recycled directly to the dryer inlet.
A particularly simple and cost-effective form of heat pump is mechanical vapour
recompression. Although not applicable to most dryers, it can be used on those where
superheated steam is used as drying medium.
12 Vasile Minea
The following section discusses some aspects of heat pumps utilization as efficient heat
recovery and energy saving devices in industrial drying processes.
3.5. Heat Pumps
In the common thermodynamic sense, a heat pump is a system in which refrigeration

components (compressors, condensers, evaporators and expansion valves) are used in such a
manner as to take heat from a source (air, water, ground, etc.) and give it up to a heat sink
(air, water, ground, etc.) that is at a higher temperature than the source. Closed compression
cycle heat pumps are theoretically based on the ideal Carnot cycle. This type of heat pump is
mostly used to recover waste heat at relatively low temperatures, and to upgrade it for process
temperatures between 50°C and 75°C. The most common type of compressor for closed-cycle
is the electric compressor which achieves an efficiency of above 90%. Rotary and screw
compressors are most suitable for large systems.
3.3.1. Principlup to state 1 before entering the compressor. At this point, the cycle is
repeated.
Heat sink
(Tsink) Condenser
3 2'
2 pcd
4 Expansion 2 Compressor 3 Tcond 2' 2s
4 Tsink
valve
T, K
pev
1 Tsource
Evaporator Tevap 1
5
5 s, kJ/kgK
Heat source
(Tsource)
Figure 3.1. Principle of heat pumps and thermodynamic cycle in T-s diagram.
3.5.2. Advanced cycles

Refrigerant circuits of heat pumps can be modified in order to minimize the irreversibility
and improving the cycle overall efficiency. It is well known that the vapour generated during
the expansion process (4-5) does not contribute to the cooling capacity of the evaporator.
However, the compressor has to compress this vapour. The role of the flash tank (Figure 3.2a)
is to extract a portion of this vapour at an intermediate pressure level, sends it to the
compressor and reducing the compressor power input. So, this vapour is compressed from
that intermediate pressure to the condensing pressure avoiding the expansion portion to the
evaporator pressure level. The loss of compression work is thus avoided.
On the other hand, the isenthalpic expansion process that reduces the pressure of the
refrigerant from the condenser level to the evaporator level is highly irreversible. Instead the
expansion valve, a work-producing expansion device can be employed (Figure 3.2b). The
work output of this device (expander) can be supplied to the compressor to reduce the
compressor power input. The expander reduces the power requirement of the compressor and,
at the same time, increases the evaporator cooling capacity. This last benefit is available even
when the work produced by the expander is not utilized. By using both benefits, the overall
efficiency gain is of about 5%.
CD CD
EX1
Flash tank C Expander C
Vapour Power
EV EV
EX2
(a) (b)
Figure 3.2. Heat pump vapour compression cycles with flash tank (a) and expander (b).
C2 C1
C2 C1
CD EV CD EV
Evaporator/
EX1 EX2 condenser
Inter-cooler EX2 EX1

(a) (b)
Figure 3.3. Two-stage (a) and cascade (b) heat pump closed compression cycles.
The two-stage closed compression cycle with one evaporating temperature (Figure 3.3a)
aims also at increasing the cycle overall performance. A flash intercooler allows increasing
the proportion of vapour to be compressed by using the superheat of the vapour coming from
the first stage compressor (C1) to evaporate some of the liquid working fluid from the first
expansion stage. The disadvantages of this cycle are the pressure drop in the intercooler and a
risk of entrainment of liquid drops in the second-stage compressor. When large temperature
lifts are needed cascade cycles are used (Figure 3.3b). They allow using different working
fluids at each stage and reasonable pressure ratios to be achieved in each compressor.
3.5.3. Working fluids

Selection of refrigerants as working fluids for heat pumps concerns their thermodynamic
properties and environmental impacts. The refrigerants are generally categorised in short-
14 Vasile Minea
(HCFC) and long-term (HFC) alternatives, as well as mixtures and pure fluids. Mixtures can
be azeotropic, near-azeotropic or non-azeotropic. Azeotropic fluids evaporate and condense at
a constant temperature, while non-azeotropic mixtures undergo the phase changes at varying
temperatures. The temperature change at the phase change is known as the temperature glide
of the mixture. If the glide fits with the temperature change of the source/sink, the glide will
contribute to a higher COP. Ozone depletion is due to a complex reaction with chlorine
compounds, such as hydro-chlorofluorocarbons (HCFCs) that are used as working fluids in
heat pumps. A useful parameter for estimating the impact of substances on ozone destruction
is the ozone depletion potential (ODP). It is a measure of the destructive potential of a
particular substance, relative to depletion caused by an equal amount of a reference substance.
CFC-11 is typically defined as the reference compound, and is assigned an ODP of 1.0. On
the other hand, the global warming potential (GWP) is an index that quantifies the capability
of chemical substances to absorb infrared radiation relative to carbon dioxide. The direct and
indirect capacities to absorb different wavelengths of infrared radiation, the residence time in
the atmosphere and the time period over which the effect on radiation are factors contributing
to this index . Since certain molecules used as working fluids in heat pumps absorb radiation
emitted by the earth, the greenhouse effect accelerates. As low-temperature refrigerants for
heat pumps currently are used, among others, HFC-134a and HFC-410A (Table 3.1), and as
high-temperature refrigerants, R-236fa and R-245fa (Figure 3.4).
The type of electricity production has also an indirect impact on greenhouse effect.
Production of electricity for powering electrical plants is associated with emissions of CO2
and has to be considered for a total picture of the global warming impact. Local conditions for
power generation vary considerably from one country to another (Table 3.2). The power
demand for a specified heat output depends on the efficiency of the heat pump. This implies
that indirect emissions can be reduced by selecting more efficient units.
Table 3.1. ODP and GWP values for some refrigerants.
Refrigerant ODP GWP (100)

HFC-134a 0 1 300
HFC-410A 0 1 900
R-236fa 0 6 300
R-245fa 0 950
R-236fa - saturated temperatures R-245fa - saturated temperatures

140 180
160
120
140
100
Temperature, °C
Temperature, °C
120
80 100
60 80
60
40
40
20
20
0 0
0 500 1000 1500 2000 2500 3000 0 1000 2000 3000 4000
Pressure, kPa,a Pressure, kPa,a
a b
Figure 3.4. Saturated temperatures of two high-temperature refrigerants; a) R-236fa; b) R-245fa.

Table 3.2. CO2 emissions associated with electricity production (indirect emissions).
Type of production CO2 emission

- kgCO2/kWhel
Coal-fired plant 1 000
Oil-fired plant 700
Natural gas-fired plant 450
Hydraulic and nuclear plants 0
3.5.4. Design of heat pumps

The heat pump evaporator and condenser generally are finned refrigerant-to-air heat
exchangers. Their heat transfer surfaces are calculated using the following equation:

Q
S
U * LMTD (3.1)
where Q is the thermal capacity of the designed heat exchanger.

The thermal capacities of evaporator and condenser (kW) are respectively calculated
from energy conservation equations with refrigerant-side enthalpy changes and flow rates, or
with the air-side temperature changes, mass flow rates and specific heat:
 
  
Q EV  EV mEV , air c p, air Ti  Ta   ha  hb    mR h1  h5   UAEV LMTDEV
(3.2)
   
Q CD  CD m R h2  h4   mCD,air c p ,air Td  Tm   mCD,air hd  hm   UACD LMTDCD
(3.3)
where τ is the operating time.

The overall heat transfer coefficients of the evaporator and condenser are expressed as:
1
U
A f 1 A f ln r f / rint  1
 
Aint hR 2kL  f hair
(3.3)

where f is the temperature effectiveness of the finned surface Af [4]. The evaporator
refrigerant-side convective boiling heat transfer coefficient can be calculated with the well
known Liu and Winterton asymptotic correlation, and the refrigerant-side condensation heat
transfer coefficient, with the Shah‘s enhancement model [5].
For both evaporator and condenser, the logarithmic mean temperature difference is
defined as:
16 Vasile Minea
Tin  Tout
LMTD 
ln Tin / Tout  (3.4)
where
Tin is temperature difference between the hot fluid inlet and cold fluid outlet (
Thf ,in  Tcf ,out Tout - temperature difference between the hot fluid outlet and cold fluid
), and
T T h
inlet ( hf ,out cf ,in
). The air-side heat transfer coefficients ( air ) of flat fin evaporators and
condensers are calculated based on Webb and Gupte correlation (1992) depending on Graetz
number values.
The mass flow rate of reciprocating compressor is related to the volumetric displacement
rate according to the following equation:

 VC
mR  a
v1 (3.5)


Where a is the actual compressor volumetric efficiency, V C - compressor volumetric
displacement rate and v1 - specific volume of refrigerant the vapor entering the compressor.
The actual volumetric efficiency is less than the theoretical volumetric efficiency due to
pressure losses in the compressor valves and the heating of suction gaz. The isentropic
efficiency of a compressor is defined as follows (Figure 3.1):
h2  h1
s 
h2 s  h1 (3.6)
Where the compressor discharge point 2s is defined by

s1  s 2 s and p2  p2 s . The
isentropic efficiency varies with the pressure ratio (   p2 / p1 ) as follows:
 s  a0  a1  a2 2 (3.7)
The compressor electrical power input (kW) is determined by the conservation of energy:
  
WC  m R (h2  h1 )  Q CD  Q EV (3.8)
The thermal power rejected by the compressor is a function of the electrical efficiency of
the compressor motor (

 cm ) and is determined by:
 1
Q C  WC (  1)
 cm (3.9)
The expansion device EX can be a thermostatic, electric or electronic valve. It controls

the amount of superheat of the refrigerant within the evaporator. The liquid receiver LR is an
inactive component for the thermodynamic cycle.
3.5.5. Performance of heat pumps

The energy performance of electrically-driven heat pumps is dependent upon the degree
of exhaust or re-circulated air cooling/dehumidification required and on the temperature lift
needed. The maximum amount of heat which can be recovered depends on the ambient
temperature (-30 to 30°C), required dryer air inlet and exit temperature and humidity. The
reported energy saving levels through heat recovery from heat pumps vary from 10 to 50%.
For heat recovery, the energy saving is equal to heat recovered, minus any additional fan
power to overcome increased pressure drop. The fan power as a function of pressure drop is
given by the formula:

Required power  V p / f (3.13)
Where:

V is the volumetric flow tare,
Δp – the pressure drop,
f - efficiency factor (typically 80-85%).
The majority of the electrically driven heat pump heat recovery systems are
dehumidifiers where the latent heat within the exhaust is removed by a refrigerant which is
then compressed and the heat released to the inlet air. The energy performance of electrically
driven heat pumps is dependent upon the degree of exhaust or re-circulated air cooling
required and on the temperature lift needed to preheat the inlet or re-circulated air.
A closed compression cycle based on the ideal Carnot cycle operates between the
temperatures T1 (heat source) and T2 (heat sink). The coefficient of performance for the ideal
Carnot cycle is defined as maximum theoretical efficiency:
T1 Tcond
COPCarnot  
T1  T2 Tcond  Tevap
(3.14)
Where:
Tcond is the condensing temperature (K), and
Tevap
- evaporating temperature (K).
18 Vasile Minea
The actual COP of a closed compression cycle is defined as the ratio between the useful

Q
thermal power output at the condenser ( CD ) and the electrical power input at the

compressor and blower ( W C  B ):
 
Q CD Q CD
COP  
  
(3.15)
W CB Q CD  Q EV
The actual COP of heat pumps is usually 40 to 50% of the COP of the theoretical Carnot
cycle. Consequently, the COP of the actual cycles is expressed as:
COP  CarnotCOPCarnot (3.16)
where
 Carnot is the Carnot efficiency.
4. DRYING HEAT PUMPS

In the 1970s and 1980s, the drying industry promoted dehumidifier concepts, but the
performance of such systems was often disappointing due to air flow and control problems,
inadequate dehumidifying capacity and inappropriate kiln structures [3]. The reliability of the
systems was often low and equipment suppliers did not provide enough information on the
actual performance of their systems. However, the best solution that would allow heat pumps
to be effective as a means of drying would be to use them in combination with a traditional
energy source at high temperatures to obtain similar drying rates and energy savings.
The majority of drying heat pump systems is dehumidifiers and the closed compression
cycle, electrical motor-driven is the most used. The electrically-driven heat pumps benefit
from high fuel prices, due to their larger absolute savings at high fuel prices. High COPs are
also beneficial. Any convection-type dryer can be fitted with a suitable designed heat pump.
Batch shelf, kilns (for wood), fluid bed, tray and rotary dryers can be used.
They recover heat from the dryer hot and humid air by condensing out the water vapor,
which is then removed as liquid. Additional heat input may be supplied by convection or by
other sources (microwave, radio-frequency or infrared). The sensible and latent heat
recovered is used to reheat the dehumidified air. Heat pump dryers improve the quality of
dried high-value products by controlling the air humidity and temperature, substantially
reduce electrical and/or fossil energy consumptions, and may reduce the drying time. Since
heat pump drying is carried out in closed systems, smell from the drying products (food, etc.)
are reduced.
Heat pumps are used in industrial dehumidification and drying processes at low,
moderate and high temperatures, but not higher than 110°C. They Heat pump dryers operate
with high coefficients of performance, normally higher than 4.
For selecting a drying heat pump and integrating it to a drier, the first step is to identify
technical and economical feasibility of installation. Factors of importance are the
characteristics if heat source (temperature, load, phase and location), as well as maximum
sink temperature and lift temperature that are of about 75°C and 50°C, respectively. The ratio
of heat source and heat sink amounts is also important. To make full use of the heat pump, the
heat source should be little more than the same size of the sink by a factor of 10% to 50%.
Drying heat pumps remove moisture from the air circulated through the drying chamber.
They are efficient systems, impose fewer restrictions on materials and construction of the
drying chamber, and can be adapted to any enclosure which is reasonably air-tight, moisture-
resistant and insulated. Initial capital investment, maintenance costs and peak power
requirements are relatively low. However, with low-temperature heat pumps, drying times
can be longer and drying rates lower as compared to conventional driers. If located outside in
cold climates, driers must be well insulated to prevent heat loss during the winter months.
Also, it is necessary to use a supplementary source of heat to maintain effective drying
temperatures.
A back-up energy source (electrical or fossil) is used to bring the drier charge up to
operating temperature at the beginning of a drying cycle, after which the heat pump supplies
the majority of required heat to continue the drying process.
4.1. Principle
Figure 4.1 represents a heat pump integrated with a drying enclosure, as well as the
refrigerant and air thermodynamic cycles in p-h and h-T diagrams [6]. The refrigerant-side of
the heat pump consists of an evaporator EV, a compressor C, a condenser CD, sub-cooling
devices as a heat rejector HR and/or a sub-cooler SC, a liquid receiver LR and an expansion
valve EX. Air is drawn from the drying chamber and forced over the cold evaporator where
the moisture condenses and the water is drained out. The sensible heat removed from the air
and the latent heat of condensation, are both transferred to the refrigerant that vaporizes. The
refrigerant absorbs heat from the air and undergoes a two-phase change from a vapour-liquid
mixture (state 6) to superheated vapour (state 1a). The refrigerant superheated vapour at state
1a is further superheated inside the internal sub-cooler/super-heater SC up to the state 1, prior
entering the compressor suction line. The refrigerant vapour is then compressed by the
compressor and absorbs additional energy equivalent to the electrical input. The compressor
electrical energy input is converted into mechanical work to raise the refrigerant vapour
pressure. This work is transferred to the condenser for heating the drying air. By raising the
vapour pressure, the condensing temperature increases to a level higher than that of the dryer
temperature. As a high-pressure superheated vapour, the refrigerant passes through the
condenser where it condenses and transfers heat back to the drying air. Within the CD, the
refrigerant vapour first undergoes a change from superheated (2) to saturated vapour (2‖), and
then condenses (2‖-2‘) and sub-cools (2‘-3). The heat rejection process 3-4 occurring in the
HR coil aims at controlling the heat pump dehumidification capacity. Sometimes, the SC is
required to further sub-cool the liquid (process 4-5). The expansion valve EX expands the
liquid refrigerant (isenthalpic process 5-6) in order to reduce its pressure to the vaporization
20 Vasile Minea
value, below the heat source level. The two-phase refrigerant enters the EV at state 6, and the
thermodynamic cycle starts all over again.
On the air-side, the air cooling (i-a) and dehumidification (a-b) processes occur at the
evaporator EV. The air at state b leaving the evaporator is mixed with moisture-laden drying
air (state i). The mixed air (state m) is directed through the condenser and heated at constant
absolute humidity (process m-d) up to a temperature higher than the dry-bulb temperature of
the drying air. When required, the back-up heater supplies additional heat to keep the air dry-
bulb temperature close to the setting point. The drying air passes through the material stack
and picks up moisture according to the process d-i at constant enthalpy. The described
process recycles the heat to maintain the drying conditions in the drier, whereas in
conventional driers considerable quantity of heat is exhausted to the atmosphere through
venting of excess humidity.
4.2. Energy Performances
Both design and economic parameters influence the overall economic viability of heat
pump dryers. The heat pump energy efficiency is expressed by the coefficient of performance
(COP) defined as the total heat supplied by the condenser (expressed in kWh) divided by the
compressor and blower electrical energy consumption (also expressed in kWh):
QCD Q
COP    CD
EC  B W *
CB (4.1)

where W C  B is the electrical power input of the heat pump compressor and blower, and  -
the heat pump total running time (s).
Coefficients of performance below 4 are normally not acceptable, but COPs of 6 have
good possibilities for economically favourable installation. The annual savings are larger at
high COPs which imply that the rapid increase of the payback period starts at lower fuel
prices for a heat pump with high COPs. A performance indicator that is commonly used to
define the performance of a conventional drier is the specific moisture extraction rate
(SMER). It is defined as the ratio between the amount of water extracted from the dried
mwater Einput
material ( , expressed in kg) and the energy input to the dryer ( , expressed in
kWh):
drier
mwater
SMER  drier
(kg / kWh)
Einput
(4.2)
This parameter depends on the heat pump dehumidification capacity, the compressor and
blower power inputs, and the total water quantity removed from the dried material. In the case
of drying heat pumps, the modified specific moisture extraction rate (MSMER) defined as the
hp
ratio between the amount of water extracted from the dried material by the heat pump (
mwater
, expressed in kg) and the energy input to the heat pump (compressor and blower) (
EChp B ,
expressed in kWh):
hp
mwater
MSMER  (kg / kWh)
EChp B (4.3)
The specific energy consumptions of the drier (SEC) and heat pump (MSEC), defined as
reciprocals of SMER and MSMER respectively, can be used to compare energy efficiency of
different type of dryers and heat pumps. The performance indicator MSMER depends on the
annually running time of the dryer. Industrial applications require at least 2000 hours of
operation per year to provide low cost per litre of water removed. Because heat is extracted
from the drier humid air instead exhaust it outside, the heat pump driers offer higher
MSMERs. These efficiency indicators range between at least 1.5 and 4, while those of
conventional convection driers are of 0.12 to 1.3 for hot air convective driers, and 0.7 to 1.2
for vacuum driers.
Air
2 C
CD 2'’
1 SC
3
Blower
HR m
1a
Back-up
heater 5
b EV
LR EX
4 6 1'
d
LV
i
i
Air Air damper

Drying enclosure
Mass enthalpy h (kJ/kg)
hi = hd
ha
Absolute humidity, kg/kg
i
a d
2s 2 m
5 3 2' pCD 2'’
4
hb
Pressure
6 pEV b
1'
1
1a
Enthalpy (kj/kg) Tb Ta Tm Ti Td
Dry bulb temperature, °C
Refrigerant cycle in p-h diagram Air cycle in Mollier diagram
Figure 4.1. Schematic of a drying heat pump, and refrigerant-side and air-side thermodynamic cycles.
22 Vasile Minea
4.3. Payback Period
The payback period is defined as the total investment cost (TIC, expressed in $) divided
by the annual net savings in energy costs (ANSEC, expressed in $/year) due to its installation,
which in the case of a heat pump is the net saving in fuel or energy cost:
TIC
Payback period 
ANSEC (4.4)
The capital costs are higher than that of conventional dryers and require additional
maintenance (compressors, blowers, lubricant, air filters, etc). These costs include design,
project management, installation and commissioning, as well as compressors, motors, control
equipment, ductwork, fans, heat exchangers, procurement, fabrication and structure supports.
The total investment cost of heat pump driers is made up of fixed and variable costs. Fixed
costs are unrelated to the amount of moisture removed from the product over the years. The
drier cost is the primary element, but it includes also the electrical demand charges. Specific
installation costs are very site-specific and they decrease as the unit size becomes larger.
Thus, the absolute increase in payback period at, for instance, 50% higher installation costs
compared to the assumed values, will be low if the payback period is short, and high if the
payback period is long. It is important to decrease the total investment cost for a heat pump
installation, i.e. the heat pump itself, its installation and other associated costs, not just the
cost of the equipment alone. Because the payback period is directly proportional to
installation costs, it is important to decrease these costs. It is sometimes argued that the heat
pump component cost will decrease when the number of installations increases. Although this
is generally true, it must be noted that only a part of the total cost (normally, between 25%
and 50%) is related to the drying heat pump itself. The rest is due to installation, piping,
control devices, engineering, etc. It is thus important to decrease the associated costs, e. g. by
developing better knowledge of heat pumps, advanced compressors, heat exchangers, etc.
Energy prices also vary, and the cost of energy saving equipment is highly specific to
each application. The simple payback periods of initial investments are generally shorter if
more product moisture is available. Higher MSMER can be interpreted as lower operating
cost, making payback period for the initial capital cost shorter. To estimate the cost-
effectiveness of a drying heat pump, it must establish the input and output air temperatures
and humidity, and estimate the amount of heat which can be recovered. The payback period
of the electric motor-driven closed compression cycles is strongly influenced by the
electricity price and the COP, and by the temperature lift.
If it is assumed that the useful heat generated by drying heat pump replaces heat from an
existing boiler with thermal efficiency  B , the payback period can be also expressed using the
equation [2]:
1
Payback period 
 FEP EEP 
   * 8760  AMC
 B COP 
(4.5)
Where:
FEP is the fuel energy price ($/kWh),

EEP - electrical energy price ($/kWh),
COP - coefficient of performance of the heat pump,
AMC - annual maintenance cost of the heat pumps ($/kWh/year).
The operating costs of electricity-driven heat pumps increase as the electricity price rises,
so the payback period increases. The larger the electricity consumption, the greater is its
influence. Hence, the drying heat pumps having high COP will be affected to a lesser degree.
The fuel price also has an impact on the payback period due to its influence on the
absolute savings. The payback periods of the electrically-driven heat pumps are very sensitive
to fuel prices. At low fuel prices, the annual savings approach zero, and the payback period
increases rapidly. They decrease at higher fuel prices, due to the larger absolute savings at
high fuel prices, although the electricity price also increase (the ratio between electricity and
fuel prices is kept constant). A high fuel price also favours this type, due to the larger absolute
savings at high fuel prices. The payback period varies from industry to industry and from
country to country and has the advantage of focusing attention on the first years of the life of
the equipment. It depends on the type of drier. Acceptable payback periods are normally
between two and three years.
4.4. Optimization Requirements
The principal advantage of heat pumps dryers emerges from the hot and humid air as well
as from their ability to control the drying air temperature and humidity. However, it is
necessary to optimize the system components and design to increase the energy efficiency of
drying heat pumps.
4.4.1. Background
Optimum operation of drying heat pumps requires a number of components and
configurations to provide appropriate thermodynamic conditions. Furthermore, for a given
dehumidification capacity of the heat pump, it is mandatory to determine the required
quantity of the dried material to provide high energy performances. A brief review of
published studies shows what design features and/or information are missing to accurately
operate laboratory and/or industrial-scale heat pump dryers. It was observed that many
research studies on heat pump drying technology don‘t specify the heat pump dryer
configuration neither the quantity of dried materials when reporting high energy performance.
They include general statements like ―the drying conditions are controlled by adjusting the
capacity of the heat pump components”, without indicating ―how‖ capacity control was
performed in each particular case. Many works also agree that the heat pump technology
allows better control of temperature and relative humidity (RH) of the drying process.
However, few of them explain how the temperature and RH control is achieved to provide
reported relatively high dehumidifying performances [6]. A number of published
24 Vasile Minea
configurations, as shown in Figures 4.3a and 4.3b either use parallel condensers [7] or
refrigerant desuperheaters [8] to reject the system excess heat outdoors. The most significant
lack in such configurations is related to the air ducts across the heat pump evaporator and
condenser. As can be seen in both examples, the air flows at the same rate through the
evaporator and the condenser. Now this is not an optimum situation because a number of
problems may occur. For example, if the heat input (sensible and latent) or the air temperature
at the heat pump evaporator inlet are too low, or if there is poor air distribution through the
coil, the unit may trip off on ―low‖ suction pressure control. If the air flow rate through the
condenser is too low, the unit trips off on ―high‖ pressure control. Also, if the air temperature
at the condenser inlet is too high, as when the back-up steam or electrical coils are starting,
the compressor may shut-down on ―high‖ limit pressure. Another concept combining a heat
pipe heat recovery heat exchanger with a drying heat pump (Figure 4.3c) [9] is also
questionable. Because the heat pipe simultaneously reduces the heat pump‘s evaporating
temperature and increases the condensing temperature, the claimed 12%-20% reduction in the
heat pump energy consumption therefore seems rather unlikely. In addition, it could be that
the overall effect on the COP is not high enough to justify the additional investment for the
heat pipe device.
Heat rejection coil Heat rejection coil

Internal heat Air
Internal heat
exchanger exchanger
(condenser) (condenser)
EX C EX C
EV EV
F Air F
Dryer Dryer
(a) (b)
Heat pipe
Evaporation Condensation
EV C Air
Dryer EX
F
Back-up
heater CD
(c)
Figure 4.2. Published configurations of heat pump dryers.

Table 4.1. Short review of published studies on drying with heat pumps.
Initial Drying Moisture

Product Moisture Drying
product temperature extraction Drying heat pump
(reference) contents time
mass (RH) rate
- - Initial Final - - - Capacity  MSMER
WC
- kg % w.b. % °C (%) - hrs kW kW kg/kWh
w.b.
Rat liver n/a 90 20 n/a 0.75 %/h 95 n/a n/a 3.2-4.5
[10]
Cheese n/a 80 20 0; 4; 8; 12 n/a 120 n/a n/a n/a
(n/a)
Vegetable n/a 30 6 30 (55); 50 n/a n/a n/a n/a 3.1-4.5
seeds [11] (50)
Paddy [12] 1 200 n/a n/a 42 & 46 8 – 15.9 15.5 n/a 4.2 2.0
(26 & 14) kg/h
Cellulose n/a n/a n/a -15 & 20 (n/a) n/a n/a n/a n/a 0.28; 4
[13]
Shredded 200 90 16 40 (n/a) 6.3 kg/h 25 15 n/a n/a
radish
Sludge 2 - 3.5 46.8 5.42 60 (36) 0.71 5.16 n/a 0.17 n/a
kg/cycle
On the other hand, the required mass of the dried materials is not mentioned in several
studies, even when the initial and the final moisture contents are indicated (Table 4.1).
Experimental results as drying curve (i.e. moisture content vs. time) for rat liver [10] or
cheese and MSMER ratios (kgwater/kWh) as a function of the air temperature at the drying
chamber inlet, are given without specifying nor the material initial mass, nor the heat pump
dehumidification capacity or – at least – the compressor power input. Without this
information it is difficult, if not impossible, to validate the pertinence of reported drying
performances. It is also not indicated how the air relative humidity has been kept constant
respectively at 40% or 50% (inlet) and 60% (outlet), and how the moisture removal rate has
been controlled. In the case of cheese drying at low temperatures between 0 and 12°C it is not
indicated how the drying system has been stabilized before starting-up the heat pump. High
differences exists between the drying air temperature leaving the heat pump evaporator
(36.5°C) and the evaporating temperature 15°C [11] may prove low input heat to the
evaporator or lack of refrigerant charge.
Up to 54% higher COPs with two-stage trans-critical CO2 heat pumps compared to those
of single-stage drying heat pumps are also reported. That gives MSMER ratios up to 43%
higher, but the authors didn‘t specify the duration of drying cycles, quantities of vegetable
seeds dried and volume of the water removed, and the energy consumed. With much
complete information [12], as the compressor nominal power input, quantity and initial/final
moisture contents of the dried product (paddy), average drying time, total water volume
removed and electrical energy consumption, it is much easier to validate the reliability of the
experimental work and of the system drying performances. In other study [13], the sulphate
and sulphite cellulose seem being dried with an ammonia two-stage heat pump system, but
the schematic layout of the fluidized dryer represents a single-stage heat pump. Moreover, no
information is given about the quantity of the dried material, drying time, volume of water
removed, compressor input power, electrical energy consumption and the heat pump nominal
26 Vasile Minea
dehumidification capacity. In return, data on MSMER values at different drying temperatures

are provided. The concept of fluidized bed heat pump drying of bovine intestine [14] at
temperatures varying from -10 to 25°C also don‘t give information concerning the heat pump
cooling capacity and the material initial quantity and properties. Other works propose
dimensioning techniques for codfish heat pump dryers, including the parameters for both heat
pump and air drying circuits [15]. However, the air thermodynamic diagram represents a
mixing air process but isn‘t represented on the drier air flow diagram. Also, it is not specified
the refrigerant for which the dehumidification capacity and MSMER and COP parameters
have been achieved.
4.4.2. System integration

The selection of the drying mode depends on the product drying characteristics and the
required loading capacity. Heat pump driers can operate in batch and continuous modes
where heat pumps are compact units installed outside the drying chamber. In the batch mode,
the product is placed on a tray that is positioned in the drying chamber and removed once the
desired product moisture content is reached (Figure 4.4a). The drying air can flow parallel or
perpendicular to the product surface. Batch drying is generally suitable for smaller production
rates but provides higher labour costs. In the continuous mode, the product is placed on a tray
positioned on a conveyor belt system of which speed can be varied (Figure 4.3b). Continuous
systems involve faster loading and unloading of drying products and are less labour-intensive.
In the case of multiple drying systems involving two drying chambers with different
conditions of temperature and humidity, heat pumps with two evaporators can be used (Figure
4.4). Such multiple (two or more) evaporator refrigeration systems are commonly used in
commercial and industrial refrigeration applications. Two air streams with different drying
conditions, i.e. high and low temperatures and humidity come from two independent drying
chambers. The overall energy efficiency of the drying process can be thus improved. The sub-
cooled refrigerant liquid is split into two streams at the exit of the condenser. One stream
enters the expansion valve EX1 of the high-temperature evaporator EV1 and the second, the
expansion valve EX2 of the low-temperature evaporator EV2. Liquid refrigerant flows
through the thermostatic expansion valves to the low and high temperature evaporators.
Because there are two evaporating temperatures, a device must be used to keep one of the
evaporators at a higher low-side pressure. A two-temperature valve in the suction line keeps
the low-side pressure of the refrigerant in evaporator EV2 at a higher pressure than in the
evaporator EV1. The evaporator temperature is governed by the evaporating pressure. A
check valve is located in the suction line coming from the colder evaporator EV1. It prevents
the warmer, higher pressure low-side vapour from entering the colder evaporator EV1 during
the off cycle. The vaporized refrigerant is returned to the compressor where it is compressed
and becomes high-pressure and high-temperature superheated vapour.
To further enhance the performance of conventional heat pump driers, several other
technologies can be incorporated to enhance the drying rates while reducing the thermal load
of the heat pump itself.
Infrared-assisted heat pump dryers could be used for fast removal of surface moisture
during the initial stages of drying, followed by intermittent drying over the rest of the drying
process. Heat for drying is generated by radiation from infrared generators (Figure 4.5a). The
infrared driers present advantages as high heat transfer rates (up to 100kW/m2), easy to direct
the heat source to the surface, quick response times allowing easy and rapid process control.
Incorporating infrared into an existing heat pump dryer is simple and capital cost is low. This
mode of operation ensures a faster initial drying rate and offers advantages of compactness,
simplicity, ease to control and low equipment cost. The are possibilities of significant energy
savings and enhanced product quality due to reduced residence time in the dryer chamber.
LV HR
EX HR
LV EX
CD M
EV D1
CD M
EV D1
D2
C D2
F Condensate C
Hot & F Condensate
Air vent Fresh air humid air Hot &
Inlet wet humid air
product Product to be dryed
Dryer fan
Dry air
Drying air
Outlet dry
Product outlet Conveyor belt product
Product inlet
Product tray
(b)
(a)
C: compressor; CD: condenser; HR: heat rejection coil; EX: expansion valve; EV: evaporator; D: air
damper; M: damper motor; F: fan
Figure 4.3. Batch (a) and continuous (b) driers with compact drying heat pumps.
LV2 HR2
EX2
M
LV1 HR1
EV2
Condensate
Two-temp
valve
EX1
M
CD
C EV1
CV
SA Air vent
F1
Dryer fan
Hot &
humid air
Product
inlet
Low-temperature drying chamber #1
Fresh air
F2 Dry air
Dryer fan
Product
outlet
High-temperature drying chamber #2
Figure 4.4. Drying heat pump with two evaporating levels (for legend, see Figure 4.3).
28 Vasile Minea
Materials that are difficult to dry with convection heating alone (as ceramics and glass
fibre) because of poor heat transfer characteristics can be candidates for radio frequency-
assisted heat pump dryers. A radio frequency-assisted heat pump dryer comprises a vapour
compression heat pump retrofitted with a radio frequency generating system capable of
imparting radio frequency energy to the drying material at various stages of drying processes
(Figure 4.5b). This arrangement can overcome the limitation of heat transfer of conventional
hot air drying systems, particularly during the falling period. The radio-frequency generator
generates heat volumetrically within the wet material by the combined mechanism of dipole
rotation and conduction effects. Radio frequency heats all parts of the product mass
simultaneously and evaporates the water in situ at relatively low temperatures usually not
exceeding 82°C. Since the water moves through the product in the form of a gas rather than
by capillary action, migration of solid is avoided. Warping, surface discoloration and cracking
(caused by the stress of uneven shrinkage) associated with conventional drying methods are
also avoided. Radio-frequency drying is a rapid drying process whereby the heat necessary to
dry a product is generated within the product itself. Product containing moisture is subjected
to an alternating electric field which causes the dipole water molecules to rotate in response to
the changing polarity of the field. This rotation causes molecular friction and at radio-
frequency drying frequencies (1 to 30 MHz) the frictional heat is sufficient to produce
temperatures exceeding the boiling point of water. In the drying chambers the alternating field
is created between two large metal-plate electrodes which are connected to a high frequency
generator. The material is dried by placing or passing it between these electrodes. The
advantage of using radio-frequency for small product items is that they can be dried to
uniform moisture content in a matter of minutes without developing defects, especially in the
case of wood.
EV
C
EV EX
CD M
C
HR EX
LV HR
LV
Blower Infrared generators Radio-frequency
generator
CD
Infrared drier
Blower Dried material
(a) (b)
Figure 4.5. Infrared (a) and radio-frequency (b) assisted drying heat pumps (for legend, see Figure 4.3).
Compared to drying at atmospheric pressure, vacuum drying reduces the vapour pressure
and allows moisture to evaporate faster at the same temperature (or at the same speed at a
lower temperature). This process accelerates the drying rate and produce good quality
product. However, the capacity of driers is very low in relation to the cost of equipment.
Consequently, the volume of product processed over a given period of time is substantially
less in comparison to that obtained from conventional driers of equal cost.
4.4.3. Refrigerant side

Significant improvements are available by modifying the cycle and adding new
components. First of all, efficiency improvement is possible by minimizing the refrigerant
pressure drop in pipes and within heat exchangers. Pipe diameters have to be selected so the
pressure drop is small, even if larger pipes contribute to an increased first cost. In the
compressor suction pipe, the vapour velocity must be sufficient high to ensure proper oil
return to the compressor.
The main features of optimized drying heat pumps are shown in Figure 4.1. On the
refrigerant side, the liquid valve LV allows controlling the refrigerant migration during the
on/off cycling and standby periods.
On the other hand, the suction vapour always has lower temperature than the condensed
refrigerant is. It can be used to sub-cool the liquid condensate and, at the same time, superheat
the suction vapour and increase the compressor work input. The internal sub-cooler/super-
heater SC improves the overall efficiency of the heat pump thermodynamic cycle by
increasing the evaporator capacity and effectiveness. The heat pump output thus increases
without any increase in the compressor work. Typical improvements of the COP and the
capacity are approximately 1% per degree Kelvin of sub-cooling. The amount of sub-cooling
of the liquid refrigerant at the condenser outlet is generally of about 5°C. If the suction gas
entering the sub-cooler (state 1a) (see Figure 4.1) is dry, the heat capacity of the suction gas is
less than the heat capacity of the liquid. For this case, the sub-cooler heat transfer
effectiveness is defined as:
T1  T1a
 SC 
T4  T1a (4.7)
If the suction gas entering the sub-cooler is saturated (wet), the heat capacity of the liquid
is less than the heat capacity of the suction gas. For this case, the sub-cooler heat transfer
effectiveness is defined by:
T4  T5
 SC 
T4  T1a (4.8)
The conservation of energy for the sub-cooler is represented by:
h1  h1a  h4  h5
(4.9)
The sub-cooler heat transfer rate (kW) is determined by the energy conservation
equation:
30 Vasile Minea
 
Q SC  m R (h1  h1a ) (4.10)
4.4.4. Air side

On the drying air side, a bypass circuit around the evaporator is recommended to vary the
air flow through the evaporator (see Figure 4.1). Firstly, in order to control the
dehumidification rate during the whole drying process. Secondly, in order to mix the air
leaving the evaporator at 80%-90% of relative humidity with the air leaving the dryer. This
supplies a sufficient air flow rate through the condenser to avoid excessive compressor
discharge pressures and temperatures. At the same time, the mixing process allows the
condenser to reach the desired air supply temperature without over-designing the heat transfer

surface. The correlation between the evaporator air flow rate ( m EV ) and the condenser air

flow rate ( m CD ) is:
 
m EV   m CD (4.11)
where the bypass factor (  ) may be less than or equal to 1.

Calculations are based on the dry air mass flow rate through the system. The volumetric
flow rate (m3/s) of the duct air blower at other conditions than nominal (i.e. at ambient
temperature) varies according the following fan low:
 
V air  V n (vm / vamb ) 0.5 (4.12)
v v
Where m and amb are the specific volumes of the air leaving the blower and the
ambient air, respectively. Consequently, the air mass flow rate (kg/s) will be:
 
m air   air V air (4.13)
The electrical power input of the air blower is converted to heat:

Q blower  Wblower

Q blower
hd  hm  
m air (3.12)
The continuity equations of water vapour and energy of the by-pass mixing process
respectively are represented by:
 m   b  (1   )i
hm  hb  (1   )hi (4.14)
Where  is the air absolute humidity (kgwater/kgdry air).

The heat pump evaporator EV may also be provided with a variable speed fan to improve
the control of the material drying rate. The combined action of the bypassing air and the fan
variable speed provides optimum air flow rate through the evaporator, independently to the
air flow rate through the condenser.
5. WOOD DRYING HEAT PUMPS
5.1. Wood drying
As for other materials, wood drying is a coupled heat and mass transfer process
consisting in moisture movement from the internal zones to the wood surface. Drying process
is an essential step in the manufacturing of most wood products. Its objective is to minimize
the development of defects to achieve a low and uniform moisture content before machining,
gluing and finishing are carried out [16]. The most widely used method for removing
moisture from wood is drying under controlled conditions of temperature and humidity.
Trees are classified under two categories: deciduous trees, known as hardwoods and
coniferous trees known as softwoods. Wood cells are composed of thin walls of wood
substance surrounding cavities. More than 90% of the volume of a piece of softwood is
composed of vertically-aligned fibre cells. They are arranged in regular rows and perform the
dual function of conducting liquid and providing mechanical support for the tree. Softwoods
have both vertical and horizontal enlarged cavities surrounded by cells serving for the
movement of moisture. The structure of hardwoods has large-diameter cells forming long
vertical tubes through which liquids can pass.
5.1.1. Moisture movement in wood

Moisture occurs in wood as free water (liquid and vapour) in the cell cavities, and as
hygroscopic or bound water contained in the cell wall saturated structure. Drying air has a
strong affinity for moisture and eventually moisture begins to leave the walls of the surface
cells. But, since cell walls also have an affinity for moisture, they will attract replacement
moisture from areas of lower affinity, i.e. the interior zones. Moisture is therefore drawn from
the interior of wood. As it reaches the surface, it is again absorbed by the dry air and the
drying process continues.
Moisture can move through wood within the cell cavities and through the pit chambers
from cell to cell, through the ray cells, through intercellular spaces and through transitory cell
wall passageways. The space available for moisture movement ranges from 25% to about
85% of the total volume of wood depending on its density. To maintain a constant drying
rate, the water molecules inside the wood must absorb additional heat in order to increase
their kinetic energy. The forces that move moisture through wood include capillary action
(adhesion and cohesion), vapour pressure and moisture content differences, and diffusion.
32 Vasile Minea
Capillary action causes the free water to flow through the cavities, spaces and pits from one
cell to another. It is due to the simultaneous operation of adhesion and cohesion. Adhesion is
the attraction between water molecules and the walls of cells and pit chambers, and the
cohesion is the attraction of the water molecules to each other. When green wood commences
to dry, free water evaporation will occur from the surface cells. Evaporation and the cohesion
of water molecules exert a pull on the water in the adjoining cell cavities. As drying
continues, and the free water in the cell cavities is progressively removed, other drying forces
become operative. Capillary action moves into the wood core and gradually disappears as the
moisture content of the core cells approaches the fibre saturation point.
When most of the capillary action ceases, the majority of the cell cavities will contain
only air and water vapour, and this establishes a vapour pressure. The higher the vapour
content, the higher the pressure. A pressure differential between cells occurs because the
amount of vapour in a given volume of air decreases as the surface of the boars is approached.
This vapour pressure gradient causes moisture in the vapour state to move from areas of high
vapour pressure (interior) to areas of lower vapour pressure (surface of the board). As
moisture begins to leave the cell walls near the surface of a board, a moisture content gradient
develops between the surface and interior cells. However, since wood has an affinity for
moisture, the drier surface cell walls will absorb moisture from cell walls of higher moisture
content, i.e. moisture moves from the wetter interior cells to the drier surface cells. A
combination of vapour pressure gradients and cell wall moisture gradients operates
simultaneously in the diffusion process. A water molecule moves through a cell wall by
moisture gradient, across the cell cavity and through openings by vapour pressure gradient,
and again through a drier cell wall by moisture gradient until it finally reaches the wood
surface. Internal diffusion of moisture takes place longitudinally and laterally, and it is the
major factor controlling the drying rate of wood.
The main consequence of wood drying is the change in dimensions that occurs with a
change in moisture content. Although wood is capable of holding large quantities of water
above fibre saturation, its hygroscopic nature becomes evident below the fibre saturation
point. The fibre saturation point (FSP) is the stage in the drying process where the cell walls
are saturated with water and the cell cavities are free of liquid water. For practical purposes it
is assumed that the fibre saturation point at which shrinkage begins is 25% moisture content.
When wood is at or above FSP, the cell wall is in a non-shrunken condition. Below FSP, as
water is removed from the cell wall by diffusion, the cell shrinks. As an approximation,
volumetric change in wood below FSP is directly proportional to the volume of water gained
or lost from the cell wall. This relationship is expressed by the fallowing equation:
Rv  Db (  MC) (2.9)
where:
Rv is the total possible volumetric shrinkage (green volume basis), %;
Db - basic specific density of wood (oven dry weight, green volume basis), kg/m3;
Ψ – fibre saturation point, %;
MC – moisture content (dry basis) below FSP, %
In the living trees, the cell wall is moist. When water is removed from the cell wall, it
may shrink resulting in reduction in cell diameter. Only the moisture contained within the cell
wall structure has an effect on shrinkage. Consequently, all water in the cell cavity must be
removed before any shrinkage of the cell wall can occur.
As wood dries, shrinkage continues until the equilibrium moisture content is reached. In
calculating shrinkage allowances, it is essential to know the wood density and the final
moisture content to which the wood will be dried.
5.1.2. Evaporation rate

Heat is the source from which the water molecules in wood acquire the kinetic energy
necessary for evaporation. The rate of evaporation is dependent upon both the amount of
energy supplied per unit time and the ability of the heating medium (air) to absorb moisture.
The moisture movement rate is governed by the capacity of the surrounding air to absorb
moisture from the surface of the wood. It largely depends on the temperature and relative
dryness of the air and wood. Water moisture evaporates from the wood surface because of the
difference in partial vapour pressures between the drying air and the surface moisture. The
evaporation rate depends on the amount of energy supplied, the mass transfer coefficient and
the air temperature, flow rate and capacity to absorb humidity. Under stable conditions, the
convective evaporation rate (or drying velocity) (

kg water /h ) may be expressed as follows:
d ( MC )
N   M wood
anh
d (2.10)
where
M anh
wood is the wood anhydrous mass (kg) and MC - the moisture content (%).
When the humid wood boards are dried under stable conditions, the moisture content
decreases at first linearly. In the case of pure convective drying, the wood surface is always
saturated with water, and consequently the boundary layer reaches the air wet temperature.
This first step is followed by a non-linear process where the moisture content decreases
until the wood board attains hygroscopic equilibrium. If the drying velocity is known as a
linear or non-linear equation (N = f (MC)), the total time required to reduce the moisture
dry
content from
MCindry to MC fin (dry basis) may be calculated with the following equation:
anh
M wood
MC dry d ( MC )
 tot    dry
fin
MCin Aev N (2.10)
A
where ev is the evaporation surface, m2.
If the temperature of the drying air is constant, the rate of evaporation will gradually
decrease to maintain a steady drying rate, the water molecules in the wood must acquire
additional energy, or the vapour pressure of the drier atmosphere must be reduced. This is
achieved by either increasing the temperature (more energy) or reducing the relative humidity
(lower vapour pressure).
34 Vasile Minea
When the vapour pressure of the moisture in the wood is equal to the vapour pressure of
the surrounding air and neither gain nor loses moisture at a given temperature and relative
humidity, the equilibrium state is achieved. Air velocity at the wood surface must be high
enough to produce rapid air change, avoid the formation of death zones and provide uniform
drying. Circulation of air is required to carry heat to the wood being dried, and to carry away
evaporated moisture. For effective drying, this circulating air must be continually conditioned
for temperature and humidity before it passes through the wood stack. The air velocity must
be high enough to provide a rapid exchange of air. Higher air velocities also minimize dead
zones and promote more uniform drying throughout the wood stack. It is essential to remove
moisture from the boundary layer as rapidly as possible to maintain the desired drying rate.
This is achieved by controlling the circulation rate of the main air stream, usually through the
use of multiple-speed fan motors. Initially, when the wood is wet, the temperature differential
between the air stream and wood surface will be equal to the wet-bulb depression. Large
quantities of heat are required to vaporize the free water brought to the wood surface, and the
rate of heat transfer is at a maximum.
Higher circulation rates are beneficial during the early stages of drying when wood is wet
and the requirements for surface evaporation and moisture removal is high. Later, as the fibre
saturation point is reached at progressively deeper levels in each piece of lumber, the wood
temperature approaches the air stream temperature, and the rate of heat transfer diminishes.
As drying progresses below the fibre saturation point, the rate of heat transfer is further
reduced and, since this affects the drying rate, the dry-bulb temperature must be increased to
maintain an acceptable drying rate. At this stage, high air velocity has little effect on drying
rate and its continuation serves no useful purpose. As the fibre saturation point approached,
diffusion to the surface becomes the limiting factor in rate of moisture removal, and air
velocity should be reduced accordingly.
Any conventional wood drier (continuous, batch, fluidized bed, rotary, etc.) that uses
convection as the primary mode of heat input can be fitted with a suitably designed heat
pump. Optimum wood drier and heat pump integration, and appropriate thermodynamic
design and control strategies are needed to provide safe operation of the system and good
quality of wood.
5.2. Wood Driers
5.2.1. Generalities
Heat pump dryers can provide efficient and cost-effective drying of low-and high grade
wood. The advantages of wood heat pump driers include efficient utilization of recovered
heat, controlled drying rates resulting in low energy cost, equivalent or reduced drying time,
enhanced productivity, and improved wood quality. When high quantity of water has to be
removed from large volumes of wood, the efficiency of drying increases and shorter payback
periods can be obtained. The economics of the installation of wood drying heat pumps depend
on how the heat pump is integrated in the process. Identification of feasible installation
alternatives (i.e. split or compact) is of crucial importance.
5.2.2. Types of wood driers

Wood driers are particular types of industrial equipments. Their selection depends on
several factors such as capital investment, energy sources, production capacity and drying
efficiency. If wood is not dried accordingly to proper procedure and methods, many losses
may result. Wood drier can be divided in conventional and specialized using solar energy,
infrared, high frequency, vacuum and heat pumps as dehumidifiers. The specialized
techniques are usually more expensive and oriented to particular high-value end products.
Air-drying reduces the moisture content of construction wood to a minimum of about
15%. Wood to be air-dried is flat-piled in open yards with stickers between each layer of
board to allow free circulation of air. Drying rate is uncontrolled and, depending on the
ambient temperature, relative humidity and wind, drying times varies throughout the drying
season. The capital cost is low and the advantages and disadvantages depend on location,
annual production, drying season, wood species and dimensions, and cost of inventory. In
cold climates, outdoor driers have en effective drying season of about 8 months.
Accelerated air-drying is similar to that of normal air-drying with the exception that air is
forced through the wood piles by large multi-blade fans. Forced-air helps reducing the drying
time, but one must consider the added cost of sheds, fans, motors and electrical power. To
minimize power consumption, controllers of relative humidity are installed to stop the fans
when higher air velocities are undesirable or ineffective. The efficiency of accelerated air-
drying is greatest with high moisture contents and thin wood boards.
Low-temperature driers operate at temperatures of 20 to 45°C and are designed to
accommodate large quantities of wood (236 to 700 m3). Interior fans produce air velocities of
2.5 to 3.6 m/s. The main objective is to produce large quantities of air-dried wood in a
comparatively short time as accelerated air-drying method, but with minimal capital
expenditure in building materials and equipment. Low-temperature driers are most
advantageous as pre-driers, particularly for slow-drying hardwoods which are pre-dried to 25
or 30% of moisture content prior to conventional kiln drying.
Conventional wood drying is a fully controlled drying procedure operating at
temperatures below 100°C. Heat losses and inefficiencies in heat transfer usually limit
temperatures to between 82°C and 93°C. Steam is the most common of heat for conventional
driers, but oil and electrical heating are also used. Humidity is controlled through the
operation of steam or water spray valves and drier vents. Spray valves, vents and heating
systems are regulated by means of controllers to maintain selected wet and dry-bulb
temperatures according to a certain drying schedule. Fan systems are designed to give
reasonably uniform air circulation at velocities from 1.5 to 3.0 m/s with or without variable
speed motors. Air circulation is periodically reversed to obtain uniform heating and drying.
The main advantage of conventional driers over pre-driers is the reduction in drying time
obtained with higher temperatures.
High-temperature wood drying uses temperatures above 100°C and is an efficient method
for drying softwood construction lumber. Driers and equipment are similar to conventional
driers, but are designed for temperatures approaching 149°C. In addition to increased heating
capacity, there are additional requirements for drier insulation and maintenance for high-
pressure steam boilers. The capital investment per unit capacity is higher, but drying times
and energy consumption are reduced by up to 25% and 50% respectively compared to
conventional driers. The reduced energy is primarily the result of the large reduction in drying
time, to better heat transfer to the wood and improved insulation of the drying enclosure.
36 Vasile Minea
Figure 5.1 shows a compact drying heat pump where all heat pump components
(compressor, evaporator, condenser, expansion valve, refrigerant piping, fans and controls)
are installed outside the drying chamber in a closed cabinet.
The hot and humid air from the drying enclosure enters the evaporator EV and the
condenser CD. A mixing process using motorized air dampers and constant or variable speed
heat pump blowers is achieved prior reheating the air through the condenser CD. The back-up
heating coil supplements the heat that the drying atmosphere receives from the condenser.
When the dryer dry-bulb temperature drops under its setting point, it raises the air
temperature before it is returned to the wood stacks being dried. The back-up heating source
may be provided by natural gas, bark or oil-fired boilers. A multi-blade dryer fan circulates
the air through the wood stack. Its rotation direction periodically changes every 3 hours at the
beginning, and every 2 hours at the end of the dehumidification process. The air vents open
when the dryer fan changes rotation direction in order to avoid air implosion hazards, and also
when the dryer dry-bulb temperature exceeds its setting point, to avoid excessive
superheating.
Split drying heat pumps also consist in two chambers thermally isolated from one
another: the drying enclosure with remote condenser CD and the mechanical room where all
temperature-sensitive components (compressors, evaporators, variable speed blowers,
electronic expansion valves with microprocessor-based process controllers) and air ducts are
thermally separated from the drying area (Figure 5.2). In this case, the air flow trough the
evaporator EV can be controlled only by using variable speed blowers. The air flow rate
through the condenser CD is provided by the drier fans. It is normally constant and designed
according to the condenser thermal capacity.
Air vent
Back-up
Dryer fan heating coil
Mechanical room
Air outlet SV
CV EV
C
Wood stack
Heat
pump Blower
Back-up
SA heat
SC Air
LR source
CD inlet
LV
EX Drying enclosure
Figure 5.1. Compact wood drying heat pump.

Air vent
Mechanical room Dryer fan

Back-up
CD heating coil
Blower
LR Air SV
supply
C
Wood stack
LV
EV
Air
inlet
SC Back-up
heat
SA source
EX
Drying enclosure
Figure 5.2. Split wood drying heat pump.
5.2.3. Ecological aspects

Condensates produced by wood drying heat pumps do not contain suspended particles,
but contain toxic quantities of volatile organic compounds (VOC) including formaldehyde
and acetaldehyde that are harmful to trout and daphnia [17]. The volumes of water extracted
from wood by drying site are generally low (<15m3/day), but the water is acid and contains
low concentrations of organic loads, as total biochemical oxygen demand (BOD) <320 mg/L
and total chemical oxygen demand (COD) <454 mg/L. Problematic pollutants are the
formaldehyde (which can exceed the acute toxicity limit by up to 5.7 times), and to a lesser
degree, the acetaldehyde (which can exceed the acute toxicity limit by up to 1.3 times). The
condensate treatment options may be selected taking into consideration factors as the capacity
to reduce the effluent COD concentration of less than 500 ppm, and the availability and
ability of equipment for treating wastewater volumes below15 m3/day.
The wood-drying facilities that are not connected to city sewer systems have to use water
treatment technologies in order to reduce the concentration of toxic substances to levels
below the limits set regulations. They also can collect and transport water to the city waste
disposal sites. If the kiln site is connected to a sewer system, the municipality may accept
condensates from wood drying heat pumps, but charge to treat them. For example, the City of
Montreal sets rates based on the volume of wastewater discharged (US$0.0442/m3) and the
contaminants that it contains (for example, US$3.44/kg of phosphorous) (2000).
38 Vasile Minea
5.3. Energy Requirements
To correctly design the thermodynamic coupled system drier-heat pump it is mandatory

to calculate the energy requirements in a given climate. In the case of wood, the total energy
required for drying with heat pumps may be expressed as follows (Figure 5.3):
Qtot, req    1 Qi  EC  B  Efan  Qbackup  Qrec

i (5.1)
where  is the heat losses parameter (0.1 to 0.3); i - thermal energies required for heating;
Q
E fan
EC+B - electrical energy consumption of the heat pump compressor and blower; -
Q Q
electrical energy consumption of the dryer fan; backup - total back-up heat; rec - total heat
(sensible and latent) recovered by the heat pump. The parameter  accounts for
uncontrollable vent-air losses of poorly insulated and tight new or retrofit dryers. The most
significant thermal energy requirements are detailed below.
If the wood initial temperature is higher than 0°C, the thermal energy required for heating
and evaporating the water extracted from the wood at each drying step j is calculated using
the following equation [18]:
  MC j
Q11   M water  c p, water Tset, j  Tin   hlv 
  100  (5.2)
If the wood initial temperature is lower than 0°C, the thermal energy required for heating
and evaporating the water extracted from the wood at each drying step j is calculated using
the following equation:
 MC  j MC  j
Q 21   M ice ( cice 0  Tin   hmelt   M wood ( cwater Tset, j  0 hlv 
 100 100  (5.3)
Vent-air losses
Efan
Qback-up
EC+B + Erec
EC+B Dried material

- Heating/evaporating water/ice
(wood) - Heat losses (walls, floor)
- Heating wood, stickers, sleepers
Qrec Qi - Overcome retention forces
- Heating dryer envelope
Drying heat pump Dryer enclosure
Figure 5.3. Energy requirements of wood drying heat pumps.

The dryer heat losses through the walls and floor are calculated as follows:
An Tn
Q2   (  j * 3.6)
U tot,n
(5.4)
where An is the heat transfer area of the dryer section n (walls, doors, ceiling, floor, etc.), m2;
ΔTn - difference between the dryer indoor temperature and the outdoor air or ground

temperature, °C; Utot,,n - total thermal resistance of the dryer section n, m2K/W; j - duration
of the drying step j. The total thermal resistance is calculated for each section n using the
equation:
 1  1 
U tot, j    n  
 hint k n hext 
(5.5)
where hint is the convection heat transfer coefficient on the interior surface,
W/m2K; δn - thickness of section n, m; kn - thermal conductivity of section n, W/mK;
hext - convection heat transfer coefficient on the exterior surface of section n, W/m2K.
The energy required for heating the wood, stickers and timber sleepers is calculated using
the following equation:
3
Q 3   mk ck Tmax  Tin 
k 1 (5.6)
where k = 1 (dried wood), 2 (stickers) or 3 (sleepers);

mk - wood, stickers and sleepers
ck - specific heat of each component k, kJ/kgK;
anhydrous mass, respectively, kg; -
Tmax
T
maximum drying dry-bulb temperature, °C; in - initial temperature of wood, stickers and
sleepers, respectively.
The thermal energy required to overcome the retention forces, is calculated as follows:
MC fin
Q4  M water  hsorpd (MC )
MCin (5.7)
where hsorp is the wood sorption enthalpy, kJ/kg.

Finally, the thermal energy required for heating the dryer envelope (structure, insulation,
slab, foundation) and other equipment has to be calculated as follows:
Q5  ms cs Tmax,s  Tini,s 

(5.8)
40 Vasile Minea
m c T T
where s (kg), s (kJkgK), max,s (°C) and in ,s (°C) respectively are the total mass, the
specific heat, and the maximum and initial temperatures of the structure component or
auxiliary equipment s.
5.4. Design issues
After calculating the energy requirements and choosing the appropriate drying schedule
based on wood drying curve, the system integration is the next step. Operation and energy
performances of drying heat pumps highly depend on the system integration and require a
number of configurations, components and parameters to provide normal and safe
thermodynamic conditions.
First of all, the heat pumps cannot accurately operate as a dryer if the total (sensible and
latent) heat extraction isn‘t large enough to provide appropriate operating parameters. The
heat pump coefficient of performance cannot be accurately determined if the evaporator heat
extraction capacity is too low compared to the material dewatering rate, and/or if the drying
heat is mostly supplied by the compressor. Appropriate correlation between the initial mass
and moisture content, the moisture migration and extraction rate of wood and the heat pump
dehumidification capacity has to be provided. This means that a sufficient quantity of dried
material has to be supplied to provide the required thermal input to the evaporator.
Insufficient quantities of wood mean small quantities of moisture and heat extraction, too low
evaporating and compressor suction pressures and temperatures, risks of humidity freezing on
the evaporator fins, and short drying cycles. In addition, such poor operating conditions may
lead to a number of heat pump operating problems. Consequently, matching the dewatering
capacity of the dried material with the heat pump dehumidification capacity is a major issue.
In the case of wood, as the moisture content drops at the surface, the water migrates from
inside the wood to the surface because of moisture content differences. The migration rate of
water depends on the liquid diffusivity, which is a function of local moisture content and
temperature. For cost-effective operation of the dryer, the migration rate of water to the
drying surface should match the moisture removal rate of the dried air. Therefore, drying
conditions need to vary with drying time or moisture content of the product for cost-effective
and safe operation of the dryer. An appropriate quantity of wood, helps attaining high drying
performances.
5.4.1. Heat pump capacity

The heat pump dehumidification capacity depends on the quantity of water to be removed
dry
from the wood. If the initial (
MCindry ) and the final ( MC fin ) moisture contents (dry basis)
M in,hum
are known, as well as the initial humid mass of wood ( ), the maximum mass of water
that can be removed can be calculated using the following equation:
MCindry  MC dry
  * M in,hum
max fin
M water
100  MCindry (5.9)
where  (0.85 to 0.90) accounts for moisture losses by condensation, leakages and vent air
openings. The initial wood humid mass depends on the dryer capacity. If the wood drying
 drying
curve and time ( ) are known, and it is assumed that the sensible thermal power

represents 25% ( sens ) of the total dehumidification thermal power, the total capacity of the
heat pump evaporator will be [19]:
  EV M water
max
hlv
Q EV 
1   sens  drying (5.10)
where  EV is the evaporator thermal efficiency. If the heat pump coefficient of performance (
COP ) is known, the power input (kW) of the heat pump compressor will be:
  * EV * M in,hum MCindry  MC dry

fin hlv
WC 
100  MCin 1   sens COP 1 drying
dry
(5.11)
The required average air mass flow through the evaporator and condenser (kg/s) will be,
respectively:
  EV M water
max
hlv
m EV 
1   sens ha  hb  drying (5.12)
CD WC  Q EV 


m CD   
c p TCD
(5.13)
h
where a and
hb respectively are the air mass enthalpy entering and leaving the heat pump
evaporator.
5.4.2. Drier capacity


Q
For a given total heat pump dehumidification capacity ( EV ) it is necessary to determine
VG
the required initial humid wood (green) volume ( ) to be dried. For a dehumidification time

of drying (generally, about 60 hours for spruce and pine, and 140 hours for balsam fir), the
latent heat (kJ) extracted is:
 
QEV ,lat  (1   sens ) Q EV * drying  * 3600
  , (5.14)
42 Vasile Minea
The total mass (kg) of water removed by the heat pump evaporator will be:
M water  QEV ,lat / hlv

(5.15)
The softwood initial and final humid masses may be calculated using the following
equations:
 MCindry 
M in,hum  1   * M wood
anh
 100  (5.16)
 MC dry 
M fin,hum  1   * M wood
fin anh
 100 
  (5.17)
For Eastern Canada,

MCindry is currently 40% for white spruce and red/white pine, and
MC dry
fin
88% for balsam fir. The minimum final moisture content ( ) is of about 18%. The
softwood anhydrous mass (kg) may be calculated using the fallowing equation:
M water *100
anh
M wood 
MCin,hum  MC fin,hum
(5.18)
The green softwood volume to be dried (m3) is defined as:
VG  M wood
anh
/ Db (5.19)
where
D b is the wood basic density. In eastern Canada, the average basic density of
softwood species are 340 kg m-3 for the balsam fir, 390 kg m-3 for the red pine and 350 kg m-3
for the white spruce. The initial softwood green volume can be calculated using the following
equation [20]:

1   sens  Q EV * drying * 3600 *100
VG 
MC in ,hum  MC fin,hum Db hlv
(5.20)
5.5. Control Aspects
Inappropriate control strategies lead to drying heat pumps operating with low
dehumidification efficiency. They may reduce the compressor technical life and cause
mechanical damages. The most significant dangers come from the liquid refrigerant entering
the compressor suction line, from compressing saturated vapour and from frequent and/or too
short compressor on/off cycles. Inadequate correlations between the dehumidification

capacity of the heat pump and the material dewatering capability can also compromise the
drying heat pump overall performance and reliability [21].
5.5.1. Refrigerant migration

In the case of high-temperature wood drying heat pumps, there is a significant
temperature difference (up to 100°C) between the drying enclosure and the mechanical room.
During the preheating step and the heat pump stand-by periods, the refrigerant migrates
towards the compressor C and the suction accumulator SA (Figures 5.1 and 5.2). These are
large metallic components with high thermal inertia. Their temperature is generally lower,
and the refrigerant condenses inside. When the compressor starts, it will first try pumping the
liquid refrigerant. Because the liquid is incompressible, the pressure differential will be too
low and the compressor automatically shuts down. After a short time it starts up again, and
several such on/off cycles may occur until the whole quantity of liquid is pumped out and the
set pressure differential is restored. Such frequent cycling reduces the useful life of
reciprocating compressors and can damage their moving parts (pistons, connecting rods) and
cylinders.
5.5.1.1. Small-scale heat pumps

To manage refrigerant migration and prevent the liquid from entering the compressor
suction line, a well known method consists in installing a liquid valve LV between the
condenser CD and the liquid receiver LR (section 4.3.3). The following control sequences
will allow safe operation of the compressor by preventing refrigerant migration and storage
inside the suction accumulator and compressor crankcase during stand-by periods. When the
compressor shuts down, the liquid valve immediately closes, but the compressor continues to
run for additional few seconds. The refrigerant from the evaporator EV is pumped-down and
stored inside the condenser CD and the liquid receiver LR. At the end of this sequence, the
evaporator is practically free of any refrigerant, and the condenser and liquid receiver store
the majority of the refrigerant charge. When the compressor restarts, the liquid valve LV first
opens for few seconds, while the compressor is still off. During this period of time, refrigerant
liquid flows from the condenser to the inlet port of the expansion valve, and then the
compressor restarts. With relatively small capacity heat pump dryers, the liquid valve is able
to totally address the problem.
5.5.1.2. Large-scale heat pumps

In the case of large capacity high-temperature split heat pumps (i.e. with compressor
power inputs higher than 100 kW), the liquid valve control may not be entirely safe for the
compressor mechanical integrity. During the preheating step, the refrigerant migrates from
the drying enclosure working at a high temperature to the suction accumulator SA and the
crankcase of the compressor C. When the compressor starts at the first dehumidification step,
it sucks up the liquid refrigerant stored inside C and/or SA. In spite of conventional
protections that may be used, too many on/off cycles with liquid entering the suction line may
seriously damage the compressor. In order to avoid such dangerous incidents, two solutions
may be used [23]. The first one suggests inserting the suction accumulator inside a thermally
insulated larger water receiver (Figure 5.4a). Cold water from the city, a lake or a river can be
44 Vasile Minea
continuously circulated through the water receiver during short periods of time prior starting
the compressor. The purpose is to cool the external surface of the suction accumulator SA.
After having cooled the suction accumulator, the cooling water returns to the city sewer, the
lake or the river. Between SA and the heat pump liquid refrigerant line, an additional cooling
circuit is provided. It includes a motorized valve MV2, a liquid pump LP and refrigerant
piping with check valve CV2. LP provides sufficient head pressure allowing pumping the
whole quantity of liquid stored inside SA toward LR. The operation of the liquid pump may
be provided by a simple liquid-level control method.
During the preheating step of a new drying cycle, few minutes prior to the compressor
starting, the motorized valve MV1 opens. The cooling water enters the water receiver at
temperatures between 5°C (in the winter) and 20°C (in the summer). The metallic envelope of
the suction accumulator is cooled and the refrigerant migrates from the compressor crankcase
to the suction accumulator SA.
After this sequence, the MV1 valve closes, the MV2 valve opens and the liquid pump LP
starts. After having pumped the liquid from the suction accumulator to the liquid receiver LR,
the MV2 valve closes, the liquid pump LP shuts down and then the compressor C starts-up
safely.
The second solution includes a modified suction accumulator (MSA1) coupled to an air-
cooled condensing unit. The evaporator of the air-cooled condensing unit is attached to the
external surface of MSA1 (Figure 5.4b) or installed inside the suction accumulator MSA2
(Figure 5.4c). In each case, the compressor B, the air-cooled condenser D and the expansion
valve E are located outside the suction accumulator. The additional cooling circuit contains a
motorized valve (MV2), a small refrigerant liquid pump (LP) and a check valve (CV2). Few
minutes before the end of the preheating step, the compressor B starts and the temperature of
MSA1 becomes lower than the temperature of the compressor crankcase C. The refrigerant
will naturally migrate toward the modified suction accumulators. After this sequence, the
valve MV2 opens and the liquid pump LP starts up. The liquid is thus pumped toward the
liquid receiver LR. After having pumped the whole quantity of liquid, the solenoid valve
MV2 closes, the liquid pump LP shuts down, and the heat pump compressor C starts up
safely.
5.5.2. Refrigerant superheating

Refrigerant superheating is required to avoid the liquid from entering the compressor
suction line. The superheating process AB (Figure 5.5a) is controlled by the expansion valve
according to the saturated evaporating pressure and temperature, and the actual suction
temperature. For most common refrigerants, a suction superheating of about 5°C is sufficient
to prevent saturated vapour from entering the compressor (process BC). This is achieved due
to the slope of the adiabatic curves (ds = 0) in the superheated vapour zone versus the position
of the saturated vapour curve. If the compressor sucks up superheated vapour at state B
during the quasi-adiabatic compression process BC, the isentropic curve doesn‘t touch the
saturated vapour line.
Remote CD
Remote CD
EV
EV LV
LV
EX
EV LR
LR
CV1
CV1
SA CV2 MSA1
CV2 C
C
MV1
Cold water
Water LP MV2 E
LP MV2 receiver
D B
(a) (b)
Remote CD
EV
LV
EX
LR
CV1
MSA2
CV2
C
LP MV2
E
B
D
(c)
Figure 5.5. Liquid management options for large-scale high-temperature drying heat pumps.
However, with some of new high-temperature refrigerants, as R-236fa and R-245fa, the
isentropic curves (EF and GH compression processes in Figure 5.5b), are more vertically
oriented compared to those of common refrigerants. This suggests that, if the superheating
amount (process DE) is too small, the compression adiabatic curve could cross the refrigerant
saturation line. In this case, the thermodynamic state of the refrigerant leaving the compressor
(F) can be located on the saturated line. To avoid such a situation, higher amounts of suction
superheating with high-temperature refrigerants has to be provided, but inside an optimum
range:
Tmin  TS  Tmax (5.21)

46 Vasile Minea
ds=0
ds=0
Critical point Critical point
Pressure, MPa
Pressure, MPa
apor
r
po
d
pCD pCD
uid
ui
C F H
va
liq
ted v
liq
ed
ed
ed
at
ra t
ra
r
ra t
tu
tu
Satu
Sa
tu
Sa
Sa pEV A B pEV D E G
TSUC TSUC
ΔTS = 5°C
ΔTmin < ΔTS < ΔTmax
Enthalpy, kJ/kg Enthalpy, kJ/kg
a. Conventional refrigerants b. High-temperature refrigerants
Figure 5.5. Comparison of adiabatic compression processes in p-h diagrams; s: mass entropy.
For high-temperature refrigerants, the optimum range is between 15°C to 20°C at any
saturated suction pressure [23]. To keep the compressor suction superheating inside the
recommended range, simple control sequences may be used. The suction pressure (pEV) and
suction temperature (TSUC) sensors send proportional signals to the system process controller
(SPC). It automatically calculates the actual evaporating temperature according to the
refrigerant saturated pressure-temperature correlation:
TEV  f ( pEV ) (5.22)
For each pair of suction and evaporating temperature, the refrigerant actual suction
superheating is calculated at preset intervals:
TS  TSUC  TEV (5.23)
With such a superheating, the SCP automatically modulates the opening of the electronic
expansion valve to keep the superheating inside the desired range. If the actual superheating
is, for example, 0.5°C bellow the lower limit, the expansion valve will slightly close in order
to increase the suction superheating. If the actual suction superheating is 0.5°C higher than
the upper limit, the expansion valve will slightly open to decrease the suction superheating.
5.5.3. Dehumidification rate

The efficiency of the heat pump drying process highly depends on the correlation
between the material dewatering rate and the heat pump dehumidification capacity.
The final quality of the dried wood depends on the heat pump moisture extraction rate.
The moisture extracted from wood condenses on the evaporator tubes and fins. This process
results from the contact between the water vapour and the surface of the heat pump
evaporator when its temperature becomes lower than the air saturation temperature. The
condensation enthalpy is released and the heat is transferred to the refrigerant. The mass of
the removed moisture depends on the air flow rate across the heat pump evaporator. Because
the evaporator heat exchange surface is constant, it is necessary to vary the air flow rate to
keep a constant dehumidification rate correlated to the material dewatering rate.
Dehumidification rate control - softwood

70 2
60
1.5
Electrical power, kW
50
Frequence, Hz
Low
40
1
30
Medium High
20
0.5
10
0 0
0 10 20 30 40 50 60 0 10 20 30 40 50 60
Time, hours Time, hours
Figure 5.6. a). Dehumidification rate control for softwood; b) Example of heat pump blower optimized
operating profile.
5.5.3.1. Evaporator air by-passing

A standard method for controlling the dehumidification rate consists in using coil face
and controlled bypass dampers. In this case, a temperature sensor controls both primary and
bypass dampers so that the temperature of the air leaving the evaporator is the same as the
temperature of the refrigerant leaving the evaporator. It is however a relatively imprecise
method for industrial applications because of low temperature reading accuracy and of low
reliability of the mixing process with air dampers.
5.5.3.2 Variable air flow rate

Toward the end of the drying cycle, the amount of moisture provided by the material
progressively decreases. The dehumidification rate can also be controlled by varying the air
flow rate through the evaporator according to time-based transfer functions. In order to
correlate the dehumidification rate to the available moisture quantity, the air velocity through
the evaporator can be controlled according to the following equation:

Q EV 1 1
  
wair

 air A c p ,air TDB ,air  hair hair  air  (5.24)
w 
where air is the face air velocity (m s-1) and air - the air absolute humidity (kg kg-1). It can
be seeing that if the absolute humidity differential decreases, the air velocity has to increase in
order to keep the heat pump evaporating capacity at a constant level. Because the humidity
content of the material is very high at the beginning of the dehumidification process, and
much lower at the end, it is necessary to continuously increase the air mass flow rate and
velocity of the warm, humid air entering the evaporator. This allows reaching the dew point
and providing quasi-constant dehumidification rates. The following time-based equation can
be used to provide low or high dehumidification rates [21]:
  K *  25 (5.25)
where ν is the blower electrical current frequency (Hz);  - drying time measured in hours
from the beginning of the dehumidification process with heat pump; K -empirical constant.
48 Vasile Minea
For softwood, such as white spruce and pine, the dehumidification cycles with high-
temperature heat pumps take up to 60 hours. In this case, the constant K varies from 0.2 (high
dehumidification rate) to 0.5 (low dehumidification rate) (Figure 5.6a). Figure 5.6b shows an
example of a heat pump blower optimized operating profile according to this control strategy.
5.5.3.3. Intermittent drying

Another method allowing providing variable dehumidification rates consists in
intermittent drying. It is well known that the drying of capillary porous materials like wood is
very often accompanied by a cracking phenomenon. In industrial drying, great care is taken to
avoid cracking as it influences the quality of the dried products. It has been stated that the
main reason for cracking, and thus destruction of the dried products, is the drying induced
stresses [22]. The greatest lack of consistency is reached mostly at the end of the constant
drying rate period or at the beginning of the falling drying rate period. Material cracking is
most likely to occur at this stage of drying, especially if the material is thick and the drying
rate is high. A method to avoid cracking is to slow down the drying rate to enable the
moisture inside the capillary pores to be consistent throughout the material, and thus diminish
the risk of cracking. Lowering the drying rate, on the other hand, increases the drying time,
which is economically unprofitable. Therefore, the drying rate has to be increased again when
the moisture distribution has become sufficiently consistent. A high drying rate is not risky at
the beginning of the drying process because the material is fully saturated and the capillary
transportation of moisture from the interior is sufficient enough to preserve the liquid film on
the surface. During this period, the drying proceeds as from an open water surface. When the
water on the surface disappears, the surface begins to shrink and drying induced stresses
appear. That is the moment for periodically increasing the air humidity by stopping the heat
pump compressor. Figure 5.7a shows, as an example, a heat pump‘s compressor hourly
percentage running profile according to the dryer set wet-bulb temperature. Figure 5.7b
represents the cumulative volume of water extraction with heat pump intermittent drying
cycle.
5.5.4. Wet-bulb temperature setting

A major issue for industrial drying heat pumps consists in increasing the useful life of the
compressors. Too short on/off cycles and many transient regimes contribute to the
compressors wear and lower overall system efficiency. To further avoid this problem, a
method has been developed to set the dryer wet-bulb temperature automatically according to
the actual measured values. It is well known that prior starting a new drying cycle, the air
wet-bulb temperature (
WBset ) is preset for each step of the dehumidification process. Also,
the air actual wet-bulb (

WBact ) temperature is monitored inside the dryer and transferred to
the system process controller (SPC) automatically. The heat pump compressor starts at the
1
WBact
end of the preheating step. At that time, the actual air wet-bulb temperature ( ) is much
1
higher than the first preset air wet-bulb temperature (
WB set ).Because the compressor is
running and the dryer is a tight enclosure with all air vents closed, the difference between the
actual air wet-bulb temperature and the first preset dryer air wet-bulb temperature gradually
decreases over time. When this difference reaches 5.5°C (or 10°F) for example, the SPC will
automatically lower the set air wet-bulb temperature, for instance, by 2.7°C (or 5°F). After
having modified the first set air wet-bulb temperature, the heat pump compressor continues to
run in dehumidification mode.
Hourly percentage of intermittent heat pump running Cumulative mass of water extraction with intermittent drying
120 900
Compressor (%) and Temperature
100% 100% 800
100
85% 700
Extracted water, litres

75% 75%
80 600
65% 500
(°C)
60
Set wet-bulb temperature 400
40 300
200
20
100
0 0
0 24 48 72 96 120 144 168 192 216 0 24 48 72 96 120 144 168 192 216
a b
Figure 5.7. a). Example of intermittent wood drying heat pump operation; b) Cumulative mass of water
extraction with heat pump intermittent drying.
The difference between the actual wet-bulb temperature and the second set wet-bulb
temperature (
WB 2
set ) continues to gradually decrease over time. When the difference between
the actual air wet-bulb temperature and the second set air wet-bulb temperature reaches 5.5°C
once again, the SPC will again automatically lower the second set air wet-bulb temperature
by 2.7°C. This control sequence will be applied several times depending on the physical
characteristics of the dried material. The wet bulb temperature is consequently set according
to the following equation [21]:
WBi set  WBi 1set  2.7C (5.26)
where i is the number of successive wet-bulb temperature settings.

During the last dehumidification step, when the actual air wet-bulb temperature reaches,
for instance, a level that is 1.1°C below the previous set air wet-bulb temperature ( set ), WB i 1
the SPC will automatically shut down the heat pump. The compressor will restart only if the
actual air wet-bulb temperature becomes 1.1°C higher than the previous set air wet-bulb
temperature. If the heat pump compressor starts and shuts down another three consecutive
times during a predetermined period of time, the SPC will stop the compressor and the drying
cycle in dehumidification mode will end. Finally, if 20 to 30 minutes after the compressor has
shut down, the actual air wet-bulb temperature does not exceed the last set air wet-bulb
temperature by 1.1°C, the dehumidification drying cycle ends for good.
5.5.5. Back-up heating

Dehumidification drying systems using heat pumps need smaller quantities of back-up
energy compared to most conventional systems. The reason is that the high-temperature heat
pump recovers enough sensible and latent heat from the dryer warm, humid air, and returns
the recovered heat to the dryer enclosure. The compressor input electrical energy is added to
this recovered heat in the form of thermal energy. Consequently, the additional quantity of
heating energy required by dryers using heat pumps as dehumidifiers is much smaller than
50 Vasile Minea
that of conventional wood drying systems. Some additional heat is however needed only to
compensate for any lack of heat pump heating capacity, dryer heat losses and air leakages,
and other uncontrollable energy losses. If the steam distribution system operates at high flow
rates (as in conventional drying processes or during the preheating stage of heat pump dryers)
the amount of heat transferred to the indoor air by the steam-to-air heat exchangers will be too
high as compared with the required additional quantity of heat.
With both compact and split drying heat pumps, the back-up heating supplying control is
a prominent issue. Appropriate strategy helps avoid useless material stresses and, especially,
saves energy. In the case of split drying heat pumps, another sensitive issue is the protection
of the remote condensers against potential thermal shocks that may occur when the back-up
heating system is activated. As a back-up heating source, steam generated by bark, natural gas
or oil-fired boilers is usually used. In large-scale conventional dryers and, also, during the
preheating steps of heat pump dryers, the solenoid valves SV are relatively large (Figure
5.8a). The steam is directed at full flow rate through the distribution piping towards the dryer
steam-to-air heating coils. These heat exchangers are located on the upper side of the dryer,
close to the heat pump remote condensers. When the dryer fans turn in one of the two
directions, the dryer warm and humid air first passes through the steam heat exchanger where
it is heated, then through the heat pump condenser.
Compressor power input Test #309

100
90
80
70
60
50
40
To the drier 30
back-up HEX 20
10
0
0 10 20 30 40 50 60
Condensate PC
Time, hours
return
SV b
MV
Compressor power input Test #397
70
60
50
40
Back-up heating source
30
a 20
10
0
0 10 20 30 40 50 60
Time, hours
Figure 5.8. a) Improved back-up heat supply; b) Example of abnormal heat pump operation with large
solenoid valve; c) Example of normal operation with small modulating valve.
If the steam distribution system operates with dehumidification heat pumps as in

conventional dryers, the solenoid valve SV fully opens and closes for short periods of time
according to the air dry-bulb temperature inside the dryer enclosure. When the solenoid valve
SV is fully open, the steam flows at a high rate through the heating coils installed inside the
dryer. Because the heat pump condenser is located close to these coils, it will be hit suddenly
by air flowing at high temperatures. During short periods of time, these temperatures are
much higher than the normal air temperatures entering the condenser. It produces undesirable
thermal shocks at the heat pump‘s condenser inlet and excessively increases the compressor
discharge pressure and temperature. Consequently, the compressor may frequently shuts-
down, as it can be seeing in Figure 5.8b. Frequent compressor cycling may damage it
mechanically. Moreover, the high compressor discharge temperatures may break down its
lubricating oil and materials. In order to address this safety issue, a second steam distribution
circuit on the steam distribution header has to be installed. It by-passes the main solenoid
valve SV and consists in a small diameter pipe connecting the steam manifold to the
distribution network. A high precision modulating valve MV, activated by the pressure sensor
PS, modulates the steam flow rate to keep the outlet pressure in a range of 5% to 10% of the
steam normal inlet pressure. With a constant steam flow rate, no sudden thermal shock will be
provided at the top of the heat pump remote condenser. The condensing pressure and
temperature will not increase suddenly and the compressor operating profile will be stable
(Figure 5.8c). When the dryer has to operate in conventional drying mode, and also during the
preheating step, the small modulating valve closes and the large solenoid valve SV controls
the steam heating supply as usual.
5.6. Performances of Improved System
Almost all described improvements have been implemented in one 354-m3 forced air
softwood dryer [23].
5.6.1. Thermodynamic parameters

Excluding the initial preheating steps (4 to 8 hours), the average duration of the
dehumidifying drying cycles lasted no more than 61 hours with softwoods as white spruce
and pine. The initial preheating steps made it possible to increase the kiln air absolute
humidity up to 0.309kg/kg before the dehumidifying cycles (with heat pumps) started up. The
air absolute humidity gradient across the heat pump evaporators varied from 0.096kg/kg,
immediately after starting the compressors, to about 0.039kg/kg at the end of each drying
cycle. The dry-bulb temperature of the air entering and leaving the heat pump evaporators
was very stable at around 80°C (Figure 5.9a). The relative humidity of the air entering
evaporators continuously decreased up to 40%, while the leaving air relative humidity was
almost constant at about 90% (Figure 5.9b).
The stability of the thermodynamic cycle was provided by supplying the heat pump
condensers with air at relatively constant temperatures. Depending on the rotation direction of
the dryer fan, the temperatures of the air entering the condenser from the top were up to 20°C
higher compared to that of the air entering the condenser from the bottom. As a result, the
heat pump compressors ran with stable suction and discharge pressures and temperatures
52 Vasile Minea
(Figure 10a and b) according to periodical changes in the dryer fan rotation direction. It can
be seen that the discharge pressures has periodically increased up to 2500kPa (absolute) only
when the dryer central fan stopped for about two to three minutes prior to each change in its
rotation direction.
All these parameters allowed evaporating temperatures of up to 45°C, and condensing
temperatures around 100°C, up to 20°C higher than the kiln‘s actual air dry bulb
temperatures.
5.6.2. Energy performances

The heat pump average coefficients of performance (eq. 31) varied from 4.6 at the
beginning of the dehumidification process to about 3.2 at the end. The average water
extraction rates of each heat pump (about 130kgwater/hour) with white spruce and balsam fir
do not include the venting losses, but account for 5% of condensed water losses (Figure
5.11a). The high temperature heat pumps (compressors and blowers) both accounted for 70%
and the dryer central fan for 30% of the system total electrical energy consumption (Figure
5.11b). The Modified Specific Moisture Extraction Rate ranged from 2.35 to 2.95
kgwater/kWhC+B. The system total energy consumption was compared with those of
conventional drying cycles using steam as the sole heating source. It was proved that the
energy consumption of the dehumidification cycles with high temperature heat pumps was
27% to 57% lower than the energy consumption of conventional drying systems. Finally, the
average reduction in specific energy costs, compared to the costs of conventional wood
drying cycles, was estimated at about 35%.
Inlet Evaporator Evaporator

100 100
Dry bulb temperature, °C
Relative humidity, %
80 80
Outlet
60 60
Outlet
40 40
Inlet
20 20
0 0
0 10 20 30 40 50 60 70 0 10 20 30 40 50 60 70
a b
Figure 5.9. Air entering and leaving the heat pump evaporators; a) Dry-bulb temperatures; b) Relative
humidity.
Compressor temperatures Compressor pressures

140 3000
Discharge Discharge
120
Pressure, kPa,a
2500
Temperature, °C
100
2000
80
Suction 1500
60
1000
40 Suction
500
20
0
0
0 10 20 30 40 50 60 70 0 10 20 30 40 50 60 70
a b
Figure 5.10. a) Compressor suction and discharge temperatures; b) Compressor suction and discharge
pressures (heat pump #1).
8000
Water extraction, Litres

HP #2 (7514 L)
6000 Dryer central
fan
4000 HP #1 (7248 L)
30%
2000
Heat pumps
0 (compressors
0 10 20 30 40 50 60 70 & blowers)
70%
Time, hours
a b
Figure 5.11. a) Cumulative volumes of condensed (extracted) water; b) Share of the dryer total
electrical energy consumption.
It was a large-scale split high-temperature heat pump driers, adequate strategies have
been used to avoid the system malfunction and equipment failures. The refrigerant migration
control strategy using a liquid valve has avoided the refrigerant from entering the compressor
suction port. Appropriate correlation between the suction superheating and the thermo-
physical properties of the refrigerant has prevented condensations during the compression
process. The dryer wet-bulb temperature setting control method according to the drier actual
wet-bulb temperature avoided frequent compressor cycling and contributed to improving its
reliability and useful life. Improved back-up heating control method avoided high thermal
shocks at the inlet of the remote condenser and provided stable compressor discharge
pressures and temperatures.
6. CASE STUDIES
About 45% of the Canadian territory is wooded representing 10% of the planet forests.
The Canadian wood industry provides 3% of the country‘s Gross Domestic Product, 10% of
the total foreign exchanges and 57% of the annual commercial surplus. In Quebec (Eastern
Canada), hardwoods such as hard maple, yellow birch, oak and white walnut represent 6.6%
of a market dominated by resinous species. Hardwood is usually dried at low temperatures
(maximum 55°C), a process that consumes up to 70% of the total energy required for primary
wood transformation. Approximately 10% of Canadian resinous wood production comes
from the province of Quebec (East of Canada). About 2% of this production is dried with
low-temperature heat pumps and the rest through other technologies such as direct fire and
the use of bark-, natural gas- or oil-burned boilers. Drying of resinous wood, highly favorable
to high-temperatures, is essential to prevent the warping and cracking of the wood. Usual
energy sources for wood drying are fossil fuels (oil, natural gas, propane) and bark.
6.1. Low-Temperature Wood Drying Heat Pump
Electrically-driven low temperature heat pumps are usually used in combination with
fossil fuels or electricity as back-up energy sources. About 25% of the hardwood dryers in
Quebec are presently equipped with low-temperature heat pumps.
54 Vasile Minea
Steam coils
Hot
dry air
Blower M
Electrical
heaters
Heat
pump Fan
Stea
m
(b)
(a)
Figure 6.1. View of the low-temperature hardwood drying experimental heat pump
6.1.1. System description

A 2.3-m3 experimental air-forced wood dryer with variable speed fans was equipped with
a 5.6-kW low temperature heat pump [24]. The heat pump (compressor and motor, blower,
evaporator, condenser, sub-cooler, refrigerating piping and controls) is installed inside a
mechanical room beside the drier chamber (Figure 6.1a).The dryer contains both steam and
electrical heating coils (Figure 6.1b). The steam is supplied at variable flow rates by a natural
gas-fired boiler (with a thermal efficiency of 80%). Inside the dryer, the air periodically flows
in opposite directions to ensure uniform heating and drying. The system‘s drying control
software includes several programmable modes, such as wood preheating and conditioning,
hybrid (with heat pump and back-up heating) and conventional (with electricity or steam
heating). This control software displays data such as the compressor, fans, heating valves and
air vents current status, elapsed operating time, dry and wet-bulb temperatures, compressor
suction and discharge pressures and temperatures, alarms, etc. About 30 electrical and
thermodynamic parameters (temperatures, pressures, electrical powers, etc.) are measured at
15-second intervals, and averaged and saved every 2 minutes.
Table1 6.1. Example of hardwood drying schedule.
Drying step Dry bulb Wet bulb Heat hump running time
- °C °C %
1 Preheating 37.8 32.2 -
2 37.8 32.2 100
3 40.5 33.9 85
4 43.3 34.4 75
5 46.1 34.4 65
6 54.4 34.4 75
7 57.2 35.0 85
8 60.0 35.0 100
6.1.2. Drying schedule and control

Dry- and wet-bulb temperatures inside the drier, as well as the compressor running times
are scheduled based on the actual wood moisture content (Table 6.1). During the first step
(preheating), the compressor is not running. The heating coils provide heat to increase the
temperature of the wood stack. When the wet-bulb temperature equals its setting point, the
heat pump starts. The heat pump compressor (and blower) runs according to empirical
correlations between the required time (in hourly percentage) and the actual operating time
[27]. For example, if the hourly percentage of operation is set at 60%, the heat pump will run
for 30 minutes and will shut down during the next 20 minutes. The heat pump runs only when
the dryer wet-bulb temperature is within a preset range. According to this range, the
compressor running time must to be increased when the actual wet-bulb temperature is higher
than the upper limit, and decreased when it is below the lower limit. When required, the dryer
uses one of back-up heating sources (electricity or steam). The back-up heating is controlled
according to the wet-bulb temperature settings. In addition, if the actual dry-bulb temperature
is higher than its set point, the air vents automatically open. If the dry-bulb temperature is
higher than its maximum preset value, the heat pump shuts down. In the conventional drying
mode, additional heating energy is supplied when the actual drier wet-bulb temperature is
lower than its setting point by a preset temperature difference, and shuts down when it is
higher. The air vents open when the wet-bulb temperature is higher and close when it drops
below the set point.
6.1.3. Energy consumption and costs

Two all-electrical drying tests with low-temperature heat pump and electrical back-up
coils are presented (Table 6.2). The initial moisture contents (dry-basis), representing the
weight of the water contained in the wood expressed as a percentage of its anhydrous mass,
were respectively 29.1% (test #1) and 40.7% (test #2). The total drying time, including the
preheating steps, was 147.03 hours (6.12 days) (with yellow birch) and 240.67 hours (10
days) (with hard maple) respectively. The final moisture contents (oven-measured, dry-basis)
were 7.4% and 7.8%, respectively. These final moisture contents are within normal ranges for
hardwood drying. The heat pump (compressor and blower) assumed 30% (test #1) and 21%
(test #2) of the total electrical energy consumption, respectively. The energy consumption of
electric back-up coils accounted for 62% (test #1) and 61% (test #2), and the drier fan, for 8%
(test #1) and 11% (test #2) (Figure 6.2).
D r ier F an: 8 %
Heat Pump : 3 0 %
Elect r ical B ack- U p C o ils: 6 2 %
Figure 6.2. All-electrical drying test #1: share of electrical energy consumption .
56 Vasile Minea
Table 6.2. All-electrical drying tests - energy consumptions.
Test # Wood Moisture content Energy consumption

- - MCin MCfin Heat pump Dryer Back-up
- - - - Compressor Blower Fan Electricity
- - % kWh kWh kWh kWh
#1 Yellow birch 29.1 7.4 747 151 229 1 828
#2 Hard maple 40.7 7.8 872 258 451 2 441
MCin : initial moisture content; MCfin: final moisture content
Table 6.3. Hybrid and conventional drying tests - energy consumptions.
Test Hardwood Moisture content Electrical energy consumption Operation rate vs. cycle time
specie
- - MCin MCfin Heat pump Dryer Heat pump Back-up
- - - - Compressor Blower Fan Compressor Steam
- - % % kWh kWh kWh % %
#3 - hybrid Hard maple 31.1 7.5 650 129 360 65 39
#4 - CONV Hard maple 36.4 7.5 - - 638 - 70
#5 - hybrid Yellow birch 75.9 7.6 902 201 400 84.7 40
Two additional drying tests with the low temperature heat pump and steam as back-up
energy (tests #3-hybrid and test #5-hybrid), and a conventional drying cycle (test #4 - CONV)
are also presented (Table 6.3).
With hard maple, the total drying time of the hybrid test #3, including its first
(preheating) step, was 16.5% longer as compared to the conventional drying test #4, while
both final moisture contents were identical (7.5%). However, the equivalent energy
consumption of the hybrid test #3 was more than 50% lower as compared to the energy
consumption of the conventional test #4 that used natural gas as heating energy source.
Compressor Running Percentage Test #5: Hybrid

120
100% 100%
Compressor % and
100
85%
Temperature (°C)
Step 1 75% 75% 85%

80
65%
60
Set Wet Bulb (Blue Line)
40
20
Actual Wet Bulb (Red Curve)
0
0 24 48 72 96 120 144 168 192 216
Time
Figure 6.3. Hybrid drying test #5: compressor running time (in %) and set and actual wet-bulb
temperatures.
D r ier F an: 9 % Heat Pump : 2 5%
St eam B ack- U p C o ils: 6 6 %
Figure 6.4. Hybrid drying test #5: Share of energy consumption .
The intermittent operation of the heat pump was preset prior each step of the drying
cycle. It was also frequently adjusted during the drying process in order to keep the actual
wet-bulb temperature close to its setting point. For the hybrid test #5, Figure 6.3 represents
the correlation between the heat pump‘s hourly percentages of operation and the setting and
actual wet-bulb temperatures. It can be seeing that the actual wet-bulb temperature has
perfectly fitted the set wet-bulb values by intermittent operation of the compressor. It can be
also seeing that when the initial moisture content (MCin) was relatively high (75.9% - test #5),
the compressor has operated 84.7% of the time. However, the compressor has run only 65%
of the time when the MCin was significantly lower (31.1% - test #3-hybrid). The heat pump
(compressor and blower) assumed 25% of the total energy consumption, while the drier fan
and the back-up heating (steam) consumed 9% and 66%, respectively (Figure 6.4). Relatively
high percentage of steam coils operation can be explained by the poor thermal insulation of
the experimental drier and its relatively high air leakage rates
Compared to the consumption of the conventional test #4-CONV (719 m3), the natural
gas consumption of the hybrid drying test #5-hybrid (350 m3) was 56% lower. Consequently,
by assuming 40 drying cycles per year, the natural gas saving would account for about 15 000
m3/year. The equivalent emissions of CO2 would consequently be reduced by 30.6 tons per
year and per drier. This takes into consideration approximately 40 kg of CO2 emissions per
year due to the heat pump additional electric energy consumption. If 500 small-scale
hardwood driers with low-temperature heat pumps would be installed, the annual reduction in
greenhouse emission would be of 15 300 tons of CO2. This estimation is done for the
province of Quebec (eastern Canada), where 97% of the electricity production is
hydroelectric. In this case, the regional conversion factor is 0.00122 kg of CO2 per kWh of
electrical energy consumed. Compared to the energy costs of the conventional drying cycle
with natural gas (test #4: CONV), the total energy costs (electricity and natural gas) of drying
cycles with low-temperature heat pump have been reduced by 20% (test #2 – all-electrical)
and by 23% (test #5 - hybrid), respectively (Figure 6.5) (2004).
58 Vasile Minea
Total Energy Cost, US$/cycle

Yellow Birch,
300
Yellow Birch MCin = 36.4% Hard Maple
250
Hard Maple MCin = 40.7% Mcin = 75.9%
200
MCin = 29.1%
150
100
50
0
Test #1: All- Test #2: All- Test #4: CONV Test #5: Hybrid
electrical electrical
Drying Test
Figure 6.5. Drying total energy costs; MCin: wood initial moisture content.
Yellow Birch Hard Maple

2500 MCin = 75.9%
Total Water Extraction, L
MCin = 40.7%
MCfin = 7.8% MCfin = 7.6%
2000
Hard Maple
1500
MCin = 29.1%
MCfin = 7.4%
1000
500
0
Test #1: All-electrical Test #2: All-electrical Test #5: Hybrid
Drying Test
Figure 6.6. Total volumes of water extraction.
2500
Hard Maple Yellow Birch
Total Water Extraction, L
MCin = 40.7% MCin = 75.9%

2000
TOTAL per Cycle
1500
Yellow Birch Below FSP
Mcin = 29.1% Above FSP
1000
500
0
Test #1 Test #2 Test #5
Figure 6.7. Total water extraction versus the fibre saturation point.
6.1.4. Water extraction

The volume of condensed water strongly depends on the wood initial moisture content
(Figure 6.6). The test #1 (all-electrical), with relatively low initial moisture content (29.1%),
has produced 2.5 times less water as compared to the test #5-hybrid, where the initial
moisture content was much higher (75.9%). This result was obtained even if the quantity of
the dried wood (hard maple) was 33.6% higher in all-electrical test #1 compared to the wood
quantity in the hybrid test #5.
The fiber saturation point (FSP) is the physical state where the cell cavities are
completely devoid of free water and their walls are still completely saturated. For hardwood,
the moisture content of the FSP is basically of about 25% (dry-basis). It can be seeing that
with an initial moisture contents lower than 41% (all-electrical test #1), the volume of water
extracted below the FSP was much higher than the volume extracted above the FSP (Figure
6.7). However, when the wood initial moisture content was approximately 41% (test #2-
hybrid), the volume of water condensed above the FSP was practically equal to the volume
removed below this point. Finally, when the initial moisture content was significantly higher
than 41% (75.9% in the test #5-hybrid), the quantity of water extracted above the FSP was
about three times higher than the quantity removed below the FSP.
6.1.5. Dehumidification rate

For every wood species and thicknesses, there is a safe drying rate at which the moisture
can be removed. In other words, there is a rate at which the wood can be dried with little or no
significant degradation or damage. On the other hand, the water evaporation rate depends on
the amount of energy supplied and the capacity of the drying air to absorb moisture. To
maintain a constant drying rate, the water molecules in the wood must be supplied with
additional energy or the partial vapor pressure of the drying air has to be lowered. This is
achieved either by raising the temperature or reducing the relative humidity of the drying air.
Exceeding a maximum safe evaporation rate increases the risks of drying defects (splits,
cracks or checks). However, drying at a rate substantially below the safe rate also causes a
risk of drying defects (increase in warp, stain and uneven drying). For example, the safe
drying rate for hard maple lumber is 6.5% of moisture loss per day. Drying rates also provide
a method of estimating drying cycle times. For conventional drying cycles, this parameter is
usually expressed in terms of percent moisture content loss per day. With low-temperature
heat pump as a dehumidifier, the drying rate is expressed in terms of average water extraction
volume (L) per hour. In this case study the average water extraction rate per cycle slightly
increased according to the wood initial moisture content. For example, with yellow birch, it
increased by 10.3% while the initial moisture content rose from 29.1% (all-electrical test #1)
to 75.9% (hybrid test #5). However, the water extraction rates below the FSP decreased by
2.4 times (test #1) and 3.4 times (test #5) as compared to their respective rates above the FSP
(Figure 6.8).The water extraction rate above the FSP is thus little sensitive to the total water
volume removed. Effectively, if we compare the tests #1 and #5 with yellow birch, the water
extraction rates above the FSP were practically similar (13 L/h and 14.5 L/h, respectively),
even if the total water volume removed during the test #1 was more than five times lower
than the volume extracted during the test #5. On the other hand, below the FSP, the water
extraction rates were proportional to the gap between the FSP and final moisture contents
60 Vasile Minea
(about 17.5%) and practically equal (4 to 5 L/h), regardless the dried specie (yellow birch or
hard maple) and the respective water volumes removed.
Hard Maple
20
MCin = 40.7% Yellow Birch
Water Extraction Rate, L/h

Yellow Birch MCin = 75.9%
16
MCin = 29.1%
AVRG per cycle
12
Above FSP
8
Below FSP
0
Test #1: All-electrical Test #2: All-electrical Test #5: Hybrid
Figure 6.8. Water extraction rates versus FSP; AVRG: average.
Table 6.4. Dehumidifying performances of the low-temperature drying heat pump.
Test MCin Specific moisture extraction ratio (SMER) Comparison -

- Initial moisture Above FSP FSP = Below FSP Above vs. Drying
content 25% bellow FSP time
- % kgwater/kWhhp Hours* kgwater/kWhhp Times Hours
#1: all- 29.1 2.06 25 0.82 2.5 138.3
electrical
#2: all- 40.7 2.5 72 1.19 2.1 220.0
electrical
#5: hybrid 75.9 2.5 104 0.87 2.9 210.3
* From the beginning of the preheating step
6.1.6. Dehumidification performances

The dehumidification efficiency of drying heat pumps is generally expressed by the
Specific Moisture Extraction Rate (SMER). This parameter represents the ratio between the
mass of water extracted and the heat pump total electrical energy consumption (compressor
and blower) (kgwater/kWhhp). The laboratory tests show that above FSP, the SMER was of 2.5
kgwater/kWhhp (Table 6.4) representing a very good performance for a dehumidification
process. Table 6.4 also indicates for each test at which moment, measured in hours, from the
beginning of each drying cycle, the FSP was attained. It depends on the initial moisture
content. When MCin was relatively low (29.1% in test #1), the FSP was reached after 25
hours. When MCin was higher (75.9% in thest #5), it was achieved after 104 hours of
operation.
6.1.7. Conclusions
A 5.6-kW low temperature heat pump coupled to a 2.3 m3 wood dryer was tested with
electricity and natural gas (steam) as back-up energy sources. The heat pump electrical energy
consumption (compressor and blower) varied between 25% and 30% of the total equivalent
energy consumption of all-electrical or hybrid drying cycles. The dryer fan generally assumed
8% to 9%, and the electrical or fossil back-up energy – between 62% and 66% of the total
drying energy consumptions. For initial moisture contents higher than 41%, the total water
quantities extracted above the fiber saturation point were up to 2.9 times higher than those
removed below the FSP. Consequently, in these cases, the dehumidification efficiency of the
low-temperature drying heat pump (MSMER) was up to 3 times higher above than below the
FSP. Finally, the hybrid drying cycles have reduced the natural gas consumption by 56% and
the equivalent energy costs by 21.5%, compared to the conventional drying cycle with natural
gas as unique heating source [25].
6.2. High-Temperature Wood Drying Heat Pump
The technology of high-temperature heat pumps is not yet available, all the more so since
early 90s environmental issues imposed the replacement of traditional CFCs used as high-
temperature refrigerants. The merits of high-temperature drying heat pumps however include
lower energy consumed for each unit of water removed, accurate control of drying conditions,
and enhanced product quality. Their limitations generally concern the need for temperature
resilient materials and fluids (refrigerants, oils, belts, etc.) regular maintenance, the risk of
refrigerant leaks and higher initial capital costs compared to conventional dryers.
6.2.1. System configuration

3
The experimental site consists in two air-forced 354 m wood dryers made of insulated
panels, each including 1500-kW steam heating coils (Figure 6.9). An oil-burned boiler of
4900-kW output capacity and 82% thermal efficiency supplies both dryers with high-pressure
saturated steam for heating and spraying. One of these dryers has been equipped with two 65-
kW (compressor nominal power input) high-temperature heat pumps [26]. A 56-kW multiple-
blade fan with outdoor motor provides forced circulation of the air at 1.5 – 2.0 m/s at the
stacks of wood outlet. Mural deflectors and inversion of the rotation of the drier fan at every 3
hours at the beginning and at every 2 hours at the end of the drying cycles contribute to
obtaining uniform ventilation. To avoid air implosion risks, three air vents open when the
central fan changes its rotation direction and, also, when the actual dry-bulb temperature
exceeds the set point. The high-temperature heat pumps HP-1 and HP-2 equipped with
variable speed blowers, compressors, evaporators, as well as electric and electronic controls
are installed within an adjacent mechanical room. However, the condensers of both heat
pumps are remote installed inside the drying enclosure. Designed for industrial processes, the
open-, belt-driven compressors are provided with oil pumps, external pressure relief valves
and crankcase heaters. The refrigerant is a non-toxic and non-flammable fluid, having a
relatively high critical temperature compared to the highest process temperature and a normal
boiling point les than the lowest temperature likely to occur in the system. Moreover, the
saturation vapour pressure at highest design temperature is not so high as to impose design
limitations on the system. Expansion valves are incorporated into the microprocessor-based
temperature/process controllers that display both set points and actual process temperatures.
62 Vasile Minea
Mechanical room
Experimental drier Conventional drier Control room
Air Supply
HP-2
C Air Return
HP-2
F
(Hybrid
CD )
21 m
EV
Drying Heat
Pumps Room Steam
HP-1
C
CD
Oil boiler
HP-1
EV
Steam
coils
5m
Drier fan
motor
Oil storage tanks

Wood stacks
Train rails
Figure 6.9. View of the industrial-scale experimental site.

Heat pump operating profile White spruce Dryer wet-bulb temperatures White spruce
80
80
70 Actual
70
Temperature, °C
60 60
Compressor input power
50 50 Setting
40 40
30
30
20
20
10
10
0
0
0 10 20 30 40 50 60
0 10 20 30 40 50 60
(a) (b)
Figure 6.10. Profiles of dryer wet-bulb setting and actual temperatures.
6.2.2. Drying schedule

This case study presents some results from two tests with white spruce (#70 and #88) and
one test with balsam fir (#176) (Table 6.5). All drying cycles included 6 to 8 hours preheating
steps at average temperatures of 93.3°C in order to destroy the micro-organisms responsible
for discoloring the sapwood. The drying conditions of each step were established based upon
moisture content, type of wood species, dimensions and quality of the wood. For white
spruce, which is normally easy to dry, at initial moisture content of between 40% and 25%,
the setting dry-bulb temperature was between 82.2°C and 85°C and the wet-bulb temperature
of 62.7°C. At a moisture content of less than 25%, the dry-bulb temperature was generally set
at 79.4°C and the wet bulb-temperature, at 62.7°C. With balsam fir, which is harder to dry,
when the initial moisture content was above 30%, the dry-bulb temperature was set at 82.2°C
and the wet-bulb at 79.4°C.
At moisture contents lower than 25%, the setting dry-bulb temperature attained 93.3°C
whereas the wet-bulb temperature was of 71.1°C. Changes in dry- and wet-bulb temperature
settings were achieved on predetermined time-based schedules. For white spruce, steps 1 to 3
generally lasted 10 hours, the step 4 held out 20 hours, and step 5, 10 to 20 hours depending
upon the wood actual moisture content. In the case of balsam fir, the first five drying steps
each lasted 30 hours, while the 6th step lasted up to 15 hours. The scope was to not exceed
the average time of traditional drying cycles for the same species of dried wood. Finally,
when the indoor dry-bulb temperature was lower than the set point value, the steam valve
opened gradually from 5% to 100% upon a time-based schedule to fully recover the set point
[27].
6.3. Thermodynamic Parameters
The control strategy allowed the heat pump compressors to shut-down every time the
actual drier wet-temperature reached the set-point (Figure 6.10a).
The drying time was of 61.3 hours and do not includes the approximately 6-hour
preheating step. The heat pump compressors ran with electrical power inputs of 65 kW, and
average compression ratios of 5.5. Stable suction and discharge pressure as well as average
refrigerant sub-cooling of 8°C were achieved. The condensing temperatures varied around
105°C, about 20°C higher than the kiln dry-bulb temperature. The evaporating temperatures
64 Vasile Minea
ranged between 41°C and 45°C. The average relative humidity of the air entering the
evaporators largely varied because of periodical changes in the rotation direction of the
central fan, and has also continuously decreased in time. However, the relative humidity
leaving the evaporators was almost constant at around 74% to 88%, except at the end of the
cycle when it dropped to 70% (Figure 6.11).
The preheating steps allowed increasing the kiln‘s absolute humidity up to 0.35 kg/kg
before the dehumidifying process start-up (Figure 6.12). The maximum gradient of the
absolute humidity across the heat pump evaporator varied from 0.214 kg/kg, immediately
after starting the compressors, to about 0.039 kg/kg at the end of the showed drying cycle.
The pick drying rate and efficiency of the heat pump were significantly higher than the
average values, due to a higher wood moisture production rate at the beginning of the cycle.
Figure 6.13 represents three dehumidifying processes based on the measured parameters at
the beginning, at the end and as cycle average values. The gradient of the air enthalpy of the
average drying cycle is of about 184 kJ/kg, while the dry-bulb temperature varied from
82.2°C (inlet) to 54.4°C (outlet), and the corresponding wet-bulb temperature, from 65.8°C to
48.8°C.
6.2.4. Energy performances

The wood moisture content prior the drying cycle was generally in the range of 35 to
45% (dry basis) (Table 6.5). The average coefficient of performance (COP) of the drying heat
pumps, defined as useful thermal power output (W) divided by electrical power input (W),
varied from 3 to 4.6.
HP-1 Evaporator: Air Relative Humidity Batch #88: Avril 9-12, 2003
100
90
Relative Humidity, %
80
70
Outlet
60
50
40
30 Inlet
20
10
0
0
8
2
0
0
8
8
6
6
4
10
18
12
20
02
12
:3
:3
:5
:0
:3
:3
:4
:5
:2
:3
2:
6:
3:
7:
4:
8:
10
14
18
23
11
15
19
23
12
16
Time
Figure 6.11. Air relative humidity entering and leaving the heat pump evaporator.
Evaporator Heat Pump-1 Batch #88: Avril 9-12, 2003

0,40
Absolute Humidity, kg/kg

0,35
Inlet
0,30
0,25
0,20
0,15
0,10
0,05
Outlet
0,00
8
0
8
2
4
2
4
10
22
36
48
40
50
04
:3
:5
:0
:2
:0
:1
:2
:2
:3
2:
6:
3:
7:
0:
4:
9:
10
14
19
23
12
16
20
13
17
Time
Figure 6.12. Air absolute humidity entering and leaving the heat pump evaporator.
h (kJ/kg)
714
460
53%
EE
START 0,23
418 EE 35%
400 LE 88% 0,136
0.134
Absolute Humidity, kg/kg
30%
AVRG EE
0.113
276
263
74%
LE END
0,0786
LE 0,074
70%
54.4 (130°F) 60 (140°F) 82.2 (180°F)

Bry Bulb Temperature, °C
EE: entering evaporator; LE: leaving evaporator; START: start of the cycle; AVRG: average; END:
end of the cycle
Figure 6.13. Dehumidifying processes.

66 Vasile Minea
kgwater / hour
The total water extraction rates were of 313 (batch #70), 263.2
kgwater / hour kgwater / hour
(batch #88) and 178.8 (batch #176). These numbers do not
kg / hour
include the venting moisture losses (on average, 90 water ), but account for 5% of
condensed water losses. The Specific Moisture Extraction Rate (SMER), defined as the
amount of water extracted by the heat pump (kg) and the total energy input (compressor and
kgwater / kWh
blower) expressed in kWh, rabged from 1.46 (batch #176) to 2.52
kgwater / kWh
(batch #70). The Specific Energy Consumption (SEC), ranged from 0.4 to
0.68
kWh/ kg water . These values not include the energy consumption during the preheating
steps, as well as allowance for the energy consumption of the kiln‘s central fan or the venting
moisture losses.
The energy consumption of the dehumidification cycles with high temperature heat
pumps was 27% to 57% lower than the energy consumption of conventional (steam) drying
systems. The average reduction in specific energy costs, compared to the costs of
conventional wood drying cycles, was estimated at approximately 35%.
The heat pump (compressor plus blower) assumed 72% and the dryer central fan 28% of
the total energy consumption. The drying time to obtain white spruce with an approximate
final moisture content of 17% to 19% was of about 2.5 days, while for balsam fir it averaged
6.3 days. Despite its longer duration, the drying of the balsam fir was less concerned with
drying speed than that of the white spruce, the main operating focus values of which were to
3
produce a high quality product. The facility specific cost for about 39,600 m of dried
3
lumber was of 14.75 US$/ m including kiln operation, electrical and fossil energy
consumption, equipment depreciation, insurance, etc. The energy cost only was of 6.86 US$/
m 3 . The objective is to reduce the energy specific cost by at least 40%.
8000
7000
Condensed water, Litres
HT-HP #2 (7514 L)
6000
5000
HT-HP #1 (7248 L)
4000
3000
2000
1000
0
0 10 20 30 40 50 60 70
Time, hours
Figure 6 14. Cumulative volumes of wood water extraction.

Table 6.5. Energy performances
Test # - #70 #88 #176

Parameter Unit - - -
Timber - White spruce White spruce Balsam fir
Drying time HP-1 61.00 61.3 151.4
(excluding preheating steps) (hours) HP-2 61.00 61.3 151.4
Average compressor power input (kW) HP-1 65.12 63.36 61.0
HP-2 62.78 58.50 57.14
Compressor energy consumption (kWh) HP-1 3,972 3,884 9,235.4
HP-2 3,830 3,586 8,651.0
Blower energy consumption (kWh) HP-1 13.42 16.6 28.7
HP-2 14.03 39.5 107.5
Water extraction (Liters) HP-1 9,454 8,263 13,550
HP-2 9,655 8,478 13,531
Final moisture content (%) - 17.2 20.6 20.7
Average COP* (-) HP-1 4.23 4.6 3.46
HP-2 3.70 4.07 3.00
kgwater / kWh HP-1 2.38 2.13 1.46
Average SMER** ( )
HP-2 2.52 2.36 1.54
kWh/ kgwater HP-1 0.42 0.47 0.68
Average SEC** ( )
HP-2 0.40 0.42 0.64
* –based on compressor and blower energy consumptions
6.2.5. Operating lessons learned

Spread over several months, the first development step of high-temperature heat pump
prototypes was aimed at ensuring a maximum operational stability of the thermodynamic
parameters, establishing the reliability of the most sensible components (compressors,
blowers, safety valves), checking the refrigerant/oil blend‘s behavior and optimizing the
critical control sequences of the system. The experimental dehumidifier dryer operated in
―extreme‖ temperature, humidity and corrosion conditions, and demonstrated specific
feedback phenomena that do not occur in traditional air-forced/heated lumber dryers. Because
some of the by-products that could be produced by chemical interactions between the
lubricant and working fluid may be acidic and lead to accelerating the corrosion of the system
components, periodical controls aimed to determine the chemical behavior of the mixture.
After more than 4000 hours of operation, the refrigerant proved to be thermally stable and
chemically inert at the highest temperatures occurring in the system and a first oil chemical
analysis proved that there were no problems with the oil breaking down or failing. The oil
still showed adequate viscosity and chemical stability as well as a good miscibility with the
refrigerant. The initial designed capacity of the heat pumps proved to be too high and
consequently, both compressors were slowed down by about 25% which finally resulted in a
more adequate capacity, reduced head pressures and improved efficiency. Because the
original employed expansion valves poorly controlled the refrigerant flowing, did not open
68 Vasile Minea
fast enough and manufacturer leakage faults were detected, they were later replaced by new
generation devices. A crack in the casting of an original pressure relief valve also indicated a
manufacturer‘s defect, and finally both these components have been replaced to 20% higher
pressure-limit valves. Other issues included the fact that the kiln was initially poorly insulated
and leaky, and some system‘s components were prematurely corroding [28].
6.2.6. Conclusions
As a clean energy technology compared with traditional heat-and-vent dryers, high-
temperature heat pump dehumidifiers offer interesting benefits for drying resinous timber.
This chapter presents the preliminary results of the development and field testing of two
prototypes accentuating their thermodynamic parameters, preliminary energy performances
and first operating lessons learned. The average measured specific moisture extraction rate of
kg / kWh kg / kWh
the heat pumps was 2.35 water (white spruce) and 1.5 water (balsam fir),
while the average coefficients of performance generally varied from 3.0 to a maximum of 4.6.
The cycle‘s duration ranged from 2.5 days (white spruce) to 6.3 days (balsam fir) including
the initial preheating steps. The refrigerant/oil mixture behaved well during more than 4000
hours of preliminary tests, proving good compatibility and chemical stability at condensing
temperatures below 110°C. Better insulated and well maintained dryers are necessary to
obtain drying temperatures higher than 100°C as well as reducing the drying duration of
resinous species by up to 25% and the total energy consumption by up to 50%. The current
goals of the study include using more corrosive resistant components, variable speed central
fan, further optimizing the drying schedules and general dryer operation and maintenance.
Finally, it is expected to help local Canadian equipment suppliers to promote research and
development of the technology and develop an appropriate market strategy. Specifications of
high-temperature heat pump dehumidifier kiln energy use and a best-practice guideline must
also be produced.
REFERENCES
Mujumdar, A. S. (2000). Mujumdar‟s Practical Guide to Industrial Drying, Principles,
Equipment and New Developments, Edited by S. Devahastin.
Mujumdar, A. S. (2008). Guide to Industrial Drying. Principles, Equipments & New
Developments, IDS2008, Hyderabad, India.
IEA/OECD CADDET Energy efficiency, (1997). Analyses series No. 23, September.
Incropera, F. P. & DeWitt, D. P. (2002). Fundamentals of Heat and Mass Transfer, Fifth
edition, John Wiley & Sons.
Rademacher, R. & Hwang, Y. (2000). Vapor Compression Heat Pumps with Refrigerant
Mixtures, University of Maryland.
Minea, V. (2008). Design and control optimizations of drying heat pumps, Proceedings of the
16th International Drying Symposium - IDS2008, Volume B, 739-745, Hyderabad,
India.
Alves-Filho, O. & Tokle, T. (1999). Heat pumps dry food products in Norway, IEA Heat
Pump Centre Newsletter - volume 17, no. 4
Alves-Filho, O. & Eikevik, T. (2007). Atmospheric freeze drying with heat pumps – new
possibilities in drying of biomedical materials, The 5th Asia-Pacific Drying Conference,
13-15 August, Hong Kong.
Wera, P., Raghavan, G. S. V. & Terdtoon, P. (2005). Loop thermo-siphon application for
heat pumping drying, IADC 2005, 3rd Inter-American Drying Conference, Montréal,
Canada.
Strommen, I., Eikevik, T. & Claussen, I. (2007). Atmospheric freeze drying with heat pumps
– new possibilities in drying of biomedical materials, The 5th Asia-Pacific Drying
Conference, 13-15 August, Hong Kong.
Li Min Xia, et al. (2007). Two-stage drying of CO2 trans-critical cycle heat pump. The 5th
Asia-Pacific Drying Conference, 13-15 August, Hong Kong.
Jinjiang, Z. & Yaosen, W. (2007). Experimental study on drying high moisture content paddy
by super-conducting heat pump dryer. The 5th Asia-Pacific Drying Conference, 13-15
August, Hong Kong.
Strommen, I., Eikevik, T., Alves-Filho, O. & Syverrud, K. (2004). Heat pump drying of
sulphate and sulphite cellulose. Proceedings of the 14th International Drying
Symposium (IDS2004), Sao Paulo, Brazil, 22-25 August, vol. B, 1225-1232.
Alves-Filho, O., Garcia, M. L., Strommen, I. & Eikevik, T. (2006). Heat pump atmospheric
sublimation and evaporation drying of bovine intestine, 15th International Drying
Symposium (IDS2006), Budapest, Hungary, 20-23 August.
Eikevik, T., Strommen, I. & Alves-Filho, O. (1999). Design and dimensioning criteria of heat
pump dryers. 20th International Congress of Refrigeration, IIR/IIF, Sydney, Australia
Liang, Z. B. et all. (2007). Experimental study on sludge drying with heat pump dehumidifier.
The 5th Asia-Pacific Drying Conference, 13-15 August, Hong Kong.
Cech, M. J. & Pfaff, F. (2000). Operator Wood Drier Handbook for East of Canada, edited
by Forintek Corp., Canada‘s Eastern Forester Products Laboratory.
Minea, V. (2008). Energetic and ecological aspects of softwood drying with high-temperature
heat pumps, Drying Technology International Journal, Volume 26, Issue 11, November
pages 1373-1381.
Shottafer, J. E. & Shuler, C. E. (1974). Estimating Heat Consumption in Kiln Drying Lumber,
Life Sciences and Agriculture Experiment Stat, Technical Bulletin, 73, University of
Maine
Minea, V. (2005). Laboratory and Industrial-Scale High Temperature Heat Pumps
Prototypes for Wood Drying, IADC 2005, Inter-American Drying Conference,
Montréal, Canada, August, 21-23.
Minea, V. (2009). Avoiding failures of large-scale high-temperature drying heat pumps,
Proceedings of The 6th Asia-Pacific Drying Conference (ADC2009) October 19-21,
Bangkok, Thailand, 179-187
Minea, V. (2007). High-temperature heat pump for wood drying – refrigeration circuits and
control improvements, Hydro-Québec patent disclosure.
Kowalski, S. J. & Pawlovski, A. (2008). Drying of wet materials in intermittent conditions,
Proceedings of the 16th International Drying Symposium – IDS2008, 951-957,
Hyderabad, India.
Minea. V. (2009). Improvements of compact and split high-temperature drying heat pumps,
Proceedings of the 8th World Congress of Chemical Engineering and Inter-American
Drying Conference IADC‟09, August 23rd – 27th, Montréal, Canada.
70 Vasile Minea
Minea, V. (2006). Hardwood Drying with Low Temperature Heat Pumps, Proceedings of the
15th International Drying Symposium, Volume C, pp. 1757 – 1762, Budapest, Hungary,
20-23 August
Minea, V. & Normand, D. (2005). Séchage de bois avec pompe à chaleur basse température,
Hydro-Quebec Report.
Minea, V. (2004). Heat Pumps for Wood Drying – New Developments and Preliminary
Results, Proceedings of the 14th International Drying Symposium, Sao Paulo, Brazil,
August 22-25, 2004, Volume B, 892-899.
Minea, V. (2005). High-temperature Heat Pumps for Resinous Timber Drying, 8th
International Energy Agency, Heat Pump Conference 2005 – Global Advances in Heat
Pump Technology, Applications, and Markets, Las Vegas, Nevada, USA, May 30 –
June 2.
Minea, V. (2007). Industrial high-temperature heat pumps for wood drying, IEA Heat Pump
Centre Newsletter, Volume, 25, No. 1/2007, 14-19, April.
View publication stats

1.industrialdryingheatpumps Minea 2011

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

1.industrialdryingheatpumps Minea 2011

Uploaded by

Copyright:

Available Formats

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

Industrial drying heat pumps

Article · January 2011

Preliminary Assessment of Deep Geothermal Resources in Trois-Rivières Area, View project

The user has requested enhancement of the downloaded file.

INDUSTRIAL DRYING HEAT PUMPS

High annual factors of utilisation;

 Higher initial capital cost;

Although heat pump-assisted drying technology is generally considered as a mature

 Uncertainty by potential users as to heat pump reliability,

Absolute humidity, kg/kg

Figure 2.1. Simplified schema of a drying system.

water over the wet solid to the equilibrium vapour pressure (

2.2. Drying Schedules

2.3. Control of Driers

Modern drying technologies make it imperative to found solutions allowing optimizing

2.4. Drying Efficiency

Because at low humidity and temperatures, energy is proportional to temperature

difference between inlet (

It is obtained when exhaust is saturated at wet-bulb temperature (Twb).

3.4. Heat Recovery Methods

3.5. Heat Pumps

In the common thermodynamic sense, a heat pump is a system in which refrigeration

3.5.2. Advanced cycles

Flash tank C Expander C

Inter-cooler EX2 EX1

3.5.3. Working fluids

Table 3.1. ODP and GWP values for some refrigerants.

Refrigerant ODP GWP (100)

R-236fa - saturated temperatures R-245fa - saturated temperatures

Figure 3.4. Saturated temperatures of two high-temperature refrigerants; a) R-236fa; b) R-245fa.

Type of production CO2 emission

3.5.4. Design of heat pumps

where Q is the thermal capacity of the designed heat exchanger.

where τ is the operating time.

Where the compressor discharge point 2s is defined by

isentropic efficiency varies with the pressure ratio (   p2 / p1 ) as follows:

 s  a0  a1  a2 2 (3.7)

the compressor motor (

The expansion device EX can be a thermostatic, electric or electronic valve. It controls

3.5.5. Performance of heat pumps

COP  CarnotCOPCarnot (3.16)

4. DRYING HEAT PUMPS

4.2. Energy Performances

Air Air damper

Refrigerant cycle in p-h diagram Air cycle in Mollier diagram

4.3. Payback Period

FEP is the fuel energy price ($/kWh),

4.4. Optimization Requirements

Heat rejection coil Heat rejection coil

Figure 4.2. Published configurations of heat pump dryers.

Initial Drying Moisture

dehumidification capacity. In return, data on MSMER values at different drying temperatures

4.4.2. System integration

4.4.3. Refrigerant side

The conservation of energy for the sub-cooler is represented by:

4.4.4. Air side

where the bypass factor (  ) may be less than or equal to 1.

The electrical power input of the air blower is converted to heat:

Where  is the air absolute humidity (kgwater/kgdry air).

5. WOOD DRYING HEAT PUMPS

5.1. Wood drying

5.1.1. Moisture movement in wood

   EV M in,hum MCindry  MC dry