03-07-2013, 04:48 PM
Experimental Compression – Resorption Heat Pump For Industrial Applications
Experimental Compression.docx (Size: 22.81 KB / Downloads: 20)
ABSTRACT
Wet compression - resorption heat pumps are a promising type of heat pumps that operates with nonazeotropic
refrigerant mixtures. The main differences, when compared to the classical Rankine cycle, are the nonisothermal
phase transition of the mixture in the heat exchangers and the compression of the two-phase mixture in
the compressor, which works as a gas compressor and at the same time as a liquid pump. Wet compression results
in reduction of the consumed power, excludes the vapour superheating and is especially attractive for high
temperature applications. For such applications estimated gain in COP is up to 20% when compared to
conventional dry compression heat pumps.
A 50 kW experimental ammonia water heat pump of this type has recently been taken in operation. It has been
designed to upgrade a water flow from 110 to 130°C, using a water waste flow with 80°C. The resorber and
desorber are of the falling film vertical shell-and-tube type, with the ammonia water film inside the tubes and a
distribution system in the top header. The intermediate heat exchanger is of the shell-and-plate type. Oil-free wet
compression of ammonia water was obtained with an especially designed twin-screw compressor. Details and
performance results of this compressor are discussed in a parallel paper.
This paper discusses the experimental set-up design and the experimental data of resorber, desorber and
intermediate heat exchanger.
INTRODUCTION
Compression - resorption heat pumps have the potential to give significant contributions to the
improvement of the energy performance of heating processes. Specifically for industrial heating processes they
allow for energy performance gains of more than 20% when compared with vapor compression heat pumps.
Ammonia-water high temperature heat pumps that are used to upgrade industrial waste heat show a number
of advantages:
There are two types of compression - resorption heat pumps: with solution circuit (liquid pump and saturated
vapor compressor are used to overcome the pressure differential) and with two-phase (wet) compression. Here the
compressor simultaneously compresses the vapor and increases the liquid pressure. Itard [1998] has shown that,
specially for high temperatures, wet compression leads to higher energy efficiencies than the solution circulation
alternative. Next to this energy advantage, wet compression also allows for higher operating temperatures. Dry
compression in the solution circulation variant leads to large superheating temperatures of the vapor and
associated problems and exergy losses.
A large number of theoretical and experimental studies have been dedicated to the solution circulation variant of
the compression - resorption heat pump cycle (reviewed by Groll [1997]). On the opposite only a few studies /
experimental set-ups have been dedicated to wet compression cycles (Malewski [1988], Bergmann [1990],
Torstensson [1991], Sixt [1995] and Itard [1998]). The main problem of the cycle is the compressor that has to be
suitable for oil-free wet compression and still show acceptable isentropic efficiencies. Torstensson and Sixt used
oil-lubricated compressors without oil recovery. Torstensson used a scroll compressor while Sixt used a wankel
compressor. In both cases the lubricant circulated together with the solution through the whole system reducing
the efficiency of resorber and desorber. Malewski used a monoscrew compressor with closed grease bearing
lubrication and separated liquid injection at intermediate pressure. Itard used a liquid ring compressor. Bergmann
used an oil-free twin-screw compressor with timing gears with separated liquid injection at suction and
intermediate pressure conditions. In the present study also an ammonia-water twin-screw compressor with liquid
injection has been used but no timing gears are used. The gaps between rotors and between rotors and housing are
smaller and the male rotor drives the female rotor. The bearings are oil lubricated and separated from the process
side by radial lip seals.
EXPERIMENTAL SET-UP
Making use of Itard’s [1998] experience, her experimental set-up has been rebuilt to allow for high
temperature operating conditions to upgrade industrial waste heat. The design operating conditions have been
selected to allow for application as heat pump for heating of waste water flows from 110 to 130°C in food
processing plants. The design conditions are listed in Table 1.
EXPERIMENTAL RESULTS AND DISCUSSION
Typical operating conditions
The working conditions of the experimental set-up are illustrated for a specific experiment in figure 4 in an
enthalpy – concentration diagram for ammonia water.
From figure 4 it becomes clear that during compression some ammonia vapor is absorbed into the liquid flow. The
liquid concentration increases from 30% at compressor injection conditions to 31% at compressor outlet
conditions.
Figure 4: Specific enthalpy – concentration diagram of ammonia water showing the operating points in
the system. For positions in cycle refer to fig. 2.
For most of the experiments reported here, the difference in concentration between strong and weak solution
ranged from 8 to 12%, depending on the operating conditions. Part of the desorption process takes place in the
intermediate heat exchanger.
Resorber performance
Two sets of experiments have been executed: one with low water flow (0.24 kg/s) and one with high water flow
(0.48 kg/s). With the low water flow the temperature glide of the water flow approached the temperature glide in
the ammonia water side of the resorber. In both cases the temperature driving forces are significant. The
temperature profiles are schematically illustrated for two experiments in figure 5.
Figure 5: Experimental resorber temperature Figure 6: Variation of the relative COP and
glide. heating power of the resorber with the
temperature driving force.
Figure 6 shows that for large water flows the COP and heating power increase significantly with the resorber
temperature driving force. The COP increases 5 to 10% per K temperature increase while the heating power
increases about 10% per K. For low water flows both the COP and heating power decrease when the resorber
temperature driving force increases. Low water flows lead to low water side heat transfer coefficients in the
resorber and consequently to higher discharge pressure for the compressor and associated lower COP for the
system. The large decrease in COP shown in figure 6 is enhanced by an increase in sink temperature for the higher
temperature driving forces.
CONCLUSIONS
The characteristics of an experimental ammonia water compression resorption heat pump have been
discussed. Experimental data of the resorber showed that the maximum system COP is not always attained for an
ideal matching of the temperature glide of the process and sink sides in the resorber. The heat transfer
performance of the resorber and desorber is poor, most probably because the liquid distribution needs
improvement and partly because the system has been operated under part load conditions. The shell-and-plate
intermediate heat exchanger shows extremely low heat transfer performance.