DOI: 10.18462/iir.gl2022.0037

Stefano FILIPPINI*(a), Umberto MERLO(a), Dario DEMURTAS(a), Ennio MACCHI(b)

(a) LU-VE GROUP, Uboldo, 21040, Italy, [email protected]
(b⁾ Politecnico di Milano, Milan, 20133, Italy, [email protected]
*Corresponding author: [email protected]

ABSTRACT

Efficient district heating and cooling (DHC) systems can play a key role in achieving the objectives of the European Union. District heating systems contribute substantially to the reduction of primary energy supplies, carbon emissions and costs.

Heat pumps are key system components when it comes to providing sustainable heat and integration of energy systems in a climate-friendly future. Heat pumps allow the use of various accessible heat sources such as ambient air, seawater, groundwater or waste heat from the industry.

Currently, the use of heat pumps for district heating is in a critical phase. Efficiency and reliability are key to the diffusion of these systems on a larger scale.

LU-VE participated from the beginning in the development of specific large air evaporators for district heating application, using NH3 as working fluid.

The paper describes in the first part the main issue characterizing the design of a large field of evaporators for heat pump application in district heating plants.

The second part of the paper illustrates the R&D activity that enables the latest product evolution, capable of lowering the energy consumption of the evaporators by 21%.

A new ventilation concept has been developed and dedicated experimental tests have been performed in order to validate the design. Air distribution has been improved as well as the aerodynamic configuration of the ventilation housing.

Keywords: ammonia, evaporators, district heating, heat pumps, low energy

INTRODUCTION

District heating is recognized as a brilliant technical solution for improving the air quality in urban areas, being capable of removing pollution produced by the multitude of replaced domestic and condominium boilers. However, at present only a small fraction of district heating networks is run on renewable energy sources (municipal solid waste and biomass combustion, geothermal wells, heat recovery from industrial processes). In most district heating networks the heat is still supplied by burning fossil fuels, either in combined heat and power (cogeneration) plants or in large boilers. These solutions do not comply with the aim of a fully decarbonized world, i.e. the scenario present in all modern energy policies. The adoption of electrically driven heat pumps as heat source for district heating can solve the problem, accomplishing the benefits of abating local air pollution as well as carbon dioxide emissions. Currently, the last point is only true for a zero-emission electric sector, a situation not yet reached in most countries, but predicted for Europe in a couple of decades. This paper describes the experience gained by LU-VE in the design of the heat exchangers for this application.

PLANT DESIGN: HEAT SOURCES AND REFRIGERANT

• Heat source

A variety of heat sources are possible for heat pumps, including ground and water, but ambient air is certainly a valuable option, due to its great advantage of being always available everywhere. Air heat exchangers are an important part of a district heating heat pump station: they require a large heat transfer area, due to the poor transport properties of ambient air (low density, low thermal conductivity) and the low temperature differences required to achieve good coefficients of performance. The heat exchange is usually organized in many modules and their footprint largely dominates the plant area (see Figure 1).

Figure 1: A view of an air heat pump district heating station: the air heat exchangers occupy most of the plant area

In recent years, LU-VE has been involved in a significant number of district heating projects, most of them located in Denmark, based on air heat pumps.

Since 2017, when the first projects started, the increase of heat exchanger units installed in the last few years for these applications has been very rapid, reaching very high numbers, as shown in Figure 2.

Figure 2: Increase of heat exchanger modules installed by LU-VE in heat pump district heating stations

• Primary refrigerant definition

As shown in Figure 3, there are two main plant scheme options:

1. The selection of ammonia as working fluid, with direct evaporation by ambient air, either fed by gravity or by a pump or
2. The adoption of a heat transfer fluid (usually a mixture of water and glycol) which exchanges heat with ambient air and transfers it to the evaporator of a refrigerant, either R134a or other suitable working fluid.

Figure 3: Pie chart of the different solutions adopted in DH plants equipped with LU-VE units

The pie chart in Figure 3 indicates a clear market preference for ammonia solutions.

• Thermodynamic simulation of three possible plant layout

To gain a better understanding of the pros and cons of these two solutions, the three cases represented in figure 4 and figure 5 below are examined:

1. The heat transfer occurs between ambient air and ammonia. The ammonia vapor is then compressed and condensed by releasing the heat to the district heating network.

Figure 4: Plant layout and TDN cycle T-s diagram for Case A

1. A first heat transfer occurs in a heat collector between ambient air and a water/glycol mixture, then a second heat transfer takes place in a R134a evaporator and finally heat is released to the district heat network. The R134a evaporating temperature is assumed equal to that of ammonia in case A. This requires a large heat transfer area for the air heat exchange, which takes place under small temperature differences.
2. Same arrangement of case B, but the surface area of the air heat exchanger is assumed equal to that of case A. Consequently, the evaporation temperature decreases, with penalties on cycle COP.

Figure 5: Plant layout and TDN cycle T-s diagram for Case B-C (same layout, just different evaporation temperature)

Figure 6: Temperature-heat diagrams of three options:

1. Direct ammonia cycle
2. Indirect R134a cycle with same evaporation temperature as case A
3. Indirect R134a cycle with same heat transfer surface as case A

The main features of the three solutions are given in Table 1, computed under the following common assumptions:

• Ambient air conditions: T=-1°C; RH=90%
• Thermal power to the district heating: 7350 kW
• District heating in/out temperatures: 30/62 °C
• Condenser Pinch point: 2°C
• Subcooling at condenser outlet: 5°C
• Compressor Isentropic efficiency: 80%
• Compressor electric efficiency: 95%

Other assumptions:

• For case A, a pressure drop of 10 kPa between evaporator exit and compressor inlet
• For cases B and C, saturated vapor at evaporator outlet

For the air heat exchangers, the same module characteristics were selected for the three solutions:

• Coil geometry: 50×50 mm
• Tube materials: stainless steel for case A, copper for cases B and C
• Fin material: Aluminum
• Tube diameter: 16 mm
• Fin spacing: 5.0 mm
• Row number: 6
• Electronic axial fan

The number of circuits was optimized to obtain the best compromise between pressure losses and heat transfer performance.

The air module performance was computed by means a proprietary code by LU-VE; the module number for case A was selected to achieve a temperature difference between inlet air and evaporation of 6.5 °C, for case B to obtain the same evaporation temperature as case A.

Table 1: Comparison of the three solutions of Fig. 6

The comparison between cases A and B highlights the merits of selecting a scheme with direct air evaporation and ammonia as working fluid:

• To obtain the same evaporation temperature, the indirect scheme requires a much larger number of modules (56 vs 24), thus a more than double plant footprint and fan electricity consumption
• NH3 thermodynamic properties allow a better cycle COP than R134a to be achieved, due to the lower irreversibilities occurring both in the heat exchange during condensation (which occurs at lower temperature, 56.2°C vs 58.7 °C) and in the expansion phase
• The lower auxiliary consumption (156 vs 359 kW fan consumption and the absence of secondary fluid pump) results in a much higher overall plant COP (3.27 vs 2.65)

The comparison between cases B and C suggests that is better not to be excessive in the search of low temperature differences between the ambient air and the evaporation temperature: the decrease of cycle COP produced by the lower evaporating temperature of case C (-10°C vs -7.5°C) is more than counterbalanced by the lower fan consumption. The result is that case C not only has a smaller plant footprint and cost but obtains a better plant COP than case B due to lower auxiliary consumption, as shown in Figure 7.

Figure 7: Electric consumption share

Another significant feature of the direct ammonia solution is its ability to maintain good performance at various ambient temperatures, as shown in Figure 8. If we compare the calculated COP with that of an ideal heat pump, capable of evaporating at the ambient temperature and releasing heat at the stipulated district heat temperatures, we find that the ratio between COP of such an ideal heat pump and the real COP, accounting for all irreversibilities of real cycle (heat exchange temperature differences, compressor losses, pressure drops, isenthalpic expansion, auxiliaries consumption) is almost constant (about 2) over a vast ambient temperature range.

Figure 8: Plant COP variation with ambient temperature, computed for the same useful thermal power

Note that the COP in the design point (ambient temperature of -1°C) is slightly out of line: this was expected, since the heat collectors circuiting was specially designed for the baseline condition and any other point of work suffers due to a little drop of efficiency in the heat exchange process.

In any case, besides the theoretical analysis, it is important to focus also on some practical problems that could affect the choice of ammonia as primary refrigerant. First of all, it is well known that the use of ammonia generates some safety issues, due to the high toxicity and flammability of this fluid: this is a significant matter in the design of a DH plant, that could face an ammonia charge of thousands of kg. Moreover, considering that the installation of these plants could be placed not very far from residential areas, we cannot exclude local citizens risings up against such a solution, making it difficult to obtain acceptance of it even if justified by a higher thermodynamic efficiency.

Another point to be considered is the plant regulation at partial load and different ambient temperature, which requires the modification of the evaporation temperature. The use of a secondary water loop certainly guarantees easier control of the system, just through the regulation of the flow inside the heat collectors. This is more critical with direct expansion application, since the thermal process is more subjected to efficiency drop caused by a flow regime that is very different from the design point.

The theoretical and thermodynamic analysis must therefore be accompanied with some more practical considerations, taking into account the demands and the uniqueness of each specific project. A higher COP does not mean that the best solution is always the same, but many aspects must be considered to always identify the right choice, always bearing in mind the advantages of ammonia in terms of global warming impact (null GWP) and the push from international legislation to the use of natural refrigerants.

THE LU-VE HEAT COLLECTOR CONCEPT

• Modular design

The LU-VE heat collector concept is based on a modular design (Figure 9), tailor-made for heat pump application.

The height of supports is selected to minimize the footprint (minimum distance between modules) and to avoid air recirculation.

Figure 9: Modular concept of LU-VE heat collector

The air flow is usually aspirated in “conventional” ammonia evaporators for the refrigeration industry (i.e. first crosses the coil, then is equalized in a vacuum chamber and finally goes through the fan blades), while in the solution adopted for heat pump application the air path is in the opposite direction: the air flows downwards, from the fan blades to the coil.

• Frost formation and defrosting

This solution allows improved behavior of the evaporator during the defrosting phase (see Figure 10), with the air flow in the same direction as the force of gravity: during the defrosting stage, the fan speed is kept to minimum, while after the defrosting has been completed, the fan speed is set to maximum for about 1 minute. The experience has shown that the complete defrosting procedure lasts for about 10-15 minutes. The large number of modules enables the heat exchange of the system while defrosting one module at a time.

Figure 10: The situation at the bottom of the air heat exchanger before and after defrosting

Further advantages of the downwards solution are the air heating before entering the coil due to fan power and the parallel flow of condensate and air.

In contrast to this, it is a common experience that the solution with the fan forcing the air onto the coil surface can produce a not uniform flow.

• Experimental activity

For this reason, an experimental activity was carried out, in cooperation with the fan manufacturer. The aim of the research was the optimization of the fan nozzle geometry to achieve an even velocity distribution across the coil. The tests helped the definition of a new fan collar made of a composite material (see Figure 11), able to improve the air distribution and lower the power consumption. The results were quite successful.

Figure 11: Original design (left) and improved design (right)

Figure 12 compares the original design (blue curve) with the new one (red curve). The improved aerodynamics coming from the new fan collar ensures significant power consumption reduction.

Figure 12: Measured absorbed power for both original and new aeraulic design

At the same time, we also performed an anemometric map of the air distribution (Figure 13). The arrangement with coil placed on fan pressure size does not ensure an even air distribution, but we can consider it acceptable.

The same test was performed also with coil on fan suction side (right picture), obtaining a better result, as expected.

Figure 13: Anemometric map with coil arrangement on pressure side (left) and suction side (right)

In any case, since the general product operation is characterized by high air dehumidification that may lead to huge frost formation, depending on the ambient temperature, a configuration with the coil placed on pressure side is considered to be the best choice.

CONCLUSIONS

Heat pumps are a key technology in the European green deal policy. This paper explained the contribution of air heat exchangers in the optimization of large heat pumps in district heating.

Different possibilities of plant arrangement with different fluids were evaluated, underlining pros and cons of each solution.

The experimental campaign on different aeraulic configuration helped the development of a new fan collar, capable of decreasing the power consumption of the heat exchanger up to 21%.

Currently LU-VE is exploiting, together with an important partner, the possibility of using CO2 as working fluid in this application, a natural refrigerant that eliminates the safety issues affecting the use of NH3.

REFERENCES

ASHRAE Handbook-Refrigeration, 2018. Ammonia Refrigeration systems, 1-28.

Colburn, A.P., 1933. A method of correlating forced convection heat transfer data and a comparison with fluid friction. Trans. Am. Inst. Chem. Eng. 29, 174-210.

Colombo, E., Macchi, E., Merlo, U., 2005. Strategy for innovation in heat exchanger design: computational approach combined with experimental tests leads to performance improvement. Summer Heat Transfer Conference – Westin St. Francis, San Francisco, CA, USA.

Gnielinski, V., 1976. New equations for heat and mass-transfer in turbulent pipe and channel flow, (1976),  359-368 (Karlsruher Institut für Technologie).

Kristófersson, J., Vestergaard, N., Skovrup, M., Reinholdt, L., 2017. Ammonia charge reduction potential in recirculating systems – calculations. doi: 10.18462/iir.nh3-co2.2017.0005.

Lemmon, E.W., Bell, I.H., Huber, M.L., McLinden, M.O., 2018.  NIST Standard Reference Database 23:  Reference Fluid Thermodynamic and Transport Properties-REFPROP, Version 10.0, National Institute of Standards and Technology, Standard Reference Data Program, Gaithersburg.

Lockhart, R.W., Martinelli, R.C., 1949. Proposed correlation of data for isothermal two-phase, two-component flow in pipes, Chem. Eng. Prog 45(1), 39-48.

Lozza, G., Merlo, U., 2001. An experimental investigation of heat transfer and friction losses of interrupted and wavy fins for fin-and-tube heat exchangers. International Journal of Refrigeration 24 (2001), 409-416.

Patankar, S.V., 1980, Numerical Heat Transfer and Fluid Flow, Mc Graw-Hill Book Company, New York.

Siegel, J., Carey, V.P., 2001. Fouling of HVAC fin and tube heat exchangers. Environmental Energy Technology Division Indoor Environment Department Lawrence Berkeley National Laboratory Berkeley, CA 94720.

Wojtan, L., Ursenbacher, T., Thome, J.R., 2005. Investigation of flow boiling in horizontal tubes: Part II—Development of a new heat transfer model for stratified-wavy, dryout and mist flow regimes. International Journal of Heat and Mass Transfer 48, 2955–2985.

Wojtan, L., Ursenbacher, T., Thome, J.R, 2005. Investigation of flow boiling in horizontal tubes: Part I—A new diabatic two-phase flow pattern map. International Journal of Heat and Mass Transfer 48 (2005), 2955–2969.