Tube Wire Vs Wall Condenser PDF
Tube Wire Vs Wall Condenser PDF
Tube Wire Vs Wall Condenser PDF
HVAC&R Research
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/uhvc20
To cite this article: P. K. Bansal & T. Chin (2003) Heat Transfer Characteristics of Wire-and-Tube and Hot-Wall Condensers,
HVAC&R Research, 9:3, 277-290
Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the
Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and
should be independently verified with primary sources of information. Taylor and Francis shall not be liable for
any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of
the Content.
This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions
VOL. 9, NO. 3 HVAC&R RESEARCH JULY 2003
This paper presents a comparative study of the performance of wire-and-tube and hot-wall con-
densers commonly used in domestic refrigerators. Both condensers were tested on the same
refrigerator in the natural convection mode. Experiments were carried out in accordance with
AS/NZS 4474-1997 on these condensers to determine condenser capacity and pressure loss.
Downloaded by [University of Arizona] at 14:46 03 February 2015
Simulation models were developed for both the condensers, which led to the evaluation of heat
transfer coefficients, resistances, and temperature profiles along the condensers. The outer heat
transfer resistance from condenser to the ambient air contributes about 80% to 90% of the total
resistance for both condensers. In general, the hot-wall condenser was found to have approxi-
mately 10% to 18% more capacity than the wire-and-tube condenser due to its larger internal
heat transfer area and configuration. This is the first investigation of its kind and may trigger
further research on the topic.
INTRODUCTION
Currently refrigeration research focuses on improving energy efficiency, reducing manufac-
turing cost, and introducing innovative designs of heat exchangers (compact, functional,
user-friendly). A well-designed condenser will not only improve energy efficiency but also
reduce the space and material for a specific cooling capacity. Two of the most commonly used
condensers in domestic refrigerators are the wire-and-tube and hot-wall condensers. These con-
densers are predominantly used in the natural convection mode by the manufacturer in New
Zealand but with distinctly different geometries.
The wire-and-tube condenser (referred to here as W&T) consists of a single copper tube and
solid steel wires that serve as extended surfaces on both sides of the tube. The tube, which car-
ries the refrigerant, is bent into a single-passage serpentine shape with wires symmetrically
spot-welded to both sides in a direction normal to the tubes as shown in Figure 1a. The con-
denser is placed at the back of the refrigerator almost vertically, where the “chimney effect”
results in heat transfer, mainly due to convection. It has been widely used and studied since the
1960s. Tagliafico and Tanda (1997) developed empirical air-side heat transfer correlations for
the natural-convection W&T condenser.
The hot-wall condenser, also known as wrapper type condenser, consists of steel tubing
(coated with copper) that is installed by direct contact on the inner surface of the outer iron plate
of side walls of a refrigerator as shown in Figures 1b and 1c. There are four panels, namely, the
left, right, back, and base panel, each with distinct tube configuration. The refrigerant flows
from the back panel to the base panel, then to the right panel and through the cross rail to the left
panel. Both the condensers have their advantages and disadvantages. A W&T condenser is
effective but prone to accidental damage and dirt accumulation, thereby degrading its heat trans-
fer capability over time. A hotwall condenser does not suffer from any of these problems, but
P. K. Bansal is associate professor and T. Chin was a graduate student in the Department of Mechanical Engineering at
the University of Auckland, New Zealand.
277
278 HVAC&R RESEARCH
Downloaded by [University of Arizona] at 14:46 03 February 2015
about 5% to 10% of the condenser heat infiltrates (Wong 2000) the refrigerating compartment
through insulation. Although, hot-wall condensers tend to have a larger heat transfer area
(spread over three or more surfaces), the increased cabinet heat load may result in the reduced
energy efficiency of the cabinet. Conductive coupling is a key consideration for both types of
heat exchangers, particularly the hot-wall condenser. The W&T configuration has key variables
of wire gauge and shunt pitch; conductive resistance per se is generally not an issue with metal-
lurgical welds. Commercial applications focus on reducing losses through the use of thermal
mastics and compressing circular tubes to flatten contact areas and increase heat transfer area.
The hot-wall condenser is a relatively new design and the lack of published studies on this
condenser is probably a consequence of the nonuniformity between each design and the com-
plex heat path. No studies on direct comparison between the two condensers could be found in
the literature. Therefore, this is the first study of its kind that presents the physics of the conden-
sation process and compares the two predominantly used W&T and hotwall condensers in
domestic refrigerators. This paper investigates the heat transfer characteristics of these condens-
ers on the same refrigerator in the natural convection mode, under identical system operating
conditions, including similar degrees of subcooling and superheat and cabinet internal condi-
tions.
EXPERIMENTAL INVESTIGATION
A domestic refrigerator-freezer (model E406B, 0.68 m wide by 1.56 m high) that was charged
with refrigerant R-134a was acquired from the assembly line of a local manufacturer. This
refrigerator was originally fitted with a hot-wall condenser. Since the other model of the refrig-
erator of similar size uses a W&T condenser, a W&T condenser (0.54 m wide by 1.31 m high)
was also acquired from the manufacturer and was installed externally to the refrigerator. Both
the condensers were compared in the natural convection mode on this refrigerator and under
identical operating conditions.
The capacities of the provision compartment (PC) and the freezer compartment (FC) were
271 and 133 litres, respectively. The fridge used a reciprocating compressor (model Embraco
FGS90HAW), an egg-crate type evaporator (Bansal and Wich 2001; Bansal et al. 2001), and a
defrost heater (350 W). It also had a nonadiabatic capillary tube as expansion device where heat
was transferred from the capillary tube to the suction line.
Both the W&T and hot-wall condensers were connected to the main domestic refrigeration
(based on vapor compression cycle) unit as shown in Figure 2. However, experiments were car-
ried out on one condenser at a time by directing the refrigerant flow to that condenser for each
experiment. The refrigerator was charged with the optimal amount of refrigerant (about 120 g)
by ensuring that the refrigerant temperatures at the inlet and outlet of the evaporator were almost
the same during the charging. Thermocouples (accuracy of ±0.1°C) were placed at the inlet, out-
let, and midpoint of the evaporator and condenser and at the outlet of the compressor. Four pres-
sure transducers (type M6450 with accuracy of ±0.1%) were installed at the inlet and outlet of
the condenser and the compressor to measure the pressure of the refrigerant. The refrigerant
mass flow rate was measured using a D6 micro-motion flow meter (accuracy of ±0.4% of the
flow measurement) at the inlet of the condenser. In order to eliminate the fluctuations in refrig-
erant flow due to compressor vibrations, an expanded section was introduced between the com-
pressor and the micro-motion flow meter. This section was spirally wrapped with a flexible
280 HVAC&R RESEARCH
Downloaded by [University of Arizona] at 14:46 03 February 2015
heater tape (having a maximum power of 99 W) so that a specific temperature of the refrigerant
could be attained before it entered the condenser. With this modification, the fluctuations in the
measurement of the refrigerant mass flow rate were reduced to less than 3% at the inlet of the
condenser; however, the pressure drop effects were found to be negligible at this point.
Closed-door experiments were carried out on the refrigerator in accordance with NZ/AUS
standard 4474-1997, where the fresh food and the freezer compartment temperatures were main-
tained at 3°C and –15°C, respectively, with the help of a thermostat. The standard specifies the
ambient temperature to be 32°C and the appliance to be placed at a minimum of 25 mm distance
from the back wall; however, the refrigerator was tested at ambient temperatures of 20°C and
25°C (to represent the real-world kitchen temperatures) and was placed at 0.5 m distance from
the back wall in order to have a uniform and consistent comparison. The test refrigerator was
placed in an environmental chamber, which was maintained at the desired temperatures and at
55% relative humidity (±5%) for all tests. The external heater (on the expanded section—see
Figure 2) was used to maintain different refrigerant saturation temperatures at five levels (40°C,
37°C, 35°C, 32°C, and 30°C) in the experiments; however, the levels of superheat and the
refrigerant mass flow rates in the two condensers varied slightly due to different evaporation
temperatures. All experiments were carried out with a 20 W resistance heater installed in the
freezer compartment of the refrigerator to provide adequate internal heat load for the continuous
operation of the compressor in order to test the condensers. The defrosting process was manu-
ally activated a few times prior to the experiments to avoid any frost formation on the evaporator
during the experiments. The refrigerator was allowed to reach steady state before starting the
experiments. Each experiment was run for five hours and the measurements of temperature,
pressure, power consumption, and mass flow rate were recorded. A number of measurements for
VOLUME 9, NUMBER 3, JULY 2003 281
Downloaded by [University of Arizona] at 14:46 03 February 2015
both condensers had to be abandoned with fixed level of superheat and subcooling because the
condenser capacity could not be evaluated due to refrigerant leaving the condenser in two-phase
condition. This was largely due to the limitation of the test rig, particularly when the condensers
were operating under natural convection mode while both the kitchen temperature and the cabi-
net internal conditions were fixed. In some circumstances, this led some of the operating condi-
tions at the condenser inlet to differ from the “real world” conditions (particularly for the
hot-wall condenser) that are normally observed on a refrigerator.
EXPERIMENTAL RESULTS
Experimental condenser capacities were evaluated from temperature and pressure measure-
ments at the inlet and outlet of the condenser. The experimental capacities (for fully condensed
refrigerant) at different saturation and ambient temperatures and superheat of 5 K are plotted in
Figure 3. The temperatures in the evaporator of the refrigerator for these experiments ranged
from –31.5°C to –23°C.
As expected, the condenser capacity increases with the refrigerant mass flow rate (Figure 3).
Since the experiments were carried out on a real refrigerator, the mass flow rate varied accord-
ing to the heat load on the cabinet and the heat input by the discharge heater in the expanded sec-
tion. For example, at saturation temperature of 32°C, the mass flow rate for the fully condensed
condition (at condenser outlet) is always less than 0.9 g/s, while at temperature above 35°C, the
refrigerant flow rate is larger than 0.95 g/s. In the experiments at higher condensing tempera-
tures, the evaporator in the refrigerator was generally operating at higher pressures, yielding a
larger refrigerant mass flow rate. It was further observed in the experiments that higher con-
denser saturation temperature yielded higher level of subcoolings. The level of subcooling var-
ied from 0.4 K (at Tsat = 32°C) to 1.5 K (at Tsat = 40°C) for the W&T condenser and 1.5 K (at
Tsat = 32°C) to 2.5 K (at Tsat = 40°C) for the hot-wall condenser. Likewise, the level of super-
heat varied from 2 to 7 K (for W&T) and 2 to 5 K (for hot-wall condenser). The hot-wall con-
denser capacities were generally found to be approximately 10% to 18% larger than those of the
W&T condenser due to heat leakage from the hot-wall condenser (through insulation) into the
282 HVAC&R RESEARCH
Downloaded by [University of Arizona] at 14:46 03 February 2015
compartment that was maintained at a set temperature. This can also be attributed to the distinct
difference in configuration between the W&T and the hot-wall condenser (Table 1) and the heat
infiltration into the refrigerating compartment. Wong (2000) reported that the heat infiltration
into the refrigeration compartment of this refrigerator by the hot-wall condenser increases the
cabinet heat load by about 5% to 10%. Figure 4 shows that the pressure loss in the W&T con-
denser was about 15% to 35% higher than in the hot-wall condenser due to smaller tube diame-
ter. The mass flux was 23% larger in the W&T condenser. As expected, the pressure loss
increases with the mass flow rate. The temperature effect on the pressure loss is insignificant.
h0 = hc + hr . (1)
VOLUME 9, NUMBER 3, JULY 2003 283
In general, hc depends on both the geometry of the heat transfer area and the surface tempera-
ture, while hr is given by the conventional radiation heat exchange equation. The overall heat
transfer coefficient based on the outer surface area Uo is computed from
U o = UA ⁄ A o . (2)
It is a measure of the average heat transfer coefficient, governed by various components of the
heat transfer resistances, as given by the UA relationship:
1 –1
UA = --------- = [ R i + R t + R o + R con ] (3)
R tot
Downloaded by [University of Arizona] at 14:46 03 February 2015
where Ri, Rt, Rcon, and Ro are, respectively, the inner, tube, contact, and outer heat transfer resis-
tances. These resistances differ for the W&T and hot-wall condenser and are given in the appen-
dix. The information on various heat transfer coefficients and other parameters is given in
Bansal and Chin (2002, 2003).
Figure 7. Overall, Radiative, and Convective Heat Transfer Coefficients Along the
Wire-and-Tube Condenser
42° is at the mass flux range of 85-90 kg/s·m2 for acceptable pressure loss and sufficient capac-
ity and subcooling at the condenser outlet.
the condensation proceeds, the proportion of inner resistance reduces slightly; thus, the tempera-
ture difference increases gradually throughout the condensation process to a maximum of 1.5°C
at the saturated liquid state. In the subcooled region (D), the temperature difference increases to
2°C due to increased inner surface heat transfer resistance.
ures 9 and 10 to mark various sections of the hot-wall condenser are abbreviated in the nomen-
clature section. The convective and radiative heat transfer coefficients are governed by the
average plate temperature, which was determined by the inner heat transfer coefficient and area
of heat transfer (Figure 9). At the base panel (BSP), where the outer surface faces the ground,
the air temperature was 5°C higher than the ambient temperature. Thus, the radiative and con-
vective heat transfers deteriorated significantly. The coefficients for the back, right, and left
panel are different due to the difference in the heat transfer area per unit length of the tube. It is
defined as the “average width,” wav, and may be shown for different panels as
The average plate temperature, Tpl,av is lower for larger wav (smaller tube density), leading to
smaller convective and radiative heat transfer coefficients. The comparison of the heat transfer
coefficients on each panel indicates the relative effectiveness of each panel in terms of heat
transfer per unit area [ q· = ho(Tpl,av – Tamb)]. Thus, the back panel is most effective for heat
transfer. The radiative coefficients display a pattern similar to the convective heat transfer coef-
ficients but contribute up to 70% of the total heat transfer. This is almost the reverse of the W&T
condenser. The Nusselt number, which is a function of ambient temperature and heat transfer
geometries, is higher for the W&T condenser due to larger area compared to any single panel of
the hot-wall condenser. In other words, the W&T condenser panel is more effective for convec-
tive heat transfer than one panel of the hot-wall condenser. However, the performance of the
hot-wall condenser may deteriorate when placed closer to the wall (in a real-world situation)
where radiant heat exchange may be inhibited. The W&T condenser is prone to dirt accumula-
tion and may also suffer degradation in convective heat transfer over time. Therefore, it is very
difficult to make a comparison of the two condensers under these circumstances; the current
study provides some insight into the real issues and direct comparison of the two condensers
under identical circumstances.
288 HVAC&R RESEARCH
For the hot-wall condenser, the tube and contact resistances are negligible, as they contribute
to less than 1% of the total resistance. Ro is the resistance against heat transfer from the tube
outer surface through the plate to the surroundings (Bansal and Chin 2002). It is a function of
plate geometry, plate area, conductivity, and outer heat transfer coefficient. Ro contributes to
about 81% and 83% to 91% of the total resistance in single and two-phase flow, respectively
(Figure 10). The magnitudes of Ro for different panels are in the following order:
The dominating factor for the outer resistance of right (RH), left (LH), and back (BP) panel is
the outer heat transfer area per unit length, wav, as given by Equation 4. For the cross rail (CR)
and base (BSP) panel, the outer heat transfer coefficient is significantly lower and, hence, results
in higher heat transfer resistance. The magnitudes of condensation heat transfer coefficients for
Downloaded by [University of Arizona] at 14:46 03 February 2015
The refrigerant, tube (outer surface), and average plate temperature vary across the hot-wall
condenser (see Figure 10). The temperature profile is distinctly different between the single and
two-phase regions, between various sections of the condenser, and between horizontal, vertical
upward, and vertical downward flows. For the condensation region, the temperature difference
between the refrigerant (Tref) and tube outer surface temperature (Tt) is generally less than 3°C,
while the maximum temperature difference between Tpl,av and Tref is approximately 5°C. The
tube and average plate temperature, Tt and Tpl,av, display similar patterns. Within a specific sec-
tion of the condenser, Tt and Tpl,av are highest for the horizontal flow, followed by vertical
downward flow and vertical upward flow. This difference in the temperature for different flows
is due to relative magnitude of the heat transfer coefficients, which is largest for the horizontal
flow.
The variation in the saturation temperature across the condensation region is around 1.2°C.
The average plate temperature is the lowest for the right panel because the area per unit length of
tube is the largest for this panel. Thus, to effectively utilize the outer heat transfer area, more
tube length should be distributed to the right panel while reducing the tube density on the back
panel. This would result in an improved condenser capacity for the same tube length.
CONCLUSIONS
The experimental results show that the hot-wall condenser has approximately 10% to 18%
larger capacity than the W&T condenser for the same operating conditions. This is due to its sig-
nificantly different tubing configuration and relatively larger inner heat transfer area. The pres-
sure losses are larger for the W&T condenser due to its smaller tube diameter compared to the
hot-wall condenser. In term of condenser weight, the hot-wall condenser is lighter than the
W&T condenser. The modeling results show that the outer heat transfer resistance contributes
up to 90% of the total heat resistance for both the condensers. In other words, the effect of inter-
nal (or condensation) heat transfer resistance is minimal. Therefore, in order to improve the con-
denser capacity, one should aim to either increase the outer heat transfer area or improve the
effectiveness of the fins. The dominant heat transfer mode for the W&T condenser is convec-
tion, which contributes up to 65% of the total heat transfer. This is contrary to the hot-wall con-
VOLUME 9, NUMBER 3, JULY 2003 289
denser where the heat transfer by radiation dominates, contributing up to 70% of the total heat
transfer. However, in the real-world scenario where the refrigerator is placed next to a wall, a
new W&T condenser may outperform the hot-wall condenser due to reduced radiant heat
exchange in the case of the hot-wall condenser, but at the same time, W&T may loose this
advantage over time when its convective heat transfer may degrade due to dirt accumulation.
This is the first study of its kind and may trigger additional research to further uncover the phys-
ics of many related issues and system optimization.
ACKNOWLEDGMENTS
The authors are thankful to Fisher & Paykel NZ Ltd for sponsoring the project and providing
necessary support. In particular, thanks are due to Messrs Ian McGill and David Thomas.
NOMENCLATURE
Downloaded by [University of Arizona] at 14:46 03 February 2015
Subscripts
amb = free stream/ambient Ho = horizontal flow
av = average LH = left-hand side panel
BP = back panel O = outside, outlet
BSP = base panel Pl = plate
C = convective R = radiative
Con = contact
Ref = refrigerant
Cond = condensation
RH = right-hand side panel
CR = cross-rail
Ele = elemental sat = saturation
Eq = equivalent t = tube
F = fin Vd = vertical downward flow
I = inside, inlet Vu = vertical upward flow
Ix = cross-sectional w = wire
REFERENCES
Bansal, P.K., and T. Chin. 2003. Modelling and optimization of wire-and-tube condensers. International J
Refrigeration 26(5): 601-613.
Bansal, P.K., and T. Chin. 2002. Design and modelling of hot-wall condensers in domestic refrigerators. J
Applied Thermal Engineering 22 (14): 1601-1617.
Bansal, P.K., and T. Wich. 2001. Optimisation of egg-crate type evaporators in domestic refrigerators. J
Applied Thermal Engineering 21: 751-770.
Bansal, P.K., T. Wich, J. Chen, and M. Browne. 2001. Design and modelling of new egg-crate-type forced
flow evaporators in domestic refrigerators. ASHRAE Transactions 107(2).
Chin, T.C. 2001. Experimental and numerical study of the wire-and-tube and hot-wall condensers for
domestic refrigerators. ME thesis, Department of Mechanical Engineering, The University of Auck-
land, New Zealand.
Standards Association of Australia/Standards Association of New Zealand. 1997. Australia/New Zealand
Standard AS/NZS 4474 – 1997, Performance of household electrical appliances—Refrigerating appli-
ances, Part 1: Energy consumption and performance. Sydney/Wellington.
290 HVAC&R RESEARCH
Tagliafico, L., and G. Tanda. 1997. Radiation and natural convection heat transfer from wire-and-tube heat
exchangers in refrigeration appliances. International Journal of Refrigeration 20(7): 461-469.
Wong, A. 2000. Study of energy consumption for household refrigerator under various operating condi-
tions. Final Year Research Project (ME 00.66), Department of Mechanical Engineering, The Univer-
sity of Auckland, New Zealand.
APPENDIX
Heat Transfer Resistance Components for the W&T and Hot-Wall Condensers
1 ln ( r o ⁄ r i ) t con
R i = ---------- , R i = ------------------------ , Rcon = 0 for W&T condenser and R con = --------------------------------- for
hi Ai 2π ⋅ k ⋅ ∆z k pl ⋅ ∆z ⋅ d t ⁄ 2
2 × h ⋅ ∆z ⋅ k A ⋅ tan h ( mL )
Downloaded by [University of Arizona] at 14:46 03 February 2015
1 o pl ix
hot-wall condenser, R o = ------------ for W&T condenser, and R o = ------------------------------------------------------------------------------
2
-
ho Ao h o ⋅ ∆z ⋅ k pl A ix [ tan h ( mL ) ]
h o ⋅ ∆z
where m = ---------------
- for hot-wall condenser.
k f A ix