Heat Transfer Coefficients

Downloaded from SAE International by Gursaran Mathur, Tuesday, April 03, 2018

2018-01-0055 Published 03 Apr 2018

Correlation for Predicting Two-Phase Flow Boiling Heat Transfer Coefficients for Refrigerant HFO-1234yf Gursaran D. Mathur CalsonicKansei North America Inc. Citation: Mathur, G.D., “Correlation for Predicting Two-Phase Flow Boiling Heat Transfer Coefficients for Refrigerant HFO-1234yf,” SAE Technical Paper 2018-01-0055, 2018, doi:10.4271/2018-01-0055.

Abstract

A

uthor has developed a correlation to predict flow boiling heat transfer coefficients for refrigerant evaporating in an automotive evaporator. This is a first correlation in the open literature for HFO-1234yf to predict heat transfer coefficients for automotive evaporator. The refrigerant mass flux was varied from 500 to 1200 kg/m2.s; heat flux was varied from 2 to 6.2 kW/m2; inlet refrigerant qualities from 0 to 40% and exit qualities of about 95%. The tests were conduct at 4.4 °C and the oil circulation ratio was maintained at 3%. Experimental data has been used with MINITAB software, Version 16.1.0 to develop this correlation. Multivariate nonlinear regression analysis has been done to develop this correlation. Experimental data along with refrigerant properties, hydraulic diameter that affects Reynolds

Introduction

T

o address the global warming concerns of R-134a, some of the OEMs around the globe have started using refrigerant HFO-1234yf as an alternate replacement for HFC-134a. In last few years, the author has conducted a number of experimental studies [1, 2, 3, 4] with refrigerant HFO-1234yf. In these experimental studies, the author has experimentally determined overall system performance with and without suction line heat exchanger; TXV optimization; both evaporator and condenser thermal and hydrodynamic performances; and heat transfer coefficients for both twophase flow boiling and condensation of HFO-1234yf for typical automotive heat exchangers and AC loop configurations. In some of the studies by the author, the experimentally measured heat transfer coefficients for flow boiling and condensation were compared to the predictions from standard correlations available in the literature. The experimentally measured magnitudes of the heat transfer coefficients were within ±12~20% of the predicted numbers. However, these correlations were not specifically developed for HFO-1234yf refrigerant. Hence, it is important that we develop a correlation specifically for HFO-1234yf that could be used by engineers and designers for predicting heat transfer coefficients for flow boiling region for typical automotive applications. The only correlation available in the © 2018 SAE International. All Rights Reserved.

number, Prandtl number and other appropriate variables have been used to develop this correlation. Details of the newly developed correlation have been presented in the paper. The developed correlation will be used to predict flow boiling heat transfer coefficients for HFO-1234yf for automotive evaporator (laminate evaporators). The following is the developed correlation for HFO-1234yf: hexp/hl = 2.8738 (1/Xtt)0.109. The developed correlation predicts the experimentally obtained data within ±23%. Further studies are planned to improve this correlation and to compare predictions with other correlations in the open literature; and to study the influence of amount of lubricant (%) on flow boiling heat transfer coefficients.

open literature is by Saitoh et al. [5] that was developed for a refrigerant flow within a single circular tube and not for an automotive application. In this paper, the author has presented a correlation that has been specifically developed for prediction of two-phase flow boiling heat transfer coefficients for HFO-1234yf for typical automotive exchangers and refrigerant mass flow fluxes. Experimental data has been used with MINITAB software, Version 16.1.0 to develop this correlation. Multivariate nonlinear regression analysis has been done to develop this correlation. Experimental data along with refrigerant properties, hydraulic diameter that affects Reynolds number, Prandtl number and other appropriate variables have been used to develop this correlation. Details of the newly developed correlation have been presented in the paper. Currently at this time, we do not have a correlation in the open literature that has been specifically developed for HFO-1234yf for automotive application.

Literature Review There are a number of correlations that have been developed in last 20+ years to predict f low boiling heat transfer

Downloaded from SAE International by Gursaran Mathur, Tuesday, April 03, 2018 2

CORRELATION FOR PREDICTING TWO-PHASE FLOW BOILING HEAT TRANSFER COEFFICIENTS

coefficients. However, these correlations have been developed for specific class of refrigerants (ASHRAE, 2013): a. Methane series: e.g., R11, R12, R22, R32, etc. A number of correlations exists for methane series refrigerant. b. Ethane series: e.g. R-134A, R-152A, etc. A number of correlations exists for ethane series refrigerant. c. Unsaturated Organic Compounds: Refrigerant HFO1234yf falls under this category. There is no correlation available in the open literature for automotive evaporators. These are laminate type design and details are provided in next section. There have been a handful of pure academics type studies but for only for single tubes (e.g., [5]). These might have been developed for chiller type applications. Kandlikar’s [6] correlation is perhaps the most widely used correlation to predict heat transfer coefficients for refrigerants during flow boiling. This correlation was developed with 5426 sets of data points for 24 refrigerants. This correlation was developed for both nucleate boiling dominant mode and forced boiling modes of heat transfers. The readers are referred to Kandlikar’s paper for the equations for predicting heat transfer coefficients. Another widely used correlation to predict two-phase flow boiling heat transfer coefficient is the one proposed by Shah [7, 8]. A summary of important correlations was compiled by Shah [9].

Experimental Set-Up and Tests Tests were conducted in the system bench test facility. The test facility is instrumented in accordance with the ASHRAE standard [10, 11, 12, 13]. The facility is capable of measuring the performance of the system using air-enthalpy and refrigerant-enthalpy methods. The test facility is fully instrumented to record the operating parameters, e.g., pressure and

temperatures of the refrigerant to plot the thermodynamic process on the pressure enthalpy diagram. Schematic of the test AC loop facility is given in Figure 1. Figures 2, 3 and Table 1 provides the detail of the evaporator that is used for testing for this study. Dry and wet bulb temperatures are also determined to compute airside enthalpies on the evaporator and condenser. This test facility consists of two rooms (See Figure 4) - one simulating the vehicle cabin (Figure 5) and the other room simulating the engine compartment as shown in Figure 6. The HVAC unit (i.e., evaporator) is placed in the room simulating the vehicle cabin while the condenser and compressor are placed in the room simulating the vehicles engine compartment. All parts are assembled in these two rooms based on the vehicle configuration. Evaporator freeze is controlled by cycling the compressor via a thermistor located in the evaporator. However these tests were conducted to ensure we do not have freezing to ensure a steady state conditions to collect test data. The airflow to the HVAC unit is supplied through an air machine that is designed based on ASHRAE standard [10, 11, 12, 13]. Cabin interior is maintained at 25 °C and 50% RH; and engine compartment room temperature is varied from 25 to 45 °C. The compressor rpm is varied from 800 to 3000 rpm; the condenser face velocity is varied from 2 m/s to 10 m/s; and the evaporator airflow is varied from 5 to 9 m3/ min. Evaporator downstream air temperatures were measured at 20 places (4 rows X 5 columns) with a 30 gauge thermocouple wire gauge, thinnest thermocouple wire available in the market. Four thermocouples are located on the upstream of the evaporator to determine enthalpy at the inlet from the airside. Outside surface temperatures were also measured at many locations. Surface temperature on the outside of the evaporator wall (and insulate the thermocouple) is generally used for measuring the inside wall temperature. This is a fair assumption as the thermal conductivity of aluminum is very high (198 W/m°K) and the wall thickness is less than 1 mm. Hence, the wall resistance is very small. For these experiments, refrigerant mass flow rates have been calculated based on the compressor characteristics (volumetric efficiency as a function of rpm) and operating condition. Table 2 shows the

© SAE International

FIGURE 1 Schematic & p-h diagram for an air conditioning system

© 2018 SAE International. All Rights Reserved.



FIGURE 2 Photograph of the evaporator used for the


current investigation


FIGURE 3 Refrigerant flow pass configuration of a typical evaporator with 6 internal passes


TABLE 1 Details of Laminate Heat Exchanger

Variable

Magnitude

Height (mm)

270

Width (mm)

300

Depth (mm)

48

Fin Design

Louvered

Tube Design

Stamped Dimpled

details of the test parameters. Lubricant compatible with HFO-1234yf was used for this study. A total of 140 cc lubricant was used in the system that resulted in an oil circulation ratio of about 3% [14]. A typical laminate type evaporator [15] is shown in Figure 2. It consists of a number of stamped plates and © 2018 SAE International. All Rights Reserved.

3

louvered fins. The plates have clad material on both sides. The plates and fins are stacked and then either vacuum brazed or brazed in controlled atmospheric brazing (CAB) process. The advantage of using CAB process is that it is a continuous process in comparison to vacuum brazing that is a batch process. The plate design is such that when brazed, internal flow passages for refrigerant are created. The plates have diagonal ribs (or multiple dimples) to augment the heat transfer and provide strength. These plates have central partitioning ribs that facilitate the reversal of refrigerant flow direction. In some of the new designs, an insert is used between plates to augment heat transfer. This also helps in improving the strength of the evaporator as the inserts braze to both sides (inside) of the plates. Also, some of the suppliers have started using micro-channel tubes [16] for evaporators. These evaporators may have tank on either ends or on one end only. For a same airflow area, a single tank evaporator has a better performance in comparison to a double tank evaporator, as the available heat transfer area is greater. In other words, the ratio of the total heat exchange area to the total volume of the core is more for evaporators with single tanks. Laminate evaporators typically have 4 to 6 refrigerant passes. For a same cooling capacity, the average refrigerant velocities in a 6 pass (Figure 3) evaporator are higher than a 4 pass resulting in a better temperature profile for the evaporator core. Figure 3 shows an automotive evaporator with a 6 pass flow configuration. The two-phase refrigerant enters the evaporator through the inlet pipe and vapor exits the evaporator through the outlet pipe. The two-phase refrigerant enters the evaporator through the tank and moves upwards in multi-flow channels (or plates) in pass 1 and then exits the evaporator as vapor from pass 6. A laminate evaporator operates as a cross-flow heat exchanger in which the air flows normal to the refrigerant. The performance of laminate evaporators is superior to the serpentine [17] or finned tube [18] evaporators. This is due to higher heat transfer coefficients. Laminate evaporators are lighter in comparison to other types of evaporators. The rate of heat transfer per unit weight (and volume) is more for laminate evaporators in comparison to other types of evaporators. The depth of laminate evaporators currently varies from 50 to 90 mm. However, the automotive industry is making an effort to reduce the volume of the evaporator core by decreasing the depth of the core in the range of 35~50 mm [19]. The evaporators are typically coated with hydrophilic coating [20] for effective drainage properties [21, 22]. Some of the evaporator designs may be prone to accumulation of lubricating oil in the plates at part load conditions [23]. The average surface temperature was used in the calculation given in next section. The average saturation pressure was determined by knowing the inlet and outlet evaporator pressures. Four thermocouples are located on the upstream of the evaporator to determine enthalpy at the inlet from the airside. And the corresponding saturation temperature was used for the calculation for equation 6. During testing, test conditions were maintained in order to have two-phase flow over the entire evaporator. In other words, there was no super heating at the end of the evaporator. Table 1 shows the design



FIGURE 4 Photograph of the test facility - the room on the left side simulates the cabin interior; and the room on the right


simulates the engine compartment.


FIGURE 5 HVAC system set-up in the room simulating the cabin interior



5



FIGURE 6 Compressor, Condenser set up in the room simulating engine compartment


TABLE 2 Parameters for the Current Investigation

Variable

Magnitude

Cabin Condition

25 °C, 50%RH

Ambient Conditions

25, 35, 45 °C

Evaporator Airflow

5, 7, 9 m3/min

Condenser Face Velocity

2 m/s

Refrigerants

HFO-1234yf

Compressor RPM

800 to 3000

Refrigerant Mass Flux

200 to 1200 kg/m2 s

Heat Flux

2 to 6.2 kW/ m2

details of the evaporator that is used for this study. The following section presents the details of the experiments. The pressure transducers have an accuracy of ±0.25% over the full scale. This corresponds to ±0.09 kg/cm2 for the high side pressures for the AC loop. Thermocouples with a measurement accuracy of ±1.0 °C were used to measure the air temperatures. The experimental uncertainty of heat transfer coefficients was of the order of ±8% from the refrigerant side measurements.

Heat Transfer Coefficients Average evaporator heat transfer coefficient [24] is calculated form the following equation:

h e = Q /  A ( Tav wall − TS,av )  (1)


Boiling Heat Transfer Coefficient Versus LockhartMartinelli Parameter Flow boiling inside of a tube typically consists of a combination of the nucleate boiling and forced convection evaporation. For forced convection evaporation, flow boiling heat transfer coefficient data can be correlated as:

h exp fn h l [1 / X tt ] (2) n

where hl is the liquid phase heat transfer coefficient and Xtt is Lockhart-Martinelli parameter. Single-phase heat transfer coefficient by Dittus Boelter [24] is given by the following equation:

h l = 0.023 ( k / D ) Re0.8 Pr 0.4 (3)

Lockhart-Martinelli [25] parameter is given by the following equation:

X tt = [(1 − x / x )

0.9

( ρ v / ρl ) ( ( µl / µ v ) 0.5

0.1

 (4) 

The above equations are used in the development of the correlation to predict flow boiling heat transfer coefficients during evaporation process. The magnitude of forced convective evaporation can be expressed by the gradient n of a linear regression of the experimental data. Figure 9 shows the measured heat transfer coefficients in the pre-dryout region plotted against (1/Xtt). It can be seen that most of the current data can be fitted to the regression:



h exp / h l = 2.8738 (1 / X tt )

0.109

(5)

The above described methodology to correlate the two phase flow boiling heat transfer coefficient through single phase liquid flow has been used by many researchers. For details, the readers are referred to Table 2 of chapter 4 of ASHRAE Fundamentals Handbook (1993). Note that the effect of mass flux is included in the correlation (equation 5) through Reynolds number assuming single phase liquid flow. The effect of heat flux is included indirectly in LockhartMartinelli parameter Xtt. Note that Xtt (equation 4) varies from approx. 11 to 0 when the refrigerant quality is varied from 0 to 1.0 by keeping the other thermo-physical and thermodynamic properties constant at the evaporating temperature. The effect or pressure drop is already included in the correlation as the refrigerant qualities are computed by using the actual test data that included pressure drop. The developed correlation is valid for the following range of operating parameters: Evaporating pressures: 0.188~0.210 kg/cm2G. Refrigerant Mass flux from 200 to 1200 kg/m2 s. Heat Flux: 2 to 7 kW/m2). The following section provide details of the test results along with detailed discussions.

Results and Discussions Figures 7, 8, 9, 10, and 11 show the performance results from this investigation. These tests have been conducted with the cabin maintained at 25 °C with a relative humidity of 50%

and evaporator airflow rate varied from 5 to 9m3/min with heat and mass flux ranging from 2 to 6.2 W/m2 and 500 to 1000 kg/m2 s. The refrigerant temperature was maintained at 4.4 °C. Figure 7 shows the experimentally measured evaporator flow boiling heat transfer coefficients for HFO-1234yf as a function of refrigerant quality, for 3 different refrigerant mass fluxes at a refrigerant temperature of 4.4 °C (40 °F). The f low boiling heat transfer coefficients increase with the increase in the refrigerant quality. The heat transfer coefficients continue to increase until dryout starts to occur. Dryout occurs as liquid refrigerant is unable to adhere to the inside wall surface due to high vapor velocities thereby shearing off the liquid from the wall At this point the wall surface temperature starts to increase resulting in lower heat transfer coefficients. This phenomena occurs beyond refrigerant qualities of 0.75. Data for all 2 refrigerant mass fluxes show similar behavior. In order to determine a correlation of heat transfer coefficient as a function of single phase liquid heat transfer coefficient and Lockhart-Martinelli parameter (equation 2), Figure 8 was plotted. This figure shows the ratio of the flow boiling heat transfer coefficient to the liquid phase heat transfer coefficient (hexp/h l). Figure 9 shows the calculated liquid phase heat transfer coefficient (h l) calculated by equation 3. Finally, the ratio of hexp/hl is plotted against Xtt (equation 4) on a log-log graph in Figure 10. This regression is given by equation 5. This is the correlation that has been developed to calculate the flow boiling heat transfer coefficient during evaporation process. Finally, Figure 11 shows the experimental data versus the calculated data with the developed correlation. As can be seen the developed correlation is able to predict the flow boiling


FIGURE 7 Evaporator flow boiling heat transfer coefficients with HFO-1234yf system at an evaporating temperature of 4.4 °C




7


FIGURE 8 Average evaporator flow boiling heat transfer coefficients for baseline (HFC-134a) and HFO-1234yf as function of evaporating temperatures at a mass flux of 430 kg/m2.s


FIGURE 9 Single phase liquid heat transfer coefficient as a function of refrigerant mass flux evaporating temperature of 4.4 °C

heat transfer coefficient within ±23% of the measured data. Only one data point is out of ±20% range. Figure 12 shows a comparison of the predictions from the current correlation with Kandlikar’s [6] correlation. The © 2018 SAE International. All Rights Reserved.

solid curve is the correlation developed by the author and the data points are the predictions form Kandlikar’s correlation. As can be seen from the graph, Kanlikar’s correlation predicts heat transfer coefficients higher by approx. +30% over the




FIGURE 10 Heat Transfer coefficient as a function of Lockhart-Martinelli parameter


FIGURE 11 Developed correlation for predicting flow boiling heat transfer coefficient

entire range of refrigerant mass flux. This is a significant difference between the developed correlation and the predictions for correlation form the open literature. However, it should be noted that even through the heat transfer coefficient is higher by 30%, the overall heat transfer rate might not be very different as the overall heat transfer coefficient form refrigerant to air “U” is really controlled by the fin side thermal resistance.

Effect of Oil and Moisture On AC System Performance In this investigation, the heat transfer coefficients have been measured with a 3% OCR. Studies performed by many researchers ([26, 27, 28, 29, 30]; Zhao et al., 2002) have concluded that there is a significant impact on the heat © 2018 SAE International. All Rights Reserved.



FIGURE 12 Comparison of current Correlation with

Kandlikar’s [6] Correlation

9

•• The following is the developed correlation for HFO1234yf: hexp/hl = 2.8738 (1/Xtt)0.109 •• This is the first correlation in the open literature that has been developed for laminate evaporators that can be used to predict flow boiling heat transfer coefficients •• The developed correlation is able to predict the flow boiling heat transfer coefficients within ±23% of the experimentally determined data.


•• The developed correlation can be used for design and test engineers in performance prediction of the automotive evaporators

transfer coefficients as a function of OCR. ASHRAE standard [14] describes the procedure for determining the oil circulation ratio (OCR) in an AC loop. Depending upon the lubricant quantity (i) evaporation and condensation heat transfer coefficients might increase or decrease that will have a direct impact on the overall system performance; (ii) a lower viscosity can help improve compressor power; (iii) pressure gradient always increases with lubricant in the system; (iv) presence of oil also helps in providing a uniform temperature of the heat exchangers; (v) decrease in the oil helps in improving isentropic and volumetric efficiencies; and (vi) very low oil might result in internal leaks resulting in lower system and compressor performance. A detailed summary of the effect of lubricating oil on the performance of direct-expansion and flooded evaporators; and condensers was presented earlier by the author [31]. The author has planned to conduct tests with different % OCR ratios in the system with HFO-1234yf that could have major influence on the component and system performance as discussed above. One more variable that tends to affect the AC system performance is moisture. If the moisture contents in the system are very low, the system performance is not affected. However, if the moisture quantity goes beyond a certain value [32] it adversely affects the system performance.

Conclusions An experimental investigation has been conducted to determine flow boiling heat transfer coefficients during evaporation process for HFO-1234yf. A methodology is presented to develop correlation by regression analysis by using experimental data with calculated liquid heat transfer coefficient and Lockhart-Martinelli parameter. Based on the current investigation, the following are the main conclusions: •• A correlation has been developed to predict flow boiling heat transfer coefficients for HFO-1234yf for automotive evaporator (laminate evaporators). © 2018 SAE International. All Rights Reserved.

Basic fundamental mechanism of heat transfer has been described in details. This piece of information will be useful for the designers and engineers.

Future Work Further tests have been planned with HFO-1234yf to determine the heat transfer coefficients as a function of percent oil circulation ratio; and to improve this correlation and to compare predictions with other correlations in the open literature.

Acknowledgments The author would like to thank CKNA’s test laboratory for helping in conducting tests; and the management (Mr. Jeff Dzidedic, Goto san) for giving permission to publish this study. Author would also like to thank Mr. Ayush Mathur, BCBS, Detroit for the regression analysis for the correlation.

Contact Information Dr. G.D. Mathur, P.E. Fellow SAE, ASHRAE, ASME Technical Specialist, Climate Control Sr. Manager, Design & Development CalsonicKansei North America 27000 Hills Tech Court Farmington Hills, MI 48331-5725 Phone 248-848-4855, [email protected]

Nomenclature A - Area (m2) D - Hydraulic diameter (m) h - Heat transfer coefficient k - Thermal Conductivity (W/m K) n - Power in equation 2 Pr - Prandtl number Re - Reynold number



Q - Heat Transfer Rate (W) T - Temperature (K) x - Quality X - Lockhart and Martinelli parameter defined by equation 4 Greek Symbols: ρ - Density μ - Viscosity Subscripts aw, wall - average wall e - evaporator exp - experimental l - liquid s, av - saturation, average av - average v - vapor

References 1. Mathur, G.D., “Enhancing AC System Performance with a Suction Line Heat Exchanger with Refrigerant HFO-1234yf,” SAE Paper # 2011-01-0133, 2011, doi:10.4271/2011-01-0133. 2. Mathur, G.D., “Experimental Investigation of the Performance of a Laminate Evaporator with HFO-1234yf as the Working Fluid,” SAE Paper # 2011-01-1170, 2011, doi:10.4271/2011-01-1170. 3. Mathur, G.D., “Experimental Investigation of AC System Performance with HFO-1234yf as the Working Fluid,” SAE Paper # 2010-01-0041, 2010, doi:10.4271/2010-01-0041. 4. Mathur, G.D., “Experimental Performance of a Parallel Flow Condenser with HFO-1234yf as the Working Fluid,” SAE Paper # 2010-01-0328, 2010, doi:10.4271/2010-01-0328. 5. Saitoh, S. and Nakamura, Y., “Boiling Heat Transfer Coefficient for HFO-1234yf Flowing in Small Smooth Diameter Horizontal Tube,” 2010 Next Gen Air Conditioning & Refrigeration Symposium, Tokyo, Feb 17-19, 2010. 6. Kandlikar, S.G., “A General Correlation for Saturated TwoPhase Flow Boiling Heat Transfer Inside Horizontal and Vertical Tubes,” Journal of Heat Transfer 112:219-228, 1990. 7. Shah, M.M., “A New Correlation for Heat Transfer During Boiling Flow Through Pipes,” ASHRAE Transactions 82(2):66-86, 1976. 8. Shah, M.M., “Chart Correlation for Saturated Boiling Heat Transfer: Equations and Further Study,” ASHRAE Transactions 88(1):185-196, 1982. 9. Shah, M.M., “Evaluation of General Correlation for Heat Transfer During Boiling of Saturated Liquid in Tubes and Annuli,” HVAC&R 12(4), Oct 2006.

10. ASHRAE, “Fundamentals Handbook,” (Atlanta, GA, American Society of Heating Refrigerating and Air Conditioning Engineers. Inc., 1997). 11. ASHRAE, “Refrigeration Handbook,” (Atlanta, GA, American Society of Heating Refrigerating and Air Conditioning Engineers. Inc., 1997). 12. ASHRAE Standard 41.2, “Standard Methods for Laboratory Airflow Measurements”, Jul 1987. 13. ASHRAE Standard 37-1988, “Methods of Testing for Rating Electrically Driven Unitary Air Conditioning and Heat Pump Equipment”, 1988. 14. ANSI/ASHRAE Standard 41.4-1996, “Standard Method for Measurement of Proportion of Lubricant in Liquid Refrigerant”, American Society of Heating, Refrigeration and Air-Conditioning Engineers, Inc., Atlanta (USA), 1996. 15. Mathur, G.D., “Simulation of Thermal and Hydrodynamic Performance of Laminate Evaporators,” Society of Automotive Engineers, SAE Paper # 2000-01-0573, 2000, doi:10.4271/2000-01-0573. 16. Grosse, G. et al., “Heat Transfer in Microchannels Tubes for Automotive Evaporators,” SAE Paper # 2006-01-1449, 2006, doi:10.4271/2006-01-1449. 17. Mathur, G.D., “Performance of Serpentine Heat Exchangers,” Society of Automotive Engineers, SAE Paper # 980057, 1998, doi:10.4271/980057. 18. Mathur, G.D., “Modelling and Simulation of Thermal and Hydrodynamic Performance of Heat Exchangers for Automotive Applications - Part II: Evaporators,” New Developments in Heat Exchangers for Automotive Design (SP-1262), SAE, Paper # 97083021-30, 1997, doi:10.4271/970830. 19. Nakamura, T., “Ultra Thin & Ultra Light RS Evaporators,” SAE Paper # 2003-01-0527, 2003, doi:10.4271/2003-01-0527. 20. Casenave, C., “Development of Surface Coating for Evaporators,” SAE Paper # 2004-01-0215, 2004, doi:10.4271/2004-01-0215. 21. Mathur, G.D., “Water Carryover Characteristics from Evaporator Coils During Transitional Airflows,” SAE Paper # 2000-01-1268, 2000, doi:10.4271/2000-01-1268. 22. Mathur, G.D., “Investigation of Water Carryover from Evaporator Coils,” SAE Paper # 1999-01-1194, 1999, doi:10.4271/1999-01-1194. 23. Mathur, G.D., “Experimental Investigation to Determine Accumulation of Oil in a Single Tank Evaporator with Tank at the Top at Different Compressor Operating Speeds,” SAE Paper # 2004-01-0213, 2004, doi:10.4271/2004-01-0213. 24. Incropera, F.P. and DeWitt, D.P., “Fundamentals of Heat Transfer, Chapter 11,” (John Wiley & Sons, 1981). 25. Lockhart, R.W. and Martinelli, R.C., “Proposed Correlation of Data for Isothermal Two-Phase, Two-Component Flow in Pipes,” Chemical Engineering Progress 45(1):39-48, 1949. 26. Baustian, J.J., Pate, M.B., and Bergles, A.E., “Measuring the Concentration of a Flowing Oil-Refrigerant Mixture with a




Bypass Viscometer”, ASHRAE Transactions, Vol. 94, Pt. 2, No. 3180, 1988. 27. Bayani, A., Thome, J.R., and Favrat, D., “Online Measurement of Oil Concentrations of R-134a/Oil Mixtures with a Density Flowmeter,” International Journal of Heating, Ventilating, Air-Conditioning and Refrigerating Research 1(3), 1995. 28. Eckel, S.J. and Pate, M.B., “In-Tube Evaporation and Condensation of Refrigerant-Lubricant Mixtures of HFC-134a and CFC-12,” ASHRAE Transactions 97(Pt 2):62-70, 1991. 29. Hwang, Y., Cremaschi, L., Radermacher, R., Hirata, T., et al., “Oil Circulation Ratio Measurement in CO2 Cycle,”

11

Proceedings of the 2002 International Conference-New Technologies in Commercial Refrigeration, 22-28, 2002. 30. Zhang, M. and Webb, R.L., “Effect of Oil on R-134a Condensation in Parallel Flow Condensers,” Proceedings of Vehicle Thermal Management Systems Conference, SAE, 369-376, 1997. 31. Mathur, G.D., “Alternatives Takes on R-12,” Engineered Systems 10(7):49-53, Sep 1993. 32. Mathur, G.D. & Goswami, D.Y., “An Experimental Investigation of the Effect of Moisture on the Performance of an Air-Conditioning System”, 31st Intersociety Energy Conversion Conference, 3, 2021-2026, 1996.

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior written permission of the copyright holder. Positions and opinions advanced in this paper are those of the author(s) and not necessarily those of SAE International. The author is solely responsible for the content of the paper. ISSN 0148-7191