Evaporation heat transfer and pressure drop of refrigerant 134A in a small pipe - Reply

(1)

manifold designs, which will aﬀect the ﬂow distribution in the channels. Explanation on this will be welcomed.

The present authors have used the Shah correlation [5] to predict the Yan and Lin data for Tsat¼ 31 °C, q00_{¼ 5 kW/m}2_{, and 100 and 200 kg/s m}2 _{mass velocity} (G). This comparison shows that the Yan and Lin data for G¼ 100 kg/s m2 _{and x}_{¼ 0:2 are approximately five} times that predicted by the Shah equation. However, at 0.8 vapor quality, the predicted values for both mass velocities are approximately equal and are 25% above the Shah correlation prediction. One may expect nucle-ate boiling to influence the evaporation coefficient at low vapor quality. However, the q00_{¼ 5 kW/m}2_{is sufficiently} small that one would not expect significant nucleate boiling enhancement at this heat flux. The authors ex-planation of this will be appreciated.

References

[1] Y.-Y. Yan, T.-.F. Lin, Evaporation heat transfer and pressure drop of refrigerant 134A in a small pipe, Int. J. Heat Mass Transfer 41 (1998) 4183–4194.

[2] C.-C. Wang, S.-K. Chiang, Y.-J. Chang, T.-W. Chung, Two-phase ﬂow resistance of refrigerants R-22, R-410A and R-407C in small diameter tubes, Trans. IChemE, Part A 79 (2001) 553–560.

[3] S.G. Kandlikar, A general correlation for saturated two-phase ﬂow boiling heat transfer inside horizontal and vertical tubes, J. Heat Transfer 112 (1990) 219– 228.

[4] Y.Y. Yan, private communication, December 2001. [5] M.M. Shah, Chart correlation for saturated boiling heat

transfer: equation and further study, ASHRAE Trans. 88 (1982) 185–196.

Ralph L. Webb Jin Wook Paek Department of Mechanical Engineering College of Engineering Pennsylvania State University 206 Reber Building University Park PA 16802-1412, USA Tel.: +1-814-865-0283; fax: +1-814-865-1344 E-mail address: [email protected]

Reply to Prof. R.L. Webb’s and Dr. J.W. Paek’s comments

Dr. Yan and I examine the comments from Prof. Webb and Dr. Paek carefully. Here is our response. We appreciate their comments to point out our mistakes.

(1) By checking all the measured raw data and the data reduction procedures leading to the evaporation heat transfer coefficient hr and friction coefficient ftp with extreme care, the results presented in Figs. 13 and 14 of the article for the comparison between the corre-lations proposed by Yan and Lin (1998) and the mea-sured data are noted to be in mistake. More specifically, the error in ftp( DPf=ð2G2vmL=DiÞ) is due to the in-correct evaluation of the specific volume vmfor the two-phase R-134a liquid–vapor mixture. The measured data for the frictional pressure drop DPf, however, are cor-rect, so are the heat transfer coefficient hr. Moreover, the correlation for hr given in the article of Yan and Lin (1998) is also incorrect. This correlation is too compli-cate to use conveniently and there are 36 values of the empirical constants involved in the equation. Some mistakes were made in the curve-fitting procedures in missing the final step to bring the data well above and well below the correlation together.

(2) A new and simpler correlation for hris proposed here. For Xm50:7 hr¼ 4:36 kl Di Pr1=3l ð1 XmÞ0:5ðC1 Reeqþ C2ÞðC3 Bo þ C4Þ ð1Þ and for Xm>0:7 hr¼ 4:36 kl Di Pr1=3_l ð1 XmÞ0:5ðC1 Reeqþ C2Þ ð2Þ

Here the coeﬃcients C1to C4are expressed as

C1¼ 0:0124G0:368 ð3Þ

C2¼ 1:49G0:514 ð4Þ

C3¼ 1166Xmþ 1028 ð5Þ

C4¼ 0:53 e0:931Xm ð6Þ

Note that the unit for the mass ﬂux of R-134a G is kg/m2_{s, and Re}

eqand Bo are respectively the equivalent Reynolds number and Boiling number, which have been deﬁned in the article. Meanwhile, a new correlation is provided here for the friction factor as

ftp¼ 0:127 Re0:1925eq ð7Þ

The comparison of the above correlations with the correct measured data for hrand ftpis shown in Figs. 1 and 2. The results show that the root-mean-square de-viations between the above correlations and measured data are 18% for the heat transfer coeﬃcient hrand 22% for the friction factor ftp.

(3) The refrigerant R-134a is sent into the 28 small pipes in a row by an upstream plenum, which is a hor-izontal large cylindrical container with two openings of 84 mm wide and 2 mm high to allow the refrigerant to

(2)

move into the small pipe from the container and to allow the refrigerant to move into the container from the re-frigerant loop. This inlet flow arrangement is not ex-pected to cause significant flow maldistribution at the inlet of the pipes.

(4) The large discrepancy between the present data for hrand the Shah correlation at low heat flux q00w, low mass flux G and low quality Xm can be made clear by visualizing the boiling flow in the small pipes. However, we do not conduct flow visualization here and hence do not have information on this question. Visualization of boiling flow in a small channel is currently in the stage of

experimental system design in our laboratory. We plan to start the experiment in about 18 months.

Y.-Y. Yan T.-F. Lin Department of Mechanical Engineering National Chiao Tung University 1001 Ta Hsueh Road Hsinchu 30049 Taiwan, ROC Tel.: +886-35-712121; fax: +886-35-720634 E-mail address: [email protected]