Analytical determination of two phase boiling heat transfer coefficient in pipes

[kfoh]

Hello all,

For the design of a parametric heat exchanger program I need to determine the above mentioned two phase boiling heat transfer coefficient.

Since the design tool shall be kept reasonalbly small in size, semi empirical correlations should be used.
I however struggle to find a correlation giving me the required heat transfer coefficient, since I also do not know the heat flux.
All the correlations for heat transfer coefficients either use heat flux to calculate it or the other way round. I found a method by Liu-Winterton, which however fails in predicting reasonable flux values.

Does anybody know an alternative possibility?

kind regards

 

[davefitz]

There are some published heat transfer coefficients ( mostly applied to the nuclear steam generators), but they are very large and do not represent a significant resistance to heat transfer compared to the normal fouling of the inside surface of the tube. As a practical matter, assume infinity for the 2 phase HT coefficient but ensure a conservative value for inside surface fouling is also used. The other essential 2 items are to ensure you do not exceed critical heat flux ( DNB or dryout) and use clean treated water so that the inside fouling does not exceed your design assumption. The critical heat flux correlations generally used for tubes are those of KonKov and Doroshchuck.

"Whom the gods would destroy, they first make mad "

 

[ione]

You can take a glance at the correlations reported on this link

http://web.iitd.ac.in/~pmvs/courses/mel725/flowboil.pdf

Handle those correlations for what they are, namely the best fit of experimental data.

 

[kfoh]

Thank you guys for your advice!
@davefitz
I will take that into account. Anyway in my calculations the boiling heat transfer coefficient is extremely high, such that the overall heat transfer coefficient is mainly determined by the shell heat transfer coefficient. The fluids I am using are air and LNG. Fouling was so far not planed to be taken into account, since the study is wanted for theoretical thermal sizing of the exchanger. Thank you also for the hint with the DNB, I didnt have a look at this yet.

@ione
Thank you for your post. I already knew that paper. However it is of limited use for me, since the boiling flow is mostly vertical flow while I need horizontal flow. Additionally exactly the problem I had described above is given for this paper: Either heat transfer coefficient is already known or heat flux. For my calculations however both are unknown. I just have the bulk fluid temperatures. With a method of Shah I estimated the wall temperature (by taking the heat transfer coefficients of the last iterative step of my program). With this I then calculate the heat flux by using the Liu-Winterton approach. However I'd like to have a second paper giving me some confirmation on the results this method gives, for it seems to me quite unrealistic.

 

[davefitz]

Horizontal tubes will have vapor layout at the top centerline of the tubes and will only have benefit of the 2 phase boiling HT coefficient on the bottom half of the tubes. Full circumferential "wetted " boiling can occur if you use internally rifled tubes, or use vertical tubes. Refer to tech papers by KWU authors Kiefer, Kohler and Hind in the Intl J or Multi phase flow, circa 1982-1995. There is not expected to be considerable fouling with pure LNG service.

"Whom the gods would destroy, they first make mad "

 

[srfish]

According to "Process Heat Transfer" by Serth the Chen correlation has been the most widely used. Serth also gives the Gungor-Winterton and the Liu-Winterton correlation.

There is also a text book "Boiling Heat Transfer and Two-Phase Flow" by Tong and Tang.