Hi Friends, There is a square matrix of 20x20 long horizontal cylinders (5 in dia) containing liquids at temperatures in the range (100-800 C). Water at 50 C is being sprayed on the upper row of 20 cylinders, Total Spray Flowrate = 100 kg/s. I want to know the heat transfer coefficient between a cylinder and the spray water. Some literature/correlations for the spray heat transfer coefficient will be highly apprecited. Thanks

Noora, I think this problem is not as easy as it looks at first glance. I work for a company that designs and manufactures commerical nuclear fuel assemblies, and worked in the testing group for about five years. One of my projects was to develop spray heat transfer coefficients for Boiling Water reactors. Our bundle was 10x10 with 10.05 mm rods, but the rods were vertical. The testing cost about $1 million for heater rods, power supply, etc, etc., and took over a year to design, build, and execute. Of course, all results are propietary.

One problem is that the cold water hitting the hot rods (800 C) rods will first sputter (from Liedenfrost phenomena) before the bundle will quinch. There is also the problem that you may have CCFL occurring at the top of the rod array so that the spray may not be able to penetrate.

To make the answer a little shorter, look in the open literature for spray heat transfer during a BWR loss of coolant accident. However, Kreith (Principles of Heat Transfer) gives some formulas for flow over a tube bank in chapter 9 that includes the effect of tube arrangement.

Dear Tom, Thanks a lot for your reply. As far as the BWR fuel bundle is concerned, it is vertical and may not work in my case, I think. I ll go through Kreith. Can u give some reasonable number for the HTC for a cylinder with 800C surface. I would appreciate if u give me your e-mail so that I can further explain u the problem.

Thanks a lot again

Noora

Noora, during a LOCA for a BWR the initial spary heat transfer coeffieicnts that are used are 3.5 BTU/hr-ft2-F for side rods and 1.5 for interior rods. These are mandated by law and are therefore very conservative. After the bundle is quenched (usually on the order of 180 sec), we apply a heat tranfer coefficient of 14000 BTU/hr-ft2-F. Again, this is an imposed value and not based on testing.

Quenching the bundle is the hardest part. We use the Yamanouchi quench front velocity model or the 2DQ model by Kahn and Steves to calculate quenching. It is during this quenching that the Leidenfrst superheat plays an important role.

I hope I have helped some rather than added confusion.

Tom

Not sure if this is useful, but take a look at:

http://www-waterloo.ansys.com/cfx/il...art_henryk.pdf