h forced convection correlation, frontal plate 2

[xj25]

Hi,

hc=25W/m2K is a value frequently used for forced convection for calculating overall heat transfer coeficients for windows, walls etc. It is referenced in various regulations, EN and ISO.

I have been looking from where this value comes, and some sensible way of linking it to the so called "mean wind speed". This "mean" speed is (from other references) usually taken as 8-10m/s as a common value.

I looked for some Nusselt correlation for frontal flux forced convection. There are several well known references for flux along a plate, but I found nothing for perpendicular or "random direction" flux (just for natural convection on horz. plate).
An idea is use a paralell flux correlation, with some "rule" for choosing a sensible charecteristic lenght, as it seems that for typical speeds of 8-10m/s, a L about 1-2m seems to work for getting the above described hc

I usually prepare overall U calculations and would like to undestand better the assumpions of the hc used. I would appreciate any inshight or correction, thanks for reading.

 

[ione]

I’m aware of such a correlation for Nu (vertical plate and horizontal flow):

Nu = 0.228*Re^0.731*Pr^(1/3)

What above is applicable for gases and for Re in the range between 4.0E+03 and 1.5E+04. Re is computed taking the height of the plate as the characteristic length.

You should anyway take these correlations for what they are, namely a best fit of data empirically collected for specific geometries.

 

[IRstuff]

There used to be a forced convection heat transfer calculator that would produce about 25W/m^2K with about 12.3m/s air velocity.

I can't find the calculator, but it was based on Reynolds and Prandtl numbers to calculate the Nusselt number and hence, the transfer coefficient

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529

Of course I can. I can do anything. I can do absolutely anything. I'm an expert!

 

[ione]

Maybe the following one?

http://www.thermal-wizard.com/tmwiz/default.htm

 

[xj25]

I run into this
Link and at the end there is a discussion about the contribution of natural and forced convection, but again there is no reference to such case for vertical plate and horizontal flow.
the thermalwizard neither have this case for forced convection.

Ione, do you have the textbook referece for that correlation?
For H=1-2m and v=10m/s gives me Nu=3800-6500 and h about 80-95W/m2K, but the Re=7,2E5 is out of the range.

Regards

 

[IRstuff]

Have you looked at

http://people.csail.mit.edu/jaffer/SimRoof/Convection/

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529

Of course I can. I can do anything. I can do absolutely anything. I'm an expert!

 

[ione]

See attached file.

Reference quoted is:

M. Jakob. Heat Transfer. Vol. l. New York: John Wiley & Sons, 1949.

 

[ione]

And again (configuration 9)

http://www.mscsoftware.com/training...dex.html#page/Thermal/th_convection.11.2.html

 

[IRstuff]

There's a nice set of correlations at

http://www.egr.msu.edu/~somerton/Nusselt/i/i_a/i_a_1_(ii)/i_a_1_(ii).html,

which are apparently derived from Incropera and DeWitt

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529

Of course I can. I can do anything. I can do absolutely anything. I'm an expert!

 

[ione]

IRstuff,

The correlation provided by Incropera and DeWitt is the one I've reported above and they too quote as reference the german scientist Max Jacob:

M. Jakob. Heat Transfer. Vol. l. New York: John Wiley & Sons, 1949.

I've not been able to find any reference dated before this one.

 

[IRstuff]

Anything before that will most likely be less accurate, due to technological challenges.

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529

Of course I can. I can do anything. I can do absolutely anything. I'm an expert!

 

[xj25]

Hi, after reading the overwhelming summary in Link, I did some calculations with various correlations and equations for Nu found there. For L in the range of my interest (about 2,5m) got the following graph.

The main conclusion is that it seems that there is no experimental data for random or frontal flow for paralell plates with high Re,
probably because the conditions of a specific experiment are dificult to apply for a general case.

It also explains that roughness affects quite a lot, but could not found experiment data that match the predictions shown. I.e. at the end of the discussion it shows a roof experiment with probably high roughness, that shows a low value of h.

Thanks for the discussion!