Convective heat transfer from fluid to surface

[borgwiser]

Been beating around with this problem for awhile and am just not happy with my results. I'm gonna venture to ask for help...

I've been trying to model a process whereby heat is used to heat a plate (I've posted this question prior but I have revised some of the info as I've encoutered problems in working out the model). This is a closed system kind of like a heat exchanger. I have air entering a box at a temperature and exiting at a lower temperature. I have a conveyor coming into the setup cold, getting heated by the air, and exiting at a warmer temp. I've built my model but it isn't valid for every situation I throw at it. I've tried to do it in a finite element method by breaking the length of the section into thousands of pieces..well, when I change the number of pieces my data for the thermo data does some things that does not happen in the real world (for example, decreasing the length of the plate by a factor of 10 should result in less heat transfer to the plate, but I'm not seeing that).

Question: Does anyone have a good source or can provide equations to model a parallel or counterflow system such as this? I would even welcome differential equations as I have already written a Runge-Kutta solver to handle these. I'm just having a horrific time making sense of this and I know it isn't that difficult. I look forward to any input.

Many thanks,
R

 

[jcloer1]

Silly question I'm sure but sometimes I overlook the simple stuff in search of a difficult answer......are you sure the thermal conductivity of the plate isn't overcoming the time spent inside and delta T of the ambient air in the box?

 

[ko99]

Decreasing the plate size (reduced mass) should cause the plate temperature to increase faster even if the total heat transfer may be less.

I'm not sure exactly how I'd solve your problem. I might look at a slice of the plate and use t = RC to estimate it's temperature rise during the time it's inside the heated box. If I recall correctly:

t = time constant (sec). It takes approx 5t to reach 95% (i think) of the steady state temperature

R = thermal resistance (C/W) from the plate to the air.

C = thermal capacitance = mass x specific heat



ko (

http://www.ecooling.biz)