Specifying heat transfer coeff for a cylindrical pipe
 

I'm setting up an analysis for an exhaust system which includes compressed airflow through a cylindrical uninsulated stainless steel pipe. To include the effects of convective heat transfer I must specify a heat transfer coefficient and free stream temperature. I know the mass flow rate through the pipe and I know the temperatures at the inlet and outlet so could easily work out the bulk temperature. But my question is in order to do a calculation to determine roughly what value the HTC should be, do I use the equations for forced convection (Re number -> turbulent so work out Nu number, rearrange for HTC)? The issue I have with this is it doesn't include the external temperature anywhere so will this work out the HTC for the fluid? Or similarly if I use the resistors analogy for a conductive calc where the two resistors would be conduction through the pipe and convective heat transfer to atmosphere? My problem with this is that I don't know the heat transfer rate so I'm not sure how this will give me the HTC or whether this is even the HTC I need?

Thanks in advance.

The fluid side heat transfer coefficient is a flow property and you do not specify this value in CFD.


What you are doing when you use the convective BC is you are specifying the external heat transfer coefficient and external reference temperature. You can use the thermal resistance analogy and lump the heat conduction through the pipe into this heat transfer coefficient if you like.

Thank you, that makes sense! However, is it possible for me to to approximate the value of the heat transfer coefficient given that I don't know a heat transfer rate? That's where I was confused because (in electrical terms) I know the voltage and 1 of the two resistors (the conductive resistor) but don't know the current so can't solve for the second convective resistor?

Yeah you always have to make some guesses to get the htc. The (theoretical) meaning of heat transfer coefficient is that it doesn't depend on the heat transfer rate. That is, the heat transfer coefficient should only be a flow property and not a depend on (e.g. whether it is a sunny or rainy day). Of course we know that the properties change with temperature and pressure and this is not true in general.

If you want to not make guesses then you have to include the external part in the simulation. The hyperbole is always you have to model the entire universe.