[TallesC]

takes personal sequinte put the boundary condition on the surface of a pipe:

Q = constant * ( Tp^ 4 - Ts ^ 4)

where:

* Tp - is constant.
* Ts - the surface temperature of the tube
* Q - heat supplied.

ie, Q and Ts are unknown, is how I do it? if yes how to call Ts in expression?

[evcelica]

HeatFlux = Constant [W/m^2/K^4] * (Tp[K])^4 - (Temperature^4))

[TallesC]

but Ts is the temperature at the surface, and the temperature is variable across the field, is not it?

my advisor, said it could be done as follows, as the condition of no slip near the wall, the heat conduction mechanism is thus the condition would be:

************************ K * dT = const * (Tp [K]) ^ 4 - (TS ^ 4))

Ts just found on the surface, it is possible to do this?

[Antanas]

Originally Posted by TallesC

but Ts is the temperature at the surface, and the temperature is variable across the field, is not it?

my advisor, said it could be done as follows, as the condition of no slip near the wall, the heat conduction mechanism is thus the condition would be:

************************ K * dT = const * (Tp [K]) ^ 4 - (TS ^ 4))

Ts just found on the surface, it is possible to do this?

You set it as BC, so CFX takes Temperature at that boundary. Do like evcelica wrote. What's the problem?

[TallesC]

Originally Posted by Antanas

You set it as BC, so CFX takes Temperature at that boundary. Do like evcelica wrote. What's the problem?

Ok!!!Boundary type is wall right???

[TallesC]

Guys, simulation show error of overflow, but when put Q =constant its working, any idea this causes the problem???

[ghorrocks]

This is an FAQ: http://www.cfd-online.com/Wiki/Ansys...do_about_it.3F

So in your case it looks like the introduction of a variable Q (assuming that is what you have done) is not numerically stable.

[Antanas]

Originally Posted by TallesC

Guys, simulation show error of overflow, but when put Q =constant its working, any idea this causes the problem???

As Glenn mentioned you have strong dependence of source term on solution (temperature). If you set your source through flux option solver most likely fail. Try to use the idea proposed here

[TallesC]

Originally Posted by Antanas

As Glenn mentioned you have strong dependence of source term on solution (temperature). If you set your source through flux option solver most likely fail. Try to use the idea proposed here

Thank you so much Antanas!!! your idea working....