XANSYS Message: 42265 [Go back to message list]
No rating yet
Rate item:

Subject: Re: THERM convection coefficient as a function of pressure
Author: Jose Torres
Date: 2002-09-27 19:01:00

Edmund,

Thank you for your response. Let me give you a little more detail about the
problem I have. The thickness of the membrane is 0.5 mm and it has a heat flux
of 100 mW/cm^2. The plate on the back of the membrane is maintained at 10 C.
The helium on the back of the membrane does not flow in or out. The
pressure in
the back is constant and very low, almost at vacuum. Also, the spacing between
the plate on the back of the membrane and the membrane is 25 micrometers, so
the helium gas transfers heat from the membrane to the cooled plate through
collisions (bouncing between the membrane and the cooled plate). This happens
because the pressure is low, so the mean free path of the helium atoms is
large. The front of the membrane is facing vacuum and it is heated from that
side. In other words, the hot side of the membrane is up. We use this type of
cooling in our lab all the time, and the cooling is very efficient. However,
I'm not sure how to model this in ANSYS. Thanks for your response and your
comments and I hope I have cleared things out a little. If anyone else has any
suggestion I will be very happy to receive them. Thanks in advance.

Jose Torres
Department of Electrical and Computer Engineering
University of Houston
Houston, Texas

At 08:55 PM 9/27/2002 +0000, you wrote:
>Jose:

>You need to supply more info on the helium side. When you mean
>static helium gas, do mean mean that it is a chamber that has helium
>in it, with no mass flow in/out?

>If that is the case there are a number of things that can be going
>on. If the system is oriented such that the hot side is "up" then
>the helium could be stratified. If the cold side is "up" then the
>helium might have natural convection movement associated with it.
>In either case, the helium might not attribute much at all in the way
>of convective losses (since it sounds like it might be rarified) and
>radiation to the cold plate and conduction to the ring might be a
>greater factor. Applying a convection coefficient might not be
>physically real.

>If the above is your case, I would suggest running a radiation (if
>needed)/conduction model. Then add in a conduction through the
>helium (if the flow is stratified) and see if that even makes a
>difference. You might find that you can ignore the helium.

>But without knowing what you are actually using the helium for and
>what it is actually doing and the type of fluxes and cold plates
>involved, it would be hard for anyone to tell you better.

>Edmund Singer P.E.
>Senior Engineer
>Applied Mechanics Group
>Fluid and Thermal Sciences
>United Defense L.P.
>email: edmund.singer@u...

>--- In xansys@y..., "Jose L. Torres" wrote:
>> Hello everyone,

>> I'm trying to model cooling of a silicon membrane that is in vacuum
>and
>> that has a heat flux applied to it from one side. The side of the
>membrane
>> that has a heat flux applied to it, is in vacuum and the other side
>of the
>> membrane has static helium gas a low pressure (26 Pa or lower)
>transfering
>> heat to a cooled plate. The membrane is cooled mostly through
>conduction to
>> the ring it is attached to, but it also is cooled through
>convection to the
>> cooled plate at the back by the static helium. I'm having trouble
>figuring
>> out a convection ceofficient to apply to the back of the membrane,
>since
>> the convection coefficient that I need to apply must be as a
>function of
>> static pressure. However, I think the convection coefficient that
>ANSYS
>> expects is a function of flow and temperature. Does anyone have an
>idea how
>> I could obtain a convection coefficient of helium as a fuction of
>pressure.
>> If anyone has any ideas please let me know. Any suggestions will be
>greatly
>> appreciated.

>> Regards,

>> Jose Torres
>> Department of Electrical and Computer Engineering
>> University of Houston
>> Houston, Texas
Posts possibly associated with message #42265AuthorDateScore
42256THERM convection coefficient as a function of pressureJose Torres2002/09/27 
42262Re: THERM convection coefficient as a function of pressureEdmund Singer2002/09/27 
42265Re: THERM convection coefficient as a function of pressureJose Torres2002/09/27