Good Morning!

I'll try to explain my scenario the best I can and provide as many details as possible.

I'm trying to model an air insulated enclosure cooled by a Peltier (TEC - Thermoelectric) Cooler.  I have a hermetically sealed Kovar package containing some glass components that need to maintain a setpoint of 300K.  I've mounted this package to an aluminum shield since aluminum does not support a thermal gradient.  The peltier cooler is mounted to the aluminum above as well as a heat sink below.  This aluminum assembly is then mounted in a plastic shield so I have a thermally isolating air barrier. 

I'm attempting to size the peltier cooler in a worst case scenario with the ambient temperature at 325K.  One of my variables is the width of the air gap between the plastic shield and the aluminum shield.  Increasing this gap should improve the efficiency of my system allowing me to use a smaller peltier cooler.  However, when I run the analysis, regardless of the air gap width, my results are the same.

I've tried including a physical model of the air between the aluminum and plastic shields, but I'm not confident in the results, or if this is a reasonable method.

Here is a list of my thermal loads:

Initial Temperature on all surfaces = 300K

Heat Power of the Peltier = -0.5W (this is a variable)

Convection = 25W/m2K (natural convection) w/ Bulk Ambient Temperature = 325K

Any help would be great!

Thanks in advance.

-Jason

P.S. In the attached picture, the tops of surfaces have been hidden for clarity and all dimensions are in mm.