Date

2008/06/12 11:35:42

Contents

1. File Report
    Table 1  File Information for module flow 2d round6_001
2. Mesh Report
    Table 2  Mesh Information for module flow 2d round6_001
3. Physics Report
    Table 3  Domain Physics for module flow 2d round6_001
    Table 4  Boundary Physics for module flow 2d round6_001
4. User Data
    Figure 1
    Figure 2
    Figure 3
    Figure 4
    Figure 5
5. Summary and Conclusions

1. File Report

2. Mesh Report

3. Physics Report

4. User Data

Figure 1.

Figure 2.

Figure 3.

Figure 4.

Figure 5.

5. Summary and Conclusions

This air cooling analysis includes both air flow and heat transfer. This is the first of the module structures to be run in high resolution mode. It is still more limited than previous runs in that the structure is shortened still more to fit memory constraints.

Figures:

1. Temperature contour at mid plane

2. Silicon surface temperature

3. Stream lines with color indicating velocity

4. Contour of flow velocities at mid plane

5. Contour of flow velocities at three planes along the module

This shows only a 3.4 deg rise in the silicon temperature over ambient. Significant cooling of the silicon is a result of the carbon support tube which conducts heat away from the silicon adding significant area for transferring heat to the cooling air.

Some input boundary conditions:

Thermal power input is generated in the inner silicon ladder at the center (100 mW/cm^2)

Air inlet velocity 8 meters/sec

Outlet boundary condition: open, relative pressure 0

In order to work with the FEE approach very thin structures have been modeled as much thicker structures. The thermal conductivity parameter has been adjusted accordingly to give the correct conductivity along the plane, but then too low conduction perpendicular to the plane.

Concerns about current version of the model:

Too short to show velocity development from initial fixed velocity boundary condition at the entrance.

Too short so that air does not see full power input.

Only one silicon ladder producing heat, so again not getting the full heating of the air, but this may not be too big an error. Consider the full volume of air at this velocity is 284 ft^3/min and the power from the silicon is 160 watts - this gives only a 1 deg C temperature rise in the air.

This low value for air heating suggests that design improvements may be possible that require less total air flow volume. The design challenge will be to maintain good surface air velocities with reduced total flow.

This work must be checked with more careful analysis and prototype testing, but if these low temperature rises turn out to be correct then this means that it should be possible to run the system at ambient temperature which is a much simpler task than operating below ambient where significant work is required to prevent accidental condensation. It also avoids the need for extensive thermal insulation of the system.