Just a real quick post on the temperature stuff before I head to work (will do more detailed testing later).  The temperature controls seem to work as expected with one slight oddity.  At least in my initial testing it appears that the fan speed is exactly inverse of what you would expect.  The fan seems to idle down when the temperature is dropping (i.e. the peltier cooler is engaged at full power) and the fan speeds up to full power when the cooler is turned off.

I think there were some other control channel id numbers that you thought were related to the temp controls.  I am wondering if the peltier power and the fan speed are set as two separate things.  If that is the case I also wonder if there isn't a second temperature sensor on the "hot" side of the peltier in addition to the one on the ccd which you are already reading.  If that is the case it might be possible to run the peltier and still keep the fan speed fairly low until the peltier itself starts heating up at which point the fan can be sped up as needed.  Of course all of this is just speculation.  As mentioned I hope to do some more detailed work on the temp control testing, but wanted to give you these initial results right away in case you had any ideas, etc.

The other thing for future work as a "nice to have" is that right now, the cooler seems to flip between 100% power and 0% power.  A better way would be setting the power based on how far away we are above the desired temp.  A formula something like

10% + 10%*x

where x is the temp (in C) that the ccd is above the setpoint.  Basically if its above it runs at at least 10 percent power, but ramps up the farther away it is.  This way as it comes down to temperature the cooler idles back so you don't "overcool" the sensor.  That should make the temp much more stable instead of "hunting" around for the correct value.

Read More...