On our first tests with Audine, we saw that the Kaf-400 sensor used to freeze when cooling the camera. So we took an inventory of all the solutions that could be used to avoid freezing:
- Create a vacuum; the most efficient mean for eliminating freezing!...but one of the most complex: it needs to modify the original housing of the camera, to buy a vacuum pump, etc...
- Dry the air inside the camera with silica gel for example; It seems not to be really efficient according to some 'specialists'; the disadvantages are a risk to introduce silica gel dust, the need to often open the camera to regenerate the silica gel,... But it's one of the most simple and cheap solution.
- Filling the camera with dry and neutral gas, at the atmospheric pressure. It's possible to buy very dry gas (99,...% of humidity); It can be nitrogen, or an other non-flammable dry gaz... However, buying nitrogen gaz cylinder and pressure reducer is often very expensive and not really profitable (a little less if you belong to an astronomy club).

So we decided to try to fill the camera with a dry gas in an aerosol can (usually used for cleaning photographic optics, lenses,... and cheap (about 50 Francs or 8 $US).
We had to fix two valves on the housing of the camera: we've used two little valves usually used in gasoline supply for model airplanes. We drilled a 4mm hole in the side and the back of the camera ans screwed the valves, with a rubber seal (in order to improve airtightness). We also applied silicone around the plugs. The valves cost me about 70-80Fr (10-12$). Recently, Claude Gragnolati such valves for 58Fr; He gave me its reference: "plateau gicleur pour OS 30, 3V FP, needle Valve assembly 22281903". 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

The filling of the camera is made with a plastic tube...Once the camera is sufficiently filled, close the valves and try to cool; fill it once again if it the CCD freezes...
It's not really easy to determine when the camera is filled...you'll have to test it; Nevertheless, I saw that the gaz is heavier than the air, so you'd better fill the camera with the admission valve under the exit valve (which is the contrary of the photo below!)
According to our tests, the camera is keeping dry during about one week (but it may certainly depend of the ambiant atmosphere...)
Thanks to this device, we can cool at -20°C (-4°F or 253°K) without any problems of freezing, and the temperature can certainly be lower... (recently we colled at -30°C).
 
 
 

An other view of the camera with the two valves

Contact: (pas de lien direct: protection anti-spam)