SoundGuy, Well in general thermal grease or potting to fill the intersticial spaces in the interface is a good idea but in this case probably not waranted. The insulation over the top of the sensor and covering the surounding area creates a volume within which the temp is pretty homogenous. If the temp to be measured was varying quickly and it was important to track these rapid changes and not have the "experimental lashup" time integrate the quick time varying deltas, then it would be time for the grease.
Potting isn't such a bad idea though but I would probably opt for spray foam insulation. I would probably do something like:
1. afix the sensor to the filter body with as small a piece of tape which is practical (over the unit not double stick or anything between the sensor and good intimate contact with the filter body)
2. spray the filter body a couple inches around the sensor with a mold release compound (can substitute silicone spray lube or similar) It is important to do step one first as the tape won't stick after applying mold release.
3. spray a little "low expansion" spray insulating/sealing foam over and around the sensor and let it cure
4. after it cured I would secure it to the filter body with a couple wraps of duct tape. I first thought of a tie wrap or hose clamp but that would lead to the premature demise of the cured foam because I am not too gentle with things.
This will insulate the sensor from the macro-environment while insuring intimate contact with the micro-environment to be measured. When it is time to change the filter, you cut the tape and remove the sensor along with its insulating foam encapsulation with a gentle twist. It is form fitted to the filter and may be reused over and over by replacing the duct tape.
Of course, heat transfer grease may be used if desired. My off-the-top-of-my-head guess is that it won't make a couple degrees of difference which would be insignificant in this application.
In moderate climates you could probably hold the sensor to the filter with a hose clamp and cover the area near the sensor with a really thick glob of RTV silicone ruber) This too would be reuseable but in a really cold winter would have more error as its is not as good an insulator.
Historical note: Most of us probably recall hearing the term... RTV and most of us know that RTV is silicone rubber but how many of us know what RTV stands for?
RTV is Room Temperature Vulcanizing. Take notes, there will be a quiz!
Patrick
