Fighting condensation in thermowell

I've got a thermowell with a dallas 1-wire probe in it. It is submerged in liquid that I keep potentially very cold (so far, as low as 27 degrees -- it's a mixture of glycol + water).

This morning I woke up to find that my system was failing to read from the 1-wire bus. Eventually I discovered that this was due to the buildup of condensation that had collected in the aforementioned thermowell.

This temperature reading is a pretty critical factor in my design. I need to come up with a solution to combat the condensation. I have a couple thoughts:

1) use some kind of epoxy / glue / cement at the bottom of the thermowell. This would prevent any of the electrical contacts from being exposed to air / water, which could also prevent oxidation and so on. The substance I use may also aid in heat transfer, which would give more accurate readings.

2) ensure that the chamber is air tight, and fill it with a gas with low moisture content. If this is really a viable idea, any thoughts on how I might do this? Would the CO2 I use for my kegs be appropriate? The only real advantage I see here is that I would be able to replace the probe if it ever needed to be done.

 

Just seal the opening of the thermowell with some silicone caulk so you can remove it later.

 

Perhaps I'm overcomplicating this a bit. I'm sure the amount of moisture in that volume of air is fairly negligible, even here in Texas

Hmmm, I am going to Colorado soon, I should take them up there and seal them

Thanks for the tip. I imagine that is perfectly adequate.