O-ring Permeation Model for Enclosure with Elastomeric Seals

That is what we do with some of our products. We fill with nitrogen boiled off from a liquid nitrogen dewar to ensure that it is dry but cylinder nitrogen should be OK. This is equipment designed to purge & fill units with dry nitrogen, no doubt there are other makes around.

Continuing N2 thoughts from comment above... Thanks for this thought line SE

If N2 is used (as suggested) and pressurized inside (not sure how to control the pressure in any controllable way especially with such a small volume), then it too will permeate through the o-ring over time. Granted permeation thru Viton is wonderful (one reason why semiconductor vacuum equipment guys use it), but I will need to take a look to see how long it will take to leak down. My guess is that an elastomer seal in a glass-filled polypropylene housing will leak down faster than the life expectancy of the product (5 years). Maybe it's not driven by the leak rate through the raw materials since N2 is such a large molecule, but more through the sealing interfaces, which will be harder to control. Maybe this will simply buy more time... say cutting the time down so that desiccants can last longer.

This is not a hermetically sealed product. I think it could be pressurized, just not sure how without assembling in a pressure vessel with automation or adding a "valve" feature that could be used for charging. This product is about the size of a computer mouse.

I am inquiring with three desiccant companies since they should have the right expertise in this area. Electronic packaging info from these guys is not readily available as opposed to desiccants and packaging containers.

The seal could be designed in such a way that the pressure increases the force on the seal adding to the integrity...the N2 could be injected via needle through a rubber nib and sealed on removal of charging needle...A charging cylinder with the measured volume of N2 could be used for precise volume control...

..."Nitrile and natural rubber gasket materials are good choices for gaseous nitrogen, although EPDM is generally considered best. EPDM gasket material has good heat resistance and a working temperature range of around -60 to 320°F (-50 to 160C). Resistance to oxidation, UV, ozone, and abrasion are all good and it also withstands water, acids, and alkalies along with ketones and alcohols."...

https://www.henniggasket.com/gasket-answers/nitrile-gaskets-2/best-gasket-material-nitrogen/

SE, interesting information. Thank you. I will investigate further. I have received first contact from a desiccant company and will post the what I learn as it comes in. I considered gaskets vs. o-rings, but the packaging prevent the use of a gasket for the sensor seal. I could have possibly used a gasket (longer permeation path) for the other seal, but the benefit didn't outweigh the design challenges due to the increase in housing size.

You could seal the enclosure in a cool, dry atmosphere below the lowest temperature the device is likely to see in the field. That way the unit will be under positive pressure at all times so no worries about moisture ingress.

Thanks for the info GM1964. I thought about what you suggested. I have chosen operating min T at -5C, a bit below freezing which could be seen at ground level in an agriculture setting (e.g. San Joaquin Valley). What gave me pause was not this temperature since you could back fill with N2 (as other have suggested). It was the typical non-operating or storage temperature most products get tested for -20C to -40C for things like solar panels, cars, etc. I think my fingers would fall off at one assembly... Will continue my thoughts on N2 in the other message...

I agree with SE,pressurization is the way to go to prevent ingress of outside atmosphere.Plenty of materials are resistant to nitrogen. The link below(cut and paste) has a formula for calculating permeability with varying differential pressures:=====================================================================================================https://www.marcorubber.com/o-ring-permeation.htm=================================================================

It remains a mystery as to why a proprietary IP68 enclosure and fittings weren't selected in the first place, thereby obviating the <...enclosure...designed...> part?

Cost, size and similarity to interface hardware... are the issues with OTS enclosures and fittings for this application. The custom polypropylene (PP) enclosure houses a sensor. The typical enclosure for this type of sensor can be found in industrial settings where neither cost nor size are of much concern. The industrial solution is stainless steel everything or at a minimum brass - granted easier to seal since metal seals could be used. But if elastomers where used, the same issue would exist. The PP enclosure costs $2 in qty of 1000. Good luck getting even to the same magnitude with stainless.

Also, I would love to use a Gore vent or similar vent from Donaldson, unfortunately, the product lies in the mud at times and can even be submerged... Thus, a vent was a no go from the beginning. Potting has always been the backup. But, I would like to avoid the cost and complexity.