Problems due to extremely high humidity

[FlaMike]

It seems that around this time of year I experience a Nixie challenge that I believe reults from very high humidity & constant fog: my Nixie freaks out probably from condensation forming on the Pi/HATs/connectors. Several times in the past couple of weeks, the system toggled rapidly & repeatedly into & out of Freeze Protection Mode in the middle of the night when the system should have been off. This has happened in previous years, too.

Last night the "real" air temperature was in the 50s but the temperature sensor on the Nixie flipped between 0 and 57 degrees, so the system oscillated rapidly between turning the pump off (as scheduled) and starting it for Freeze Protection. Interestingly, the pool lights also turned on & I could not turn them off. Because the system was out of control and it was 2:00 AM, I flipped all of the breakers to shut it down. Other sensors, like flow, pressure, & water temperature were a mess as well.

This morning the sun popped out following 100% relative humidity overnight so I opened the door to the case where Nixie lives until it was warm & dry. I turned on the breakers & all is well... for the moment. I'll probably kill the breakers again tonight.

I assume condensation on the boards/connectors is the cause of this behavior.

I'm thinking at least part of the solution is to protect Nixie with a silicone conforming coating. To do so, I'll need to disassemble the Pi stack, clean & dry the PCBs with 100% isopropyl alcohol carefully applied, protecting the off-board connectors with Kapton tape, & then applying a few thin coats of aerosol silicone conforming protectant, & reassembling the stack once cured.

Has anyone run into this problem & used conforming protection? Did it work? Do you have a different solution that worked for you?

Thanks in advance for your thoughts & suggestions!

[tagyoureit]

That sounds like a good plan. Maybe a 5W heater strip that turns on with appropriate environmental conditions, as well? The conformal coating will only protect what's exposed but probably not the headers and other interior connections.

[FlaMike]

I've played around with the building automation hat. FIrst I moved the three input sensors (air & water temperature and flow switch) from inputs 1, 2, & 3 to 5, 6, & 7 on the hat & reconfigured REM. No love--after briefly reporting the correct temperatures and flow status, all 3 inputs went to zero & no flow respectively. The temperatures then oscillated between reasonable readings 70ish for the air & 60ish for the water, & then all went back to zero & stayed there.

I had a spare building automation hat, so I took the Pi stack apart, reflashed the Pi & pool controller software from a backup, & tried again. I thought I was in fat city... Everything looked reasonable & stable... for about 5 minutes, then BAM! Back to zero readings & into freeze protect it went & stayed there. Ugh!

When I was taking the stack apart & reassembling it there was one fascinating (I think) observation. All of the Sequent boards (fan, building automation, & 8 relay) looked almost factory fresh, after about 3 years of operation. The Pi, however, looked like a 10 year old Chevy Vega--if you're "senior" enough to remember them. Or a Yugo, if that helps. Much of the expopsed metal on the board was somewhat corroded, though no swiss cheese holes. I can't help but wonder if the Pi, though generally appearing to be functional, just isn't up to the task anymore, especially for sensor inputs via the automation hat.

Does this make sense? I don't know if Sequent protects its boards with a very light conforming coat which the Pi doesn't appear to have. Could the sensor mess be due to Pi issues & not the automation hat or stacking issues?

I've ordered a replacement Pi that should get here in a couple of days & will replace the "old beast." I just don't know what else to try at this point. All of the GPIO pins in the stack looked fine, just the Pi appears to be weathered.

Any thoughts & suggestions are welcome! If the new Pi appears to be the trick, I'll give it a light coat of silicone conforming spray & hope it gives some additional life to the Nixie.

Does anyone else in the deep south have similar experiences?? I guess the Florida Panhandle may be too moist for a long life, at least for Nixie:-(

Thanks for your thoughts & help!!!

[FlaMike]

I was wondering about getting a small heater... Do you have any recommendations? Thanks!

[tagyoureit]

I haven't dug that far into it.

[FlaMike]

OK, I was just wondering where to start looking--a bit beyond anything I've ever purchased or had experience with.  Also wondering how to trigger such a beast: humidity only, temperature and humidity, something else, what thresholds, etc. All 3 megabase inputs are messed up even though the board is dry: air temp, water temp, & flow switch.  I'm going to move them to different inputs on the megabas & hope the next terminal block & the associated circuitry is ok.  If all else fails, I have a spare megabase to get the pool up & running.  If I take that route, I'll disassemble the Pi stack & spray everything. Thanks for your help & insight!     Mike

…

On 3/4/26 10:15 AM, tagyoureit wrote: I haven't dug that far into it. — Reply to this email directly, view it on GitHub <#1155 (reply in thread)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AJ3NIBDK2Q3EJHDU4FXN34D4PBJBZAVCNFSM6AAAAACWCXFYRWVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTKOJZHE3TONI>. You are receiving this because you authored the thread.Message ID: ***@***.***>