Maybe a Gas leak?

[IslandLizard]

Modern day scales use strain gauges, and thus need temperature corrections. So a local temp sensor must be included too.

Relatively small deviations are to be expected, and (internal) calculations may be rounded on the display. The order of .33 oz (~9.36 grams) precision in the readout sounds a bit high, but may be reasonable given the measured object is in the 50# range, while looking at deltas (drafts) from an oz to a pint or more.

 

[Michele Craft]

Modern day scales use strain gauges, and thus need temperature corrections. So a local temp sensor must be included too.

Relatively small deviations are to be expected, and (internal) calculations may be rounded on the display. The order of .33 oz (~9.36 grams) precision in the readout sounds a bit high, but may be reasonable given the measured object is in the 50# range, while looking at deltas (drafts) from an oz to a pint or more.

ok so it sounds like the weight readings I'm seeing that CLEARLY change with temp do make sense and I should just expect them. I think given that leak problem above, I might also be over thinking things and can just except that the up and down will happen.

Now do I worry about the freezer cycling on and off about every 3 hours? Is that more often than should be expected? And is air vs liquid expected to be THAT different in temp (about 20F)? I have the collar siliconed well and have pink foam board, plus an RV seal on the top of the collar at the lid. No real sweating inside and my EVA dry has been in there for weeks without having to be recharged.

 

[IslandLizard]

And is air vs liquid expected to be THAT different in temp (about 20F)?

Air is the carrier of the cold (negative heat). It takes a while for the jar with water to chill down. So yeah, there is a sizeable delay factor.

Now do I worry about the freezer cycling on and off about every 3 hours?

Not too crazy short. My upright keezer comes on like that, or even more frequently.
How long is the on-cycle?

Are you chilling (recirc fan) your tap tower all the time?

 

[Michele Craft]

Air is the carrier of the cold (negative heat). It takes a while for the jar with water to chill down. So yeah, there is a sizeable delay factor.

Just so I'm clear, this isn't just the original setup with the chill down of the water jar but now the chest has been closed for 24 hours straight and we see the scale reading about 20F more than the water bottle consistently. Scale temp reading goes up about 20, water goes up about 2, triggers freezer to run but scale never gets down to the 38 - 40 range that the water bottle does. Air goes from 48 - 58F based on graph where inkbird goes from 37 - 40. Honestly, I think I might trust the inkbird readings more but expected the air reading to be somewhat closer to water reading when things are closed up.

Not too crazy short. My upright keezer comes on like that, or even more frequently.
How long is the on-cycle?

That's good to know, I didn't know what to expect going into this. As for the length of cycle, I'm not sure since I don't have inkbird wifi but I can now anticipate when it will kick on based on plaato temp so I can keep an eye out tonite for a cycle time.

Are you chilling (recirc fan) your tap tower all the time?

Nope. I bought a small fan to put on the collar but I've found I don't need it. I haven't temped the pour since we moved the unit to the basement (its final home will be there with a full bar around it) but when it was in the non air conditioned garage with temps way warmer (75'ish) than in the basement (65ish), I was pouring at no more than 44, sometimes 42, with the inkbird set at 38 and a 2deg differential so I didn't feel the fan was needed. To compensate, when we built the tower using 2" black iron pipe (2 legs up to elbows, bunch of Ts for the 4 taps), I put 1" copper pipe up into the legs of the tower on both sides and extended them about 10" down into the chest freezer and then put pool noodle looking pipe insultation up in the legs cut flush at the floor flange on the top of the lid and then 2 lines to the taps up each side through the cooper pipe. The theory was the copper would act as a heat sink and given my pour temps, I think it's working as planned (or it's just dumb luck )

 

[Gozie Boy]

It can yes, and I'm considering it BUT, our kegerator is being built into a bar so the cord on it didn't look long enough to sit up on a counter and I don't know if I WANT something sitting on the counter, that's the debate right now. Also, I don't have the clearance I don't believe because of my heat sink copper pipping that comes down from the tower, it'd interfere with the gauges IF they'd even clear the lid jacked up 3"

I also needed a longer cable between the scale and readout unit. Here is what you need:

https://www.ebay.com/itm/283613838681
This is a slim-style 4-conductor phone line. Make sure your get the 'straight' connector version, and not the reversed one. They are cheap and available in whatever length you need. Of course you can also get a longer USB power cable if you need one.

This scale stays on for as long as you keep the power adapter plugged in ... and you don't get a power outage. In the latter case, you will need to get someone to re-tare the scale as you lift your tank or keg off of it. Sadly, my power has a lot of glitches, hopefully yours is better!

 

[Beholder]

Just so I'm clear, this isn't just the original setup with the chill down of the water jar but now the chest has been closed for 24 hours straight and we see the scale reading about 20F more than the water bottle consistently. Scale temp reading goes up about 20, water goes up about 2, triggers freezer to run but scale never gets down to the 38 - 40 range that the water bottle does. Air goes from 48 - 58F based on graph where inkbird goes from 37 - 40. Honestly, I think I might trust the inkbird readings more but expected the air reading to be somewhat closer to water reading when things are closed up.



That's good to know, I didn't know what to expect going into this. As for the length of cycle, I'm not sure since I don't have inkbird wifi but I can now anticipate when it will kick on based on plaato temp so I can keep an eye out tonite for a cycle time.


Nope. I bought a small fan to put on the collar but I've found I don't need it. I haven't temped the pour since we moved the unit to the basement (its final home will be there with a full bar around it) but when it was in the non air conditioned garage with temps way warmer (75'ish) than in the basement (65ish), I was pouring at no more than 44, sometimes 42, with the inkbird set at 38 and a 2deg differential so I didn't feel the fan was needed. To compensate, when we built the tower using 2" black iron pipe (2 legs up to elbows, bunch of Ts for the 4 taps), I put 1" copper pipe up into the legs of the tower on both sides and extended them about 10" down into the chest freezer and then put pool noodle looking pipe insultation up in the legs cut flush at the floor flange on the top of the lid and then 2 lines to the taps up each side through the cooper pipe. The theory was the copper would act as a heat sink and given my pour temps, I think it's working as planned (or it's just dumb luck )

You may want to put in a fan anyhow, as the temperature stratifies without any movement of the air. The coldest sits at the bottom where the beer is drawn from anyhow, but it will flow through lines going to the top. This would explain why your air temps never drop to your measured liquid temps.

If you get good pour temps without excessive foaming, then RDWHAHB, but a fan could help with final tweaks to dial
it in.