Auber PID: Display rises 10F when compressor on, drops 10 off

[rocketman768]

I have the 1/32 PID from Auber. I am having an issue with the temperature reading that I want to see if anyone else has experienced before. I have it wired to directly drive the compressor since it only draws 1.3 amps through the onboard J1 relay. It seemed to work fine until yesterday. It took approximately 20 minutes to cool down from room temp to 45F, and would cycle on about every hour with a 5F hysteresis band.

Today, I see something strange. Whenever the compressor kicks on, the readout rises about 10F over the course of a minute or two. Then, it takes forever to get back down to the set temp (maybe a couple of hours). Whenever the compressor turns off, the readout drops 10F over a minute or two.

I should also mention that I got impossible readings of 106F at one point when I put the probe into a bit of water that I knew the temperature of.

I called Auber about this, and the guy I spoke to told me to try a shielded thermocouple, or basically to do anything I could to reduce electrical interference, which he thought was causing the problem (he sounded like he's heard this problem at least a few times before). I happen to have a shielded one, so I tried it and no difference (now reading 66F when it is actually 43F).

I have noticed that the tip of the probe is carrying a couple of volts of AC, and I don't know if that's normal or not. Since the PID is mounted in the door of the fridge, could there be condensation on the internal board that's conducting some voltage and screwing up the thermocouple voltage?

Anybody had this problem or know what's going on? Any suggestions?

 

[Hawgbranch]

I do not own one rocketman but I have read about them a lot here. I seem to remember another member having an issue like this. I believe it turned out to be a grounding problem. You can try looking through some of the RIMs build threads. Lots of great info.
Good luck!!

 

[klyph]

Ever thought of using an RTD?

 

[Hermit]

If you are worried about condensation, then try drying the unit out. You have tried a second probe so that rules that out? No? Out of the warranty period?

 

[rocketman768]

Alright, I may have found the proximate cause. I noticed when the compressor was on, the controller was getting very hot. Measured 120F with my IR thermometer. By shorting the thermocouple input, the internal temp reference for the thermocouple was 130F or so. I had a feeling this extreme reference temp or something else related to the heating was throwing off the display. I also have another pid that did not heat up, so I suspected this was the culprit.

This told me either I had a wiring problem or that there was some non-zero resistance between relay pins 4 and 5 that was dissipating heat. I also noticed that while the manual said that the polarity of power pins 1/2 did not matter, it suggested the neutral be hooked up to pin 1 (+) while I had it hooked to pin 2 (-). I rewired it and made sure all the connections were tight, and heating issue and fluctuation issue went away. Thank goodness; it was driving me crazy. I don't think the switching of pin 1/2 had any effect, because it doesn't make sense that it would, but rather I suspect that either pin 4 or pin 5 wasn't getting a good connection, causing those pins to heat up and screw up the internal temp reference.

 

[Hermit]

Glad you found your problem AND it didn't cost you any $.