Solar thermistor not working after replacing twice

[Ryano2121]

Solar hasn't been working consistently. Solar temp in the ET was erratic going from a very low temp like 9* to a very hot temp like 255*.

Switched the wires for the air and solar on the ET and the odd temperature followed to the solar thermistor, leading me to believe the issue is the thermistor rather than the circuit board.

Bought and installed 2 new sensors. The first one read wrong immediately so I returned it. The second one read correctly for a week and today it's back to being erratic.

The thermistor itself is not touching anything on the roof, I have it suspended in the air.

What is the next step?

 

[needsajet]

Is the thermistor protected from sun with a coating, or inside a bit of small black tubing? That can help them last longer. They do wear out pretty fast, but you should get a few years out of it. If totally protected they live the longest.

 

[Ryano2121]

It's not protected, I can do that. But this one lasted one week. This makes me think something else is causing the big swings in resistance. Could it be the wire? I have a 2 story house so there is a long wire going to the roof which the thermistor is connected to.

 

[needsajet]

If you have a multi-meter, you can short the ends of the wire at the thermistor end by twisting them together, then measure resistance (ohms) at the controller end which should be zero, or at least tiny tiny practically zero.

 

[needsajet]

By the way, a very high temperature means a short circuit. When they fail, they usually read a gradually higher (incorrect) temperature.

 

[Ryano2121]

I have a meter. With the 2 thermistor wires twisted together, where am I putting the second probe?

 

[Pool Clown]

Are the sensors you purchased OEM?

You bought two, one failed immediately. The other followed shortly thereafter. Maybe a bad batch? Perhaps return the other, and start over with a different vendor?

 

[dschlic1]

The insulation on the wire between the sensor and your controller might be bad.

 

[Bama Rambler]

Also keep in mind that we're talking resistance here, so even a small amount of corrosion in a connection or terminal block can lead to erratic readings. I'd clean any connection even on the board well and see what happens then.

 

[needsajet]

Cleaning everything up is always good for sure (we get insects making nests inside boxes down here), and I've seen bad batches in the past in the greenhouse industry where we used hundreds of thermistors, so I should have thought of that too!

The two probes on your meter go on the two wires at the controller, after disconnecting from the controller. You're measuring to see if there's a short in the wire, so the meter is sending electricity out to the twisted connection on one wire, and back to the meter on the other. You'll have to check it often through different conditions for a day or two, such as when it's cold and also when warm/hot because it may be intermittent by the sounds of it. But if it's zero (or practically zero) resistance, and a consistent reading, then it's not the wire.

 

[Pool Clown]

Removed.