Home HVAC overcooling (artificial high sensed temp)

[2ndGearRubber]

Very ignorant of HVAC, so excuse my lack of terminology.

My home has intermittently become VERY cold, in the 60s, despite the thermostat being set to 72F as usual. Issue being the thermostat is displaying the sensed interior temp as being 75F, and is running the AC aggressively in an attempt to meet the 72 degree target.

Originally this was happening last month, and I changed the batteries in my White-Rodgers 1F86-244 thermostat. My guess was that the sensing element the thermostat uses to sense current interior temp was inside the thermostat, and perhaps old/weak batteries were contributing to drift in the reading.

The issue has since come up again, the thermostat being set to 72, and reading an interior home temp of 75 despite me being very cold. I removed the batteries again, and the system defaults back to 78F as the sensed temp when they are reinstalled. Then the system corrects downwards towards what is probably the correct temp within the home at 71F (don't have an IR thermometer handy). This occurs within less than 5min of removing the batteries so there's no chance the temp of the hose dropped 4 degrees.




Previously the batteries I removed were leaking slightly with minor corrosion on the terminals, which I cleaned. There seems to be no damage to the terminals and corrosion/acid did not appear to reach anywhere else. It appears the thermostat sensing element is drifting. Does this seem like a logical conclusion? I have not checked for voltage drops between the battery area and the circuit board, but expect to find none as the contacts are still bright an shiny. Thus it seems to me the correct option is to replace the thermostat assembly? It appears all of the temp sensing elements are contained within it, and if it has good batteries it's simply a matter of the unit having failed?

Thanks in advance.

 

[rooster59]

Is there a bigger than needed hole in the drywall behind the thermostat letting hot air into the back of your thermostat?

 

[chinboys]

Perhaps a faulty thermostat?
The batteries are for retaining the cooling or heating programing.
The 24 V AC transformer in your blower motor assembly provides the power on the Rh terminal that the thermostat switches (passes) to the cooling terminal (Y) and the fan (G) terminal during a cooling call.
Or the set back value is too much.
How is the actual temperature being determined?

Try a different thermostat.

 

[2ndGearRubber]

Is there a bigger than needed hole in the drywall behind the thermostat letting hot air into the back of your thermostat?

It appears the hole is 1/2" or smaller.

The thermostat is mounted on the main floor, HVAC unit in the basement. So even if I did have an air leak issue, the basement is typically a degree or two colder anyway. I don't think that section of wall on the main floor is directly connected to any section above, but that's just a guess.

Perhaps a faulty thermostat?
The batteries are for retaining the cooling or heating programing.
The 24 V AC transformer in your blower motor assembly provides the power on the Rh terminal that the thermostat switches (passes) to the cooling terminal (Y) and the fan (G) terminal during a cooling call.
Or the set back value is too much.
How is the actual temperature being determined?

Try a different thermostat.

Thank you for the lesson on the layout. I was unsure exactly how the transformer was involved.

The issue I have, I don't know how temp is determined. The thermostat is new enough to have a circuit board, and looking around inside the case I don't see a blatantly obvious external thermistor.

I have to ASSume the temp sensing is within the thermostat itself? Perhaps I will test this with a hair dry this afternoon. If the batteries are for powering the display and settings, perhaps I should confirm presence of a good power feed and ground from the transformer and accept that as enough to replace the thermostat.

 

[OccupantRJ]

Catch it doing wrong, then tap the thermostat with your fingers. If it stops immediately, the relay in the thermostat is likely sticking closed. On my thermostat, you can hear the relay click as it works.
it does sound like the temp sensing has gone nutty though.

 

[Aileron]

Getting any heat coming down the wall from the attic maybe?? Dont believe it would be a relay sticking since the sensed temp is enough to call for cooling. Is the sun shining on that stat?

 

[BillK]

and reading an interior home temp of 75

If the thermostat is reading 75 but the temperature in the room is actually less than that it sure sounds like the thermostat has failed. Somebody else mentioned the relay but if it was that it would not be showing the wrong temperature.
If nothing else has changed over the years then it sure sounds like a bad thermostat to me. I am betting the corrosion from the batteries did it in. They can give off corrosive fumes when they leak.

 

[fitter30]

Here's the manual. Page 4 is configure. Need to go through the menu write down all the settings then page 5 clear the program to factory. Re enter in configure settings. See what happens. Still having problem. Replace t stat.

 

[Wrench97]

Does your thermostat need the batteries to run or does it run on the 26v hvac transformer?
IE it has a wire in the C connection and the R connection, or if the batteries are removed the thermostat still works?

 

[2ndGearRubber]

Getting any heat coming down the wall from the attic maybe?? Dont believe it would be a relay sticking since the sensed temp is enough to call for cooling. Is the sun shining on that stat?

No sun on the unit, it's on a center wall near a staircase. This is occurring one time every 1-2 weeks, so I would ASSume that such an issue isn't a concern. I've lived here 6 years, same thermostat, just happened about 2 months ago the 1st time.

If the thermostat is reading 75 but the temperature in the room is actually less than that it sure sounds like the thermostat has failed. Somebody else mentioned the relay but if it was that it would not be showing the wrong temperature.
If nothing else has changed over the years then it sure sounds like a bad thermostat to me. I am betting the corrosion from the batteries did it in. They can give off corrosive fumes when they leak.

Ah, fumes. Didn't consider but yes that would be relevant. I have a new thermostat coming tomorrow via Amazon. I'll disassemble the old one and look for signs of corrosion. Corrosion would certainly skew a thermistor element if that's what's used to determine temp.

Does your thermostat need the batteries to run or does it run on the 26v hvac transformer?
IE it has a wire in the C connection and the R connection, or if the batteries are removed the thermostat still works?

If the batteries are removed, it lingers for maybe 15 seconds, then dies and the screen goes blank and the unit unresponsive. I would ASSume the capacitor is bleeding out.

 

[BrandonV]

This is occurring one time every 1-2 weeks

So this isn't a continuous issue? I assumed it was happening all the time... I had that happen once with a Honeywell thermostat. It would just "stick" at a temperature regardless of what was happening.

Time for a new thermostat.

 

[2ndGearRubber]

So this isn't a continuous issue? I assumed it was happening all the time... I had that happen once with a Honeywell thermostat. It would just "stick" at a temperature regardless of what was happening.

Time for a new thermostat.

No, intermittently the temp drifts severely. Resetting power the the thermostat corrects it, for a time.

 

[BrandonV]

No, intermittently the temp drifts severely. Resetting power the the thermostat corrects it, for a time.

Odd. It sounds like you took the thermostat apart down to the printed circuit board. You might want to give it a good cleaning with IPA. Might be some residue on there from the battery leak which isn't visually apparent.

 

[Wrench97]

Odd. It sounds like you took the thermostat apart down to the printed circuit board. You might want to give it a good cleaning with IPA. Might be some residue on there from the battery leak which isn't visually apparent.

I usually drink a good IPA didn't know it was a cleaning solvent.

 

[Wrench97]

No, intermittently the temp drifts severely. Resetting power the the thermostat corrects it, for a time.

When you have the thermostat off check to see how many wires you have in the bundle if you could power on the red and common on the C it'll save you from have to change bats as often and reprogram(the bats will just be used during power outages).

 

[BillK]

Go to Walmart and buy one of these:

Robot or human?

Sit it right on top of your White Rogers thermostat. When the problem happens look at the two thermometers. The one on the thermostat should be the same (or very close) to the Acurite. If not you know for sure that the Thermostat is bad.

That being said, to me it is obvious your thermostat is messing up. I would have already been to Home Depot, bought a new one and had it installed

 

[2ndGearRubber]

Odd. It sounds like you took the thermostat apart down to the printed circuit board. You might want to give it a good cleaning with IPA. Might be some residue on there from the battery leak which isn't visually apparent.

I haven't fully disassembled yet, but I will.today when the new unit arrives.

I don't like guessing. I'm bad at guessing. So I always try to get to the root of a problem. If I have to guess that's what it is, but it's not my preference. If the new thermostat does not correct the issue, I'll reinstall the old one and use an oscilloscope to figure out how it actually works, then troubleshoot.

Removing the batteries from the unit seems to kill it's function entirely, and the HVAC won't run, or at least I haven't heard it try to run with the thermostat INOP.

Go to Walmart and buy one of these:

Robot or human?

Sit it right on top of your White Rogers thermostat. When the problem happens look at the two thermometers. The one on the thermostat should be the same (or very close) to the Acurite. If not you know for sure that the Thermostat is bad.

That being said, to me it is obvious your thermostat is messing up. I would have already been to Home Depot, bought a new one and had it installed

Amazon delivers today.

 

[BrandonV]

I haven't fully disassembled yet, but I will.today when the new unit arrives.

I don't like guessing. I'm bad at guessing. So I always try to get to the root of a problem. If I have to guess that's what it is, but it's not my preference. If the new thermostat does not correct the issue, I'll reinstall the old one and use an oscilloscope to figure out how it actually works, then troubleshoot.

Removing the batteries from the unit seems to kill it's function entirely, and the HVAC won't run, or at least I haven't heard it try to run with the thermostat INOP.



Amazon delivers today.

Yup. Around here, the C wire is NOT connected on most new installations for some reason to the thermostat. The installers leave you at the mercy of some batteries.

Thermistors rarely drift, in my opinion. When they fail, they tend to fail completely. I suspect that if you trace the circuit between the thermistor and the microprocessor, you'll find the issue. It's probably related to reference voltage problems or irregular resistance somewhere in between.

It's too much of a coincidence for the issue to be caused by the leaking batteries.

By the way, were the leaking batteries Duracells? I had them damage a Honeywell thermostat. After I took pictures of the damage with the batteries, Duracell sent me a check without any questions. I'm not sure if their policy has changed, but they paid for a brand new thermostat. It's a good thing to keep in mind if alkalines damage something again.

 

[dogdog]

First step would be getting an external thermometer and place it around where the thermostat sits and verify if it is actually the same temp as the thermostat reading?

Maybe airflow issue ?

 

[2ndGearRubber]

Yup. Around here, the C wire is NOT connected on most new installations for some reason to the thermostat. The installers leave you at the mercy of some batteries.

Thermistors rarely drift, in my opinion. When they fail, they tend to fail completely. I suspect that if you trace the circuit between the thermistor and the microprocessor, you'll find the issue. It's probably related to reference voltage problems or irregular resistance somewhere in between.

It's too much of a coincidence for the issue to be caused by the leaking batteries.

By the way, were the leaking batteries Duracells? I had them damage a Honeywell thermostat. After I took pictures of the damage with the batteries, Duracell sent me a check without any questions. I'm not sure if their policy has changed, but they paid for a brand new thermostat. It's a good thing to keep in mind if alkalines damage something again.

I don't remember.... they looked pretty old. Already in the trash. My expectation would be board damage, not the thermistor itself. Of course, I ain't an EE, so unless it's physically screwed up I don't think I'll bother to pull out my board probes and the scope...... maybe I will.


EDIT: I buy energizer lithium for my torque wrenches. I think I will either start using those on the new thermostat, or simply change them twice a year like my smoke alarms.

 

[BrandonV]

EDIT: I buy energizer lithium for my torque wrenches. I think I will either start using those on the new thermostat, or simply change them twice a year like my smoke alarms.

Alkaline batteries have no place in my home. I either use the Energizer Lithium batteries you get in a bulk pack at Sam's Club or I use quality NiMH rechargeables with a good microprocessor controlled charger. Neither leak unless they've been woefully abused.

I had a $10k piece of test equipment ruined from a Kirkland Signature/Duracell battery. Never again.

 

[2ndGearRubber]

Issue is fixed. I identified the thermistor element on the board, and it indeed had damage on the trace from the leaking batteries.

Of course, I did this after installing the new $45 thermostat.