Intermittent Furnace Issue

Our furnace has recently begun intermittently failing to start. As best I can tell, it seems to more specifically be that the flow inducer motor is intermittently failing to start.

It starts up fine 95% of the time, maybe getting closer to 90% of the time now. Obviously the motor isn't seized, and says not to oil so apparently has sealed bearings. I would have thought the capacitor for starting the motor might be going bad, but this motor doesn't have one.

Rotating the motor, wiggling the wires, and staring disapprovingly at it does nothing. Manually adjusting the thermostat down below room temp and back up to desired temp causes it to fire right up. Cycling the furnace power switch causes it to fire right up. Disconnecting the motor from the control board and reconnecting it cause it to fire right up. 

None of the DIY furnace troubleshooting guides I've found seem applicable to the situation. What causes this type of issue, and how do I check for it?

Can't really tell you how to diagnose it further, but if you want to play parts hanger, the inducer can probably be bought for under $150.

Mine had a bad bearing which made it obvious where the problem was.

In case svrex or Curtis don't know the answer, I'd head over to https://www.reddit.com/r/hvacadvice/. 

Driven5 said:

Manually adjusting the thermostat down below room temp and back up to desired temp causes it to fire right up.

This part has me a bit perplexed as the call for heat from the thermostat should be no different when you manually cycle the thermostat vs when it auto cycles other than maybe the interval at which you are cycling it.  

If the furnace hasn't cycled for a while and the temps have fallen in the house below the thermostat kick-on, can you get the furnace to cycle successfully just by turning the thermostat down and back up or do you need to do something to the furnace?

In reply to Driven5 :

It might not be the draft inducer.  Cycling power may be resetting a safety, look at the board if it happens again and see if there’s a flashing LED (typically red).  If so, count how many times it flashes and then look at the wiring diagram to see if the fault codes are listed by flash count somewhere.

The only flaw in my theory is the part where you unplug the motor from the board and plug it back in / starts.   Then again if there’s a current sensor or flow switch built into the motor to prove it’s running before allowing the ignition sequence that could be reset by unplugging the motor too.  

I had a similar weird problem with my 20-year-old Rheem 93% efficient furnace a couple years ago.

I 2nd 11GTCS's note about looking for flashing lights and reading the manual.  My manual also had a flow chart of what happens at startup, so I could follow along the sequence of what it was doing.  You might have to tape the switch that turns everything off when you take the access panel off so that you can *see* the flashing light.

TLDR: I never did figure out exactly what the issue was, but it had something to do with a pressure switch that was supposed to tell the controller that the draft inducer fan was making enough of a draft.  If I bypassed the switch while it was starting, it would start (and would continue to run even if I put it back in the circuit).  The controller was too smart to leave it bypassed, it expected to see it change state from not-enough-draft to enough-draft.

TL: it all started when the draft inducer fan stopped turning because it was 20 years old and sticky.  I got it working again by taking the fan out and spinning by hand.  I ordered a new inducer fan, and when it came in I installed it.  After a day or two, the furnace wouldn't start at all with the above symptoms.  On a whim, I put the old fan back in and it worked.

I tried a new pressure switch, I checked that the exhaust pipes weren't blocked, I even tried a new controller board.  There was something about the new fan that it didn't like, just at startup.

We had that same issue,  call the power company to come and test the meter/regulator on the outside of your house.

Ours was failing and they replaced it with a new one. There was no charge.

We had similar symptoms on our furnace, our HVAC company replaced a bad zozzle and our furnace now works as it should.

Our Trane furnace needs a little cleaning at the burner tips/nozzles about every three or four years.  There is a wee slit that the fire travels between the burners to light them all.  Eventually one gets corroded, the flame doesn't pass through, the next burner doesn't light, and the sensor doesn't see flame, so it shuts the gas off.

Just cleaned the crud out recently....

SkinnyG (Forum Supporter) said:

Our Trane furnace needs a little cleaning at the burner tips/nozzles about every three or four years.  There is a wee slit that the fire travels between the burners to light them all.  Eventually one gets corroded, the flame doesn't pass through, the next burner doesn't light, and the sensor doesn't see flame, so it shuts the gas off.

Just cleaned the crud out recently....

Had something similar happen to our old heater and water heater as well. Quite common if you are on propane vs natural gas.

mtn said:

In case svrex or Curtis don't know the answer, I'd head over to https://www.reddit.com/r/hvacadvice/. 

Sorry. Definitely not SVreX. 
 

Take it away, Curtis!

Of course now it has worked flawlessly since posting this. LOL

I do see the red LED on the control board, so I'll check that next time it does fail to start.  In fact, now that you mention it, I do have some vague recollection of noticing it on or blinking one of the times, and probably thought something dumb to the effect of 'why is the power-on light red instead of green'. Duh.

I can't see how the issue could be at the burners, as my understanding is that it doesn't initiate anything with them until after the pressure sensor detects the flow inducer running. So if either of those don't work right, it doesn't even try the burners. But it is a good heads up to check them out in general.

Thanks for the suggestions!

Failed to start again tonight.

Red light flashing... Yay!

1 Flash. And the winner is...

.

.

.

.

Lockout - Check Furnace 

Not sure what type of heat you have, but my oil heaters have done that too.  You hear the click, and the motor comes on.  Even hear the whoosh as the oil ignites.  Then in a second it shuts down.  Usually a bit later it does it again and stays lit.  I used to mention it at the annual service with no solution.  One guy told me some sensor was not getting the right signal and it was a safety shut off.  I stopped worrying about it, house is still warm.  Better to have a sensor shut down the system than a failed sensor let the boiler blow up!

In reply to Driven5 :

Well that's disappointing.   With a lockout alarm being proved now, if the draft inducer is the first thing that starts in the light off process it's probably time to order one up. 

My furnace had the draft inducer fan's pressure tap nipple get plugged up.  The fan itself was fine, the pressure switch was fine, the rubber tube connecting 'em was fine.  Yay for all the components checking out but the system not working.  

In reply to 11GTCS :

Yeah, that's what I'm thinking.

In reply to Oapfu :

If I'm understanding it right, that should still kick on the inducer but not ignite the burner? Mine isn't even starting the inducer.

In reply to Driven5 :

Right, with my furnace (Heil something from ~2001): the inducer would start up and run for a short amount of time (I don't remember exactly how long, this was like 10+ years ago); the control board would decide that the inducer couldn't possibly be running b/c the pressure switch hadn't closed; abort mission and shut down the inducer.

Well, it's probably not the inducer motor. I swapped in the best condition used one I could find, and the exact same problem persists... However a new problem has cropped up too.

So sometimes it will fail to start the blower motor, just as before. But other times the blower motor will turn on, the pressure switch will activate, and the burner will ignite... For maybe 5 secodns then turn off. It'll repeat this short cycling a number of times before giving up. A few tries in things apparently get hot enough to engage the main fain, alebeit briefly and only every third or so attempt. When it cycles out, it's just the same 'lockout - check furnace' error code. I did try removing the air filters during this to see if it made any difference... It didn't.

I replaced the main power switch, just because, and not surprisingly no change.

I'm mostly down to either the control board or thermostat, so I have the best condition 'guaranteed working' used control board I could find on it's way.

I'm also planning to check, and possibly replace, the flame sensor and limit switches while I'm digging around in there.

Any other thoughts or ideas?

In reply to Driven5 :

Oof, sorry it wasn’t what seemed to be the obvious issue.  You can eliminate the thermostat from the equation with jumper wires if you’d like to give it a try.   

Typical thermostat wiring is color coded; R / red is 24 v positive, yellow is Y (cooling if you’re so equipped), W / white is heat, G / green is the fan.  Some thermostats also have a common terminal often a blue wire that allows the furnace to power the thermostat if it’s a programmable one or a digital display.

Turn the power switch on the furnace off and carefully remove the thermostat from the sub base.  Once you’ve done this, you can turn the power back on.   If you have a meter that can test low voltage (24 v AC ) you should read 24 V between R and any of the other terminals.  If you take a jumper wire (alligator clip on each end is the best) and jump from R to W, the furnace should try to light and if it lights off and once it warms up enough to make the fan operating limit switch the main fan will start.

If you do the above and the problem goes away it’s the thermostat.  If not it could be an issue with the control board.  Either way, don’t forget to take the jumper off and turn the power off to reinstall the thermostat.   Good luck and let me know if you have any questions.  

Have you checked or replaced any of the pressure switches?   When my furnace had a somewhat similar problem, it had something to do with the pressure being bad or the pressure switch (though I never figured out exactly what, and that furnace has now gone to a better place).

Anyway, when I was trying to diagnose it, I discovered that the controller would check for lack-of-pressure signal from the pressure switch before it would start the inducer fan, then once the fan was running it checks that the pressure switch is now closed.  So I couldn't just jumper out the pressure switch, the controller looked for it to change.

So maybe you have a flaky pressure switch?  Sometimes it fails open and sometimes closed?

In the past day it had gotten bad enough that it would may heat for one cycle, then run into the short cycling issue before giving up. Cycling the power to reset the fault doesn't seem to fix this problem, but occasionally it would work for one more cycle. 

So I tested continuity on the two limit switches tonight, and apparently I need to pony up for a new multi-meter and stop using the Hazard Fraught one I got for free years ago. IF I can trust the ~4 ohm resistance shorted across the 2 leads as my 'zero', then the limit switches *should* be ok. Of course, I also noted this:

I'm not sure if that's recent or old, but either way I figure a new one is cheap insurance anyway. While I'm at it I included a replacement for the other limit switch too, because why not? 

I also cleaned the flame sensor. While I was opening the fire box up to get to it, I noticed all but one of the screws holding the cover on were stripped to finger loose... Including the one for the ground wire to the control board.  Hmmm...

Well after reassembling everything, including screwing fresh threads into previously unused holes and ensuring a good ground connection, it has now gone 3 cycles without a hiccup.  As far as I'm aware the original problem may still persist, unless it was just a grounding issue, but with any luck the new short cycling issue seems like it *might* be resolved.

11GTCS: Thanks for the heads up on how to test if the thermostat is the issue! That'll likely be next if any problems continue to arise.

alphahotel: I hadn't come across anything before about the pressure switch being checked open by the control board before starting the inducer. I'm a little skeptical that it would be designed to allow a fail-closed condition, but I suppose stranger things have happened. That's a thought I'll have to keep in my back pocket.

In reply to Driven5 :

I'm betting it was the intermittent ground...    Good sleuthing and keep us posted.

SV reX said:

mtn said:

In case svrex or Curtis don't know the answer, I'd head over to https://www.reddit.com/r/hvacadvice/. 

Sorry. Definitely not SVreX. 
 

Take it away, Curtis!

Oh, crap.  No pressure  :)

Seems like you're on the right track.

They work on a pretty simple if/then flowchart.  If inducer flow, then fuel and ignite.  If heat detected, then continue.  If no heat detected, then shutdown.  rinse, repeat.

If you're getting inducer, and ignition, and it shuts down a few seconds later, I would suspect the thermocouple/sensor.  If you're getting intermittent inducer start, I would be highly suspect of the circuit board.

Also, is the duct blower coming on?  If you get fire then shutdown, it could be because it's getting too hot because the duct blower isn't providing flow to cool it off.

So far so good. Which means it's still looking like the ground was to blame for the inducer intermittently not starting, and either the ground or the flame sensor was to blame for the short cycling. I'll be replacing the limit switch with the burnt insulation regardless. Unfortunately, I still have the purchased control board on the way, but I think I'll just keep it (and the original flow inducer) as a spare to keep nursing this furnace along for another couple of years. Despite unnecessarily buying both a flow inducer and control board, if a pro would have at any point missed the grounding issue, this still came out cheaper... And even if they wouldn't have, the education was well worth the price.

Thank you all for helping!