Home furnace problem, need help.

[gismo]

My furnace has been acting up lately. It uses spark ignition and a single wire "flame sensor" probe. The symptoms are that when the thermostat calls for heat, the induction blower runs, the pilot gas valve opens, the ignitor lights the pilot and then instead of the main gas valve opening, the main gas valve goes "clunk, clunk, clunk..." a few time about half a second apart with several second pauses between groups of "clunks". The first time this happened, I cleaned off the "flame sensor" a little (scraped it with a screwdriver) and it worked OK for a while but the problem came back. Then I pulled the flame sensor out and cleaned it up real good with a wire brush leaving it shiny. This didn't seem to help much but then I discovered that there was some resisitance between the ground on the ignition/flame sense controller and the furnace so I added a jumper wire between them. This worked great for a while but several days later it began to act up again in the same way.

The big question is whether this is a problem in the controller (RobertShaw 715U) or the flame sensor itself. The sensor appears to be a single wire rod about three inches long with two of those three inches inside a ceramic insulator. It's definitely not a thermocouple so I assume it senses the flame by conduction. I traced out part of the controller's schematic and it looks like there is a low current 120 v source source appled to the probe and something that looks like a synchronous detector connected as well.

I'm wondering is there is supposed to be some kind of coating or plating on the sensor that has burned off over the years (the furnace is getting close to 20 years old) that's causing my problem or if it should work with a freshly cleaned probe for more than a week. In that case I would assume the controller is the problem. It's definitely not the gas valve as I've measured the voltage on it's coil and found little or nothing when it's off but supposed to be on. Also I can jumper from the pilot valve terminal to the main valve terminal and make it work.

I suppose I could just get a probe and try it, but I'd rather not waste the money and worse yet the time to go to the distributor across town for the wrong part. The probe costs something like $20 and the controller is more like $150.

 

[Ken Ibold]

Can't help you, Lance. My air handler blew up two weeks ago. We used a space heater for a couple of days, and that got us through a cold snap, but eventually I will buy a new air handler.

 

[Dave Siciliano]

Is there a thermal coupler? I've had to deal with those several times and have always just found it better to replace them.

Best,

Dave

 

[gismo]

Is there a thermal coupler? I've had to deal with those several times and have always just found it better to replace them.

I'm guessing you meant "thermocouple" and the answer is no. This furnace uses an electronic flame sensor instead.

 

[astanley]

I'm guessing you meant "thermocouple" and the answer is no. This furnace uses an electronic flame sensor instead.

I ran this by my uncle, who owned a large HVAC company for years. He said "replace the flame sensor, if it still has a problem, you have a backup flame sensor!"

A bit tounge in cheek, but...

Cheers,

-Andrew

 

[Pjsmith]

My furnace has been acting up lately. It uses spark ignition and a single wire "flame sensor" probe. The symptoms are that when the thermostat calls for heat, the induction blower runs, the pilot gas valve opens, the ignitor lights the pilot and then instead of the main gas valve opening, the main gas valve goes "clunk, clunk, clunk..." a few time about half a second apart with several second pauses between groups of "clunks". The first time this happened, I cleaned off the "flame sensor" a little (scraped it with a screwdriver) and it worked OK for a while but the problem came back. Then I pulled the flame sensor out and cleaned it up real good with a wire brush leaving it shiny. This didn't seem to help much but then I discovered that there was some resisitance between the ground on the ignition/flame sense controller and the furnace so I added a jumper wire between them. This worked great for a while but several days later it began to act up again in the same way.

The big question is whether this is a problem in the controller (RobertShaw 715U) or the flame sensor itself. The sensor appears to be a single wire rod about three inches long with two of those three inches inside a ceramic insulator. It's definitely not a thermocouple so I assume it senses the flame by conduction. I traced out part of the controller's schematic and it looks like there is a low current 120 v source source appled to the probe and something that looks like a synchronous detector connected as well.

I'm wondering is there is supposed to be some kind of coating or plating on the sensor that has burned off over the years (the furnace is getting close to 20 years old) that's causing my problem or if it should work with a freshly cleaned probe for more than a week. In that case I would assume the controller is the problem. It's definitely not the gas valve as I've measured the voltage on it's coil and found little or nothing when it's off but supposed to be on. Also I can jumper from the pilot valve terminal to the main valve terminal and make it work.

I suppose I could just get a probe and try it, but I'd rather not waste the money and worse yet the time to go to the distributor across town for the wrong part. The probe costs something like $20 and the controller is more like $150.

Lance: I'm a little confused: You hear the gas valve make a noise, but it does not open to allow gas into the burners, while at the same time there is no voltage at the gas valve? Seems like both could not be happening simultaneously. Does it still 'clunk' when you try to light it, or has it stopped entirely now?

If there is truly no voltage, the first place to look is to the flame tip sensor. Here's a page that shows a pic of one: http://www.heatpro.us/service/ignitor.htm These things die regularly, since they measure in millivolts and therefore it doesn't take much to get them out of spec.

Try cleaning it again and resetting the power to the furnace and kicking it on. If it works at least one time, I'd say that's the culprit. They are usually relatively inexpensive (mine was $20).

Good luck!

 

[gismo]

Lance: I'm a little confused: You hear the gas valve make a noise, but it does not open to allow gas into the burners, while at the same time there is no voltage at the gas valve? Seems like both could not be happening simultaneously. Does it still 'clunk' when you try to light it, or has it stopped entirely now?

Actually I think it's working right now, but I haven't done anything to "fix" it since last night when it was misbehaving.

The sound the valve was making was like it was briefly opening but mostly staying shut, and the voltage across the coil was consistent with that (i.e. there were brief "blips" of voltage when it was "clunking").

If there is truly no voltage, the first place to look is to the flame tip sensor. Here's a page that shows a pic of one: http://www.heatpro.us/service/ignitor.htm These things die regularly, since they measure in millivolts and therefore it doesn't take much to get them out of spec.

Try cleaning it again and resetting the power to the furnace and kicking it on. If it works at least one time, I'd say that's the culprit. They are usually relatively inexpensive (mine was $20).

Good luck!

OK, this is what I was hoping to learn, that the sensor could degrade in a way that a thorough cleaning wouldn't fix. I'll get a new one.

 

[kthompson2k]

Maybe try kicking it really hard!
Kidding, ours did something similar last year and it was a valve? that was stopped up on the inside, repair & repairman was like $150, which again shows why I do not do plumbing or hvac.....I am an electrical contractor though.

 

[Ghery]

My furnace quit a couple weeks ago. For a $250 charge, the following things were done:

1. Cleaned the oriface to the pressure switch that tells the controller that the draft fan is running (that got the furnace running again).

2. Checked the resistance of the igniter. Should be about 50 ohms, was closer to 100 ohms. About 14 years old, so it got replaced. Old one still works, it is now an emergency spare (and I know how to replace it).

3. Cleaned the flame sensor. I think he used sand paper on it. It was still working, but now should have a number of years before it needs servicing again (or replacing).

Anyway, it works great again and the house is warm.

 

[MSmith]

I have had similar problems with my old furnace in the past. It was always either the igniter (in my case 2 of them cracked clean through) or the flame sensor (twice it was cleaned - once by the gas company with steel wool and once by me with sandpaper).

The multiple clunks of the gas valve makes me wonder if it's the valve itself.

Supposedly, the expected life of a gas furnace is only 10 years these days. Nobody makes them like they used to.

Oh, we replaced the furnace mid-summer when the A/C condenser got a hole in a coil after a near-miss lightning strike {venting Freon into the basement}. It was time to replace it anyway so we got a new one with a two-stage gas burner and a bigger condensor.

Does anybody know if there is a lot of difference between using a two-stage furnace in timer mode with a one-stage thermostat (low stage for 15 minutes, then high fire) vs. using a two-stage thermostat?

 

[flyingcheesehead]

Lance:

Tech support folks LOVE people like you... You've got all the info, you've already tested several possibilities, yet you're open to suggestions rather than being dead-set on it being a particular problem.

Is there such a thing as "tech support" for furnaces? I'm guessing you've gotta know somebody who knows stuff about 'em.

 

[ScottM]

I had a very similar problem last winter. Did the same stuff with the flame sensor as you, still had problems so we swapped out the controller and it has worked perfectly every since. BTW the new controller kit came with a new flame sensor so that got replaced as well.

 

[wsuffa]

Lance:

Tech support folks LOVE people like you... You've got all the info, you've already tested several possibilities, yet you're open to suggestions rather than being dead-set on it being a particular problem.

Is there such a thing as "tech support" for furnaces? I'm guessing you've gotta know somebody who knows stuff about 'em.

Tech support SHOULD love people like Lance. IMHO, they don't. Most tech support people have to read the script and walk you through the script, step by excrutiating step.

I had SBC Internet group tell me one time "we cannot help you at all" because I had a BSD Unix machine on the circuit rather than Windows-only. Their router (RBACK3) was messed up - it finally took getting the local sup involved by pulling some strings from work.

 

[flyingcheesehead]

Tech support SHOULD love people like Lance. IMHO, they don't. Most tech support people have to read the script and walk you through the script, step by excrutiating step.

Those folks aren't really tech support. They're robots a lot of the time. But, that's what you get when you pay close to minimum wage for tech support. Good tech support requires a bit more cash.

However, the good ones still go through the script these days. The difference comes at the end of the script. Bad tech support will tell you it's someone else's fault (Dell: "Call Microsoft!" Microsoft: "Call Dell!"). Good tech support will continue and use their brain, provided the person on the other end of the line has one too.

For instance, when I call Apple tech support (yes, it has happened), I can usually start out with "Hi, here's my problem, I used to do Mac tech support and I've already zapped my PRAM, reset the power manager, disabled extensions, and reinstalled the system." That kills pretty much the whole script, so we can get right down to business.

You are right, though... Too many idiots calling themselves tech support these days, and their ego gets in the way when someone like Lance calls.

 

[gismo]

I had a very similar problem last winter. Did the same stuff with the flame sensor as you, still had problems so we swapped out the controller and it has worked perfectly every since. BTW the new controller kit came with a new flame sensor so that got replaced as well.

That's probably going to be my final solution. I got a new flame sensor today and it didn't make any difference, although the thing has been working since I took the controller apart to see if anything was obviously wrong there. I hate to put that much money into it since I plan to replace it in a year or so, but I suppose it'll be worth it so I don't have to get a new one right now. This will be the second replacement controller for this furnace, the last one quit sparking a few years ago.

 

[gismo]

Tech support SHOULD love people like Lance. IMHO, they don't. Most tech support people have to read the script and walk you through the script, step by excrutiating step.

I had SBC Internet group tell me one time "we cannot help you at all" because I had a BSD Unix machine on the circuit rather than Windows-only. Their router (RBACK3) was messed up - it finally took getting the local sup involved by pulling some strings from work.

Actually it's kinda frustrating when you just want an answer to a specific question and "tech support" can't go off script. It doesn't help my emotional state when it becomes obvious that I know more about the stupid product than the support person either.