DIY Chatroom Home Improvement Forum

DIY Chatroom Home Improvement Forum (
-   HVAC (
-   -   Lennox Whisperheat runs normally but shuts off after 5-30 minutes (

sumguy 01-27-2013 09:10 PM

Lennox Whisperheat runs normally but shuts off after 5-30 minutes
This is a 6-burner natural-draft Lennox whisperheat that probably dates to 1981 - 1985. It has a Johnson Controls G776RGD-11 Ignition Control Module that I took apart a few years ago and re-soldered some connections on the PC board to fix an intermittent spark problem.

The furnace is in the basement of a 2-story building - it only heats the basement so it's no big deal that it's not working reliably for the past month or so.

It sparks up just fine, the burners come on, the fan comes on, I walk away and come back later to find that it's shut down. The burner air-intake door is still open but the green LED on the ignition module is out. The thermostat is still calling for heat.

The power switch for the furnace is near by, and when I turn it off the damper door closes. When I turn it back on, the door opens, the green light comes on, it sparks up, burners come on, and then the fan. I go upstairs, come back a short time later and find that it's shutdown again.

When I start doing this, it's maybe 58f ambient temp in the basement, and I keep power-cycling the furnace every time I find that it's turned itself off, and slowly the temperature in the basement comes up to 65f by 5 pm when I leave the office. It seems that as it gets warmer in the basement, the furnace seems to run longer between shutdowns.

I verified that the 2 over-heat sensors and the damper sensor are not the problem (I bypassed them with a jumper wire and still had the problem).

I completely removed the fan-door switch (so it's not the problem) and I checked all the connections on the fan control board and the fuse.

About the only thing left is something wrong with the gas valve or the flame sensor.

I wanted to measure the voltage on the SENSOR input on the ignition module - but funny thing the module won't spark-up when my volt-meter is connected (!?). So I connect the meter between the SENSOR and ground after the module sparks up and I get an AC voltage of about 7 volts (before the main burners come on) and the voltage quickly drops to 5 volts after the main burners light up.

Does anyone know what the voltage should be on the SENSOR input?

The sensor seems to be a short rod made of some type of metal - it looks clean, but it's spade-connector is very rusty-looking.

I'm going to take the ignition module off and have a close look at it, clean the contacts, make sure it's got a good ground and try it again.

Could this problem be in the gas valve?

One thing I notice is that when it sparks up, I'm seeing some sparking between the damper-door and a couple of the burner intake openings (the door practically touches the round open ends of the burners when its in the open position). Should I be seeing that? I guess it means that the burners don't have a good connection to ground...

how 01-27-2013 09:54 PM

A couple of things to check

I think you should park youself in front of the furnace in a position where you can observe the pilot flame and watch it until it shuts down again.
I'd be watching to see if the pilot flame starts to wave around after the heat exchanger has really heated up. (crack in xchanger widening with the heat and allowing the fan pressure to blow the pilot flame away from the sensor.) That would make it non repairable!

If it shuts down with no pilot wavering....

Clean the flame rod, spade connection, the pilot assy and the surrounding burner faces to give that sensor the best conductive service to work with.

The sensor is measured in DCMA

Focused2 01-27-2013 10:28 PM

If you are seeing sparking all over the place that is not normal, you may have a cracked ignitor which would cause it to not arc at the proper location to light the pilot. You may need to take a microammeter reading at the ignition module. Ammeter is normally placed in series with the circuit , looking at the images of the controller you mentioned it is hooked up to the SENSE terminal. Read up on flame rectification and reading flame signal.
On your multimeter microamp symbol is uA or pretty close. Like someone mentioned , best thing to do is keep an eye on it.

sumguy 01-27-2013 11:05 PM

Regarding the pilot and the flame sensor -> I would have thought that when the pilot gets lit, and then the main gas comes on, and it gets ignited by the pilot - that the pilot would be shut off and the flame sensor would continue to operate by sensing the flames from the burner and tell the module that everything is ok.

But if you're saying that the pilot stays on after the main burners kick on, well ok, I don't see the point of doing that.

Thing is, if the feeble flame from the pilot is enough to tell the sensor that there is a flame, then when the main burners come on then there sure as hell is going to be enough flame to tell the sensor that yup - there is a flame - regardless if there is a breeze from a cracked exchanger causing the pilot to blow around a little - No?

Regarding the sparking between the burner and the door - strange because the door pivots on what looks like plastic dowels on both ends (so the door shouldn't be making electrical contact with ground) yet there is sparking between the door and a couple of burners.

I did take the burners out and vacuum up a bunch of rust in the galleys and knock a bunch of rust out of the burners, so when I put them all back in they must not be making good ground contact. I suppose I could connect a ground to one of the screws that holds the pilot/sensor to the 5'th burner.

In spite of this sparking, the furnace lights up just fine - no problems getting the burners going.

Besides the flame sensor, there was also another sensor (mercury-filled tube?) that is part of that assembly - it runs to the gas valve. I suppose that it's a regular old-fashioned mercury switch used to shut off the main gas supply if it gets cold?

how 01-27-2013 11:44 PM

My memory of that unit is that while the pilot flame reaches the main burners, the main burners do not impinge on the flame sensor. Usually the old mercury sensors just reside in the pilot flame. That system also usually takes from 30 to 85 seconds to move from initial pilot ignition to the main burners coming on. The wavering of the pilot will shut down the main burners whether it is being caused by a pilot flame that's too soft, a cracked exchanger or excessive chimney draft.
If you observe the main burners running happily after the pilot flame has shut down, then the mercury sensor is controlling the main flame and it usually has a pin system that plugs into the gas valve. Oxidation on the pins of that plug were one of the weak links to failures and I see few of them still operating today in the field.
Perhaps a picture would clarify the system you have but I would still watch the pilot flame through to shut down.

Marty S. 01-28-2013 06:26 AM

Multiple problems. First is it's 30 year old furnace with a duracurve heat heat exchanger,they almost all cracked around 15 years old. Second is enough rust you could vacuum it out, that's likely a venting problem. Third is the shut down problem. No sense in dealing with the last if the first two conditions are present because the furnace will need to be replaced. Your best bet is to call a good tech for a diagnostic and inspection. One who has a combustion analyzer and knows how to use it to check venting would be a big plus. Around here that would cost $90 and if you needed a new furnace they would apply that towards the purchase.

sumguy 01-28-2013 08:14 AM

If there was a crack in the heat exchanger in one (or more) of the galleys, there would be a breeze that blows the flame from that galley out towards (and maybe through) the combustion-air intake damper (and thus triggering the over-heat sensor to open). I observe the flames from all galleys to be smooth and uniformly rising upwards.

If there was any blockage in the flu stack, then the combustion gases would tend to flow out the upper passive intake vent (and thus triggering the overheat sensor in that location to open). In my case, I feel no heat coming from the upper intake vent, so the flu gases must be rising unobstructed up the stack (which is very hot).

I don't think this is a duracurve, but I'll check today...

Marty S. 01-28-2013 12:37 PM

All of your assumptions are incorrect. It is your furnace though and you're free to do what you want.

BTW I'm a 16 year lennox tech,nate certified, nci co and combustion analysis certified and hold a certificate from heat exchanger experts training.

sumguy 01-28-2013 01:54 PM


Originally Posted by Marty S. (Post 1104128)
All of your assumptions are incorrect.

Unless you provide details as to why they are incorrect, how can your statement be vetted?

A crack in the exchanger would form a passage between the blower-side and the combustion-side of the exchanger. The air on the blower-side is at a higher pressure because that's what the blower does - it pulls return air and forces it through the exchanger at high pressure to the distribution ductwork. A hole in the exchanger would allow this high-pressure air to leak into the combustion galley and disrupt the natural convective flow of the combustion gases and flame and cause the gases and even the flame to flare out of the galley to the most convenient low-pressure path - which is out the combustion air-intake vent on the furnace cabinet.

If you refute that explanation, then why don't you explain what a hole in the heat exchanger would do.

sumguy 01-28-2013 02:11 PM

Just to clarify, the furnace in question is Lennox G20Q5/6E-150-5. I believe I see "1995" somewhere on the large decal inside the cabinet, which could mean the furnace is newer than I thought.

In one of my previous posts I indicated that there was a small tube (mercury switch) running from the ignitor/pilot assembly to the gas valve. I was wrong - there is no such tube. The only tube or line running from the gas valve to the ignitor/pilot is the 1/4" tube (aluminum?) that supplies the pilot with gas. Other than that, there are two wires - one for the ignitor and one for the flame sensor.

When the burners are going, the tip of the flame sensor (the top 1/4" of the rod) glows red, and seems bathed sufficiently in flame.

I don't see the term "Duracurve" printed anywhere on this furnace.

The gas valve is Honeywell VR8204H 2327.

This is the furnace for the lower floor. The furnace for the upper floor is an older model Lennox (G81-220V-2) and it must be original to the building (1981/1982). It has standing pilot, no electronics, natural draft, and it does have a "Duracurve" sticker. It gives me no trouble - it just works...

yuri 01-28-2013 04:12 PM

That blue Johnsons control is known to develop intermittent problems where it dies for no reason and acts up. I have changed LOTS of them. Replace it with the new improved Honeywell kit from Heat exchanger is probably cracked and needs a experienced Lennox tech with a video camera/bore scope to find it.

Marty S. 01-28-2013 05:56 PM

You're confusing a crack with a gaping hole. Both condemn a furnaces heat exchanger but a crack will not show any flame disturbance,most of the time. Think of it like a fractured arm vs a broken arm. The reason the put a cast on both is because a fracture can easily turn into a break. A furnace can't heal itself so a crack will turn into a hole or rupture at some point in time. Since we have no way of knowing if that will take an hour or a decade the furnace is legally deemed unsafe to operate.

Excessive carbon monoxide in the gasses leaving each cell or the heat exchanger walls being half as thick as they should be from rust(both signs of the combustion gasses not leaving the cells correctly ... venting issues) are also reasons that the unit is legally deemed unsafe to operated . Note that venting includes the heat exchanger, it is not limited to the flue. The only way to check for proper venting on that furnace is with a draft gauge in the flue in combination with multiple readings from each cell using a combustion analyzer.

Now I'm not saying your furnace absolutely has cracks and/or venting problems but years of experience with the G20 and the conditions you described sure have me leaning that way. As stated earlier it's your furnace and house so do what you want. My suggestion is to get it checked out before sticking any money into getting it operational though. The G6,G8,G12,G16 and G20 all had duracurve heat exchangers.

BTW a hole might or might not cause roll out. I've seen some itty bitty pin sized holes cause roll out because of the location in relation to the blower and others where I could stick four fingers in with no flame disturbance. It just depends on how the air from the blower hits the breach.

Focused2 01-29-2013 08:43 AM

Know a bit of talking is going on about cracked heat exchangers. In California we have a requirement to have a CO detector in the home. Basically what happens is what these folks are saying: intermittent issues with firing etc. for 20 bucks you can pick one up at local home depot and install per instructions.
You may have gotten a decent unit, but I would play it safe as it is a silent killer. Once the heat exchanger develops a crack it will not be evident until it runs for some time and as the area around the crack gets hotter the opening widens and flue draft starts to affect the burning characteristics in the main flame/pilot area. So now the CO with the air going into the living space and a hazard is created. Headaches, stomach cramps etc. Myself , if you don't want to get it checked out at a minimum install a COalarm. "Keep the customer informed and you keep the customer" a saying from one of the big companies IBM ? So you have been informed.

techpappy 01-29-2013 10:32 AM have the right detector but, just for future reference Carbon Monoxide is CO not CO2 which is Carbon Dioxide. just sayin' :whistling2:

Focused2 01-29-2013 01:16 PM


Originally Posted by techpappy have the right detector but, just for future reference Carbon Monoxide is CO not CO2 which is Carbon Dioxide. just sayin' :whistling2:

My bad. Whistling also. But we get the message I hope....

All times are GMT -5. The time now is 03:26 PM.

Copyright 2003-2014 Escalate Media LP. All Rights Reserved