O.T. Furnace problem

A while back we had a power outage, I thought no biggie get the generator's out to keep the furnace's, refrigerater's, ect running.
No problem with the house but the furnace in the shop for some reason won't run off a gen set.
It's a Lennox G51mp 135,000 btu 90% with electronic ignition.Installed new January 04 the LED diagnostic code is alternating fast flashes. The troubleshooting guide say's

1) Improper ground
2) Open ignitor circuit
3) Low line voltage
Called my installer he checked polarity then installed a 6' ground rod just for this unit still no go,unit runs fine on utility power.
my generator's are a Coleman 3200,Honda 3800X, and a Miller Big 20. Like I said the house run's fine same Lennox 110,000 btu but pre LED diagnostic installed September' 98.Checked voltage on all three generator's they check fine at 123 volt's.Both furnaces are set up on pigtail connections so I can just run extension cord's when occasions like this come up.Called Lennox was told to check the three items above.
Any ideas? thank's.

J.Ramsey

The Miller, is it a welder/generator? If so it, like all of their ilk, produce dirty power. Poor frequency regulation as well as a lot of noise riding along with the power. The electronics in the furance won't tolerate it. Little generators like the Coleman, usually aren't much better, mechanical engine governors and poor voltage regulation. One single generator, like a bigger Honda or Onan is probably the answer.

reggie' the Miller Big 20 is 200 amp 100% duty cycle welder with aux power of 3.0 kva driven by a 4 cyl contintental.I have a buddy who has a furnace like mine and runs just fine off a gen set. talked to a Weststar electric guy this afternoon he said it could be dirty electric also.Don't know what to think.Thank's

J.Ramsey

You know it's not the ground or ignitor crcuit because it runs off utility. That leaves voltage. Where did you check the voltage? At the genset or inside the furnace? Check as close to the furnace or inside (under load) as you can. A modern power supply should tolerate a pretty wide voltage range. Do you know if it is a switching type supply?

I'd agree that dirty spikes in the power output of the generator is the key, and evidently the furnace controller is more tolerant in one building, or else the length/inductance/capacitance of the cable acts as a noise bypass to some extent in one building

One thing you might try is a ferrite type "line conditioner". I'm not talking of a surge arrestor, or a UPS, but a ferrite type filter. I'm not up to date on current brands models, but these are basically just a box with a large ferrite core choke inside, and possibly some caps.

You might also get a couple or a few of .001mfd disc ceramic caps. put them across the input and output of the 24V control transformer, which feeds the electronics package, with as short leads as you can muster.

Apparently you aren't the only one. From Bob Vila's answer board:

Lennox G51MP Furnace (Generator Power Requirements?) 02/03/2007 01:13 PM OwenDC
My Lennox Dealer said that the power from my 3000watt generator my not be clean enough to start the heat cycle? The generator a Yamaha, has a power "inverter" to deliver consistant clean power. Has anyone experience with this problem? Fan runs fine using 700 watts. When call for heat the power jumps to 1100 watts on the power grid so why won't my generator run the furnace?
Member Since
01/28/2007

Total Contributions
2 Posts


Post Reply | Watch this Topic
Control board 02/03/2007 02:42 PM HKestenholz
This has been posted several times these past months. Seems the inverter power pulses don't work well with generator output cycling. Nobody has gotten back with a successful attempt. I would think a thermocouple or powerpile pilot system would be successful, as they gas valve isn't controlled by an intermediate control board.

www.heatpro.us energy businessmen's knowledge

Yes, you really have to find out the MAKE and MODEL to get good answers. There IS more than one machine made.
Member Since
05/02/2001

Total Contributions
3093 Posts


Post Reply | Watch this Topic

If you have access to a 'scope or a frequency meter you can see the output of the generator and then check it on the power companies service.

As stated before some of the furnaces are pretty intolerant of spikes and frequency drift.

The gen set may put out 115/220 V.A.C. but at what cycle per second? and is it stable under load?

Your motor on the furnace should have no problem with noisy AC.

The electronic control might be bothered from high frequency noise, but it requires very low power, maybe a few watts. I do not have one and don't want one. So I am not familiar with details except hearsay of failures. Mine are all pilot light.

If you isolate the 120 V AC power leads to the electronic control, and supply these from a low pass filter (appropriate inductors and capacitors), then you could clean up the waveform into the electronic control. Another way would be to use a Sola Constant Voltage transformer, preferably the so-called sinewave type. This device is a very good filter as well as providing good voltage regulation. You would want the smallest size one which I believe is about 120 VA. These should not be run unloaded so a 50 W bulb load might be desirable. You won't want to buy one new because of the price.

There is probably a 120 V to 28 V transformer that supplies the electronic ignition, and thermostat. This is usually less in size than a 3" cube mounted on a plate on an electrical box inside the furnace.

.

I have a Wacker 5.6kw construction grade gen and I ran my whole house for 12 days after huricane close call. I ran 2 window AC's, fridge, DISH Satilite receiver/TV, and my computer with no problems. Dirty power will show up on your TV as lines or static. I had none of that.
I would suspect the wiring. Is your patch cord wired properly? Could you switch legs on the control?

"Dirty power will show up on your TV as lines or static"

That statement is absolutely untrue.

I CAN tell you that I've seen some strange things with modern day, sensitive electronics. One example, is various transformer type regulated DC supplies, such as what Astron builds.

I've seen two situations where these regulator supplies were the ONLY device affected. In one case, an Astron supply was being used to operate an amature radio transeiver. A nearby floorlamp with a common dimmer control would cause that supply to go completely nuts

In another case, a variable temp soldering station on an electronics workbench would cause a similar regulated 12V DC bench supply to go nuts. It would hum and grunt, and the voltage would become unstable.

440Roadrunner, A lot of common (now) dimmers use SCR's that will affect many power supplies...dirty power, check with an O Scope.... WL7WW

The Miller is primarily designed to be a welder and has a generator for convenience only. Just to run a few tools, drill, grinder, etc. at the job site. The welder portion produces noise and harmonics which are being induced into the AC windings.

I bet the issue is the electronics in the furnace does not "see" the neutral side as it does with utility power. This may be due to a ground isolation circuit kinda like what a GFCI does. Since the generator does not use an isolated neutral plus a ground but instead relies on "transformer" action to generate 120v power that is probally where the inherit issue lies. If the generator puts out 220v, then a leg could be split to provide one hot side like the utilities do. I don't know what is code there but usually some kind of isolation switch would be required so generator power does not travel back up the line( can kill lineworkers) and then you could tie directly into your breaker panel.(of course if you have 220v gen power)

"I bet the issue is the electronics in the furnace does not "see" the neutral side as it does with utility power. This may be due to a ground isolation circuit kinda like what a GFCI does."

No. The control board/ ignition module in those types of furnaces are powered either directly off line power--IF they are the type known as "hot surface" igniton, or else they are powered (I believe in this case) off a common, ordinary 120-24V AC control transformer.

None of these units have any such circuitry as GFCI or anything similar. They ARE however microprocessor controlled boards, with lots of low power solid state components. As such, they are susceptible to spikes and noise.

While I got "out of the game" before the model that the original poster has came along, I spent a number of years servicing heating/AC, and have seen hundreds of modern, electronic controlled units.

One example of how sensitive these are:

Typically, flame "safety" is operated by what is known by a "flame rod" or "flame rectification."

What happens is, the controller applies an AC voltage to a rod positioned in either the pilot or the burner. When a "good" flame is present, the circuit of the AC-to the rod-rectified through the flame-to burner ground returns a current to the controller of some 1 to 3 or as much as 5 MICROAMPS. This is an EXTREMELY small current, and these things can become "twitchy."

Sometimes all it takes is "dirty" conductive condensation on the flame rod insulator. A burner that is not adjusted correctly, or draft problems, etc. In the case of the Lennox Pulse, they used an 8mm spark plug body (like old Pontiacs used) with a rod instead of the plug electrode.

It turns out, that inside the insulator, there was a spring loaded "expansion" joint--the rod was NOT solid from the wire connector to the rod electrode. Sometimes these detectors would get intermittent, and not have low enough resistance to operate when cold. The furnace would try to fire, and sometimes lockout before the rod got warm enough to make contact. I fought several of these, and argued with Lennox support, until I finally found first one, then two, and proved it to myself.

The point is, these controllers ARE microprocessor, low power devices, and subject to these very types of problems.

experienced this issue myself-found solution

The ice storm that caused major power outtages all over Oklahoma also left me without power for 4 days. I experienced the same issue with my rheem criterion furnace when I tried to run it off a generator set. The draft fan would run, ignitor would glow,gas valve kick on,burners light for about 4 seconds then poof. burners quit, draft motor still running. dang. tried several times. turned off various loads ( running on gen set) tried again. same.I was using a heavy soow cord set from the gen to inside then a second cord ( factory made ends)12-3 sjtw type cord to the furnace (25") > I switched to a soow 12-3 cord ( much longer even) and isolated the ground from the house-at the furnace and grounded the furnace directly to the generator set this time ( no ground to the gen set before-left it grounded to the house wiring) worked great for the whole time I was without power. I even ran another furnace of this type the same way ( at another house).

Yes, it sounds like a flame rod problem.

Either with flame rod and its power, OR with the detector and micro that uses the signal.

As it happens, I have recently been working with flame rods and detector circuits on a non-furnace project. They are very fussy items.

As mentioned, they act on the principle of flame rectification, which depends on the relative area of the rod and the "ground" around the fire. If the power is funky, with dirty power, spikes, high frequency interference, etc, they may be messed up, they may not rectify right, and so they won't work.

if the furnace starts, runs 4 sec and shuts down, that's right. The system didn't get its flame rod signal and assumes no fire. It then shuts down to avoid pumping the furnace full of explosive mixture. Residential systems run a bit longer than big boilers before a flame rod fault shuts them down, because of lower gas volume.

If it refuses to start, it may be because it gets a signal that shows flame rod current when there is no fire. that is a fault that should cause it to refuse to start, since it can't depend on its sensors.

The items listed in the original post to check are items that would relate to non-ignition. They probably don't want you messing with the flame rods, OR the ignitor may BE the flame rod. That can be done also.

You're having a neutral/ ground problem. Generators and typical household wiring handle the neutrals differently.

hybidder said:

You're having a neutral/ ground problem. Generators and typical household wiring handle the neutrals differently.

Click to expand...

Not if done to code.....

And if wired into the system, and a 115 tap exists, there should be a default neutral established.

However, if NOT to code, etc, there could be a problem with flame rods etc with no neutral and no ground established.

...and you're assuming it's done to code?

hmmm, simply connecting a generator without a means of correctly altering the neutral/ ground configuration WILL NOT be to "code"

While we are discussing NEC violations, this:

...then installed a 6' ground rod just for this unit ...

Click to expand...

may be a pretty serious one in and of itself. ALL grounding electrodes for a structure need to be bonded together. Otherwise, there may be a significant potential difference between the electrical service ground and the "local ground" established by the separate rod.