No gas ignition while HSI is hot

I am back on here again far too soon with another problem.

I have a nearly 7 year old natural gas furnace located in the attic of my 2 story home. It's a builder-grade furnace made by Goodman. I have other gas appliances (water heater, stove) that work fine so the house's gas supply is ok.

The problem is that my furnace does not light up at all. There does not seem to be any gas flow to the ignitor.

* control board status light is ON (normal operation)
* thermostat calls for heat
* inducer blower kicks on
* HSI glows bright
* blower and HSI turn off and status light turns OFF. No gas ignited.

The status light does not blink any failure code. It's just OFF.

I read the York furnace thread which seems to describe the same problem.

Because of my previous problem I had a spare gas valve on hand. Swapping the valve did not change anything. While it is possible that both are bad, I think that is unlikely, especially since the status light did not have 1 or 5 blinks which would tell me to check the gas valve.

Is there any reason to think that my problem may not be a bad furnace control board?

 

Steady “OFF” - No power or internal fault
Steady “ON” - No fault found unit functioning properly


Sounds like a bad board.

 

What is the model of the furnace? How many wires are there to the pressure switch?


Try measuring the voltage at the connections for the electric gas valve. You should get 24 VAC for 3-5 seconds or so after the HSI has had 20-30 seconds to heat up.

Check for any loose or poor connections from the gas valve going back to the ignition control, and see if you are getting 24 VAC at the ignition control for the wires going to the gas valve under the same conditions listed above.


You may be able to get a job replacing perfectly good gas valves. Some significant number of repair services seem to specialize in replacing good parts!

 

It's a GMT090-4. There are two wires to the pressure switch. I checked all of the wire connections (not just to the gas valve) to make sure they are tight. I checked the hose from the pressure switch also.

One thing that I forget to mention is that the unit clearly goes into some failsafe mode because I have to flip the power switch before I can initiate the startup sequence again.

I have a multimeter in the toolbox, but I hate to admit that I don't know how to properly use it to check voltage.

 

Try pulling a wire off the pressure switch and start the furnace. See if you get a diagnostic code involving the pressure switch.

 

Check the spill switches around the burner box. The spill switches are wired in series with the gas valve. If one is tripped the gas valve will not energize. Also if a spill switch has tripped it could indicate a faulty heat exchanger. Should be three or four switches around the burner box. There will be a reset in the middle of the switch, between the two, probably purple, wires. If it is tripped you will feel a click when you push the reset.

 

Ok, I pulled one wire off. The inducer ran for 2 or 3 minutes before shutting off. The control board's status light blinks three (3) times. Referring to the service guide:

status light: 3 blinks
equip. problem: pressure switch failure to close
check: venter pressure switch; vent blockage

So it looks like the control board got that one right. The sequence never progressed to sending power to the HSI.

 

I assume spill switch is another name for the flame rollout switch? There are two of those in series with the gas valve. I pushed the reset button in the middle of each of them but didn't sense any click. They are loose--I can pull them up a little bit and then let them fall back down. I'm guessing that indicates neither of them were tripped. The service guide says there should've been either one (1) or four (4) blinks depending on model. In my case there was jack for blinks.

 

You can pull one of the wires off a flame rollout switch --- that will probably produce another diagnostic code error indicating the the rolout switches are OK when the wire is connected.

Unfortunately, I can't find a wiring diagram for the furnace on line.

Apparently you can't check for the presence of the correct voltage to the gas valve.

It sounds like a bad ignition control, but without a wiring diagram to verify what's going on and your ability to conduct suitable tests there more guesswork going on than there should be.

Best to call a repairman who can check things out correctly.

 

Try disconnecting and isolating the control wires to your gas valve, and giving it a heat demand. Does the LED still fail to operate?

http://www.icmcontrols.com/downloads/ICM280-ig.pdf

 

Houston, this is Apollo 11---


Thanks for posting the description of the ignition control for the furnace. I hadn't been able to find it, and I'll add it to my list of saved furnace manuals.

But I don't see how removing a wire from the gas valve is going to tell you anything useful. The way I read the list of diagnostic codes, there is nothing there to detect that condition.

And generally speaking, that's true on most ignition controls in my experience. A failure of the gas valve should be detected as a flame rectification failure, with one flash of the diagnostic light according to the manual you supply.

Or perhaps you are trying to find out if you can get the one flash code to operate? That might be worth a try. My guess is that the answer will be no.

I've requested measurement of the AC voltage to the gas valve when it should be turning on, twice, with no reply on that point.

If that voltage is present when it should be, it suggests a defective valve or bad gas supply.

If it's not present, it suggests a bad ignition control. Indeed, a bad ignition control is suggested by the lack of the one flash code, which ought to be there even if the gas valve or the gas pressure is bad.

That's what occurs to me so far, anyway.


This is Apollo 11 clear---

 

Or considering the topic we are dealing with, perhaps that should be:


"Houston, we have a problem"


Apollo 13

 

Can you feel the gas valve click if you put your fingertips on it?

IF the gas valve click and HSI glows, ohms test the HSI. If too much resistance it is burning out and weak. If that's the case, then the glow will not set off the gas, and the HSI and gas valve will shut down.

If no gas valve click is felt, test for 24 volts ac there. Should get volts to occur in about 30 second after ignitor starts to glow.

If no volts suspect bad control module.

 

I was surprise to see that no LED could indicate a gas valve error with this board. I wondered if an error would display without the valve connected or if it would blank out again.
As ECM stated measuring for 24VAC to the valve would be a good idea.

 

Yes --- with these ignition controls most of the diagnostic lights are checked with small currents to see if various witches are open or closed.

So--- if a limit switch opens a diagnostic code is generated.

If the flame sensor detects a flame when the gas valve isn't energized --- ditto.

If the gas valve fails to open, the flame sensor isn't going to detect a flame when there should be one, so you'd get a diagnostic code for that.

But the gas valve isn't tested directly, only indirectly as just described.

 

What does the diagnostic code chart on this furnace list for no LED is what I'd like to know. I see what this ICM replacement board is, and I know what it commonly is.

 

Apollo 13 to Houston---


Good question!

Got a link for a furnace circuit diagram that uses that ignition module? I'd like to see it.

Thanks for pointing something out to me I've never seen before.

 

Apollo 13 to Houston---


Do you know what's being tested by that diagnostic code and how it's being tested? I could certainly imagine a small current being passed through the gas valve coil to verify that it hasn't burned out.

But turning the light OFF sound pretty non intuitive. Why not add more blinks?

If you turn the power off to the furnace, aren't you going to get no light which then suggests replacing the gas valve?

Just doesn't sound like a good diagnostic code. But there it is!

I'm guessing that I've seen that list of codes but never noticed the additional codes at the top which you point out.

Again, thanks for an education.


This is Apollo 13 signing off!

 

I agree 100%. That is not a good method for displaying a gas valve fault.