Intermittent Failure to Heat - Part 2

Started by SnakeDoctor, December 13, 2009, 06:07:29 PM

Previous topic - Next topic

SnakeDoctor

Well, I'm back with the same problem I've had in the past but this time I have more information.

The problem is that my furnace (Goodman GMNT080-4) will occasionally fail to ignite (most of the time is works fine).  When it fails, I get one flash of the LED on the control board.  Flame sensor is clean and the igniter is glowing fine.  I had the unit serviced and the tech could not find a problem.  I told him it was intermittent and he said to call when it wouldn't start consistently otherwise he would have to replace parts on a trial-and-error basis.  The failure is still intermittent but seems to be a little more frequent.  Today, I decided to go in the attic and sit there until it failed and observe what was happening.  Here's what I saw when it failed...

Blower turns on
About a 20 second delay
Faint click - probably the igniter relay
Igniter starts glowing
About a 10 second delay
Faint click - probably the gas valve relay - no flame
A few seconds later the igniter turned off
About a 30 second delay
Faint click - probably the igniter relay
Igniter starts glowing
About a 10 second delay
Faint click - probably the gas valve relay - no flame
A few seconds later the igniter turned off
About a 30 second delay
Faint click - probably the igniter relay
Igniter starts glowing
About a 10 second delay
Faint click - probably the gas valve relay - no flame
Rapped on the gas line connected to the gas valve with the heel of my hand and GOT A FLAME!!!
Furnace performed as expected from this point.

So, I have a few questions...

1.  Sounds like a sticking gas valve.  Would you agree?
2.  It looks like it would be easy to replace the valve.  I am quite handy and would feel comfortable replacing the valve myself.  Anything out of the ordinary I should be aware of?
3.  Are there any adjustments that may be required after the valve is replaced and if so, what are they and are any special tools required?

Thanks for your help.

Bill

Admin

Sounds like the pressure switch is opening.  Check the drainage system for blockages.  Pull the pressure switch tubing and look for water.  The pressure switch may be damaged if it got wet, and may need to be replaced.  I saw this exact problem when the installers forgot to remove the drain plug from the collector box.  The build up of water was causing the pressure switch to trip, and the water caused the pressure switch to rust and stick.

SnakeDoctor

I checked all of the drainage system and everything is clear (no water).  I also poured water down both drain pipes to ensure there were no clogs.

So I tested the pressure switch by pulling the tubing off that connects it to the collection box.  I did this test twice - once before the thermostat called for heat, and once while the furnace was running.  Both tests resulted in a blink code of 3 blinks (Pressure switch failure).  My failure has always resulted in a blink code of 1 blink so now I'm pretty sure it's not an issue with the pressure switch.

This takes me back to my original 3 questions.  What do you think?

Thanks again for your help.

Admin

Does the furnace have a sticker with the codes?  The newer models list 1 flash as system lockout.
Goodman told me that the ignitor resistance affects the ignition sequence.  You could try switching the ignitor.  You can't bypass the pressure switch until the exhaust motor starts, but if it happens again just jumper out the switch and see if you get ignition.  When I came accross this problem it didn't flash a pressure switch code, it took 5 times of ignition failure and gave me a 1 flash system lockout code.

It would be rate to be the gas valve, hitting the valve may have caused the pressure switch to close.

No signs of water in the pressure switch tubing?
Did you pull the larger drain tube off the collector box?
Could there be a blockage in the venting?

How old is the furnace?  Did you ever have a drainage issue?
The blockage could be in the furnace, not the drain line.
Check all the black tubing in the furnace and the ptrap.  Pull the big black tube off the collector and make sure the plug was removed.  It sounds like a drainage issue... The colder it is the more condensate the furnace produces , which would cause more frequent lockouts.  The pressure switch can trip and reset so fast the board won't lockout until to many attempts have taken place.


Admin

The gas valve is simple to change, nothing to do but check for gas leaks an verify manifold pressure is set correctly.

SnakeDoctor

Yes, the furnace does have a sticker with flash codes: 1 Flash - Fail to Ignite (NOT System Lockout), 3 Flashes - Pressure switch failure.

This is the 6th winter this furnace has been in use.  This problem actually started 3 years ago but it was very infrequent when it first started - maybe once a week.  I've never had a drainage issue.

The ignitor seems fine.  When I watched the failed sequence, the ignitor was glowing bright red.

I also noticed in my testing today that if the pressure switch is disconnected (pneumatically - not electrically) that the igniter won't even come on during the startup sequence so it seems unlikely that hitting the the valve had anything to do with the pressure switch since the igniter was glowing at the time I hit the valve.

No sign of water in the pressure switch tubing.  I could hear the switch in the pressure switch operate when I lightly sucked on the tube.  I did this 5 or 6 times.

Yes, I pulled both drain tubes off of the collector box (one large and one smaller one) and checked them.

I also made sure there were no blockages at the exit of the drain box.  I was able to insert a Q-Tip into each outlet which came out damp but mostly clean.  The plug was removed.  Actually, it wasn't a plug, it was a cap.  The cap is still installed on the outlet that is to be used when the furnace is mounted horizontally.



Admin

This furnace is in an attic and not installed horizontally?  Does it have 2 plastic vents?

The ignitor won't come on if the pressure switch tubing is disconnected, the switch won't close.  I'm saying the switch closes as it should, but something is causing it to open and interupt the ignition sequence.  Can you verify if 24 volts gets supplied to the gas valve when you see the ignition fail?
A faulty control board would be a more likely cause of the problem than a gas valve.  Next time it starts failing, install a jumper wire between the 2 pressure switch terminals.  Don't jumper the switch until after the ventor motor starts.  This is only to diagnose the problem, don't run the furnace with anything jumpered.  The safety would lock out the next cycle for pressure switch stuck closed or did not open, just incase you left the jumper installed by accident.

Intermittent problems are never fun.  Let me know what you find out!

Hgye

Take an ohm's reading on the HSI and tell us what it is.

Take a dual port manometer and hook the positive side to the combustion box tube coming from the pressure switch, and hook the tube going to the inducer motor to the negative side of the manometer (both tubes must come from the pressure switch to the manometer.).  Find out the difference in pressure and compare it to the rating on the pressure switch.  Let us know what you find out.  Also, take an electrical meter and take a reading of the pressure switch and see if you have 24 volts on both sides of the switch with the inducer motor running.  Let us know the results.

SnakeDoctor

Admin:

Yes, it's mounted vertically in the attic.  I'm not sure what you mean by "2 plastic vents".  Can a picture be attached?  I couldn't figure out how.

I'll try to catch a failure while measuring the voltage across the gas valve but it may be a while before I can catch it.

I don't know if this tells you anything or not but as a test, I set the system up to call for heat.  Each time the HSI started glowing, I disconnected the tube to the pressure switch (simulating a pressure switch failure).  When I did this, the HSI would stop glowing at which time I reconnected the tubing.  About 20-30 seconds later, the HSI would start glowing again.  I would once again disconnect the tubing which would once again cause the HSI to stop glowing.  I repeated this sequence over 20 times and the system never stopped trying!  Does this mean that the pressure switch does not cause an interruption in the ignition sequence that results in a flash code of 1 flash?


Hgye:

Resistance across the HSI measures about 82 Ohms according to my very old analog VOM.

The layout of my pressure switch is a single tube from the inducer motor box to the pressure switch.  Thus there is only a single tube rather than two as you describe.  I don't have a manometer but I do have tubing and water so here's what I did:
I filled the tube with some water, connected one end to the inducer motor box and left the other end open.  I observed a 21mm displacement (one side 21mm up and the other side 21mm down) in the water which by my calculations is a delta pressure of 0.0306 psi (0.0622 inHg or 1.58 mmHg).  Now, as for the rating of the pressure switch, I'm not sure.  Here's what is written on the label:

MPL-9300-V-1.10-DEACT-NO/VS
813701-59 (E)
DC 40/01/CR

Also printed in white on the body of the pressure switch is "SJ" and "1.10" (no units indicated).  There are also contact ratings that I did not take note of.

Measuring voltage across the pressure switch terminals:  While the inducer motor is running, 0 VAC. While the inducer motor is running with the tubing removed from the pressure switch, 24 VAC.

SnakeDoctor

Oops.  An error in my pressure calculation.  I think that since my manometer showed 21mm up and 21mm down that I should have used 42mm in my calculation resulting in a reading of 1.65 inH2O.  Sorry about the units of Hg previously used.  That was just dumb.  I'm still learning.

Admin

The newer models lock out after 5 attempts, but if you did it 20 times with no lockout then it looks like a different problem.  After you pull the tubing off, do you hear the same click as when you first witnessed the misfire? ... Just before the ignitor stops glowing?  Try pulling the tubing off after the burners fire, do that 5 or 6 times and see if you get a 1 flash lockout.  It looks like your switch is rated 1.10, which you exceed @ 1.65 "wc.  Try getting that reading at the gas valve when it fails again.

Hgye

It appears that your pressure switch requires 1.10" Water column (W.C.) to operate, and you say that you have 1.65" W.C., so it appears that you don't have a venting issue.  You say that you have 0 volts across the two terminals with the inducer motor running, and 24 volts with the tube unplugged, so that means that your contacts in the pressure swirch are closing.  The resistance on your HSI is 82 ohms.  Depending on the make of the HSI they run about 50 ohms brand new.  I always change them when I find one at 100 ohms or higher.  Yours is getting up there, but still has a couple of years life in it, and it should still work fine.  You say that the flame sensor looks fine, but you cannot tell the shape of a flame sensor by sight.  You have to measure the amps it produces to be sure.  Brand new they measure between 5 and 10 micro Amps D.C.  I always change them at 2 microAmps or less.  Measure yours by putting your meter in series between the flame sensor and the control panel and letting us know what it is.  Also, make sure the ground wire which goes between the burner(likely attached with one of the burner screws) and the furnace is secure.  The flame sensor relies on the ground to complete the curcuit.

If all the above tests well, I would put a manometer on the manifold side of your gas valve when you are having the problems to see if the valve is opening.  If it is not, you have found your problem( gas valve), if it is opening, then I would have to side with Admin, that it might be your control board.  Let us know what you find.

Hgye

Sorry for repeating you Admin.  You posted while I was typing!

SnakeDoctor

Admin / Hgye:

I tried pulling the tube from the pressure switch after the burners fire.  Here's the sequence of events:
-Burners are fireing
-Removed tube from pressure switch and fire goes out immediately
-Replaced tube on pressure switch
-About a 20 second delay
-HSI glowing for about 10 seconds
-Gas valve open and burners are once again fireing
-HSI off

I repeated this sequence seven times.  It looks like it would go on forever (as before).

When I pull the tubing off, I cannot hear the click of the pressure switch contacts.  I'm pretty sure the clicks I hear are the relays on the control board for the HSI and the gas valve.

I keep a spare HSI since they always seem to fail at about 6:00PM on Friday just before a nor'easter arrives.  The resistance of the new one is 53 ohms.

The current reading on my flame sensor is somewhere between 3 and 4 microamps.

I could not find a ground wire connecting the burner to the furnace however, the burner case (which the flame sensor is mounted to) is bare sheetmetal that is screwed to the furnace case which is also bare sheetmetal.

As for connecting the manometer to the manifold side of the gas valve, I am now in new territory.  I cannot find a place to make the connection.  Any hints?

BTW, thanks for sticking with me on this.  I really appreciate the help.

Admin

Click Here to download the installation manual for the GMNT080-4.

I assume your furnace is installed in the downflow position, since it is in the attic.  I would refer to pages 9 & 10 of the installation manual, and verify the drainage is hooked up correctly.

There are a couple service bulletins on this model of furnace,

Click Here to download Service Bulletin #S-159, regarding nuissance vent switch tripping.

Click Here to download Service Bulletin #SF-008, regarding a new ignitor kit.  When you saw the furnace fail, I'm assuming you didn't hear or smell the gas, so this probably isin't the issue.

Click Here to download Service Bulletin #S-218, regarding manifold pressures.  Also look at page 14 in the installation manual for the gas valve layout.  Most gas valves have a plug (with a allen key) you remove and then thread in an adapter to connect to your manometer.  I believe with this valve, you just open the plug (using an allen key) and insert a 5/16" hose over the connector and connect that to your manometer.  I just put my 1/4" manometer tubing and insert it inside a 5/16" tube.

At this point I would suspect the control board is causing the intermittent problem.  You just have to catch it failing again, and see if you're getting 24V to the gas valve.  If you don't get 24V the control board could be the problem, because it's not sending 24V to the gas valve.

Hope this helps...

SnakeDoctor

Thanks Admin for all of your help.

My furnace is installed in the Upflow Left as shown on page 10 of the installation manual.  The drains are installed as shown in the manual.

Nope, no gas smell so I think you're right, this one is not the problem.

As has happened before, after messing around with the unit as I have done over the past few days, the failure rate seems to go down.  It has also become warmer so who knows.  I'll try to catch the failure with a meter across the wires connected to the gas valve.

Take a look at my Avatar - it's a picture of the gas valve.  The black dot to the right of (and slightly above)  the slotted brass screw is smooth - no slot or allen key hole.  Is it possible that it's a cap that must be pried off to gain access to the connection for the manometer?  Or is that something else?

Thanks again for your help.

Bill


Hgye

Bill, you are smarter than most when it comes to home repair, but I think that you should call a profesional to finish the job.  After some thought, I cannot advise you to break open any gas connections.  If done incorrectly the results could be deadly.  Good luck.

Admin

Here in Ontario a homeowner of a detached home can work on the gas equipment, as per O.Reg 215/01 Section 55.  If it's good enough for the government, then it's good enough for me!   :angel:

Your avatar is small, but I think you'll find the allen key screw just below where the black manifold leaves the gas valve.  You connect your manometer here.  Gas will only flow out of this port when the furnace is running in heat mode.  Refer to the rating plate in your furnace, but usually the manifold pressure should be set at 3.5" wc.  You should soap this fitting after it's reinstalled to ensure you haven't left a gas leak.

SnakeDoctor

Admin/Hgye,

Thanks for all of your help and feedback.  It looks like the saga will continue.  When the failure rate increases, I'll check the voltage to the gas valve and try to observe what happens upon failure.

I think I don't need to mess with the gas connections since the furnace seems to be working fine most of the time.  As a side note, I actually have two of these furnaces.  The second one is also a GMNT model but has a much smaller BTU rating.  I've had no problems with it.  One idea I had was to swap the control boards (if they are the same part number) and see what happens.  Any thoughts on that?

Admin

Let me know the model #, and I'll tell you if the boards can be swapped.  Or just look at the boards and see if the part #'s are the same.0

SnakeDoctor

The two furnaces I have are:

GMNT080-4  S/N 0202623875

and

GMNT040-3  S/N 0201633983

Thanks,
Bill

SnakeDoctor

Well, recently I've noticed the return of a (near) regular failure.  In the morning when the thermostat automatically sets the temperature up, the heat comes on with no problem.  Then within the next 2 or 3 cycles, one of them fails.  After a manual reset, no more failures until the next morning when it all starts again.

I compared the control boards in the two units I have, and found them to be the same, so I swapped them.  The failures have continued so it seems unlikely it is the control board.

Any ideas where to look next?

B-Techy

This is my first post here and I hope I'm not stepping on anyone's toe's.
Haven't worked on many Goodman's, but after reading through the posts and then rereading your first post I would say that the gas valve is the problem.
Your need to be there when it fails to see if the valve is getting power, and if it is,  have a BBQ lighter at the pilot assembly to try and light the pilot (DON'T touch the HSI). If you have gas there it will light, replace the HSI. If it don't light, then there is no gas, have a tech replace the valve.

SnakeDoctor

I spent yesterday setting up to determine if the gas valve is getting power when the furnace fails to ignite.  Long story short:  Yes it is.  I have my meter connected across the gas valve terminals and this morning I observed the meter indicate 24VAC with no ignition during the startup cycle.  I observed it attempt ignition 3 times (I think) before the control board indicated one flash (failure to ignite).

Now the strange part...

After the "Failure to Ignite" flash code occured, I set the Tstat down so there was no call for heat.  About 15 seconds later, I set the Tstat back up so that heat would be called for.  The furnace went through it's startup cycle normally (ignition occured) and the heat came on.  It's been cycling normally ever since.  Based on recent history, I will be able to repeat this again in a day or two.

bster352

Does your HSI look kind of like a paper clip or is it solid? I have had problems with the solid HSI not igniting the gas intermittently.

Hgye

Just for the heck of it, put your spare, new, HSI on the furnace and see if it fixes the problem.  You already own it and it won't cost you anything to try it.  You said it reads 83 ohms, so it should still be good, but I have seen stranger things before.

Admin

Yah I agree.  Sometimes the ignitor resistance can be off and cause this kind of ignition problem.  From what you say it sounds like a faulty gas valve, but an intermittent gas valve problem is very uncommon, with this model anyway.

Try to test how long you have 24 volts at the gas valve when the unit is working properly.  You may find that even though you are getting 24V at the gas valve when the ignition fails, it may not be present long enough to open the valve.

Make sure you let us know what fixes the thing when you get it figured out :)

SnakeDoctor

Quote from: Admin on January 27, 2010, 05:59:16 PM
Try to test how long you have 24 volts at the gas valve when the unit is working properly.  You may find that even though you are getting 24V at the gas valve when the ignition fails, it may not be present long enough to open the valve.

?? When the unit is working properly, I have 24V from the time it fires until the time the Tstat quits calling for heat.

When it fails to ignite I would say (guessing from memory because I didn't time it) that there is 24V for around 5 seconds.

Don't forget, on two different occasions, while there was 24V at the valve and no flame (failure mode), I tapped on the gas line to the valve and then got a flame.

I'll give the new HSI a try and let you know.

Admin

5 secs should be enough to open the valve.  Sounds like it is the valve after all!

SnakeDoctor

OK, it seems like the valve to me too.

I'd like to replace it myself because it seems like a no-brainer.  I've worked with natural gas lines before so I feel comfortable doing it.  However, before I do, I have one question...

Assuming I get the appropriate replacement (same specs as the current valve), is there any "tweaking" required after the new valve is installed or is it simply replace the valve (checking for leaks of course) and you're done?

Admin

The valve should be adjusted to 3.5" wc, which the manufacturer does.  I use a manometer to verify the manifold pressure and I clock the gas meter to verify the input is correct.  Just double check the rating plate to ensure the manifold pressure rating.
You should be alright just changing the valve and soap testing the gas connections.

SnakeDoctor

Just out of curiousity, have you ever changed a valve and had to adjust it to make it be the factory settings?

Admin

I used to find alot of water heater gas valves that were out of range, usually lower than rated.  I wouldn't worry about it.  Your replacement valve will most likely be set correctly.  I'm assuming you're in a 0 - 4500 ft altitude.

To be safe, I would time how long it takes the slow dial on your gas meter to make one revolution.  On the .5 cu ft dial I would expect it to take about 22 sec for an input of 80,000 BTU.




peanut2004

What was the outcome??? I had the same problem, changed everything, board, pressure switchs..ended up being the flame sensor...The furnace would not lock out, it would always try to light, again and again..

SnakeDoctor

Well, a new heating season is upon us and the saga continues.  Nothing really new is happening.  I have not changed any parts on the furnace and the problem still persists.  We have pretty much learned to live with it.  I may change the gas valve this year if it gets worse.  This is the eighth season on this furnace - what is the typical lifespan of a unit like this?  I'm just trying to decide if it is worth putting the money into it.

I checked my gas meter while the furnace is running and it takes about 25 seconds to make one rev on the 0.5 CuFt dial (I timed it for 6 revs which took 151 seconds).  So I guess my 80k BTU furnace is closer to a 72k BTU furnace.