Gas and Oil Home Heating Furnaces - Ignition Flame Rollout - Question for the Brain Trust

I am working with a roof mounted 1984 model Gaffers & Satlers forced air furnace/AC combo unit with an open combustion chamber and flame rectification and need to enlist the wisdom and experience of the brain trust moderating and participating in this forum.

The furnace is in good shape but annoyingly wants to rollout on first ignition when the furnace and firebox are cold, e.g., after being off overnight or all day, especially when there is no breeze or wind.

Pilot, flame sensor and ignitor are all working normally.

Gas valve is Robertshaw 720-070 staged opening;
ICU is White Rodgers 50D50-843

When getting the furnace ready for winter yesterday everything worked and lit great (no rollout) when the shroud was off. There was a nice breeze at the time.

But, after buckling it up and letting it shut down (satisfied heat call) when it tried to relight later in the evening the telltale "poof" reared its ugly head.

In your experience, what are the possible causes of flame rollout and have you experienced flame rollouts in furnaces where the gas pressure was set too low?

Thanks for your thoughts.

John

The usual cause of flame rollout is that the draft through the furnace is impaired in some way.

Normally, a current of air enter the furnace through the burner area, then carries the combustion gasses through the heat exchanger, into the vent out out the the vent cap/chimney.

In natural draft furnaces this is done by convection --- the fact that hot combustion gasses are lighter than cool air and they rise through the furnace, vent and chimney.

With more modern furnaces, a draft inducer motor provides the motive force to circulate the combustion gasses through the furnace.

Is you furnace a natural draft furnace or does it have an inducer motor and pressure switch? If the latter, does the inducer mtotor appear to be starting and coming up to speed before the burners turn on?

I might add that some small amount of "poof" when the burners light isn't necessarily a problem. How MUCH poof and have you observed the burners when they light to determine whether flame is actually rolling out the front of the furnace?

I might add that some small amount of "poof" when the burners light isn't necessarily a problem. How MUCH poof and have you observed the burners when they light to determine whether flame is actually rolling out the front of the furnace?

Thanks for responding.

The "poof" actually occurs when the burners don't stay lit because their lighting "blows out" the pilot, shutting down the gas flow. When the pilot relights any accumulated gas is ignited. Last night the poof was visibly discharging from the unit chimney.

The unit draws air through louvers into the combustion chamber. There is no supercharger to force air into and pressurize the chamber.

I read on a Fire Safety Training Institute web page that flue obstruction from soot and scale might block gas trying to leave a chamber. If I remove the top of the furnace unit, would the heat exchanger flue likely be exposed? Maybe that part of the furnace needs to be cleaned out. I have cleaned from below annually, but never from the chimney/flue going to the chamber. I did not want to disturb its asbestos wrapping.
Maybe it's time to put on a mask and do it.

I'd be suspicious that the pilot burner needs to be disassembled and the pilot burner and pilot orifice cleaned.

The pilot burner should look like a small blowtorch, which has a sharp blue flame that is fairly resistant to being blown out by burner ignition.

When dirty, the pilot flame gets soft, more like a candle flame, oftentimes with yellow tips and is subject to being blown around and blown out like a candle.

I'd check that first.

You aren't really describing flame roll out in your last post.

I'd be suspicious that the pilot burner needs to be disassembled and the pilot burner and pilot orifice cleaned.

The pilot burner should look like a small blowtorch, which has a sharp blue flame that is fairly resistant to being blown out by burner ignition.

When dirty, the pilot flame gets soft, more like a candle flame, oftentimes with yellow tips and is subject to being blown around and blown out like a candle.

I'd check that first.

Thanks for reminding me about the pilot clogging up. I now remember that over the Summer before last heating season, a spider set up residence in the pilot tube, Heaven only knows why. Maybe one of its relatives moved in over this most recent Summer.

I'll check it out.

BTW, If what I described was not a flame rollout, what would it be called?

Thanks again!

If my guess as to the cause is correct, it's simply an affect of having a dirty pilot burner.

You apparently have an intermittent pilot ignition system for your furnace. The pilot is lit each time the thermostat calls for heat, and is shut off completely when the thermostat is not calling for heat.

On a call for heat, the ignition control turns on the gas to the pilot burner and the spark to light the pilot. When the pilot lights, this pilot ignition is detected electronically by the ignition control, The ignition control then usually shuts of the spark and turns on the main burner gas.

However, if the ignition control detects that the pilot light goes out while the main burner is lit, it shuts off the gas to the main burner and starts the ignition process over again by turns on the gas to the pilot mburner and the spark.


Based on your description, it sounds like the ignition control isn't detecting that the pilot is lit reliably. Whenever it fails to detect that, it shuts off the main burner and turns on the spark again.

A dirty pilot is the most common cause of that problem, although other things can cause it as well, such as a bad ignition control box.

If my guess as to the cause is correct, it's simply an affect of having a dirty pilot burner.

<SNIP>

Based on your description, it sounds like the ignition control isn't detecting that the pilot is lit reliably. Whenever it fails to detect that, it shuts off the main burner and turns on the spark again.

A dirty pilot is the most common cause of that problem, although other things can cause it as well, such as a bad ignition control box.

I took the shroud off this morning and pulled the pilot assembly for inspection. The pilor was clean as a whistle.

So I did a little birthday candle test on the reinstalled pilot with the MV disconnected. I tried to blow it out with my breath and was able to do so. I cranked up the pilot adjustment for a stronger flame and can't blow it out now with my breath. With a large cardboard flap I can blow it out if I make a big gust right on top of it. The pilot relights immediately because there is virtually no delay between the pilot going out, the ICU detecting it and sending out spark to the ignitor.

After reconnecting the MV lead to the gas valve, the burners lit with a whoosh.

Following my own advice and pulling the unit chimney revealed a fair amount of accumulated soot and scale above the heat exchanger so I vaccuumed it out and would up with about a half cup of scale and soot. Not a huge amount, but it may make some difference in the breathing of the combustion chamber. I'm keeping my fingers crossed.

The real test will be when it automatically lights tomorrow morning. Problems always seem to manifest themselves most noticeably when it's cold and the air is still as it usually is in the morning.

I'll report back on what happens.

Did you disassemble the pilot burner by removing the brass ferrule that holds the pilot tubing in place and remove, inspect and clean the pilot orifice?

The pilot burner flame should look and act like a small blowtorch if it's clean. It should be resistant to being blown out, but it can be blown out with an effort.

I can't evaluate whether the pilot is clean based on your description.

Some pictures would help if you care to post some.

Did you disassemble the pilot burner by removing the brass ferrule that holds the pilot tubing in place and remove, inspect and clean the pilot orifice?

The pilot burner flame should look and act like a small blowtorch if it's clean. It should be resistant to being blown out, but it can be blown out with an effort.

I can't evaluate whether the pilot is clean based on your description.

Some pictures would help if you care to post some.

The pilot orifice was clean too. I pulled the compression fitting and made sure nothing was plugging up anything. Put a new sleeve on when reassembling.

It's a little dark for photos now but I'll try to take one and post it tomorrow.

Would a pilot flame impinging on the burner have anything to do with this?

Due to warmer weather, I did not have cause to work on the furnace until a couple of days ago.

After adjusting and pilot flame and gas pressure, the furnace lights reliably.

In monitoring the burning activity in the firebox, I am unable to obtain the kind of "sharply defined inner flame" recommended by a web reference for setting the air shutter to optimum burn. I envision something like the flame in a gas stove burner. If I set the gas pressure low enough to seat the inner flame atop the burner tube, there is insufficient heat produced to start the fan blowing via the thermostatic fan switch.

The flame I am using now seems to work well, but I thought I would take a movie of it for the experts to review and comment upon, if justified. Only ten seconds long, it is representational, having been shot after the firebox had heated up. The great majority of the movement of tubes/flames up down results from my hands wobbling. Flame dancing is at a minimum.

How does this look guys? BTW, the maximum height of flames in the movie does not exceed 3/4", and probably averages 1/2".

Gaffers_and_Sattlers_furnace_burn_2 on Flickr - Photo Sharing! (http://www.flickr.com/photos/44613514@N06/4100217388/)

Thanks

John

Sorry, but setting the burner input and promary air on a gas burner isn't a DIY job, in my opinion and experience.

And you can't tell how it should be by looking at it. The burner input needs to be set correctly and the combustion gasses need to be checked for carbon monoxide with an instrument.

My suggestion is to call a repairman to have the burners set up correctly, check the heat exchangers for cracks and inspect the venting system.

Unfortunately, doing routine maintenance and inspections on a gas furnace isn't a DIY task. Too many things require inspection and need experience to be able to interpret properly.

A good many repairs can be done by DIYers, because the skill set and judgment needed is limited. Routine maintenance is often more difficult than doing repairs that have caused a furnace to quit working.




Seattle Pionner

You could probably get some idea if gas flow is too low, by timing the gas meter and converting cubic feet to btus(based on 1 ccf (100 cu.ft.) = approx. 100,000 btus, and comparing to your furnace's input btu rating on your mfg. identifcation label, that will list that specification. Crank up the stat and time (your cell phone may have a stop watch) any of the fractional cubic foot dials on the gas meter, then do the math.

You could probably get some idea if gas flow is too low, by timing the gas meter and converting cubic feet to btus(based on 1 ccf (100 cu.ft.) = approx. 100,000 btus, and comparing to your furnace's input btu rating on your mfg. identification label, that will list that specification. Crank up the stat and time (your cell phone may have a stop watch) any of the fractional cubic foot dials on the gas meter, then do the math.

Thanks for the suggestion. The gas valve outlet pressure setting is 4.5" on a water column.

I will do a count on the meter consumption the next time the furnace is turned on. Thanks again.

I timed the consumption of gas this morning with the furnace running continuously and the math yields a use rate of 37.05 cu. ft. / hr. I also climbed back up on the roof to reinspect the mfr. specification label and the timing was almost right - I can see that there are typed numbers on the foil label but the sun angle was not sharp enough to read the shadows completely.

The purchase invoice referred to it as a 3 ton AC/ 100,000 btu furnace unit, but I am unsure about the furnace rating as the label appears to be only five digits ??,000 in the input btu hr cell.

The heat gain information shows a rise from ? F to 100 F at 5.? INSP

Again, I want to get this information right, but it appears that pressure may be part of the problem. If it turns out I am 1/2" or 1" short of pressure, that would change a lot of things and explain much of the lighting and heating behavior of the furnace in the recent past.

I will keep you updated.

I can see that there are typed numbers on the foil label but the sun angle was not sharp enough to read the shadows completely.

I am unsure about the furnace rating as the label appears to be only five digits ??,000 in the input btu hr cell.

The heat gain information shows a rise from ? F to 100 F at 5.? INSP

I will keep you updated.

Here is my update:

G & S Magic Chef
spec label reads:

Input 80,000 btu hr

AT TEMP RISE OF

70 F to 100 F

AT 5.0 IN S.P.


Unless you folks tell me not to do this, I am going to interpret that the foregoing means 5.0 " pressure on a water column and I will reset my gas valve to 5.0" using my manometer.

Then I will adjust the air shutter for the increased pressure and see how the furnace lights and runs.

Given the low gas consumption encountered (37,00 btu hr +/-) following ecman51's formula and my timing of 94.6 seconds for two turns of the 1/2 cu ft dial on the gas meter (less 1 cu ft per hour for four range pilots). Upping the gas pressure from 4.5" where it is now would probably also reduce fan cycling which occurs in the morning.

I will report back to the forum what my results are.

This afternoon I adjusted the gas valve to 5.0" on a water column and let the firebox heat up completely before adjusting the air shutter. The gas consumption checked on the gas meter has gone up about 4 cu ft/hr (less than I expected).

The higher pressure has caused the burner tubes to produce flames which don't dance the way previous pressure settings had. Air shutter adjustment produced nice, symmetrical ribbons of flame on each tube and all tubes were lit and stayed lit.

But best of all, I watched a startup after the pressure change and it was beautiful.

The pilot lit, the gas valve turned ON and shortly thereafter a line of blue flame walked back down the burner tube to the flame spreader, then across to the other burner tubes before the higher stages of gas pressure kicked in. When the final stage was reached, everything was burning nicely.

No poofs, no booms, no pops.

Thanks for all of your help gentlemen.

:)

Justed wanted to let you know I read all your posts.

It sounds like what you have done has made an improvement in the right direction.

It does seem odd on the surface, that by upping the pressure by over 10% that the gas volume would not go up accordingly. But because the gas is going through restrictors, one could liken the effect as to what happnes to cars consuming gas at X mph and then what they consume at 3X mph, due to the wind resistance. Same with engine horsepower. Ever notice that cars with 700 horsepower can run the quarter mile in say 9 1/2 seconds. But cars with 1400 horsepwoper can't run it in 4 3/4 seconds. It rather takes a vehicle with 7-8,000hp to do that. Or with water - how if you up the pressure by double does not equate to having the water shoot out the hose twice as far, nor put out twice the water volume, due to those same forces of restriction and friction. Those sorts of things.

Anyone else?

Here is my update:

G & S Magic Chef
spec label reads:

Input 80,000 btu hr

AT TEMP RISE OF

70 F to 100 F

AT 5.0 IN S.P.


Unless you folks tell me not to do this, I am going to interpret that the foregoing means 5.0 " pressure on a water column and I will reset my gas valve to 5.0" using my manometer.

Then I will adjust the air shutter for the increased pressure and see how the furnace lights and runs.

Given the low gas consumption encountered (37,00 btu hr +/-) following ecman51's formula and my timing of 94.6 seconds for two turns of the 1/2 cu ft dial on the gas meter (less 1 cu ft per hour for four range pilots). Upping the gas pressure from 4.5" where it is now would probably also reduce fan cycling which occurs in the morning.

I will report back to the forum what my results are.




Well ecman ---


This post is a perfect example of why it's a mistake (in my opinion) for DIYers to replace a gas valve.

This guy has entirely the wrong idea about what he is doing, and is dangerous.

Well ecman ---


This post is a perfect example of why it's a mistake (in my opinion) for DIYers to replace a gas valve.

This guy has entirely the wrong idea about what he is doing, and is dangerous.

Without being disrespectful to Seattle Pioneer, I would like for him to justify his comments about what in my post indicated that I have the wrong idea about what I am doing and therefore am dangerous.

I am always willing to learn from others.

Like an automobile engine which exists for producing torque, inadequate fuel means inadequate everything else. A furnace only differs in that heat is the primary product, not byproduct. The torque in a furnace turns the fan which drives the air past a heat exchanger.

Please tell me what you think I am doing wrong. Thank you.

Sorry, but I'm not going to try to coach you through doing a task it's perfectly apparent you lack the competence to do.

That's no insult. You just don't know how to do the task and you don't know what you don't know.

My suggestion is that you have a competent repairman install and adjust the valve and make sure the furnace is safe to operate.


Generally speaking, it's my opinion that replacing a gas valve is not a task DIYers should try to do.

I am very suprised that the temp rise specifys anything concerning the gas valve.most manufacturers give a temp range,and static pressure of airflow .5 (example).since the temperature rise deals with airlflow across the exchanger,why would it give gas valve data? normally,the gas valve has info stamped on it.I think seattle is right,I think you missed the boat on this one.I could be wrong,but every manufacturer I have seen,has not given any data on temp rise concerning the gas valve.does it specify natural gas? what if the unit was installed with propane,would your 5.0 data apply? probably not.but .5 static would still apply to airflow no matter what fuel is used because temp rise has to do with airflow across te heat exchanger...

and just to add some additional info......you could have a serious problem with the heat exchanger...crack...hole.....etc....the furnace should light all the time.....if you remove the shroud and it lights,you replace it,and it does not.what does that tell you? the unit was designed by an engineer to have the shroud on at all times.by increasing the gas pressure,it may be enough to overcome the rollout,but you could still have a problem with the exchanger....you could also have a problem the the draft and chimney.....i suggest you get very good carbon monoxide detectors,one on each level.....also have the venting,chimney system checked.......good luck

Yes, seabee, you are seeing some of the reasons this guy made me very queasy.

There are a number of things that need to be checked and properly adjusted when replacing a gas valve. In theory it might be possible to talk someone through all those steps, but odds are people aren't going to do them correctly.

The most important decision a DIYer needs to make is: "Is this a job I can do safely and reasonably, or should I have a pro do it?" It's difficult for a lot of DIYers to make that decision for themselves.

It's telling that EVERY manufacturer recommends that only qualified people work on heating equipment.

I'm probably foolish to try to help people as much as I do. I do draw the line when I have doubts about whether people can safely do a job, and I'll tell people when I think they should get competent help.

Of course, a good many people will ignore that. But at least you've tried.

Well now he has jacked his manifold press up to 5". Wanna bet this unit is seriously overgassed. Please reread the rating plate. Look for the rating, should probably be 3.5 I am willing to venture. Something tells me you mistook static pressure for gas pressure and now you have cause for serious concern.

Are you able to call your local gas co and have them clock meter properly and do a combustion test? Also not all gas meters use the same formula when clocking.

If the underlying cause was a bad heat exchanger, why should that affect his meter clocking?

And we are not talking some number like 2,000 btu's off. We are talking a huge number discrepency (80,000 rating, but only 37,000 clocked). And the paltry burner flames bear this out.

He did not say the flames were big and yellow. They were blue, but teeny. I agree the pressure is now likely off too the high side. But it still sounds like something is plugged somewhere, rather than an exchanger issue as cause for poor flame. He claims the orifices are clean. meter? regulator? In the gas valve? Manifold?

I guess it would not be good if he had the valve adjusted too high, even IFall is fine for now, - but then something cleared itself out by itself over time and then he had too much gas and flame.

I have a feeling that at this time, he does not have too much flame. If he was the unit, would run noticably hotter and possibly high limit if he screwed things up to the extent claimed here. He could check that and see what his temperature is above the burners.

I can't recall that he said anything about his flames dancing specifically as caused when the blower came on. I may have to reread this entire thread though to see.

It sure can't hurt if gas company willing to check pressure, that is for sure.