Furnace flame sensor and possible fuel problem

[Bill]

I have a furnace that keeps locking out for an hour at a time when it returns to normal until the next lock-out occcurs. My furnace contractor has replaced the control board twice, the gas-valve once, and put in three new flame sensors. They now are wondering if my LP gas could be at the root of the problem because the flame detectors have a green tinge to them.

I have two questions:

1) The ignition failures that lead to the lock-outs do not involve any flame. That is, they occur before ignition; the gas just doesn't seem to flow and the ignitor is definitely on. If the flame sensor was the problem, wouldn't the failure occur after the flame was present, that is it would not detect the flame even though one was actually present? Isn't the flame detector's only function to detect for absence of flame when gas is flowing into the burners and to shut off the gas at that point?

2) Assuming it is the LP gas, what could be wrong with it? What sort of adulteration or impurities should be looked for? If it's a problem with the gas, why would it work most of the time and not others? The first failure followed the last fill date by three months. The second failure followed the next fill date by three weeks. Now, on the same tankful, lockouts are increasing in frequency to four or more a day.

I'd greatly appreciate any suggestions on what could be wrong. I'm at a loss.

Thanks.

[Greg O]

Most all gas furnaces cycle in this manor, the stat calls for heat, the inducer motor starts and satisfies a pressure switch. After a short time, 15 seconds perhaps the igniter starts. Now on a hot surface igniter, it needs to heat for a few seconds before the gas valve opens, where a spark ignition will fire at the same time the gas valve opens. The gas valve will open and stay open for a couple of seconds before the control board even looks for flame. After a couple seconds for an ignition trial, the board will "look" for flame sense and keep the valve open, or if no flame is sensed go into lockout, or retry.

If your furnace locks out on flame sense failure and no flame has been present, you have other problems other than the flame sensor itself. It could be many, board, gas pressure, valve, even wiring could be a problem. Greg

[Bill]

Thanks for your informative and quick response, Greg.

Your description of my furnace cycle sounds exactly like what is happening. I don't know, however, that I have never noticed a spark ignition at the point the gas valve opens--even when it is functioning normally. I do have a distinct orange glow, though, that seems to describe a "hot surface ignition." That glow subsides at ignition or ignition failure.

Gas pressure has been measured both at the point it enters the furnace cabinet and inside the valve. Both pressures have read normally.

Is their a way to isolate whether or not the intermittent failure occurs from the gas side or the spark side?

Thanks.

Bill

[Bubba]

Are you sure you are telling us the whole story bill? You've let this supposed "furnace contractor" do all these parts replacements and still letting him back for more? Are you paying for all this? Bill, do yourself a favor and find a real company because the one you found isnt. Bubba

[Greg O]

Other than to throw parts at it, no. This is one of those times when patience is required. Somebody may need to just baby sit the darned thing for a while. Usually by jumping the stat connections at the furnace, and watching it go through it's ignition sequence with a voltmeter or two connected at the appropriate places a tech will find the problem. Sometimes it may take some time though! I had one I spent the better part of the day watching! A couple other techs had looked it over and gave up. Those intermittent problems can really be a problem sometimes. Greg

[Bill]

Greg, again thanks for your input. They have had meters on several of the circuits, but never had anything but normal readings. At the time, though, the furnace was functioning normally. It did fail twice while they were here, but they didn't have a meter hooked up at the time. One tech looked at the other at the first of the failures and said, "no gas." That's what caused them to put in a new valve. But within a day, I was having lock-outs again.

I think they're now grasping at sraws, however. They suspect a problem with the composition of the LP gas, and pointed to the green residue on the flame sensor. I don't see how that has anything to do with the ignition failures, particularly after you described the cycle to me. There certainly is no problem with the gas water heater sitting right next to the furnace.

Bubba, I have not paid anything for any of the parts or labor except for the first service call six weeks ago, or so. The furnace is only

28 months old and I have a ten year-warranty. I am concerned, though, that with their expressed concern about the quality of the LP gas, that they may be very frustrated and starting to "bail out" on me.

Thanks.

Bill

[Joseph]

Model# Serial# may be of help.

that keeps locking out for an hour at a time when it

The flame sensor does exactly what you would think it does. It senses current rectification through the flame to ground. The control module looks for that current after it opens the main valve. If it is not there then the system goes into lock out. Each unit is a little different but the basics are the same for all.

Could have other inert gases mixed in with it. None, have the LP guy come out and check it. Pockets of air in the line (wag). It sounds like permanent failure is imminent.

Is the guy working on it the one that installed it? You might want to call the Mfg and explain the situation to them and see if they can remedy the problem. It's hard to tell what other possible solutions there may be w/o being able to put my hands on it, I can only make a WAG. This may be a bad Molex plug, wire connector slowly deteriorating, piece of debris in gas line... etc...

Joseph

[Bill]

Joseph, thanks for your input on this. The furnace is a Trane XV80, TUD100R9U5K; Z291YLRIG

module looks

Is the current in the sensor all of the time, or only after the valve opens?

Yes, the same folks that installed it are the ones trying to repair it. What is a Molex plug?

Thanks.

Bill

[Bob Pietrangelo]

There is no spark sent anywhere on a Hot Surface System, there is not even a device that creates a spark on the furnace. The HSI may need to burn up to

45 seconds before voltage is sent to the gas valve. Is there voltage to the gas valve or does the sequence of ops end before that?

[Bob Pietrangelo]

GET A NEW CONTRACTOR

[Bob Pietrangelo]

You have a train furnace it it sounds like a pressure problem in the inducer or the pressur switch. Tell the company to come out with their manometerand t fittings if they come out again. This will have to be fixed by a qualified Trane dealer, not someone who is not versed in service. So far it seems you have had parts changers out there

[Bill]

Bob,

There is correct voltage to the gas valve at least 80% of the time and everything performs correctly; it is an intermittent problem. To the best of my knowledge, though,a reading has ever been checked during an ignition failure, though.

when this problem started, I strongly suspect that there is a signal problem between the board and the valve; the board, itself, has been replaced twice, the valve once. Are there connectors or switches between the valve and the board?

Thanks for your response. If you think of anything more, I'd really appreciate hearing about it. It locked out eight times over the weekend. I'm guessing that number represents almost a third of the total lockouts since early January.

Bill

[Bob Pietrangelo]

Sorry, TRANE

[Bill]

The Trane dealer's technicians came out again today. This time, luckily, they had a meter hooked up to the new valve when the furnace failed to ignite. They listened to the sequence of clicks and discovered that the solenoid didn't "click" when it should have when the voltage to the solenoid was normal during this period . They replaced the five day-old valve with another new valve.

I'm keeping my fingers crossed that they have finally solved the problem.

Thanks to everyone who helped me with this.