Results 1 to 11 of 11
  1. #1
    Join Date
    Oct 2011
    Posts
    8

    Furnace Intermittent Pilot Problem

    I am troubleshooting an LP-fired furnace (Dayton 3E384) with electronic spark ignition. The unit uses a Honeywell S8600M Control Module and a Honeywell VR8104A Gas Valve. When the thermostat calls for heat, the control module initiates continuous sparking at the pilot and the pilot valve opens. But the flame is unstable: flickering, and repeatedly extinguishing and re-igniting. During this time, I can occasionally hear some faint clicking from the gas control valve. This continues until the flame sensor finally gets hot enough to signal the control module to open the main valve. As soon as that happens, the main burners light-off, the control module stops the sparking, and the pilot becomes a steady normal flame.

    I suspect that, while the control module is generating the spark, the pilot valve is not getting enough power to open fully and that the valve is rapidly pulsing open and closed as the spark generator draws current to produce each spark. As soon as the control module stops the spark generator, then there is sufficient power to hold the pilot valve fully open and the pilot flame becomes normal. But I am unsure if the problem is in the control module or the gas valve. The system is rated for nominal 24VAC and I am measuring greater than 24VAC at the control module's power input terminals and also at the gas pilot valve terminals (PV & MV/PV). However, I am making those measurements with a Fluke digital multimeter, and although I set the meter to use manual ranging, it's not able to detect/display high-frequency voltage transients (i.e., rapid drops in power each time a spark is generated). I have also measured the resistance between the PV & MV/PV terminals, and the MV & MV/PV terminals, and both readings were essentially identical (i.e., both the PV and MV solenoid coils have the same resistance).

    To eliminate the possibility of problems with the wiring, high-limit switch, etc., I connected jumper wires directly from the power supply transformer to the PV terminals on the gas valve (bypassing everything, including the control module). The symptoms are identical: intermittent & irregular flame while the control module is generating spark; normal flame after the MV opens and the spark generator ceases.

    So either the gas pilot valve solenoid is "weak" (and thus too sensitive to fluctuations in the power), or the control module is defective (and is drawing too much amperage during spark generation, thus robbing the gas control valve of enough power to stay fully open). I suppose a third possibility is that the transformer is "weak" and not able to keep up with the current demands. But I think that is unlikely.

    Any assistance in narrowing down this problem would be greatly appreciated.

  2. #2
    Join Date
    Aug 2007
    Location
    Northern VA 38 degrees N by 76 degrees W
    Posts
    5,058
    Quote Originally Posted by SpyGuy View Post
    I am troubleshooting an LP-fired furnace (Dayton 3E384) with electronic spark ignition. The unit uses a Honeywell S8600M Control Module and a Honeywell VR8104A Gas Valve. When the thermostat calls for heat, the control module initiates continuous sparking at the pilot and the pilot valve opens. But the flame is unstable: flickering, and repeatedly extinguishing and re-igniting. During this time, I can occasionally hear some faint clicking from the gas control valve. This continues until the flame sensor finally gets hot enough to signal the control module to open the main valve. As soon as that happens, the main burners light-off, the control module stops the sparking, and the pilot becomes a steady normal flame.

    I suspect that, while the control module is generating the spark, the pilot valve is not getting enough power to open fully and that the valve is rapidly pulsing open and closed as the spark generator draws current to produce each spark. As soon as the control module stops the spark generator, then there is sufficient power to hold the pilot valve fully open and the pilot flame becomes normal. But I am unsure if the problem is in the control module or the gas valve. The system is rated for nominal 24VAC and I am measuring greater than 24VAC at the control module's power input terminals and also at the gas pilot valve terminals (PV & MV/PV). However, I am making those measurements with a Fluke digital multimeter, and although I set the meter to use manual ranging, it's not able to detect/display high-frequency voltage transients (i.e., rapid drops in power each time a spark is generated). I have also measured the resistance between the PV & MV/PV terminals, and the MV & MV/PV terminals, and both readings were essentially identical (i.e., both the PV and MV solenoid coils have the same resistance).

    To eliminate the possibility of problems with the wiring, high-limit switch, etc., I connected jumper wires directly from the power supply transformer to the PV terminals on the gas valve (bypassing everything, including the control module). The symptoms are identical: intermittent & irregular flame while the control module is generating spark; normal flame after the MV opens and the spark generator ceases.

    So either the gas pilot valve solenoid is "weak" (and thus too sensitive to fluctuations in the power), or the control module is defective (and is drawing too much amperage during spark generation, thus robbing the gas control valve of enough power to stay fully open). I suppose a third possibility is that the transformer is "weak" and not able to keep up with the current demands. But I think that is unlikely.

    Any assistance in narrowing down this problem would be greatly appreciated.
    Is it lighting within 4 seconds of the pv receiving current? Check gas pressure.

  3. #3
    Join Date
    Jan 2009
    Location
    Central Iowa
    Posts
    213
    It sounds more like a gas problem than a control problem. Check your pressures with the valve open and closed. Also you might pull apart the pilot assembly and clean it out to make sure you dont have a chunck of dirt in the pilot nozzle.
    The most obvvious thing to check though... Has the furnace been converted for use with LP. If it's new, it may not have been.

  4. #4
    Join Date
    Oct 2011
    Posts
    8
    Quote Originally Posted by second opinion View Post
    Is it lighting within 4 seconds of the pv receiving current?
    The pilot lights off almost immediately after the PV receives current. But it is not a stable, continuous flame during the time that the spark generator is functioning. It's as if the pilot is sputtering the gas.


    The main burner lighting time depends on how cold or hot the pilot ignition sensor is when the thermostat calls for heat. For example, if the sensor is cold, then the system may go through more than one 90 second ignition trial cycle to get the sensor hot enough for the control module to "think" that the pilot is lit. Realize that it takes a while for the sensor to accumulate enough heat because the pilot flame is constantly flickering on and off during the pilot ignition cycle. As soon as the sensor is heated enough by the flickering pilot for the control module to "think" that the pilot is lit, then the control module signals the MV to open and then the main burners light off and the spark generator ceases. When that happens, the pilot becomes stable and normal. If the pilot flame sensor is already hot (e.g., the unit has been operating, I shut it off, then call for heat again), then the main burners will light off almost immediately after the thermostat calls for heat. The longer the pilot flame sensor has to cool off (more rest time between shut down and next call for heat), then the longer it takes for the sensor to again get hot enough to signal the control module to open the MV.

    Quote Originally Posted by second opinion View Post
    Check gas pressure.
    Quote Originally Posted by blabath View Post
    It sounds more like a gas problem than a control problem. Check your pressures with the valve open and closed.
    When I first saw the flickering pilot, my initial thought was that the gas pressure was too high. But as soon as the MV opens and the spark generator turns off, the pilot becomes steady and normal. This could be explained if the high gas pressure were being reduced by the opening of the MV. But it was my understanding that the pilot gas pressure is independent of the main burner pressure (otherwise, when the MV opens, the sudden drop in pilot pressure could cause a normal pilot flame to extinguish). Furthermore, when the main burners are lit, both the pilot flame and the main burner flames are all normal and correct. But I will double-check the gas pressures.

    Quote Originally Posted by blabath View Post
    Also you might pull apart the pilot assembly and clean it out to make sure you dont have a chunck of dirt in the pilot nozzle.
    An obstructed pilot would not explain the symptoms I'm seeing. During the ignition sequence, the pilot is not weak or lazy: it's more like puffing little spurts of gas that "pop" when the spark hits them. The puffing seems to be occurring at the same frequency as the spark generating cycle, which is why I'm thinking that the PV solenoid is fluttering while the spark generator is operating. I did check the pilot when I performed the system tune-up before resuming operation of the furnace, but I'm willing to double-check the pilot.

    Quote Originally Posted by blabath View Post
    The most obvvious thing to check though... Has the furnace been converted for use with LP. If it's new, it may not have been.
    The furnace is not new, however it has been out-of-operation for some time. This is installed in an industrial building and was installed by a previous tenant. After that tenant vacated, the propane tank was removed and the system was not used for some time (I don't know how long it was out of service, but I believe it was more than a couple of years). The new tenant wanted to use the furnace and recently had a new LP tank installed at the site.

  5. #5
    Join Date
    Mar 2010
    Location
    Arnold, Mo
    Posts
    395
    If you are getting more than 24V to the pv/mv then I think your transformer is fine. It sounds to me like the pilot valve solenoid is bad.

  6. #6
    Join Date
    May 2004
    Location
    SE Pennsylvania
    Posts
    898
    The flame sensor does not work on the heat of the flame, it is only a steal rod with a ceramic insulator. The flame sensing circuit depends on a current passing through the flame to the rod basically closing a circuit. If the flame is lazy the current will not be constant. When your main burner ignites it is by chance that the flame stayed steady long enough to provide the flame sensing circuit with a 2΅A or greater flame signal. The pilot flame needs to be steady. Rig up a tee on the pilot line between the pilot and the valve and check the pilot pressure to see it if it is steady. If it isn't steady check the inlet pressure to the main valve (not manifold pressure), should be 11" WC. If pilot pressure is steady, clean the pilot assembly and the orifice or replace, adjust the pilot pressure until it is steady without the main burner lit.

  7. #7
    Join Date
    Apr 2008
    Posts
    71
    If your pressure's are good and you wired it directly to the transformer and its still acting the same its a weak solenoid.As long as your pilot and oriface is clean too.

  8. #8
    Join Date
    Oct 2011
    Posts
    8
    Quote Originally Posted by hvac69 View Post
    The flame sensor does not work on the heat of the flame, it is only a steal rod with a ceramic insulator. The flame sensing circuit depends on a current passing through the flame to the rod basically closing a circuit. If the flame is lazy the current will not be constant. When your main burner ignites it is by chance that the flame stayed steady long enough to provide the flame sensing circuit with a 2΅A or greater flame signal. The pilot flame needs to be steady.
    Thanks for the clarification. I was curious how the flame sensor worked and had assumed that it contained a thermocouple in the rod that was back-feeding DC current over the ignition cable to the control module. The fact that it would consistently sense the pilot faster when the pilot assembly was hotter reinforced my assumption.

    So if I understand your description, the control module is trying to sense electrical current flow through the ionized gas of the pilot flame in the time interval between ignition sparks? And the fact that the hotter assembly would allow faster detection is because the hot electrode is more conducive to electrical current flow?

    I would not characterize the flame I'm observing as "lazy"; that would describe a pilot with insufficient gas pressure. I think the best way to describe it is "pulsed" (rapid little on/off jets of gas) that seems to be caused by fluttering of the pilot valve as the control module draws power to generate each spark. As soon as the control module stops generating spark, the pilot becomes steady and continuous with a proper flame profile. I'm still not sure how these observations could be explained by improper gas pressure, although I will test the pressures. Two other tests I would like to perform to try to isolate the problem:

    1. Disconnect the ignition lead from the pilot assembly and connect to a spark gap (so the control module will not be detecting the pilot). Disconnect power from the control module and disconnect the MV. Jumper power to the PV, bypassing the control module. Manually light the pilot and observe the flame. If normal (as I suspect it will be), then apply power to the control module to start the spark generator and see if the pilot flame becomes "pulsed" again.

    2. If the above test repeats the symptoms, then apply an independent 24VAC power supply to the PV (while the control module is generating spark) and see if that returns the pilot flame to normal.

    Neither test will give me a definitive answer as to whether the PV solenoid is weak or if the control module is drawing too much power. But my bet is on the gas valve being bad.

    Because of Thanksgiving and other commitments, I probably won't be able to get back to this job until next week.

  9. #9
    Join Date
    Nov 2006
    Location
    Southeastern Pa
    Posts
    17,921
    I don't believe the pilot solenoid is being starved for power while a spark is being created.

    The module uses one metal rod to make a spark, and a second to sense a pilot flame using rectification.

    Are you a tech or a building owner?
    [Avatar photo from a Florida training accident. Everyone walked away.]
    2 Tim 3:16-17

    RSES CMS, HVAC Electrical Specialist

    AOP Forum Rules:







  10. #10
    Join Date
    Oct 2011
    Posts
    8
    Quote Originally Posted by timebuilder View Post
    Are you a tech or a building owner?
    I'm a tech. (Wishing I was the property owner! )

    Quote Originally Posted by timebuilder View Post
    I don't believe the pilot solenoid is being starved for power while a spark is being created.
    It does seem strange, but it's the only thing I can think of at the moment that would result in the symptoms I'm observing. Granted, the gas valve may not actually be starved for power per se; it could be just that the PV solenoid is marginal and is not fully opening when the control module is drawing extra power to generate the spark.

  11. #11
    Join Date
    Nov 2006
    Location
    Southeastern Pa
    Posts
    17,921
    You should be able to see enough change in the pv voltage during spark if your hypothesis is correct.
    [Avatar photo from a Florida training accident. Everyone walked away.]
    2 Tim 3:16-17

    RSES CMS, HVAC Electrical Specialist

    AOP Forum Rules:







Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  
Comfortech Show Promo Image

Related Forums

Plumbing Talks | Contractor Magazine
Forums | Electrical Construction & Maintenance (EC&M) Magazine
Comfortech365 Virtual Event