Guesses on what this component "was"... "VR" on PCB

[fastline]

Of all wonderful things, I am working on m furnace tonight. The ignitor module is acting up. This unit has the" internal" sensor by using the spark wire and checking for uA on the ground strap. I verified that the current is there but for some reason, pilot will start, flame is good, will start to light the main burner, then just shut down and repeat this endlessly.

I finally just jumpered over to 24V for the main gas valve so I can get some heat in the house but I will get no sleep babysitting that.


I pulled the PCB for the ignitor and all testing fine except some unknown component. All I could find on it was "68A" and PCB is marked as "VR" with no polarity. Device is a typical black axial lead component probably 4x8mm or so. The component tested open with continuity and diode tests. I decided to just break it open to try and understand what it was and looks like a thin wafer between axial leads. Like a black/green disc about 1mm thick.

I am going to have to guess what the heck this was and try something. I put the PCB back in so I can by pass it and get heat and still running like before without that component. Not sure if that was really the issue or not. I was about to just put a 10K resistor in there but something tells me I should check with sharper electronic minds before possibly toasting it. Just cannot tell what it was supposed to do... It seems to have had connection between primary and secondary coils for the boost coil for the spark. There is, however, another high voltage boost coil as well.

 

[tunedwolf]

Just going by your description, I would hazard a guess and say it was a MA68A Varistor.

A quick search on the web finds a datasheet: https://www.littelfuse.com/products...tors/Littelfuse_Varistor_MA_Datasheet.pdf.pdf

If it looks anything like that, then I would say it's a fair bet

 

[fastline]

I did find that data shet earlier but these look like typical barrel type cylinders very similar to a diode but there is no banding or anything that would indicate a diode.

IIRC though, a Varistor should NOT conduct unless somthing shorts them which takes down the whole circuit?

 

[ronsimpson]

There is a chance that the original design used a MA68A and in production a different part was substituted in.

 

[fastline]

I probably don't have anything else to go on so might as well accept that as fact....lol

so, any idea how that varistor should bench test? IIRC, they should read open? Will only conduct at high voltages?

I am still at a loss as to what the thing would even do in this circuit. Seems to work exactly the same without it. My only experience with them as MOVs is for surge suppression and this appears to be smack in the middle of the circuit. IIRC, they usually fail exploded and this one is intact.

 

[kinarfi]

I was recently having problems with my furnace and tried a new thermostat, nope, cleaned the igniter, nope, so I pulled the circuit board again and use a weller soldering gun to liquefy all the solder joints again, they all looked good but if I wiggled the wires that plugged into it, it would drop the flame. After liquefying EVERY solder connection on the board, I have had no further problems.

 

[dr pepper]

When you switch a coil of nearly any type there is an inductive kick back, which can be 1000's of volts, instead of using an active device that can take 1000's of volts its easier to clamp the kickback to a maximum known voltage with a varistor.
During its life it has to soak up lots of energy spikes so it is something in my opinion that fails.

 

[cbennett]

I would look for a diagram/schematic of the furnace. The service manual for my furnace shows a variety of safety measures that will shut down the furnace if tripped, such as duct air back pressure sensors, thermocouple to make sure the flame is lit, (main burner+pilot?), access panel open switch, etc. Maybe Google can help if you don't have a manual.

 

[fastline]

Thanks for the toughts guys. Kinarfi wins this one I guess. I was highly suspect that the module was not really failing but was losing contact to the gas valve for some reason. Wiggling wires was doing nothing. However, sometimes would work, sometimes not. After the post of solder joints, I told myself I had already looked but I have nailed several board issues over the years by spotting very small solder joint cracks. Sure enough, there it was, relay pins. they typical round cracking from cold or weak joints. Ran out to the bench an touched it up. Seems to be running perfect now. I feel confident in th repair. However, that varistor is now missing. Not sure if that will eventually take out the module or not. I hate to put something in there as a guess and kill the circuit but... The whole system is old so probably only about 2yrs left before full replacement anyway if it can last that long.

 

[kinarfi]

Glad to hear I helped, thanks for come back