[FrFtr28]

I have an older, what I think is early 90's, Armstrong Air natural gas furnace. Everything seems to be working until the spark to ignite the pilot gas is called. I have good spark, but it only lasts for maybe a half second and fails to ignite the pilot gas. I have removed the pilot assembly and cleaned everything, so I don't think that's the problem. I am able to manually light the pilot, which I know can be dangerous, after a few attempts.

Any ideas on what I should test? I am handy with a multimeter, I just don't know what exactly I should be testing.

 

[yuri]

Post some pics of the burner area and the control board as we need to know what system it uses.

Does it have a Honeywell S86 ignition control with a red wire running to the pilot assembly?

Go Advanced>Manage attachments to post pics

 

[FrFtr28]

Post some pics of the burner area and the control board as we need to know what system it uses.

Does it have a Honeywell S86 ignition control with a red wire running to the pilot assembly?

Go Advanced>Manage attachments to post pics

I'm at work right now, but I can post what pictures I have on my phone already. Here's one from when I was doing some troubleshooting before. In the photo you can see the side of the control board with the wires connected, I had it removed when I was troubleshooting. The top orange wire runs to the spark igniter on the pilot assembly. The control board is a Robertshaw Model SP735.

 

[yuri]

Yeah I am VERY familiar with that control and the S86 which does the same job.

As long as you got 24 volts steady to it from W then it should produce a good spark. I am 99% sure at that age the spark transformer inside the control has died from use/old age. If you feel brave and wear gloves you can take the ignition wire off and hook a piece of wire to the probe and hold it close to metal and it should give you a strong 1/2" spark or a jolt if you are not careful. That rules out your pilot burner. 99% of the time the control is faulty. You should be able to easily get the same one from americanhvacparts.com or supplyhouse.com. You can convert it to the S86U universal but it canbe tricky to figure out the new wiring. I would get another Robertshaw.

 

[FrFtr28]

Yeah I am VERY familiar with that control and the S86 which does the same job.

As long as you got 24 volts steady to it from W then it should produce a good spark. I am 99% sure at that age the spark transformer inside the control has died from use/old age. If you feel brave and wear gloves you can take the ignition wire off and hook a piece of wire to the probe and hold it close to metal and it should give you a strong 1/2" spark or a jolt if you are not careful. That rules out your pilot burner. 99% of the time the control is faulty. You should be able to easily get the same one from americanhvacparts.com or supplyhouse.com. You can convert it to the S86U universal but it canbe tricky to figure out the new wiring. I would get another Robertshaw.

I am getting a good spark now, it just doesn't spark for long enough I believe. Before it would spark for maybe 2 seconds, whereas now it only sparks for a fraction of a second. When I test voltage on the wire that controls the pilot valve, it only gets voltage for maybe a fraction of a second as well. Is that long enough?

 

[user_12345a]

edit: this was written b4 your reply.

check for 24v incoming to the module and if it's consistent change the module.

-----------------------------------------------------

The ignition module should be mounted and if u post the model number we could see what the trial period is. (would be more than half a second on any module)

When it's trying to ignite you have to check voltage between common and pilot on the valve, should be ac volts 24 +/-.

If the voltage applied to the pilot valve drops off when the spark does, something is going on with the module or it's power supply. If not could be an issue with connections.

If the valve is dropping out check the power supply to the module (terminals will vary - could be something like th and thc)

 

[FrFtr28]

edit: this was written b4 your reply.

check for 24v incoming to the module and if it's consistent change the module.

-----------------------------------------------------

The ignition module should be mounted and if u post the model number we could see what the trial period is. (would be more than half a second on any module)

When it's trying to ignite you have to check voltage between common and pilot on the valve, should be ac volts 24 +/-.

If the voltage applied to the pilot valve drops off when the spark does, something is going on with the module or it's power supply. If not could be an issue with connections.

If the valve is dropping out check the power supply to the module (terminals will vary - could be something like th and thc)

The ignition module is a Robertshaw SP735. It is mounted currently, I just had it off when I took the picture. The voltage applied to the pilot valve is so quick that it doesn't register on my digital multimeter, but I can see the needle move when I use my analog.

Also, I can get the burners to light by applying heat to the flame sensor on the pilot assembly bypassing the pilot. Once the main burners light, the pilot also stays lit with a nice blue flame. That's why I am thinking the problem is either the control board or the igniter itself. I also read somewhere that a bad ground could be the culprit.

 

[user_12345a]

Ground affects spark and flame sensing.

If it was a grounding issue or even an issue with the spark circuit outside the module , the pilot valve would stay open for the full trial period the module permits regardless.

Check the power to the module for the heck of it, but i'm betting it's the control.

You don't need to use a direct robert shaw replacement, just an equivalent with the same connectors such as -> http://www.americanhvacparts.com/p-...s-robertshaw-obsolete-sp734-sp735-lennox.aspx

Also, I can get the burners to light by applying heat to the flame sensor on the pilot assembly bypassing the pilot.

It's not the heat that does it, current passes through the flame from the flame sensing rod to ground.

Further proves it's not the ground and has to be the module.

 

[FrFtr28]

Ground affects spark and flame sensing.

If it was a grounding issue or even an issue with the spark circuit outside the module , the pilot valve would stay open for the full trial period the module permits regardless.

Check the power to the module for the heck of it, but i'm betting it's the control.

You don't need to use a direct robert shaw replacement, just an equivalent with the same connectors such as -> http://www.americanhvacparts.com/p-...s-robertshaw-obsolete-sp734-sp735-lennox.aspx



It's not the heat that does it, current passes through the flame from the flame sensing rod to ground.

Further proves it's not the ground and has to be the module.

I'm leaning towards the control as well, but I wanted to get some guidance before I spent that kind of money on such an old furnace that is only a backup to my wood furnace as it is.

When you say check power to the module, I'm looking for a constant 25 volts to it, correct?

 

[yuri]

The control has a 30 second trial trial for ignition. It needs to be powered continuos so it goes thru that trial for ignition. While it is sparking it has to energise the pilot valve or it won't have gas to light. If it is getting a steady 24 volts from W thru the pressure switch then it should spark and keep the pilot valve open. The relay for the pilot valve and what turns on the spark may be failing. Looks like the control to me.

Has nothing to do with grounding.

The control needs to be mounted, however if you have a green wire from the pilot burner to the control it will sense the flame.

 

[FrFtr28]

The control has a 30 second trial trial for ignition.

Does that mean the pilot gas valve should be open for 30 seconds? If not, how long should the pilot valve be open? It seems like the pilot valve does not open long enough for gas to reach the spark.

The control needs to be mounted, however if you have a green wire from the pilot burner to the control it will sense the flame.

No ground wire, just the orange wire from the control and a white wire from the flame sensor to the control. There is a ground wire from the control to the body of the furnace, however.

 

[yuri]

The control opens the pilot valve and turns on the spark at the same time and goes for 30-60-90 seconds depending on module and brand B4 it locks out or retries.

The control must be mounted and there is a screw which goes thru it on a copper ring or tab ( depends on brand etc ) to ground it. How it works is the signal on those goes thru the hood of the pilot burner thru the furnace metal and back to the control thru grounding. If it has a green wire from the pilot burner to the control that helps. Not all do. Some have a green wire from the control to ground near it.

However if hanging in the air it will still spark and light the gas it just won't sense the flame and will lockout.

At the end of the day if you got steady 24 volts to it from W it should spark for 30-60-90 seconds and open the pilot valve. If not replace the module with a Roberstshaw. . There are different wire terminals and terminology between the varies S86 series and Robertshaw and you have to get the timing settings proper. Not easy to decipher that from the Honeywell info unless you want a headache. If it is obsolete you got no choice. Very rare I run into those old units where I am as most have been replaced due to cracked heat exchangers.

Like I said try run a wire from the spark terminal and see if it sparks strong to ground. If it does your igniter porcelain may be cracked or the pilot burner badly corroded or warped. Not often that happens but it can. Most of the time it is a bad control. Should still open the pilot valve though.

 

[user_12345a]

that's an 80% induced draft furnace - if the heat exchanger is fine it's worth the cost of the module.

diy it and it's not that much money - could probably find one for $150 or something. A pro would charge closer to $500.

 

[FrFtr28]

Thank you both for your help. I'm going to try an do some more troubleshooting tonight if I get a chance.

I did measure the voltage at the TH connection on the module and it was getting a constant 24 volts and never dropped off. I will run a wire from the spark terminal to ground and see if it sparks. I will post back with any new info.

 

[user_12345a]

don't even bother checking the spark, with pv not getting voltage for the entire trial period the module is bad for sure.

no need to go further.

 

[user_12345a]

You can look up to robert shaw model and cross reference it.

They all work the same way, have the same connections which do the same things (with different terminal designations - have to match them up)

If the terminal sizes are different you can buy the connectors and crimper for not much. Would have to cut, restrip the ends, crimp the new ones on.

No big deal but takes time.

 

[user_12345a]

They all work the same way, have the same connections which do the same things (with different terminal designations - have to match them up)

Not sure what i was saying - actually not quite the same, as yuri stated.

Compare the pre-purge delay specs; the inducer has to run a bit before the burners light.

You can get a module with a delay equal to or greater than what u have now.

 

[supers05]

They will be listed as replacements for specific models of igniters. This is what keeps the csa or ul listing valid. If this matters to you, and it should, make sure yours matches the replacement.

Cheers!

 

[user_12345a]

The same could be said for motors, relays, hot surface ignitors, etc, etc.

i don't personally care as long as the part itself is a safe fit - works in the same way (delays, etc), isn't a down-grade and has the approvals itself. (ul, csa, whatever they use)

a matter of personal choice

You could end up shelling out a lot more than you have to for oem style parts.

 

[supers05]

ESA (and related organizations) are much more flexible when it comes to replacements. Match ratings, and have a listing number, and you're good.

For fuels, those organizations are anal. Insurance companies are just as bad when it's fuel related. Robert Shaw, along with a bunch of others offer cheap(ish) listed aftermarket stuff. You just have to find the right one that matches. Some of those Universal ones come with a list of model numbers longer then this thread as being listed to replace with.

Sent from my SM-G870W using Tapatalk