Pilot sensor corrosion - chronic problem?

prwood

I've previously posted about a problem where our boiler failed to ignite, and with help from the forum, we tracked it down to apparent corrosion on the pilot flame sensor, which prevented the boiler from detecting successful pilot flame ignition. I cleaned it off using steel wool, and it was successfully able to ignite once again:

https://forum.heatinghelp.com/discussion/181965/boiler-occasionally-fails-to-ignite

The boiler had been installed in 10/2016 and had been running fine up until the initial issues in 12/2020. My initial cleaning of the sensor was on 12/20/2020, and it had been running fine up until 3/6/2021 (roughly 7 weeks). However, we just had a couple of issues again this weekend where it failed to ignite for long periods. Basically the heat would eventually kick in after 3-4 hours of the thermostat calling for heat. Upon inspection I found that the pilot sensor was mostly covered with light orange deposits. I cleaned it off with steel wool again and it has been igniting properly for the past two days.

Are there any potential causes for this type of corrosion on the pilot flame sensor, especially ones that might crop up after just 7 weeks? If there's something that could cause this to happen I'd like to address it, rather than just cleaning the sensor off when it happens.

JUGHNE

You could try to clean the pilot burner itself and the flame "hood" on it.

Jamie Hall

Is it really corrosion, or could it be a surface deposit? How are you for dust? Any sodium containing chemicals around (pool disinfectant, road salt, that sort of thing)?

Hap_Hazzard

Can you post a picture of what the pilot flame looks like?

prwood

Jamie Hall said:

Is it really corrosion, or could it be a surface deposit? How are you for dust? Any sodium containing chemicals around (pool disinfectant, road salt, that sort of thing)?

Based on the orange color I assumed it was rusty corrosion, but it could be something else.

The basement where the boiler is located is quite dusty.

I don't know of any sodium-containing chemicals, but I will check the area.

prwood

Hap_Hazzard said:

Can you post a picture of what the pilot flame looks like?

I'll see what I can do.

SteamingatMohawk

What is the texture of the orange material, hard, crusty or soft like a dust coating?

There are stories on the internet about contamination in the air that can affect the flame, which might also pertain to the pilot thermocouple (assuming that is what you are referring to). Things like paint fumes from paint drying out either in projects or open cans drying. Another story about kitty litter (ammonia) affecting the flame and when the person moved the kitty litter away the color changed. Is the laundry nearby? If so, has anything changed there?

Some boilers (mine is an HB Smith) use a high temperature thermocouple, You might want to check that out, too.

Contamination in the gas, but that's probably unlikely.

SteamingatMohawk

Before I knew about needing the high temperature thermocouples, mine failed in about a year.

JUGHNE

This is not a thermocouple, it is a flame sensor in a flame rectifier safety system.
The flame sensor is one side of the circuit and the pilot hood is the other side.
The hood and especially the edges of the hood that face the sensor must be clean also.
And as said in your last post the grounding circuit from the pilot to the ignition module must have good contacts. Just loosing and resetting screws or push on connections will cut any oxidation away.

prwood

SteamingatMohawk said:

What is the texture of the orange material, hard, crusty or soft like a dust coating?

There are stories on the internet about contamination in the air that can affect the flame, which might also pertain to the pilot thermocouple (assuming that is what you are referring to). Things like paint fumes from paint drying out either in projects or open cans drying. Another story about kitty litter (ammonia) affecting the flame and when the person moved the kitty litter away the color changed. Is the laundry nearby? If so, has anything changed there?

Some boilers (mine is an HB Smith) use a high temperature thermocouple, You might want to check that out, too.

Contamination in the gas, but that's probably unlikely.

The orange coating was very thin, smooth, and light. Not crusty. It wasn't light enough to come off with an air blast, but it did come off easily with steel wool.

Actually it's interesting that you mention paint fumes, because we do use the basement for painting projects several times per year. For example, over the past several weeks we have been removing shutters from our house in batches, bringing them to the basement, painting them and letting them dry down there before reinstalling. That being said, just prior to the initial incident in 12/2020, we weren't doing any similar projects, so I'm not sure if that's the culprit. I don't think we generally keep open paint cans but we do let things dry out down there. It's one big room so it would share the same air with the boiler.

Laundry is on the floor above the boiler on the other end of the house, although it is at the top of the basement stairs, and the door is frequently open.

prwood

Hap_Hazzard said:

Can you post a picture of what the pilot flame looks like?

I tried a few angles, but I don't have any way to get a direct photo or video of the pilot flame without getting my hand or my device scorched, since the main burner is igniting right away at the moment. I took a few videos from the closest angles I could get and posted them here:

https://photos.app.goo.gl/vm9NbnjGHpCAqiQy5

I'm not sure how helpful they'll be since you can't see the flame directly. There's no way to get my phone up between the flame tubes to have a view of the pilot flame and get it out in time before the phone or my hand gets scorched by the main burner. :-/

unclejohn

Does the dryer vent leak air into the basement? That could be the cause.

mattmia2

It is probably better to use scotchbrite or emery cloth. Steel fibers from steel wool will embed in the probe and i suspect that is what your orange coating is. If it is the problem or if you just wiggled something else that is making a poor connection, i'm not sure

prwood

unclejohn said:

Does the dryer vent leak air into the basement? That could be the cause.

Possibly. The dryer, the vent hose, and the vent are all at street level, but the dryer does sit right next to the stairs to the basement, so if air was leaking from the dryer, some of it could potentially make its way down the stairs.

SteamingatMohawk

A couple of ideas.

If your dryer discharges outdoors, it is pulling outside air into the house somewhere. With it close to the basement, the air could be drawn into it, but unless you changed things since you put in the boiler, it probably isn't a factor.

How about increasing air tightness elsewhere in the house?

How about increasing the moisture content in the house?

I hope the steel wool you are using is not a Brillo pad.

Stainless steel steel wool is fairly easily obtainable. It won't rust and you can use it to keep animals out of openings to your house without it falling apart or staining anything.

Here's a wild idea. If the casing on the thermocouple has corroded and there is pinhole or crack in it, it may be possible for something to get inside and short it out. Probably pretty unlikely, but installing a new thermocouple may eliminate the problem. If you do change it, you could take a resistance reading on the new one before you install it and the old one after you take it out. Who knows what it will tell you.

If it's the thermocouple, I suggest you get a spare, so if you have a problem some time down the road, you have it already on hand...these things tend to happen at the worst time. I learned that the hard way.

Hap_Hazzard

Is this a standing pilot or automatic ignition? I think we've all been assuming it's a standing pilot, but those videos look and sound more like an auto-ignition system.

prwood

SteamingatMohawk said:

A couple of ideas.

If your dryer discharges outdoors, it is pulling outside air into the house somewhere. With it close to the basement, the air could be drawn into it, but unless you changed things since you put in the boiler, it probably isn't a factor.

How about increasing air tightness elsewhere in the house?

How about increasing the moisture content in the house?

I hope the steel wool you are using is not a Brillo pad.

Stainless steel steel wool is fairly easily obtainable. It won't rust and you can use it to keep animals out of openings to your house without it falling apart or staining anything.

Here's a wild idea. If the casing on the thermocouple has corroded and there is pinhole or crack in it, it may be possible for something to get inside and short it out. Probably pretty unlikely, but installing a new thermocouple may eliminate the problem. If you do change it, you could take a resistance reading on the new one before you install it and the old one after you take it out. Who knows what it will tell you.

If it's the thermocouple, I suggest you get a spare, so if you have a problem some time down the road, you have it already on hand...these things tend to happen at the worst time. I learned that the hard way.

I'm using fine grade stainless steel wool. I could switch to using emery cloth, as someone else had mentioned that steel wool could leave deposits on the sensor.

This house was built in the 1860's and there isn't much airtightness anywhere, so air could be getting into and out of the house from anywhere. However, I could probably take a look and see if there are any major leaks that could be easily plugged.

Is the thermocouple the same thing as the flame sensor? I was just reading that the term 'thermocouple' is typically used on systems that have a standing pilot light that needs to be monitored, while 'flame sensor' is used for electronically controlled ignitions to monitor whether a flame had started. (reference) In any case, if my boiler has both of them, I'm not sure where the thermocouple is. The last time I looked at it, the flame sensor housing didn't have any cracks in it and it looked in very good shape, other than the deposits I cleaned off.

Hap_Hazzard

Yeah, I think we've all been assuming you have a standing pilot and a thermocouple. Thermocouples are bathed in flame all season long, so any contaminants in the air can get deposited on the thermocouple. Normally, though, they stay pretty clean. I've never cleaned mine. I don't have any experience with electronic ignition systems, but hopefully our pros can help now that we know what you're working with.

prwood

I apologize for the confusion. The sensor I am talking about is called the "pilot sensor" in the documentation for the boiler, which may have misled people into thinking my boiler has a standing "pilot light". However, what it's sensing is the pilot flame which is controlled by the electronic ignition. Once the sensor picks up that the pilot flame has started, it opens the valve for the main burner, then shuts off the pilot flame. So it's only exposed to flame during the ignition process, and otherwise it sits just underneath the main burner flames.

mattmia2

A thermocouple generates a voltage form the heat of the pilot flame and both terminals connect at the gas valve so it is very unlikely to be affected by dirt or corrosion at the pilot burner end. Corrosion or dirt can rather easily affect the pilot burner itself but not the themocouple.

This isn't what you have.

You have some sort of intermittent pilot spark ignition system. It appears you have a separate spark electrode and flame rod. The flame rod uses the plasma state of the air in the flame as a rectifier to pass a current to the control to sense when the pilot is lit. All the connections from the control to that rod and from the pilot burner back to the ground of that control need to be clean and tight.

Some pictures of the gas valve and the control would let us say exactly what you have. There is a button at the top of the editor box to attach images, if you attach them inline more people will look at them than if you post a link.

Hap_Hazzard

Not your fault. We tend to think in terms of what we're familiar with. Since this is such a short-duration flame, it makes me think that you really are seeing corrosion instead of combustion by-products. The only thing I can think of that would cause that is chlorine, but you would have noticed a smell if that was the case. I'm straining my brain to think what might be doing this to the flame sensor but not the flame hood or any other bare steel in your basement, but I'm drawing a blank.

Do you know what the humidity is down there?

mattmia2

I think the problem is elsewhere, either one of the connections in the circuit or something like a solder joint in the control itself, you just happened to wiggle whatever it is in to conducting when you cleaned the rod. Also make sure the flame rod and its wire isn't shorting to the burner or the cabinet anywhere.

Is the pilot supposed to turn off when the main burner lights? The system needs to prove flame somehow the entire time the main burner is on. There could be a port in the main burner that is clogged or the main burner could be underfiring causing it to no longer encompass the flame rod when the pilot burner turns off if the pilot burned is supposed to turn off once the main burner lights.

JUGHNE

Did you get and change the pilot assembly as you mentioned in your last posting?
It might be worth having a spare on hand.

The pilot IIRC, will always stay on.
The flame sensor must always be in the flame to keep the gas on.
If, when the main gas valve opens, it "snuffs" out the pilot it may then lose the flame signal and shut down.

Can you post pictures of the controls and the wiring diagram included?

prwood

mattmia2 said:

I think the problem is elsewhere, either one of the connections in the circuit or something like a solder joint in the control itself, you just happened to wiggle whatever it is in to conducting when you cleaned the rod. Also make sure the flame rod and its wire isn't shorting to the burner or the cabinet anywhere.

Is the pilot supposed to turn off when the main burner lights? The system needs to prove flame somehow the entire time the main burner is on. There could be a port in the main burner that is clogged or the main burner could be underfiring causing it to no longer encompass the flame rod when the pilot burner turns off if the pilot burned is supposed to turn off once the main burner lights.

TBH I was just assuming the pilot flame was turning off, but maybe it's not. I can't get a good view of it. Once the main burner ignites all I can see is flame.

prwood

Hap_Hazzard said:

Not your fault. We tend to think in terms of what we're familiar with. Since this is such a short-duration flame, it makes me think that you really are seeing corrosion instead of combustion by-products. The only thing I can think of that would cause that is chlorine, but you would have noticed a smell if that was the case. I'm straining my brain to think what might be doing this to the flame sensor but not the flame hood or any other bare steel in your basement, but I'm drawing a blank.

Do you know what the humidity is down there?

I haven't measured the humidity down there, but I can.

prwood

JUGHNE said:

Did you get and change the pilot assembly as you mentioned in your last posting?
It might be worth having a spare on hand.

The pilot IIRC, will always stay on.
The flame sensor must always be in the flame to keep the gas on.
If, when the main gas valve opens, it "snuffs" out the pilot it may then lose the flame signal and shut down.

Can you post pictures of the controls and the wiring diagram included?

I did not get a spare pilot assembly, once it appeared that cleaning the sensor was solving the problem.

I'll get some pictures soon.

prwood

mattmia2 said:

I think the problem is elsewhere, either one of the connections in the circuit or something like a solder joint in the control itself, you just happened to wiggle whatever it is in to conducting when you cleaned the rod. Also make sure the flame rod and its wire isn't shorting to the burner or the cabinet anywhere.

The first time this happened, I checked all visible connections and wiring to make sure there wasn't a short or a loose connection, per the manufacturer's diagnostic flow chart. I guess one easy way to test this theory would be, the next time this happens, go through the motions of removing the pilot assembly to clean the sensor, but don't actually clean it, just reinstall it, and see if the system comes back on.

prwood

Here are some photos:

Front of boiler:


Wiring diagram:


Ignition controller:


Main valve:


Pilot assembly side view:


Flame hood closeup:


Also, I figured out a way to get a better angle of the pilot flame ignition, and shot a video of that sequence. I couldn't find a way to upload a video to the forum, so here's a link to the video:

https://photos.app.goo.gl/8MPySvMBDJrjuJta7

The humidity in the vicinity of the boiler is 28%.

JUGHNE

Was the ignitor probe/spark always bent that much? The insulator on that looks cracked....sometimes the spark probe is somehow connected to the flame sensor....strange but I have seen it.
I would try the new pilot assembly first.

Also, it seems that your gas pressure could be high.
That could lift the main flame off the burner and snuff out the pilot burner.

You can disconnect the MV red wire at the control module (isolate to not short it anywhere) and view the pilot only burning.

Hap_Hazzard

From the video, the pilot flame looks good, and, from the picture, it appears to be hitting the sensor right in the middle. I'm not sure if the spark electrode should be bent that much or not. The wider the gap, the bigger the spark you get, but, depending on the voltage, you might not get any spark at all if the gap is too big. It's similar to gapping a spark plug.

prwood

I'm not sure about the bend in the electrode. The first time I saw it was in 12/2020 when I removed it to clean the sensor, and it already had that bend in it.

The company that makes the pilot burner assembly has a number of assemblies that seem to have various bends in the electrode. I am not sure if my exact model is on there, but PSE-NA231 seems to be the closest

https://pse-usa.com/product-types/pilot-burner-assemblies/



SupplyHouse has my model, but unfortunately due to the angle of the photos, I can't tell whether there's a bend or not:

https://www.supplyhouse.com/Weil-McLain-511-330-218-Pilot-Burner-Assembly

ethicalpaul

I wonder if during manufacturing they spark test each one and then "gap" them as necessary by bending the electrodes.

Hap_Hazzard

ethicalpaul said:

I wonder if during manufacturing they spark test each one and then "gap" them as necessary by bending the electrodes.

It's possible. It's also possible that the installer might have fiddled with it. Since there aren't any instructions in the setup guide for doing so, that might be a problem.

Before we focus too much on the gap though, remember that the issue had to do with corrosion on the sensor probe. In the latest video, you can see a slight delay between the time the pilot lights and when the burners fire up. This is to be expected because the sensor probe has to reach a certain temperature before it opens the main gas valve. If there's a problem caused by corrosion on the probe, I'd expect this interval to get longer and longer until the burners fail to fire before the pilot shuts off. (I assume the pilot cuts out at some point, but I don't know how long it takes.) So this is what I'd watch for, and I'd look very carefully at the probe before attempting to clean it. If it's glowing red-hot, the problem would be in the electronics or connections rather than the probe itself.

If, instead, it starts just buzzing and failing to light the pilot at all, then I'd suspect a weak spark, and at that point I'd try widening the gap.

mattmia2

Does the sensor electrode need to be hot? i don't know enough about the physics of flame rectification, but I thought it just needed to be in the ionized gas of the flame to conduct.

mattmia2

If you replace the control, 2 wires and the pilot burner your problems will probably go away if you have already made sure the ground from the pilot burner to the control is good. You can test with a voltmeter at the gas valve to make sure it isn't failing intermittently.

Hap_Hazzard

It appears to be a thermal sensor, and the bluing in the picture indicates part of it has been red hot.

prwood

Hap_Hazzard said:

If there's a problem caused by corrosion on the probe, I'd expect this interval to get longer and longer until the burners fail to fire before the pilot shuts off.

I was able to observe the behavior of the ignition several times before I removed the pilot assembly to clean the sensor. Here's what I observed in the failure condition:

1. Pilot burner ignited immediately when spark occurred.
2. Pilot burner continued along with spark, but main burner did not ignite.
3. Pilot burner and spark stopped after a few seconds.
4. After a few minutes, the above cycle repeated.

Essentially, when failures occurred, the above cycle would repeat until the flame was eventually detected (which took about 4 hours in one case) or until I got down to the basement to remove and clean the sensor.

BTW, here is the diagnostic flowchart for "main valve does not come on" from the boiler manual. Interestingly, cleaning the sensor does not appear as a troubleshooting step on this chart (I didn't check to see if it was on the chart for other problem conditions), which makes me think maybe the assembly does need to be replaced:

Hap_Hazzard

So it sounds like the only remaining question is, did cleaning the probe really fix it, or is it an intermittent contact issue. Were you able to see if the probe was getting hot? Was there a lingering reddish glow coming from the probe after the pilot turned off after 15 seconds?

Hap_Hazzard

BTW, do you have a microammeter?

prwood

Hap_Hazzard said:

BTW, do you have a microammeter?

I have the cheapest multimeter from Harbor Freight. This is the DCA range: 200mA/2000mA/
20mA/ 10A. Not sure if that is sufficient for the required testing.

JUGHNE

Temperature has nothing to do with this method of proving presence of flame.
Yes, things will get hot as they are in the fire.

Flame rectification will prove a flame within 1-2 seconds and realize loss of flame within 1 second. This obviously is not reliant on temperature.
The module sends out an AC signal thru the FS/ground circuit.
The flame conducts this signal thru the fire and to the hood.
When passing thru the the flame the AC is rectified to DC, and if the correct minimum of microamps is sensed by the module it will keep the gas valve energized.
The electrical pathway must be well maintained....good ground connection from pilot to module. Follow the green wire and insure all connections are reset.

The simple thermocouple system mentioned above may need up to 30-60 seconds to shut a gas valve off in the loss of flame.

I believe that the pilot burner stays on in operation. A volt meter on PV and PV/MV terminals while in normal operation would tell....24 VAC means PV is on.

Have you removed the burners for cleaning? There seems to be just a short hesitation for full burner ignition. If any delay in lighting there could be enough "poof" to snuff the pilot flame out.

IIWM, I would remove all the burners. Clean and brush all passageways. Air pressure applied.
Then replace the pilot assembly at the same time.

If the problem persists, your gas pressure on the outlet of the gas valve should be checked for the correct inches of water (WC) .