There was a thread some time ago in 2011 from another member and I am having an identical issue I think. This is the old thread:

http://www.doityourself.com/forum/bo...cement-pn.html

I've got the same White Rodgers 8F42 in my unit. Here's the story: 4 months ago the boiler kicked on on a slightly chilly September night. At two in the morning we were awakened by the water-bug alarm on the floor of our furnace and laundry room. The boiler relief valve was spitting out hot water and the pressure was around 30psi. I assumed a bad auto-feed valve or fill-trol tank. I shut off the boiler and a few days later, after turning it on shut the supply of city water to the boiler figuring that I could live that way and watch the pressure daily to avoid letting the system get low.

Upon reopening my fill valve for the heck of it a month or so ago I was surprised to find that everything was fine. And a friend of mine who is a plumber took a look and tapped on the fill-trol tank and was sure that the bladder was fine. It was NOT full of water. So everything seemed fine. Although we had been noticing all fall and winter so far that rooms seemed hotter than we wanted them to be. I blamed my wife, she blamed me, we both blamed the kids for perhaps changing thermostat settings etc.

Until yesterday when our basement was 88 degrees. All three stats were down, none calling for heat and yet the circulator was still running and occasionally the burner even fired. I shut the furnace down and once it was powered back on the circulator stayed off . . . until it stuck again.

We have 3 zones and each is controlled with a Taco zone valve. I don't believe that any of the valves are stuck on - as that was my first guess after reading of some issues online. My voltmeter reads 24 volts across the top and bottom contacts all the time on each zone valve and the middle contact is 0 volts to the top one (and 24v to the bottom) until the zone turns on, then it reads 24 volts of potential to the top contact and of course 0 volts to the bottom one. At that point the wires leading from the zone valves to the aqua stat (which have 24v between them when all zone are off, show a voltage difference of 0 and the relay kicks on with a snap. The circulator turns on and then the gas valve opens if needed, as it should. When a zone is satisfied, and after a 2 minute delay, the voltages switch back and the circulator shuts off . . sometimes.
This is how those readings are supposed to look, correct?

Every 3 or 4 times, though the circulator does not shut off. The zones seem closed, the voltages read correct. This concerns me since I feel like it's hurting the circulator. And here is the absolute evidence I would think - the mere hint of a tap on the White Rodgers aquastat cover immediately shuts off the circulator. It's sensitive enough in fact that i can't take the cover off while the thing is stuck to see the relay in the stuck position. That's enough to trip it. And if I leave the cover off, it never seems to stick! Yet there is no binding at all - that was my first thought. It's happened enough times so that I can't believe it's been coincidental to have the cover off the aquastat and have it work properly. Yet what is also strange is that I don't seem to hear the relay click off when I tap the box. Maybe I'm just obscuring the click with the sound of my tap - but it's a real gentle tap. The circulator just turns off. It's like the relay has already snapped but somehow those leaf contacts within the part of it that swings are still making contact until I tap the box? I'm even starting to wonder if that steel cover on the thing has acquired some magnetism which is interfering with the solenoid release somehow.

Before I go and try to replace that whole unit I am grasping at straws to see if there is some little trick to salvaging it.

So a few questions
1. What exactly is the aquastat. Is it the whole White Rodgers 8F42 unit or is it just that part that you can see in the original posters pics that's black and plastic and has the high and low temp settings on it?

2. Is the big solenoid-looking relay with the 775 on it the only mechanical moving part in the box? Or is there another small relay in or behind the black temperature control block inside this box? Because I'm finding it hard to believe that that big relay is really sticking. It is very clean and moves effortlessly manually.

3. If I do end up needing to replace this whole thing - and maybe this is a question for the OP of the other thread - and I need to drain to change the immersion well too, is that draining the boiler or the whole system? With zone valves shut and returns shut I should be able to isolate the boiler right? Although I wonder what kind of tricks I have to do to bleed air out of the system later.

Thanks to anyone that can help.

 

For starters, the 'tap test' is BS. I defy anyone to prove to me in a 'double-blind' test that they can tell a good tank from a bad one by tapping on it.

Yes, but this won't tell you if the valve is MECHANICALLY stuck open, yet that would generally not call the boiler to fire.

Yes, the whole box.

Sort of... the high limit switch has some (very slightly) moving parts in it but unlikely to be cause of what you are seeing.

Odds are well in your favor that you won't have to drain the system, or the boiler.

Zone valves generally can't be relied on to be 100% shutoff.

But I don't think you'll have to drain.

 

Thanks NJ Trooper. Yeah, I wondered that about the tapping on the tank thing. Although the pressure in my system is fine if the boiler and circulator don't get stuck on so I'll take it for now that that part is OK. Since I posted I've noticed that I can actually hear the zone valves closing if I listen carefully. I hear both a mechanical motor sound and a swooshing thereafter which is I guess the actual water flow in the pipe being shut off. It reminds me of a much quieter version of what I hear when my outside sprinkler system valves open and close in the summer. If I cool the system, the lines, post-zone-valve all remain cold except on the one that's supposed to be open. So they seem to be pretty good. Plus, playing with the lever on the valves acts as it should - serious resistance to manually open when the valve is supposed to be closed; very easy to open all the way when the valve is already supposed to be open.

I noticed, though, a very loose wire and screw on the bottom terminal of one of my zone valves. I tightened it and now I haven't had the system get stuck on. What I'm wondering is if that bottom wire was ever so slightly shorting to the terminal above it which would effectively keep the aquastat relay open. Maybe a little arc keeping the potential dropped just a bit so that it was not achieving a full 25 volts across those thermostat wires to the aquastat which is what I'm supposed to see when the relay is open. It just wasn't quite able to let go. My tap on the relay was just the extra bit. I think this is unlikely, though.

If I let it go for a few days, is my circulator in danger, running with closed valves sometimes?

 

As far as you know, given the track record of boiler gauge INaccuracy!

Yes, I agree it is not part of this problem. Maybe an issue later on, but for now let's focus on the immediate problem.

I suppose it is remotely possible. Relays need quite a bit of current to get them closed from an open, but not nearly as much to hold them closed.

I too think it a bit unlikely... but plausible...

Short term 'dead heading', probably not... long term, well, it could overheat itself.

What you could do is choose one zone to use as a 'dump' for the pump... (it rhymes!) ... disconnect and tape the wire from terminal 1 on that valve, and lock that valve manually OPEN.

It won't call for heat by itself, so the zone may cool down... or you might get enough 'gravity flow' to keep it somewhat warm... but in the event that it DOES stick on again, the water has someplace to go.

That would be a somewhat typical failure mode. Old pitted relay contacts sometimes have a tendency to 'weld' themselves together. Very weakly of course, but enough to hold the contacts together.

Can you SEE the contact points? Are they pitted and burned?

 

Wait... I should ask what circulator pump is on your system before answering the question about hurting the pump...

What circulator pump is on your system?

 

Its a Taco cartridge pump. I know that it's water cooled. When the stays on, the high temp limit in the aquastat does shut off the boiler itself shortly after it kicks on. So the water temp isn't really getting any hotter than it usually does when circulating I would think.

The good news is that I think I've fixed it. Funny story though. I called the Gas Company since my wife reminded me that we have a service contract with them. They came rather quickly (the next morning). The guy was plenty nice but as I expected, was unwilling to simply change an aquastat at their time and cost if he couldn't actually see it stick. He kept suggesting that maybe it was coincidence that my tapping happened at about the same time that a zone valve finally shut. I kept insisting that that kind of coincidence happening 10 times is comparable to winning Lotto.

This kind of experience is similar to what I've had a few other times dealing with service companies who pretty much assume that everyone is clueless. Well he ended up sanding the relay contacts - what he thought were the contacts anyway. I suggested that the contact did not consist of the relays throw arm and the end of the solenoid. That's where he sanded. He insisted that that was the contact and that sanding would solve the problem. OK. At this point I had to leave for work. On my way, it hit me.

When I got home, I shoved a piece of card stock between the solenoid and the relay arm and closed it manually and guess what, the circulator and boiler still turned on. To be extra sure, I took a voltmeter and read 120 volts across two leaf-like springs embedded in the arm of the relay that actually make the connection when they bend a little when the relay closes. There are actually two sets of these, one on top of the other. So when the relay was snapping open, those poorly designed little leaf contacts were of course staying stuck together (which is why I heard the relay open before I tapped on the aquastat and heard nothing when my tapping actually released them). Twenty years of line voltage with a little arc here and there right before and after contact would lead to those things sticking. And they are literally only a few millimeters apart anyway when the relay is open.

How a guy with 27 years experience as he claimed would not know this . . .
So I got a small piece of sandpaper in between those things, pressed the relay closed for some good friction and sanded a bit. Bingo. Everything good for the last few days.

It won't last forever. At this point I feel that I can get a free new aquastat out of this service contract by pointing this out to someone. But I'm actually wary of having someone else install it after my experience. And I'm itching to get that guy back and say . . watch this . .

 

No... nothing does...

Thing about 'sanding' contacts though, in GENERAL, those contacts are plated with a material that is supposed to resist the tendency to stick together from a 'welding' arc.

Over time, a 'self-wiping' contact will wear the plating out and the result is that they start sticking.

You probably did no further damage to the contacts by sanding them since the plating is probably already worn off, but it is never recommended to sand any relay contact points. They should be cleaned only with non-abrasive tools (card stock) saturated with cleaning solvent (De-Oxit Gold).

Just a 'heads up' for future reference.

BTW, you won't find them for general use like on a boiler, but TUNGSTEN contacts will almost never stick. Those contacts were probably silver-nickel plated.