Another Boiler Overfill Question - Nightmare Easter Long Weekend!!!

Powe

I have read all of the boiler overfill discussions that come up on a search for "boiler overfill" and they contain a wealth of good information, but they do not directly address the "event" that I just had happen a couple of weeks ago.

On the Easter long weekend, I was away from the house for about a day and a half, and my boiler overfilled right up to the third floor of my house. Water was pissing out the radiator vents and there's been a lot of damage to the hardwood floors, and some ceiling damage from water leaking through the second floor.

I am still scratching my head about this overfill event. Sorry for the long post but I am trying to give the full story and background. (Thanks for reading)

The boiler is (estimated 30 years old) but seems to be in good shape. You can see some pics over on this discussion. I re-did the plumbing on the control side of the boiler in 2021 to replace some leaking rusted out fittings, and it's been operating flawlessly since then. I have not changed anything on the boiler system since then either, except for doing regular maintenance.

My maintenance routine includes flushing the LWCO and the water level float switch (McDonald Miller 150 or 150S), as well as flushing the boiler drain, every two weeks. A little sludge comes out but not much. I also check the LWCO and manually rotate the fill valve in the condensate return tank to make sure it itsn't getting stuck (but I think this is exactly what happened...). I refill the water from the HWT. Except for this event, there have been no leaks, and it uses hardly any water.

The condensate return tank (CRT) has a capacity of about 15 gallons. The CRT has a float switch that keeps it full (~80% full) by operating a solenoid valve with supply water which I plumbed from the HWT. The 150s on the boiler operates a pump on the CRT to return condensate to the boiler when the boiler water level gets low. Under normal operation, the CRT does not call for water refill except when I flush it out a bit every couple of weeks.

When I returned after being away for a day and a half, I had soggy wet floors around the radiators on the first two floors of a three story house and the basement was quite wet where it had been dripping through. A bit of water escaped on the third floor, but nothing to worry about. (My hardwood floors are now badly cupped and I'm having to consider replacing them. But that is another nightmare ... er ... story.)

Of course when I saw all the water in the house, I ran down and shut the system power off and began madly draining the boiler and cleaning up. All the make up water for the boiler comes from the hat water tank which was firing and it was completely cold. I did not take time to wander around and see what was malfunctioning, so I am not entirely sure what went wrong. However, I thought I heard the CRT pump run once or twice before I ran down to shut everything off, but I am not sure.

Here is my hypothesis of what went wrong ...
I think what happened is that the CRT float switch activated, which means the CRT level had fallen to about ~60-70% is where it cuts in. This might be the first time it's needed water as I always manually fill it after my bi-weekly maintenance which includes flushing a bit of water out the drain of the CRT.

When the water level in the CRT rose, I think the float switch must have stuck on, and the CRT continued to overfill. It probably backed up through the steam trap into the dry return and (probably) also filled the boiler through the line from the boiler feed pump on the CRT. Water may have risen in both the return line and the boiler header - I do not know.

I have a manual valve that limits the fill rate for the CRT to a slow trickle otherwise the pressure from the water coming in will force water out the CRT vent tube. This would have been a lot worse if I had that valve full open.

The CRT tank has a copper vent tube (1" copper) that rises to ceiling level in the basement, and then two 90's turn the end of the pipe down. There was water coming out the vent intermittently (like a surge) every 10 seconds or so. (The odd thing is I don't know why the boiler feed pump would be cutting in (as I said above, I think I heard it running) to feed water to the boiler, when the boiler was overfilled?!?!?!) The 150S should not have been turning on. But I don't think the system would have filled to the third floor unless the pump was running???

The boiler was definitely overfilled. There was a LOT of water pressure in the boiler when I started draining it. When I started draining it, I had no idea that the system was full to the third floor. If water got up to the third floor, the first and second floor radiators would have been filled to the level of the vents I assume.

I drained the boiler almost empty as I assumed there could be a lot of sludge returning (as discussed in other boiler overfill discussions here), and I had visions of water hammer and plugged pipes, etc. I also filled and flushed it a couple of times before trying to restart it.

I spent a couple of hours trying to clean up the water before returning to the boiler to try to turn it back on. We were still in sub-freezing outdoor temps so I had to see if the boiler would fire up again and luckily it did - and without any water hammer or drama at all. It seemed to be working perfectly.

So here (finally) is my question: Could the CRT float switch being stuck cause the entire system to fill almost to the third floor? OR would the boiler feed pump ALSO have to have been pumping to force the water up that high in the system.
OR
Maybe it wasn't a stuck CRT float switch and the 150S float switch was stuck on and calling for water so the CRT obliged and gave it all the water it could and continued to refill as long as the 150S was still calling for water?

Since the event, everything has been operating perfectly. I have closed the water fill valve for the CRT and have been checking the boiler and CRT levels daily and everything seems to be operating perfectly.

After the event, I was a bit confounded when I realized how high the water got in the system when there is a vent in the CRT at basement ceiling height that essentially opens the system to the atmosphere, and I'm wondering if I'm missing something?

A couple of other facts that may be relevant here:

1. There is a check valve on the boiler feed line from the boiler fill pump so the boiler cannot force water back into the CRT. As far as I know it is working correctly.
   
2. There is a steam trap on the dry return line before it drips into the CRT.
   
3. The steam trap is not plugged, as I checked it after draining the system (I have a drain valve on the steam side of the trap to be able to drain and check the condition of the condensate before it goes into the trap.) I have checked it several times and the steam side of the trap has not flooded since the overfill.
   

If anyone has any thoughts on what could have gone wrong, besides the stuck CRT float switch or a stuck 150S float switch, I would appreciate any thoughts as I do not want a repeat of this event. I am now planning to get a water, temperature and humidity sensor that will alert me when any of these are out of acceptable range.

Thanks in advance!

109A_5

Hello @Powe,

It seems to me your system (for its size) is way too complicated, causing issues you don't want or need.
Is the condensate return tank, pump, check valve, CRT float switch all actually necessarily ? If so I would shut off the system water feed if you go away for more than a few hours. Or add a Hydrolevel VXT-24, however they need maintenance too, and are not foolproof.

Seems to me there are too many different things that can fail, and if just one fails it floods your system with nothing to stop it, if unattended.

CRT float switch sticks closed = Flood
McDonald Miller 150 or 150S thinks the water in the boiler is too low (when it is not) = Flood
Pump control, if there is any additional devices besides the 150 that could keep the pump running = Flood
Solenoid valve leaks = Flood
HWT, a source of sediment so the solenoid valve leaks = Flood

I only have a gravity condensate return, M&M 67 LWCO, a manual fill valve and no mystery problems. I don't need to touch the system for weeks if I don't want to.

Hydrolevel VXT-24 for one example;


At least the VXT recognizes certain abnormal issues and stops the fill (assuming its valve does not leak).

Jamie Hall

I can see any number of failures which could cause your condensate return tank to overfill. A @109A_5 says, your system is much too complicated to be reliable.

However, the vent on the condensate tank should have limited the water level rise to the elevation of the 90 degree downturns at the ceiling. There's something else going on here.

PC7060

Regarding the cupped hardwood, unless it’s an engineered hardwood most cupping will work itself out in a couple weeks as the moisture moves out of the floor. Best to do nothing until it stabilizes.   

ScottSecor

Is the M&M 150 mercury ball type or the snap switch type?  We have seen many of the snap switch types (especially the early models) fail and flood systems.    We have also witnessed the 150 float chamber get loaded with rust and have the float hang.

Another item that concerns me is that it appears your using hot water to feed the system.   In most occasions,  cold water is preferred.  

Powe

@109A_5  and @Jamie Hall, thanks for your comments. I agree about the system being complicated. It is as it was, when I bought it in 2020 and I plan to make changes but have not had time ... or perhaps courage ... yet.

Regarding how the water got up so high...
How fast does water flow through a steam trap?  The steam trap would have been water logged and with +/-30 psi of water pressure above it. Would it be able to function at all or handle that pressure of water? Would it be wrecked?

If water was being pumped into the boiler and couldn't get past the steam trap fast enough (or at all) that would explain how the water could get higher than the vent.

But this would seem to mean two failures - the 150s stuck on, calling for water and the condensate tank float switch also stuck on, and calling for water. Or perhaps the pump was emptying the condensate tank and it was happily refilling normally?

Another possiblity that seems unlikely, but I'll throw it out there ... The house was chilly (probably in the mid 60s degrees) when I arrived and so the thermostat would have been calling for heat. If the boiler was successful at generating any steam the boiler pressure could have pushed the water that high. But this would mean the high limit switch is not functioning because that height of water is over 30psi and the high limit switch would have prevented the boiler from firing. I haven't tested the high limit switch since last fall so if that's not working it's a huge potential problem and another thing on my list to do. I cleaned out the pigtail and flushed it out last fall and the switch was totally fine - not stuck or full of crude and it operated freely.

Powe

PC7060 said:

Regarding the cupped hardwood, unless it’s an engineered hardwood most cupping will work itself out in a couple weeks as the moisture moves out of the floor. Best to do nothing until it stabilizes.   

Yeah I haven't touched it yet. Still running dehumidifiers. But it's not improving yet. I'm not going to rush into any work on the floors. Too much else to do anyway.

Powe

ScottSecor said:

Is the M&M 150 mercury ball type or the snap switch type?  We have seen many of the snap switch types (especially the early models) fail and flood systems.    We have also witnessed the 150 float chamber get loaded with rust and have the float hang.

Another item that concerns me is that it appears your using hot water to feed the system.   In most occasions,  cold water is preferred.  

The 150S is the snap switch type, not mercury. It gets flushed every week or two, along with the LWCO and a couple gallons out the boiler drain. I'm fastidious about the regular flushing. However I am currenly leaning toward the 150S being the culprit for this overfill event. 

Re hot water for makeup water, I've heard numerous times that using hot water is preferred because a lot of the free oxygen gets driven off in the HWT thus saving it from being released in the boiler.

I'm certainly a nube when it comes to steam and boilers, with only 3 years owning this one, but this is the first time I've heard that make up water should be cold. I stand to be corrected obviously but it makes a lot of sense to me to use hot - not just the oxygen, but also not shocking a hot boiler with cold water. Can you explain why cold is preferred? 

109A_5

Hello @Powe,

If I had to guess I think this scenario is most probably with the description you provided.

Powe said:

Maybe it wasn't a stuck CRT float switch and the 150S float switch was stuck on and calling for water so the CRT obliged and gave it all the water it could and continued to refill as long as the 150S was still calling for water?

I think it had little to do with the F&T trap, which both valves would be open anyway (no steam and the float floating). Also why couldn't the water just go up the main pipe and the risers.

As @ScottSecor suggested I would be very suspicious of the M&M 150S, either the snap switches (microswitch) or rust accumulation in the float chamber causing unwanted behavior. I've cleaned the rust out my M&M 67 LWCO, not hard to do, just messy.

This would also explain why the water was not simply draining out of the Condensate tank vent until it was way too late as @Jamie Hall noted.

If you get the number off of the snap switches (microswitch) you probably can replace just them inexpensively, the whole assembly is kind of pricey.
Also the gaskets are available for the M&M 150S, or you can make them with gasket paper or the like.

109A_5

Hello @Powe,

Eventually with the height of the water (Head Pressure) the Pressuretrol may have shut the boiler down.
You could use an additional Pressuretrol or the like to electrically disable the water feed. 1.5 PSI would be about 3.45 feet of Head.
Adding to the sophistication of the system.

Powe

109A_5 said:

Hello @Powe, Eventually with the height of the water (Head Pressure) the Pressuretrol may have shut the boiler down. You could use an additional Pressuretrol or the like to electrically disable the water feed. 1.5 PSI would be about 3.45 feet of Head. Adding to the sophistication of the system.

Lol! I bet you said "sophistication" with a completely straight face. Right?  😄

I could just tie in (with a relay for isolation) with the high pressure limit that shuts the boiler off. If the boiler is overfilled it wouldn't come back on and might be safer than shutting off the water when the boiler might still be able to fire.

109A_5

Hello @Powe,
Yes, there is many ways to prevent boiler flooding in the future. As much as I like automation and the like, for me and my boiler I like to keep it simple and just use a manual feed. I am usually in the basement at least once a day so I give the boiler a look see. I realize most folks don't want to do this. And I do test and maintain the LWCO.

As far as Hot water for the feed water, according to "The Lost Art of Steam Heating Revisited" page 356 lime can settle out on the feeder's valve seat, causing a leak and a flooded boiler.

Powe

109A_5 said:

Hello @Powe, Yes, there is many ways to prevent boiler flooding in the future. As much as I like automation and the like, for me and my boiler I like to keep it simple and just use a manual feed. I am usually in the basement at least once a day so I give the boiler a look see. I realize most folks don't want to do this. And I do test and maintain the LWCO. As far as Hot water for the feed water, according to "The Lost Art of Steam Heating Revisited" page 356 lime can settle out on the feeder's valve seat, causing a leak and a flooded boiler.

Well, there it is. Pulled out my copy of the book and read the section. Cold water it is! Must have missed that when I read the book. I guess I'll add that to my list of things to do when our heating season is over. I don't think it's the problem in this instance as the solenoid valve is only about a year old. But until I check it ...

109A_5

As far as cold or hot water I looked at the VXT water feeder install manual and it did not specify cold or hot. As far as cold shock to the boiler with a VXT maybe the orifice limiting the water flow for their water metering system strategy minimizes the cold shock issue.

I manually feed my boiler very slow, and I resist the urge to walk away.

Jamie Hall

A bigger issue with the feed water than hot or cold -- it mustn't be softened. If your water is very hard, spring for some distilled or deionized for the fill. Otherwise -- straight out of the tap.

30 feet of head is about 20 psi, which is why your pressure relief valve didn't open. However, it is enough pressure that it may have damaged a steam trap or two, depending on what kind of trap it is. It may also have damaged a vent.

I am still puzzled as to why it didn't come out of the vent on the condensate tank...

JUGHNE

If there was 30' of head it would be pushing 15 PSI, (from looking at the old tridicator gauge I keep on my desk). The pressure relief valve would have been close to opening.

OP stated he has water coming out of radiator vents, so I assume this is a single pipe system.

Maybe only 1 F&T at the feeder pump and no rad traps??

My best guess is that the 150 stuck on keeping the pump running, filling the boiler and all piping up to the 3rd floor.

The EOM F&T could pass only so much water back to the pump tank, allowing the pressure to build up to flood the system. And also keeping the overflow/vent from passing water out in the basement.

Had the 150 shut off the pump for any length of time, the check valve in the pump supply to boiler would have kept the water from flowing back to the tank to overflow the tank vent.

The tank fill valve probably did it's job to add water until the F&T returned the water and may have added more water as the F&T was slow return of water. ( I have had one stuck in the up position (Only 3 years old) to where water would not turn on and boilers shut down with LWCO)

I would open the 150 and make sure the chamber was clean and also that the float was not waterlogged (leaking) which would hold it down and call for water continuously.

I had success in adjusting the microswitches in a 150 that had a very close differential which would shut down the boiler whenever there was a call for feeder pump. (The input flow from the pump was not throttled down and caused the boiler water level to bounce.....both the switch and ball valve were adjusted to correct this.)

Just a WAG.

Powe

@109A_5, the cold shock wasn't a big concern for me because the feeder goes into the condensate tank and mixes with ~15 gallons of water before getting pumped into the boiler. The only way water is fed directly into the boiler is if I am manually filling. 

@Jamie Hall, no water softener here. Our water is moderately hard at around 9 grains (160 mg/L) and no iron.  We had a water softener at another house for a while until it broke down and I never bothered fixing it. This house used to have one but it was already gone when I bought it. There was a labyrinth of piping to supply soft water to some fixtures and not others and to bypass the softener when they wanted. Thankfully the boiler was plumbed to always get un-softened water.

Powe

Also I estimate it only got to about 24-25 ft of head when I discovered it. I did mention it was coming out the vent but not continuously. It was surging every 5 to ten seconds or so. And there was no water coming out of the main vent either which is strange I think as well.

JUGHNE

Most main vents have a seal off float for this instance happening.

Some rad vents also have floats that may seal off the water.

The surge may have been the float moving up and down like a fishing bobber as the water passed thru the F&T in the basement.

Powe

@JUGHNE, thanks for your insights. The 150S seems to be the primary suspect in this discussion and I'm now more curious to get it opened up to see what I find.
That's going to have to wait til next weekend at least. For now the water feeder is off so if it is the 150 it can only get the 15 gallons from condensate tank to overfill the boiler. 

JUGHNE

Do you have a sight glass on your feeder pump tank?

Not something you want to run dry--- out of water.

mattmia2

The feed pump is the only way it could feed that high. If the fill valve for the tank wasn't holding it would not have filled above the vent. The pump pumping water in the system is the only way it could have filled above the vent. The LWCO must have stuck or a relay for the feed pump if it has one.

It is possible the feed pump was cycling on its overload as a result of exceeding its duty cycle or because it was working against too high a pressure.

I'd give the floors several months or more to fully dry before I tried to refinish them or otherwise repair them, see how much returns to the proper dimensions when it dries out.

109A_5

The scenario that @JUGHNE and @mattmia2 paint makes sense. Which is why I wonder for a system that size why is all the extra hardware needed, it just seems to be problematic and more to maintain. If the system needs a bit more water than what is normal found in that particular boiler it can be done without a pump. The tank just needs to be at the correct height in relationship to the boiler's normal water level.

I think I would study the system thoroughly and try to figure out why all that 'extra' equipment was installed, and if actually not needed remove it. Life may be simpler.

And if you decide to keep it all, using a relay in the Pressuretrol (1.5 PSI Cut-Out) circuit that disables the whole electrical part of the water feed / pump power would add some safety margin so it does not happen again. Just wire up the relay correctly.

At 1.5 PSI Cut-Out, 3.45 feet of Head and correct Pressuretrol operation the water may not even come out of the condensate tank vent, you would just find a flooded boiler that was cooling down.

Still If the feed valve leaks and overfills the condensate tank the Pressuretrol and relay may have little control over this scenario unless the water leaks by the pump too. I would expect the water to come out the condensate tank vent but go no higher.

mattmia2

109A_5 said:

The scenario that @JUGHNE and @mattmia2 paint makes sense. Which is why I wonder for a system that size why is all the extra hardware needed, it just seems to be problematic and more to maintain. If the system needs a bit more water than what is normal found in that particular boiler it can be done without a pump. The tank just needs to be at the correct height in relationship to the boiler's normal water level.

The condensate tank and feed pump were a misguided attempt by someone that doesn't understand steam to solve some problem, likely one with the near boiler piping or water quality that was causing the water to exit the boiler in to the mains, or maybe just some sort of elevation problem they didn't know how to solve otherwise.

If you are in an area where one of the great steam contractors is, paying them to take out the condensate tank and fix whatever problem they were trying to solve will be a bargain compared to cleaning up the water damage.

Powe

JUGHNE said:

Do you have a sight glass on your feeder pump tank?

Not something you want to run dry--- out of water.

@JUGHNE , Thanks for your posts here. The main vent is very old, and I have purchased a new Gorton #2 to add to the system this year to solve a problem that I have only one vent for two heating loops.

No sight glass on the condensate tank, but I check it almost daily by rotating the float switch manually and ensuring it returns to the off position. I've thought of installing a sight glass but I'd have to jury rig something and it doesn't seem necessary at this point. I am thinking of getting rid of the tank this summer and going with a wet return.

Since I rebuilt the boiler "control" plumbing (where the sight glass, pressure switches, the 150S and the LWCO are located) in summer 2021, the boiler uses very little water. This was a HUGE improvement over my first heating season (2020-21). There were several leaks in the plumbing and the condensate tank pump gasket and seals leaked like a sieve. I had to add several gallons of water every day, which probably shortened the life of the boiler.

At that point I had not found Heating Help yet and was trying to learn everything I could by watching Youtube videos and reading whatever I could get my hands on. I bought The Lost Art of Steam Heating Revisited in the spring of 2021 and learned what I could. Here's the "before" and "after" shots of the boiler when I did the work on it.

I basically rebuilt it as it was, and at that point, I still didn't know a lot about steam heating (and I'm still learning). Since the "after" pic, I've added an emergency shut off switch by the stairs, and moved the shut off that is behind the boiler (high up on the green board), to the front near the pressure gauge. I have more changes planned for this summer. Forgive the mess. I still have a lot of work to do on the system and the (111 year old) house.

Powe

mattmia2 said:

109A_5 said:

The scenario that @JUGHNE and @mattmia2 paint makes sense. Which is why I wonder for a system that size why is all the extra hardware needed, it just seems to be problematic and more to maintain. If the system needs a bit more water than what is normal found in that particular boiler it can be done without a pump. The tank just needs to be at the correct height in relationship to the boiler's normal water level.

The condensate tank and feed pump were a misguided attempt by someone that doesn't understand steam to solve some problem, likely one with the near boiler piping or water quality that was causing the water to exit the boiler in to the mains, or maybe just some sort of elevation problem they didn't know how to solve otherwise.

If you are in an area where one of the great steam contractors is, paying them to take out the condensate tank and fix whatever problem they were trying to solve will be a bargain compared to cleaning up the water damage.

All of you guys are making a lot of sense, and in another discussion, @Jamie Hall has also mentioned that the condensate tank is probably unnecessary for this system. Like I said, I rebuilt it the way it was, after my first heating system because I didn't know any better. Now that I've learned a bit more, I will probably try to make some of the other changes that have been suggested here before. I need to increase the size of the steam main for the back half of the house as it is way too small at only 1.5" and it serves 7 radiators. Taking out the condensate tank and changing to a wet return is a fairly big job that I want to do this summer as well. Anyway, I'm off topic on my original post for this discussion.

I'm glad I'm hearing from you guys that all the extra hardware is unnecessary as I wondered that myself, but didn't know any better.

Edit: About the floors - the dehumidifiers I have running have already made the main floor hardwood settle down substantially to the point it may not need refinishing.

109A_5

Hello @Powe,

By the way on YouTube.com there is a bunch of videos about the McDonnell & Miller #150 and the McDonnell & Miller #67 if you think you would benefit from them.