Hammering on newly installed boiler

Fred

STEAM DOCTOR said:

Be careful when LWCO probe is not installed in boiler tapping. I have seen boilers where installer used a nipple and coupling. Boiler was on bit of tilt toward probe. Coupling and nipple were always filled with water because water doesn't flow uphill. LWCO never cut out. This is always why I don't like pressure relief valves on sides of boilers.

I assume you mean on the sides of Hot water boilers?

Charlie from wmass

But it can clog from sediment and not function. That is the issue Steam Doctor is making.

Abracadabra

KC_Jones said:

Am I the only one that goes crazy when I hear about this stuff? Is it really that hard to size a steam boiler? I mean seriously?!

@KC_Jones I know... but like @Charlie from wmass said, 15-20 years ago, even I thought 30% additional for piping was "standard". The guy I worked for did it, my suppliers spouted it, etc,etc..

KC_Jones said:

So is there any plan? Are you going to work on it now, or is this information now going to be presented to the original installer? That LWCO setup sounds odd at best.

@arbalest said he'll try to continue to work with the original installer for now and have him make corrections.

Charlie from wmass said:

I am curious is the LWCO for a steam system? It sounds like a hot water one.

@Charlie from wmass I believe it's was a MM PSE801 @arbalest should be able to confirm.

arbalest

First of all, a HUGE THANKS to Abracadabra! That is a courtesy above and beyond, to say the least. Thank you! I will rearrange the order of his comments to reply here.

Abracadabra said:

But pretty sure that main issue with hammering was due to the lack of skimming. It seems installer really did no significant skimming.

Right, knowing what I know now I can say that the installer hardly did any skimming. After reading and a few YouTube videos I skimmed the heck out of it. At one point I closed the drain and forgot to turn off the manual feed. The system got quite full and you could say it was like a final rinsing. :-)

I have not heard a single instance of hammering since then but I am not going to let me think that everything is OK.

Abracadabra said:

Returns need to be dropped to floor and hartford loop properly piped...

I am going to start with the original installer to get this done at a minimum. You mentioned that since my returns were the dry type the may not need to be dropped further but that a correct connection to the equalizer was not optional.

Abracadabra said:

Proper header needs to be built.

I would like the whole nine yards done here - two risers, header (or even drop header), proper reducing elbow connection to the equalizer, etc. I don't expect the installer would do this and I am not sure I want him to. I know Abracadabra is booked right now but if the current configuration can get me through his busy season he would be my first choice to redo this.

Abracadabra said:

More main venting required.

Abracadabra explained in detail how I can do this and I know I can handle it. Once again -- thanks so much!

Abracadabra said:

Makeup water supply needs a backflow preventer.

There are two or three more things I want done with the water supply. I'd like to consider them as all part of the same reworking.

Abracadabra said:

Installation of a vaporstat and low pressure gauge (0-30" w.c.)

I understood this to be optional but maybe it will be done at the same time as the header work.

Abracadabra said:

I also suspect that boiler maybe be oversized as I was told that 30%?ish? was added to system EDR and a boiler with 130% of EDR was selected.

I will find my notes on the radiator sizes and put them up here. The process here was that I got four quotes and 3 were for 350K BTU boilers. The old one was 375K. The fourth guy, who I went with, quoted 300K BTU.

I suspected that he made a mistake and when I inquired he said that he passed his measurements to the supplier and they recommend this one. Base on my inquiry he used the dimensions to confirm himself that it wasn't too small (my concern given the other quotes). Then I carefully measured and used online calculators and compared that to the boiler specs and saw that it was indeed big enough, if not a little over. I was okay with it being a little over based on my own reading of the 130% guidelines.

I tried to do my homework -- but look at what the others were suggesting!! I will post the radiator info here when I locate it.

Abracadabra said:

I'm also not thrilled with the installation of the probe LWCO...

I don't know about this stuff but I will be open to having it all redone when the other work gets done.

Abracadabra said:

That's all I can remember off the top of my head.

You mentioned a concerns about a smoke detector in the room (I have two CO detectors), a fresh air source with needs calculated to assure the minimum, adding a valve to the bottom of the h-loop when it is redone and some other minor details that I jotted down and can address on my own.

Thanks again to everyone here.

arbalest

KC_Jones said:

To the OP for your reference and a perspective... ...sounds like you are for sure sucking water out because of that piping... ...you are going to flood the boiler. All that make up and subsequent draining will destroy the boiler. Make up water brings oxygen...

Good to know. On the bright side I only hear the boiler pause to add water once in a while. That is compared to pausing for water every other minute a few days ago.

While I don't hear hammering now it sounds like I will have to remove water regularly (to prevent flooding?) until the piping is all correct? Maybe a bit-o-skimm'n will be part of my morning coffee routine until it is fixed.

vaporvac

If your boiler is flooding, can't you just shut off the auto-fill and do it manually?

arbalest

vaporvac said:

If your boiler is flooding, can't you just shut off the auto-fill and do it manually?

The unit pauses in the heating cycle to add water (controlled by the LWC) so if I turned that off (close the supply valve) the boiler would just stop running altogether.

arbalest

Well, in spite of all of the skimming and a quiet day I woke up to hammering again. By the time I got to the basement it was quiet. Maybe that is when the over-fill from the previous cycle finally gets thrown up into the pipes? As I sat near the unit for the whole cycle water was added twice.

The site glass had been clean but it looks like a layer of gunk has accumulated so even more skimming might be in order.

Today I am going to contact the installer armed with the information posted here and the helpful information from @Abracadabra.

KC_Jones

arbalest said:

vaporvac said:

If your boiler is flooding, can't you just shut off the auto-fill and do it manually?

The unit pauses in the heating cycle to add water (controlled by the LWC) so if I turned that off (close the supply valve) the boiler would just stop running altogether.

Just to clarify, this shouldn't happen. It doesn't need water the water is most likely in the system, just not getting back to the boiler in a timely fashion. The water is being sucked out through the header, part of the function of the equalizer (when piped correctly) is to allow that water to immediately get back to the boiler. If you shut off the auto feed the boiler will shut down on low water, but as soon as the water returns it will start up again. If it was mine that's what I would so. As said the excessive make up water is bad for the boiler and I wouldn't want my brand new boiler getting off on the wrong foot. If your installer says anything about that being normal and "I see boilers take on a lot more than that", tell him he is wrong it's not normal. We have heard these stories on here many times. The hack installers will try and explain things away and say they know and been doing this for X number of years etc. I also think you are correct in not asking them to improve the piping, they wouldn't do it and can't be expected to go beyond minimum spec (a good steam man doesn't worry about minimums). Of course this installer hasn't gotten to minimum spec yet. lol I wish you best of luck with the installer.

Abracadabra

arbalest said:

Well, in spite of all of the skimming and a quiet day I woke up to hammering again. By the time I got to the basement it was quiet. Maybe that is when the over-fill from the previous cycle finally gets thrown up into the pipes? As I sat near the unit for the whole cycle water was added twice.

The site glass had been clean but it looks like a layer of gunk has accumulated so even more skimming might be in order.

Today I am going to contact the installer armed with the information posted here and the helpful information from @Abracadabra.

#1 - Where in the cycle does the hammering happen? Beginning? Middle? End?

#2 - Use a permanent marker and place a mark at the level of the water in the gauge glass. If the water level stays the same after days and you have not drained water out of the system, yet the LWCO keeps feeding water, you have a leak somewhere.

arbalest

KC_Jones said:

Just to clarify, this shouldn't happen.

I realize that. So maybe both parts of the piping are contributing to the problem 1) wet steam due to lack of a proper header 2) condensate not returning fast enough due incorrect connection between the loop and the returns.

While the boiler is running the water level is in the middle of the glass and maintained by refilling. At the end of the heating cycle the condensate returns and site glass is full -- I don't see the top of the water level.

arbalest

Abracadabra said:

#1 - Where in the cycle does the hammering happen? Beginning? Middle? End?

#2 - If the water level stays the same after days and you have not drained water out of the system, yet the LWCO keeps feeding water, you have a leak somewhere.

I should have clarified for #2. During the heating cycle the water level is either correct or lowers to the point of adding make-up water. A while after the cycle I can see that the level is over the site glass top, so I remove some via the valve on the skimming tap. I don't think it's leaking out, it's seems like it just up there in the pipes thanks to the wet steam or improper equalizer connection. After it cools off the water level is too high.

To answer #1, now when I am not around to remove some water (I assume) the hammering starts at the beginning of the next cycle. Before, there would be a few mini-cycles of heating/water-adding/heating/water/... then the hammering would start. So I would say at the beginning or the middle depending on the context.

I have a question to ask but I am going to post that in a separate reply.

arbalest

You've seen the pictures of the horizontal stuff that joins the returns the the equalizer. I know that the proper configuration is a vertical to an elbow then a close nipple to the equalizer. The question I have is that since this is going on below the water line in both cases isn't that section full of water either way? I would like to understand how this connection affects the operation and what is not happening due to it being wrong.

Thanks to everyone here I not only know that the header/main stuff is wrong I know exactly how and why a proper set-up solves that part of the problem. But I don't understand the particulars of how the other mistake affects the system.

Also, I know I don't have enough venting on the returns. Can this conceivably contribute to this particular problem? I know it is a problem and it will be fixed but I am asking if it can be contributing to this particular problem.

Jamie Hall

If the water level is at or near the top of the glass when the boiler starts, that's high enough. Period. Now, if it drops way down -- down far enough to trip the LWCO -- that simply tells you that that water has gone out into the system somewhere, no? Now if you turn the system off at that point and wait patiently, does the water level go back up to where it started, or very nearly so?

If yes, you don't have a leak -- or at least not a serious leak. And you don't need to add water to the boiler.

However, clearly water is going out into the system. Bad piping is part of the problem. High pressure could be a huge part of the problem -- have you checked the pressure? Keep in mind that pressure in the boiler will push water out of the boiler into the returns; if the pressure is high enough you could be filling one or more of what should be dry returns with water -- and they can hold a lot (as well as hammering like mad).

How long does it take the water level to recover after you have shut the system off? A few minutes is fine. Longer and you may have a badly clogged wet return somewhere.

Now going back to square one here --- if the water level does not return to where it started, but stays low, you have a leak. A significant leak. You need to find it and fix it!

arbalest

Jamie Hall said:

Now if you turn the system off at that point and wait patiently, does the water level go back up to where it started, or very nearly so?

No -- after the heating cycle ends and the water returns, the level is above the site glass. Too high rather than too low.

When I drain it down to the middle of the glass and restart the system it pauses once or more during the heating cycle to take on more water. After a couple of these it would hammer.

This pattern has slowed down a lot since I have done the skimming but it hasn't stopped.

Abracadabra

arbalest said:

You've seen the pictures of the horizontal stuff that joins the returns the the equalizer. I know that the proper configuration is a vertical to an elbow then a close nipple to the equalizer. The question I have is that since this is going on below the water line in both cases isn't that section full of water either way? I would like to understand how this connection affects the operation and what is not happening due to it being wrong.

Thanks to everyone here I not only know that the header/main stuff is wrong I know exactly how and why a proper set-up solves that part of the problem. But I don't understand the particulars of how the other mistake affects the system.

Also, I know I don't have enough venting on the returns. Can this conceivably contribute to this particular problem? I know it is a problem and it will be fixed but I am asking if it can be contributing to this particular problem.

Going to try to reply from phone while waiting for an appt... excuse typos. Here goes.. While boiler is off, the return manifold is full of water along with that long horizontal cludge which creates the bad hartford loop which should be a short nipple. The incorrect near boiler piping is causing wet steam taking with it a bunch of water out of the boiler. oil on the surface of the water will exacerbate this situation. The water level drops as boiler continues to steam. Water level now drops below the horizontal cludge. Steam is now able to push back into the returns. Now you get hammering. As water continues to leave boiler level drops which triggers the LWCO. When boiler shuts off, all that water eventually makes it's way back to boiler flooding it.

arbalest

Hatterasguy said:

Can you clarify the height (relative to the glass) where the LWCO probe sits?

The line that is cast into the side of the LWC (MM67) lines up with the bottom of the site glass. The glass is 6" tall.

That seems to put the line in the manual (almost 24") right in the middle of the glass. I don't know if this answers your question.

arbalest

Abracadabra said:

Water level now drops below the horizontal cludge. Steam is now able to push back into the returns. Now you get hammering.

Thanks. I will get both fixes done one way or another but I like getting an understanding of what is going on. And by the way, I spoke to the original installer this morning and it went well. Now I have to see if he actually does something about it. The h-loop part at least.

Is it fair to say that even though the connection is 2 1/2" too low the bigger problem is that it is all horizontal to the return manifold rather than forming a loop below it (like a p-trap under a sink)?

Fred

To your question about the horizontal Hartford loop. That essentially is the same as not having a Hartford loop. The purpose of the Hartford loop being vertical is so that should the wet returns develop a leak (because they always have water in them, The vertical part of the Hartford loop will not let water, in the boiler drop below the Hartford loop, which is typically a couple inches below the normal water level of the boiler. It is there to protect the boiler from running dry. Once the wet returns are dropped, as Abracadabra suggested, the Hartford loop should go vertical up into the equalizer.
Another reason for some of the Hammering is the fact the boiler is overfilled with the water the water feeder is adding plus water that is returning to the boiler that was already in the system.

vaporvac

arbalest said:

Jamie Hall said:

Now if you turn the system off at that point and wait patiently, does the water level go back up to where it started, or very nearly so?

No -- after the heating cycle ends and the water returns, the level is above the site glass. Too high rather than too low.

When I drain it down to the middle of the glass and restart the system it pauses once or more during the heating cycle to take on more water. After a couple of these it would hammer.

This pattern has slowed down a lot since I have done the skimming but it hasn't stopped.

I think Jamie meant where does the water level return to AFTER you've shut OFF the auto-fill. With that on, one can't diagnose if there's a leak. Shut it off for one cycle and check. It's not really a big deal if the boiler shuts off on LW once during a cycle.
I suspect you need more skimming also. It takes quite a while to get out all the oils.
Also, I don't understand why you're not asking the installer to redo the near boiler piping to give you a proper header and equalizer. This is minimum to the instruction manual. Besides all the problems you can hear, you're using a lot more fuel than you should be.

arbalest

I've attached a sketch of three configurations. The top drawing (1) is how it is now. The connection is horizontal.

The middle drawing (2) is how it would be with just the kludge oriented vertically.

The last drawing shows the returns dropped and the kludge fixed.

I thought I understood @Abracadabra to say that the middle drawing might work since the returns are of the dry type. Did I understand that correctly or do I need to insist on something like drawing #3?

arbalest

Hatterasguy said:

Does the water get all the way down to the bottom of the glass before the 67 stops the boiler and feeds water?

Yes. And repeated tests blowing down through the mm67 shows that happening and the burners being interrupted.

arbalest

vaporvac said:

With that on, one can't diagnose if there's a leak.

All of the lines (steam and returns) are along the basement ceiling and up into the apartments. If there was a leak I would surely know it by now. I can do the recommended test anyway.

vaporvac said:

I suspect you need more skimming also. It takes quite a while to get out all the oils.

Okay, I will do more of that. It would be nice if a pre-installation step included some sort of internal pressure washing out in the alley before it is brought into the basement. I saw a YouTube video of this being done in a basement. I think it was Gerry Gill.

vaporvac said:

Also, I don't understand why you're not asking the installer to redo the near boiler piping...

I am going to ask him to do it but I somewhat don't trust him to do it right and I fully expect him to argue with me about it not being necessary. That is what he believes so his quote was based on that. But I will ask, even insist, but I expect that someone else will wind up doing it. For Hartford Loop I will really insist most of all since there is no excuse for that and it shouldn't even be that much work.

Looking at Dan's video it seems that I should want a valve in the riser to help with a really good clean-out.

Update: Just watched the video again and I guess I need a valve on the h-loop too if I want to use that clean-out method.

Fred

Your Picture #3 is the right way to do it. Picture #2 may or may not work but #3 will, for sure.

arbalest

After reading another thread about skimming it seems like what I thought was a lot is not even enough and what the installer did was virtually nothing at all.

I am going to skim even more today and maybe go out and get a fitting that will tie the 1 1/2" valve to a hose end (which looks like 1").

Fred

arbalest said:

After reading another thread about skimming it seems like what I thought was a lot is not even enough and what the installer did was virtually nothing at all.

I am going to skim even more today and maybe go out and get a fitting that will tie the 1 1/2" valve to a hose end (which looks like 1").

The standard hose end is 3/4"

Jamie Hall

arbalest said:

Jamie Hall said:

Now if you turn the system off at that point and wait patiently, does the water level go back up to where it started, or very nearly so?

No -- after the heating cycle ends and the water returns, the level is above the site glass. Too high rather than too low.

When I drain it down to the middle of the glass and restart the system it pauses once or more during the heating cycle to take on more water. After a couple of these it would hammer.

This pattern has slowed down a lot since I have done the skimming but it hasn't stopped.

Please please do this and get back to us:

Turn off the power to the autofeeder. Close the valves to the auto feeder. Drain the boiler down to 2/3rd up the sight glass. Now leave the autofeeder off. Run the boiler. It may or may not shut off on the LWCO. If it does, turn it off completely and time how long it takes the water to return to the level in the sight glass that it started at. If it shuts off on the thermostat or pressure switch, well and good -- time it anyway.

There are three possibilities:

If it takes more than 5 minutes, but does return, you have some badly clogged wet returns somewhere.

If it takes less than 5 minutes, you need to find out where it is going when the boiler is running. As a start, check your pressure and any low hanging dry returns.

If it doesn't come back after an hour or so, you have a leak somewhere which you need to find.

Get back to us...

vaporvac

Sometimes leaks are sneaky, because the condensate evaporates so quickly. Please get back with us with the results, so that we can eliminate some possible problems.

arbalest

I ran a modified version of the test above. The water level dropped 1.5 inches from the start and didn't seem to be going anywhere so I shut off the boiler and started the timer. About 1 inch came back in 6 minutes or so. It is 20+ minutes now and the level is almost back to the original level.

There was no hammering on this run and I happen to have been skimming a lot today.

BobC

It sounds like your returns could use a cleaning and you should probably do a few more skimings to make sure you get all the oil out of the water.

I would leave the water feed to the automatic water feeder off for a few days (when the boiler is being used to create steam) to check for the possibility of a leak.

Bob

Jamie Hall

arbalest said:

I ran a modified version of the test above. The water level dropped 1.5 inches from the start and didn't seem to be going anywhere so I shut off the boiler and started the timer. About 1 inch came back in 6 minutes or so. It is 20+ minutes now and the level is almost back to the original level.

There was no hammering on this run and I happen to have been skimming a lot today.

Good! Just what we needed. A little long on the return, but not bad. A little more drop when steam started than I'd really like to see -- but again not bad.

Progress! Now -- and inch and a half or even two inches of drop should leave the water within the sight glass and, more important, should trip neither the low water cut out nor the auto feeder. So as Bob suggested, leave the auto feeder cut out for a while and keep an eye on the water level in the boiler; give it 15 to 20 minutes after it shuts off to check. Mark the level -- I use an ordinary spring clothes pin on the sight glass for this. If the level drops slowly over time you may well have a leak somewhere.

The LWCO shouldn't operate while this is going on. If it does on some runs, so be it -- it's not the end of the world. Let the system recover and see what the level recovers to. Do NOT add water until you see what it recovers to! However, if it does operate on some runs, try and figure out what is different about them. Longer running time? Do you sometimes have some radiators shut off, but on at other times?

Abracadabra

@arbalest #2 or #3 will work. Regardless, add tees with a plug on one end of the run of the tee at the bottom of the header instead of elbows as you have shown in #2 to be able to clean it out.

arbalest

BobC said:

It sounds like your returns could use a cleaning... I would leave the water feed to the automatic water feeder off for a few days...

Cleanings? I guess that means opening the tees at the bottom of the returns?

As for shutting off the feeder -- the only control I have is closing the supply valve ahead of it. There is no separate switch and I don't want to disconnect wires.

arbalest

Jamie Hall said:

...if it does operate on some runs, try and figure out what is different about them. Longer running time? Do you sometimes have some radiators shut off, but on at other times?

No radiators valves are being opened or closed. As for what is different -- the pattern seems to be that the more skimming that I do the less often make-up water is added during a heating run.

It is warm here in Chicago and will be for a few days. So, to get longer runs the thermostat has to be in the mid to upper 70's.

BobC

Perhaps the returns can be flushed out with a hose from the main vent locations? Is there a drain at bottom of the returns?

Bob

arbalest

BobC said:

Perhaps the returns can be flushed out with a hose from the main vent locations? Is there a drain at bottom of the returns?

There is no drain now but when this h-loop problem is corrected there will be a drain. Thanks for the idea of using a hose in the vent ports.

Jamie Hall

arbalest said:

Jamie Hall said:

...if it does operate on some runs, try and figure out what is different about them. Longer running time? Do you sometimes have some radiators shut off, but on at other times?

No radiators valves are being opened or closed. As for what is different -- the pattern seems to be that the more skimming that I do the less often make-up water is added during a heating run.

It is warm here in Chicago and will be for a few days. So, to get longer runs the thermostat has to be in the mid to upper 70's.

Sounds as though maybe the water level is bouncing enough to trip the feeder. Leave the thing shut off -- that valve will do -- for a while (but keep an eye on the water level!).

Abracadabra

arbalest said:

BobC said:

It sounds like your returns could use a cleaning... I would leave the water feed to the automatic water feeder off for a few days...

Cleanings? I guess that means opening the tees at the bottom of the returns?

As for shutting off the feeder -- the only control I have is closing the supply valve ahead of it. There is no separate switch and I don't want to disconnect wires.

You have dry returns. You don't need to worry about flushing out any wet returns. The only area the returns are wet is the 1-2' return header at the boiler.

arbalest

Last update (for a while at least).

I haven't heard hammering in a while and the original installer is scheduled to return to correct the h-loop problem per the diagram I drew based the suggestion @Abracadabra gave me when he stopped by. I'll post a photo when he is done just to close out the issue.

Also, @Abracadabra will probably install a second riser along with a proper header and connection to the equalizer at some point in the future. I have a few more general questions about this new boiler but they are separate from this issue so I am going to ask them in their own threads. Thanks again to all who provided feedback.

Fred

Best of Luck! You're in good hands with Abra.