Weird problem in second story zone - water hammer 'clunk' on shut off

Reply

  #1  
Old 03-31-10, 09:53 PM
A
Member
Thread Starter
Join Date: Dec 2007
Location: New Jersey
Posts: 55
Received 0 Votes on 0 Posts
Weird problem in second story zone - water hammer 'clunk' on shut off

First, thanks so much to the helpful folks here - especially NJ Trooper, who has helped me to understand and maintain my baseboard system with extreme patience!

The background - oddly installed (in my opinion), 15+ year old system that uses old steam pipe holes to route the copper lines from the basement to the second floor.

There are four zones, 2 up, and one down + a boiler mate. The risers to the second story have supply/return running next to each other. They enter the baseboard at one end, run around the perimeter of each zone doubled up - meaning the supply and return lines are running parallel inside each stretch of baseboard. At the opposite end, there are elbows that terminate the run and send it back to the riser.

I'm not sure that description made sense, or if it is important to the problem at hand, but there it is...

Now, the problem:

Since we moved into the house, one zone has behaved strangely. After the water has been circulating for a while, and is good and hot, there is a "bubbling" sound that seems to come from the riser. The bubbling builds stronger until the thermostat shuts down the circulator. Then, there is a loud clunk - like a water hammer - that shakes the full run of baseboard.

This condition ONLY happens when no other zone is circulating. If there is circulation elsewhere, the bubbling sound never starts.

Things that I've done as maintenance/repairs over the past couple years have not affected this condition at all. Here's what's been replaced/maintained:

The circulator on this zone was replaced about 3 years ago.

All of the flow check valves on the entire system have been replaced, two of which just got swapped a couple weeks ago.

The expansion tank and purge valves have all (but one in a crappy location) been replaced.

New pressure relief valve

New pressure regulating valve

Now - my original thought was that it was a failing check valve that was helped along by the pressure of other zones circulating. A recent loss of flow (failed cirulator) on the downstairs zone offered the opportunity to drain the system and swap the remaining original check valves. This didn't fix the clunking problem.

My next guess - the risers for this zone are on an outside wall. Perhaps there is some convection current happening in the riser that is caused by the temperature differential. The convection current creates unequal flow on the supply and return risers, which leads to the water hammer when the mechanical circulation stops.

This seems logical to me, but I have no idea how to fix it. If you've read this far, thanks! Hopefully you've got a clear picture of what's happening.

Any input would be greatly appreciated.

Thanks again,

Ken
 
  #2  
Old 04-01-10, 07:36 AM
L
Member
Join Date: Mar 2010
Location: MD
Posts: 113
Received 0 Votes on 0 Posts
Hi alken

This condition ONLY happens when no other zone is circulating. If there is circulation elsewhere, the bubbling sound never starts.
Things that come to mind when i hear bubbling sounds are.
1 system is low on water
Check system pressure with all pumps off and if you can a cool boiler. then compare it to system up pumps running and a warm boiler.
2 That zone has air in it, see if there is manual bleeder valves on your base board radiation. Shut off all pumps then bleed air from baseboard i start from the highest basemboard and work down.
3 Pump is cavitating suction pressure of pump is below NPSH Is there a valve on the discharge of the pump ?
If so run this pump only to create the bubbling sound.
now slowly throttle the discharge valve toward closed. Stop when and if the noise goes away. Note position of valve, Shut pump off and turn back on listening for the banging sound that you described. Regardless of outcome put valve back to position it was in prior to moving and post results.
 

Last edited by NJT; 04-01-10 at 05:51 PM.
  #3  
Old 04-01-10, 07:53 AM
A
Member
Thread Starter
Join Date: Dec 2007
Location: New Jersey
Posts: 55
Received 0 Votes on 0 Posts
Hi Luke -

Thanks for the reply. I'm going to try and address your suggestions to the best of my understanding (amateur at best!)

1. The system pressure seems good, and this problem persists after MANY instances of me replacing various components, draining the system, refilling, and bleeding. The pressure gauge reads between 12 and 15 Lbs (cold/hot).

2. I am doubtful that there is air - I've bled that zone several times over, hearing no air escaping. All the purge valves are new (couple years old at best), and operating well as indicated from the most recent bleed and recharge (each zone blew off air as it was being filled). Also, the circulation is silent (aside from the noise of the circulator vibrating) until the bubbling starts late in the period of circulation.

3. I'm losing you here... The ciruclator on this zone is new, and just happens to be a bit over sized for the zone. This was a topic of debate here a few years ago when I did the replacement - the plumbing house sold me a pump too big (too powerful) for my needs. I'd be quick to blame this part if the problem didn't exist before that replacement as well.

There is no valve at the discharge of the pump - just a standard cartridge-type circulator. The closest shut-off valve is after the boiler on that zone.

Again, the bubbling noise and clunk ONLY occur when the zone is at it's highest temperature. It's funny that the noise and clunk almost always come within seconds of the thermostat cut off. Just coincidence, I realize, but it's incredibly predictable. Nice silent heat, bubbling starts, within 30 seconds the zone reaches temperature and cuts off - CLUNK!!!
 
  #4  
Old 04-01-10, 11:16 AM
L
Member
Join Date: Mar 2010
Location: MD
Posts: 113
Received 0 Votes on 0 Posts
3. I'm losing you here... The ciruclator on this zone is new, and just happens to be a bit over sized for the zone. This was a topic of debate here a few years ago when I did the replacement - the plumbing house sold me a pump too big (too powerful) for my needs. I'd be quick to blame this part if the problem didn't exist before that replacement as well.

There is no valve at the discharge of the pump - just a standard cartridge-type circulator. The closest shut-off valve is after the boiler on that zone.
This test is to check for cavitation. Most pumps have a NPSH that increases as the gpm increases. To decrease flow should decrease the point at which it cavitates.
Is that shut off valve on the discharge of the pump or suction. IF the zone valve is down stream of the pump you can preform the test.
Pumps being oversized may have escessive velocity going through that zone so throttling closed the zone valve. may also decrease the noise. Mark where the valve is prior to moving incase it has be balanced
 

Last edited by NJT; 04-01-10 at 05:50 PM.
  #5  
Old 04-01-10, 11:20 AM
A
Member
Thread Starter
Join Date: Dec 2007
Location: New Jersey
Posts: 55
Received 0 Votes on 0 Posts
Would this cavitation be dependent on the temperature of the water, or would it be constant?

Edit - nevermind - just did some quick reading on cavitation.

This is starting to make sense to me. If there is an adjustment to be made on the pump, where would it be? Also, is there any problem with using the ball-valve shut off to reduce the flow after the boiler? It's currently wide open.

Perhaps the original pump was oversized to begin with (it was a monster), and I've simply perpetuated the problem. Maybe a smaller pump is the answer.
 
  #6  
Old 04-01-10, 01:53 PM
S
Member
Join Date: Nov 2009
Location: MA.
Posts: 99
Received 0 Votes on 0 Posts
Alken, If you think the pump being oversize is causing the problem try using the pump to control 2 zones by simply manually opening up one of the other flocheck valves to see if it stops the problem. Then do the same thing with another circulator running to see if there's a difference. Control 2 zones with 1 t-stat and then 2 zones with another t-stat and maybe you can narrow down what's going on. You can do this by just opening 2 folcheck valves.
 
  #7  
Old 04-01-10, 04:14 PM
L
Member
Join Date: Mar 2010
Location: MD
Posts: 113
Received 0 Votes on 0 Posts
Originally Posted by alken View Post
Would this cavitation be dependent on the temperature of the water, or would it be constant?

Edit - nevermind - just did some quick reading on cavitation.

This is starting to make sense to me. If there is an adjustment to be made on the pump, where would it be? Also, is there any problem with using the ball-valve shut off to reduce the flow after the boiler? It's currently wide open.

Perhaps the original pump was oversized to begin with (it was a monster), and I've simply perpetuated the problem. Maybe a smaller pump is the answer.
Did you preform the test ? You can reduce the gpm going into the zone by closing off the valve on the discharge of the pump, this is actually how they water balance systems. If this solves your issue, then at your leasure you can working on selecting a proper pump for your head/ GPM requirements.
There is no problem in shutting off the ball valve on the discharge of the pump. It is actually how you determine what Impeller is in the pump and if your impeller is damaged. ( You are not preforming this test . but used it as an example. Long term will cause seal damage. Friction converts to heat damaging the seal. SO you dont want to leave the ball valve closed for long durations. You are actually Going to close the ball valve until the marble sound disappears, and we are hoping that the valve will still be partially open. You wont have Kinetic energy directly converting into heat damaging your seal. Let us know at what point the sounds stops in valve position. Some people dont like ball valves i love them, very linear flow characteristics with minimal pressure drop in the open position.
I suspect you have a constant speed pump there is no adjustment other than water balancing as we described
 
  #8  
Old 04-01-10, 05:18 PM
X
Member
Join Date: Nov 2005
Location: USA
Posts: 2,458
Received 0 Votes on 0 Posts
Is the expansion tank upstream or downstream of the circulator?
 
  #9  
Old 04-01-10, 06:03 PM
A
Member
Thread Starter
Join Date: Dec 2007
Location: New Jersey
Posts: 55
Received 0 Votes on 0 Posts
Spot8 - Thanks for the suggestion - I may try that one as a test.

Luke - I appreciate your patience with explaining this to me. I'll look for a way to adjust the pump - clearly using the ball valve as an adjustment long-term is a bad idea.

xiphias, the expansion tank is downstream from the circulator. The basic configuration is:

Return (and pressure reducing valve from the street supply) > Circulators > Boiler > Expansion Tank/Air Purger > branches to each zone supply > individual zone ball valves > flow check valves > risers and baseboard runs

Given that it was 70 degrees here today, I won't be testing anyting immediately. The next cool/cold night when the heat is called for, I'll go to the basement and play around with things. I have a strong feeling that a simple pump swap using a less powerful model is going to solve a couple things:

1. The clunking/cavitation
2. The hum of the pump vibrating the lines when running.

As I mentioned, the original install of the copper lines is questionable. There's all kinds of movement throughout the house as the pipes expand and contract. I'm clearly not an installation expert, but I worked as a residential architect for 14 years. I've been in a lot of homes with baseboard heat that don't creak and groan like my house!

Thanks to everyone for the input!

Ken
 
  #10  
Old 04-01-10, 06:11 PM
L
Member
Join Date: Mar 2010
Location: MD
Posts: 113
Received 0 Votes on 0 Posts
Luke - I appreciate your patience with explaining this to me. I'll look for a way to adjust the pump - clearly using the ball valve as an adjustment long-term is a bad idea.
I am all about learning so please explain?
If your pump isnt a multiple speed this is how the professionals adjust the pump.
Switching out to a smaller pump should be more efficient.
 

Last edited by NJT; 04-01-10 at 08:02 PM.
  #11  
Old 04-01-10, 06:14 PM
A
Member
Thread Starter
Join Date: Dec 2007
Location: New Jersey
Posts: 55
Received 0 Votes on 0 Posts
Originally Posted by LukeP View Post
Can you explain why this is a bad idea?
I thought that's what I read in your last post:

There is no problem in shutting off the ball valve on the discharge of the pump. It is actually how you determine what Impeller is in the pump and if your impeller is damaged. ( You are not preforming this test . but used it as an example. Long term will cause seal damage. Friction converts to heat damaging the seal. SO you dont want to leave the ball valve closed for long durations.
Am I misunderstanding something?

Ken
 
  #12  
Old 04-01-10, 06:25 PM
X
Member
Join Date: Nov 2005
Location: USA
Posts: 2,458
Received 0 Votes on 0 Posts
Possible you are depressurizing the top of the system due to expansion tank placement and overly large circulator.
 
  #13  
Old 04-01-10, 07:00 PM
A
Member
Thread Starter
Join Date: Dec 2007
Location: New Jersey
Posts: 55
Received 0 Votes on 0 Posts
Originally Posted by xiphias View Post
Possible you are depressurizing the top of the system due to expansion tank placement and overly large circulator.
Would that happen for very isolated periods when the zone is at the highest temperature? The bubbling typically only lasts for 30 seconds or so, as the coincidental shut-down via the thermostat stops the flow. The rest of the period of time when it's running, there is no bubbling.

It still seems replacing the pump with a smaller one may do the trick for that condition as well, though...
 
  #14  
Old 04-01-10, 07:15 PM
L
Member
Join Date: Mar 2010
Location: MD
Posts: 113
Received 0 Votes on 0 Posts
Originally Posted by alken View Post
I thought that's what I read in your last post:



Am I misunderstanding something?

Ken
I could have worded it a little better
I should have said leaving the valve closed for extended period of time is not good. And will create alot of heat causing your seal to fail. But if the valve is partially open it will prevent heat build up .
This is more of a test to see if the pump possibly is cavatating.
Also as xiphias has said or is refering to
Your expansion tank should be on the suction side of your pump ( or upstream of it) When a pump comes on the suction pressure drops quickly possible below the NPSH causing the pump to cavitate. IF its on the suction it acts like a cushion preventing radical suction pressure drops. NPSH is a fancy word for net positive suction head.
In lamens terms its your suction pressure. Your pump will have a curve that will show you at what point your pump will cavitate.
 
  #15  
Old 04-01-10, 07:57 PM
A
Member
Thread Starter
Join Date: Dec 2007
Location: New Jersey
Posts: 55
Received 0 Votes on 0 Posts
That all seems to make sense. I'll definitely run that test next time the zone is circulating.

Thanks again, and for clearing that all up.

Ken
 
 

Thread Tools
Search this Thread
 
Ask a Question
Question Title:
Description:
Your question will be posted in: