Air in hot water pipes (worry free solution?)

Reply

  #1  
Old 12-01-08, 09:12 AM
L
Member
Thread Starter
Join Date: Nov 2008
Posts: 16
Received 0 Votes on 0 Posts
Air in hot water pipes (worry free solution?)

Hi guys,

This site is really awesome. I've got most of my questions answered as I am trying to do it myself, thanks to all the people who are monitoring this site and others who are just contributors. It is fun.

Anyway, here is one question I have been asking myself. Air in the pipe is very common as a lot of other threads have been providing solutions. The question is, I've done my homework to purge the pipe for a long time, is there a worry free solution to this issue?

My problem is that I am at the end of the street. The water pressure is really low (for purging air out). Plus, 3/4' pipe line going to the second floor is long so that the pressure drop is big. When the circulator pump is running, if I open the cap on a tee (I do not know what this thing is called) upstatirs, sometimes I hear air getting into the pipe instead of going out (well, you may say my pump may be weak).

I am living in New England. My house is about 2200 sf and a regular colonial style. The heating system is in the basement. It has 3 zones (2 downstars and 1 upstairs), with a TACO 007-F5 pump and 3 TACO zone valves. A Honeywell Aquastat is controlling the pump and the gas burner. I have an expansion tank on the main line coming out of the burner.

I have been thinking of an air scoop with or without an additional expansion tank. Is this something I need to pursue? Do I have to install one for each zone. Or, is there any other solution?

Thanks in advance!

Long
 
Sponsored Links
  #2  
Old 12-01-08, 09:58 AM
X
Member
Join Date: Nov 2005
Location: USA
Posts: 2,459
Received 0 Votes on 0 Posts
Originally Posted by longmian View Post
When the circulator pump is running, if I open the cap on a tee (I do not know what this thing is called) upstatirs, sometimes I hear air getting into the pipe instead of going out
You have just described an absolutely textbook case of what happens when the near-boiler piping is not configured as "pumping away."

Search around here for threads on pumping away. There's also a fantastic, simple book called Pumping Away by Dan Holohan that provides explanations and diagrams.

What is happening is that your circulator is probably mounted on the return to the boiler. Meanwhile, the expansion tank is mounted on the bottom of the air eliminator, which is on the supply. The expansion tank is the point in your heating system where the pressure is constant. Usually around 12-15 psi.

When the circulator runs, it creates a pressure differential. Because the expansion tank is on the outlet side of the circulator, the pressure differential required to move water is achieved by creating negative pressure at the top of your system. That's why you hear air sucking in when you loosen a cap upstairs. This also creates long-term air problems that don't go away. Every time the pump turns on, the pressure drops and any microbubble of air in your system expands (because of the decreased pressure) into a macrobubble, which then gurgles around seemingly forever.

In a 'pumping away' configuration, the point at which the expansion tank connects to the system is on the inlet side of the circulator. In this configuration, the circulator must create its pressure differential by adding positive pressure to the system. This has the effect of keeping microbubbles as microbubbles, and moves them around the system to the air eliminator.

There are two ways to fix the problem:

1) move the circulator to the supply side, just downstream of the point where the expansion tank connects to the system. Or

2) move the expansion tank to just upstream of the circulator. For example, remove the tank from the bottom of the air eliminator and install a plug. Cut in a tee just before the circulator and attach the expansion tank off the tee.

Both require some thought and attention to how you want the finished product to look, but both achieve the same result (my personal preference is to move the circulator, but sometimes that requires a lot more work). Both require draining most or all of the system. Plenty of help here should you choose to attempt it, or call a local professional.
 
  #3  
Old 12-01-08, 01:59 PM
G
Member
Join Date: Oct 2008
Location: USA
Posts: 2,954
Received 10 Votes on 10 Posts
"Pumping away" is good advice and the preferred solution. However, there may be another, simpler fix.

Although "pumping away" is accepted as good practice, it was not always so. For example, the Bell & Gossett Engineering Manual from the early '50s recommends the arrangement that you seem to have, and many systems were installed that way, and work satisfactorily - particularly for 2-story houses with the hot-water boiler in the basement. The modern book by Siegenthaler, while recommending pumping away, does recognize that your system's approach was followed for years with satisfactory results.

Has your system ever worked satisfactorily? Was the pump recently replaced with one that has greater capacity (possibly resulting in higher pressure drop around the circuit)?

Three zones for a 2200 sqft house will cause quite a bit of pressure drop when just one zone is calling for heat. Have you always had three zones?

What is the pressure of the system when it is cold and when it is hot? Try jacking it up to about 18-20 psi (cold), and see if that helps.

Bleed the radiators again with the pump off.

You should have an air removal device installed in your system, no matter what piping arrangement you use.

I'm hoping that maybe you can get by at least until spring.
Doug
 
  #4  
Old 12-01-08, 02:19 PM
L
Join Date: Nov 2008
Location: Connecticut
Posts: 50
Received 0 Votes on 0 Posts
I agree with Gilmorie, try increasing the pressure in you heating system to 16-18 psi. (Your water supply pressure may be low but I am sure it's not lower that 18 psi)
 
  #5  
Old 12-01-08, 04:00 PM
NJT's Avatar
NJT
NJT is offline
Member
Join Date: Oct 2006
Location: USA
Posts: 23,539
Received 3 Votes on 2 Posts
Yes, excellent physical description, but you didn't give us any running data ... as the others have suggested, what is the pressure on the boiler when cold, and again when hot ... and tell us the temperature that you took these pressure readings.

If your city water pressure is less than the pressure in the boiler, I'd sure ask them why I'm paying a water bill ... water will just 'piddle' out of a tap at 15-20 PSI ...

One clarification I'd like to make though ... the more zones that are open, the lower the pressure drop ... or 'head' ... just like resistors in parallel (hey Xiphias, there's the electrical analogy again!) the total resistance (head) is always lower than the lowest one ...

It's probably not 'linear' as resistors would be, but the basic premise is true.

Let's say there's two zones, with valves... one zone is open, lets say 5 feet of head. Second zone opens, also 5 feet of head ... total head would be somewhere around half ... or 2.5 feet ...
 
  #6  
Old 12-01-08, 08:02 PM
X
Member
Join Date: Nov 2005
Location: USA
Posts: 2,459
Received 0 Votes on 0 Posts
We've got a 007 here. Typical residential circulator. Nothing outrageous. There's a long 2nd-story circuit. Let's suppose it flows at 4 gpm, which might be about 10 ft of head in a long 2nd story circuit of 3/4" piping. The pressure differential required to do 4 gpm at 10 ft of head is about 4.3 psi.

Two things are happening. One, that 4.3 psi differential is being achieved by dropping the pressure at the top of the system. Two, that pressure reduction is allowing small bubbles to expand into larger ones that get pulled through the system. When they get back to the air eliminator where the expansion tank is, they are seeing the point of maximum system pressure and are compressed once again.

It's a real pain to get a system like this purged of air. Every purge introduces fresh, oxygen-rich water. Lather, rinse, repeat....

Yes, many, many, many older systems (and even new ones) work fine without pumping away. But old or new, there is a limit to how far you can push it. Sounds to me like this system is pushed too far.

Jacking up the system pressure to 18-20 psi cold won't help much, if at all. You still have a 4+ psi pressure drop, just from the circulator. When hot, the system is also probably in the 22-24 psi range, which is getting close to starting the relief valve weeping.
 
  #7  
Old 12-01-08, 08:09 PM
L
Member
Thread Starter
Join Date: Nov 2008
Posts: 16
Received 0 Votes on 0 Posts
Wow, you guys are throwing a lot of stuff at me - good stuff.

Okay, here is the data. When system is cold, the pressure on the burner gauge is around 10 psi. When hot (around 180 F), it is a little less than 20 psi.

The expansion tank is on the outlet main line just above the burner - just as you guys guessed. The circulator is at the return line going into the bottom of the burner. I do not know if I have an air removal device as gilmorrie has mentioned. All I can see is a Hy-vent mounted on top of the burner - it is parallel (coming out of the burner) to the main pipe which has the expansion tank installed.

I have recently replaced the circulator pump and the aquastat. They are both exactly the same models as the previous ones.

Increasing cold water pressure is probably a good idea to try but I am a little nervous about leaks this may introduce.

I moved into this house 5 years ago.It always has air in the pipe. I thought I could purge the air out so I read a lot on this site on how to get air out. I think I am doing the right thing but still there is air!

As xiphias has said, the 2 approaches he mentioned will require a lot of work in my setting. I'd ask if adding an air scoop will help in my current configuration.

Thanks again!

Long
 
  #8  
Old 12-01-08, 08:17 PM
X
Member
Join Date: Nov 2005
Location: USA
Posts: 2,459
Received 0 Votes on 0 Posts
The Hy-vent is the air eliminator. It is usually mounted on a tapping on the boiler, or from your description, mounted on an air scoop in-line with the boiler piping, which often also has the expansion tank attached at the bottom. Or maybe it's just sitting at the top of an elbow.

If you have had air problems for 5 years, but the system otherwise heats the building OK, then you most likely have a 'not pumping away' problem causing the air noise.

10 psi system pressure cold is a bit low. Bump it up to 12-15. If the hot pressure gets into the 22+ range, keep an eye on the relief valve. If it starts to weep, drain a small bit of water to get the pressure back down. But this still won't cure the air noise problem, unless there's something else going on with your system that we don't know about.

Moving the expansion tank connection is usually pretty simple. Consider snapping some pics of the boiler and adjacent piping. Host them at photobucket of similar hosting site and provide links here.
 
  #9  
Old 12-01-08, 08:18 PM
L
Member
Thread Starter
Join Date: Nov 2008
Posts: 16
Received 0 Votes on 0 Posts
Okay, now I understand xiphias's suggestions a bit better. My expansion tank is attached to a vertical pipe - there is no air eliminator (or air scoop?).

I am sort of telling myself to install an air scoop to the return line of the upstairs zone (prior to the circulator). Should this help? Do I need to add air scoops to all 3 zones? The reason I think of this is that it is much easier to do than moving the circulator around.
 
  #10  
Old 12-01-08, 08:25 PM
L
Member
Thread Starter
Join Date: Nov 2008
Posts: 16
Received 0 Votes on 0 Posts
I currently do not have air scoop in my system. The Hy vent is mounted directly on the pipe that comes out of the top of the burner [ I believe Long means the BOILER here - ed. ]. The pipe also leads to the relive valve. I think the configuration of this Hy vent (air eliminator) does not help much in venting the air as there is no other mechanism to trap the air leading to this vent. That is why I am thinking of an air scoop on the returning line.
 

Last edited by NJT; 12-02-08 at 04:59 PM.
  #11  
Old 12-02-08, 06:04 AM
L
Member
Thread Starter
Join Date: Nov 2008
Posts: 16
Received 0 Votes on 0 Posts
You are right, xiphias. I have had air problem for the last 5 years - my house generally heats okay otherwise. It is just the noise and the feeling that there is still air (quite some) in the pipe make me uncomfortable.

As NJ Trooper explained, I now understand why sometimes I hear air sucked into the pipe through the bleeder upstairs while other times air actually coming out - the 'head' thing (pressure drop). I have all 3 zones on at times during the day and only the zone upstairs on during the night. The 3 zones might call for heat at different times (rare except for the upstairs zone) as they are independently controlled.

I think the air is moving along the water, not trapped. I think a inline air scoop (air purger as others call it) like this (http://www.quality-stuff-for-you.com/info/AP400CAT.pdf) might help, but I need to know the best location to install this thing - on the returning line close to the circulator, or on the supply line going to the baseboards. And, do I need one for each zone?

Thanks.

Long
 
  #12  
Old 12-02-08, 06:35 AM
O
Member
Join Date: Dec 2007
Location: A Galaxy From Afar
Posts: 421
Received 0 Votes on 0 Posts
Longmain, one item not mentioned is what type of expansion tank is installed: bladder or non-bladder? There is a difference in how the air gets vented from the system. With a non-bladder tank is should be returned back to the tank.

But at the same time, I think the air problem is a little more basic. You mention that when the circulator is running there is negative pressure on the upstairs rad's (or the 2nd floor Tee). IOW, air will enter the system when the bleeder is opened.

This may be the real issue, as it will allow air to be pulled into the system. Air being less dense then water can seep into a system while the water can't seep out.

It may be worthwhile measuring the lengths of pipe along with the diameters on each of the zones. Then calculating each zone's head loss, then a combined head-loss for each mix of zones (1 & 2, then 1 & 3, then 2 & 3, then all 3).

You may find that the 007 pump isn't the best choice. Or that the upstairs zone has a head loss that doesn't match the pump. Which could cause problems when only that zone is active.

TACO has several technical documents (their name) on circulator sizing. TD10 and TD09 seem to cover the requirements and how to figure them.

Al.
 
  #13  
Old 12-02-08, 06:37 AM
X
Member
Join Date: Nov 2005
Location: USA
Posts: 2,459
Received 0 Votes on 0 Posts
If you really do hear air being sucked in at the top of the loop, then an air scoop won't help much. If you have a Hy-vent near the boiler, then it is probably adequate. The problem originates from the pressure drop expanding the air bubbles.

I had exactly this problem and it was my first post on this forum almost exactly 3 years ago. We fixed it by moving the point at which the expansion tank connects to the system. That system (since replaced for other reasons) had a very nice Spirovent air eliminator, but it didn't help at all. The laws of physics are hard to break.

Take some pics of the system and host them at photobucket, flickr or similar site. Let's see what we're dealing with and we can suggest the most straightforward fix.
 
  #14  
Old 12-02-08, 01:04 PM
L
Member
Thread Starter
Join Date: Nov 2008
Posts: 16
Received 0 Votes on 0 Posts
I can confirm that there is air sucked in on the second floor when I open the bleeder cap, and when only the top level is calling for heat.

As to the head loss, here is a rough calculation: From the basement (where the pump is installed) to second level (where the baseboards are), it is at least 14 feet - this already exceeds the spec of 007-F5 (0-11 ft). Loss in pipes (~50 ft) is 40 ft, or 17 psi, according to Hazen-Williams Equation - calculating Friction Head Loss in Water Pipes. The water pressure is ~20psi. So, it is almost certain that at the top of the system there is negative pressure when only the top zone is calling for heat. When more than one zones call for heat, it is better (hey, I like the resistor analogy. Well, we are talking about the worst scenario here.

I will be working on those pictures tonight and post back.

Forgot to mention, the tank is bladder type - I believe.
 
  #15  
Old 12-02-08, 05:09 PM
NJT's Avatar
NJT
NJT is offline
Member
Join Date: Oct 2006
Location: USA
Posts: 23,539
Received 3 Votes on 2 Posts
When system is cold, the pressure on the burner gauge is around 10 psi.
What temperature is 'cold' ?

A pressure difference from cold to hot of 10-ish PSI suggests that the expansion tank should be given some attention ... probably close to waterlogged (compression tank), or low on air charge (bladder tank).

Before you start troubleshooting anything else, make sure the rest of the house is in order first.
 
  #16  
Old 12-02-08, 05:12 PM
NJT's Avatar
NJT
NJT is offline
Member
Join Date: Oct 2006
Location: USA
Posts: 23,539
Received 3 Votes on 2 Posts
this already exceeds the spec of 007-F5 (0-11 ft).
The physical HEIGHT of a system is irrelevant to a pump operating in a closed loop. The water is a ferris wheel, for every gallon that you push up, one gallon is pulled back down by gravity. The ONLY thing that matters to a pump in a closed loop is the FRICTION of the piping and valves et al.

In an OPEN system, physical height has everything to do with pump head, but a heating system is after all, a CLOSED system.
 
  #17  
Old 12-02-08, 05:42 PM
X
Member
Join Date: Nov 2005
Location: USA
Posts: 2,459
Received 0 Votes on 0 Posts
longmian, if you really want to do the head calculation, here's the cookbook:

http://www.taco-hvac.com/uploads/Fil...irculators.pdf

But I've been there and done that. I'll guess your 2200 sf Colonial has a longish 2nd story loop with a total equivalent length (TEL) of 300-350 ft (see the above link for what TEL is...).

A 007 is probably the most common circulator in the U.S. Many boiler manufacturers ship them with their boilers. It can move a lot of water and is just fine for this kind of application. It will likely do just fine in your home, once you either repipe the circulator to be downstream of the expansion tank, or move the expansion tank to the inlet side of the circulator.

It's physics.
 
  #18  
Old 12-02-08, 07:52 PM
L
Member
Thread Starter
Join Date: Nov 2008
Posts: 16
Received 0 Votes on 0 Posts
Sorry that my calculation is off a lot, but I did hear and feel air sucked in - to assume negative pressure scenario.

Water "cold", I mean room temperature, around 60 F.

The house had around 1800 sf (built in 1982) when it was built, but an expansion by the previous owner added about 400 sf. I saw they did some work on the burner but not sure what it was. The configuration of circulator on the return line and expansion tank on the supply side is original - from what I can tell.

I replaced the expansion tank last winter with an identical one because I suspected a bad tank leads to air in the pipe.

Anyways, these are the pictures of my system:

http://i485.photobucket.com/albums/r.../HW-2008-1.jpg
http://i485.photobucket.com/albums/r.../HW-2008-2.jpg
http://i485.photobucket.com/albums/r.../HW-2008-3.jpg
http://i485.photobucket.com/albums/r.../HW-2008-4.jpg
 
  #19  
Old 12-02-08, 08:50 PM
NJT's Avatar
NJT
NJT is offline
Member
Join Date: Oct 2006
Location: USA
Posts: 23,539
Received 3 Votes on 2 Posts
Long, jack that cold system pressure up to at least 12 PSI ... and please do double check that the proper air charge is in the expansion tank. That pressure should match the system pressure when cold. Remember that to check the air charge on the tank, you must have zero on the boiler gauge.

Does the bleeder ever 'inhale' when the circulator is NOT running? If it ever sucks air in when the system is cold and at it's lowest pressure, it's proof positive that you don't have enough pressure to get the water to the highest point. Pumping away or not, if you don't have enough cold pressure, you are going to have problems...

If you wanted to do some experimenting, replace that manual bleed valve with a 0-10 PSI pressure gauge ... wait, better use a combination vacuum/pressure gauge !
 
  #20  
Old 12-03-08, 05:59 AM
L
Member
Thread Starter
Join Date: Nov 2008
Posts: 16
Received 0 Votes on 0 Posts
NJ, the bleeder never sucks air in when the circulator is not running. I can see water drips out from the tee under this circumstance.

When the second level and one or two of the lower level zones call for heat, the bleeder does purge air out from time to time.

I checked again early this morning after purging the air again for a while, the "cold" pressure is at about 13 psi, not 10 as I mentioned before - sorry. The "hot" pressure is still a bit less than 20 psi.
 
  #21  
Old 12-03-08, 06:40 AM
X
Member
Join Date: Nov 2005
Location: USA
Posts: 2,459
Received 0 Votes on 0 Posts
Originally Posted by longmian View Post
NJ, the bleeder never sucks air in when the circulator is not running. I can see water drips out from the tee under this circumstance.
OK, this confirms that the circulator is creating negative pressure at the top of the system. This is cured by moving the circulator or moving the expansion tank.

Looking at your pictures, you have two fairly simple options. The first is how I would do it.

1) move the circulator. HW-2008-4 shows that there is probably enough room above the expansion tank connection to cut out a section of the vertical supply piping, install a pair of shut-off flanges, and move either the existing or install a new 007 new circulator. The plain flanges where the circulator is now can be removed and replaced with a length of straight black iron pipe and couplings.

2) move the expansion tank. Remove the tank from its existing tee, and install a new tee on the return line, right about where the white line is pointing to "Return line" in HW-2008-1. Plug the old tee. Install it off a short nipple and elbow so it hangs vertically. Support the short nipple and tank with a long pipe strap to the ceiling joist.


The second option requires less time and materials, but the first option is the way to do it absolutely right.

If you move the circulator, you also probably do not need to install an air scoop / air eliminator (and have thus save even more money on time and materials). Plus the way your zones are piped it would be somewhat difficult to add one.

Here's why. By moving the circulator, the lowest pressure in the system will be right before the expansion tank. In other words, inside the boiler. This will allow the bubbles to expand and escape through the existing Hy-vent at the top of the boiler. Because the boiler is also heating the water, the hot water will drive the air out of solution.

In this case, the Hy-vent on the top of the boiler should be adequate. I have exactly the same setup on my boiler as yours, with the boiler-top tapping serving as the pressure relief and the air vent point.

It looks like you have sufficient valves on the zones that the entire system would not need to be drained, just the boiler.

I would also add a short nipple (enough to clear the boiler by a few inches) and an elbow to allow the tank to hang vertically. Support with a pipe strap to a ceiling joist.

In an ideal world, you would also repipe the autofill to enter the system where the expansion tank is connected, but it's not essential at this point in time.

So move the circulator and enjoy an air-free system!
 
  #22  
Old 12-03-08, 03:10 PM
NJT's Avatar
NJT
NJT is offline
Member
Join Date: Oct 2006
Location: USA
Posts: 23,539
Received 3 Votes on 2 Posts
OK, this confirms that the circulator is creating negative pressure at the top of the system.
Absolutely ... the _static_ pressure is high enough to lift the water, but the _dynamic_ pressure is being pulled down by the pump.

And 13 PSI cold is about perfect... your house does seem to be in order, and Xiph's last post nailed it!
 
  #23  
Old 12-03-08, 07:34 PM
L
Member
Thread Starter
Join Date: Nov 2008
Posts: 16
Received 0 Votes on 0 Posts
Thanks, guys.

Just did a little inspection on the pipes. They are mostly copper, only the curved part going into the burner from the circulator is iron. So, not bad - I am afraid of threading on iron pipes as I've never done it before and I heard from some people on this site that this can be quite challenging. I have been playing with soldering for a while now and feel better working with copper.

I will take xiphias's advice and move the circulator. It looks like it'd be tough to get it done for this winter - we will see.

A couple of more questions though. xiphias, what do you mean by adding a nipple and an elbow for the tank? You mean, extend the tee, then install the elbow and hang the tank from the elbow, or sit the tank on the elbow? Hanging the tank vertically is better than my current installation?

All in all, it's been very educational for me. I am a software engineer (mostly numbers stuff, you know) but fascinated about this kind of work. Thanks again!

PS: shoot me an email to: ***sorry, email no-no***, if you have computer related issues and want to play around.
 

Last edited by NJT; 12-03-08 at 08:44 PM. Reason: email OK in PM ...
  #24  
Old 12-03-08, 08:28 PM
X
Member
Join Date: Nov 2005
Location: USA
Posts: 2,459
Received 0 Votes on 0 Posts
For the iron part where you remove the old flanges, you should be able to find a pair of black iron (not galvanized) couplings, and a pre-threaded length of black iron piping (otherwise known as a 'nipple') that will do the job. Wrap with teflon tape, then use the good pipe sealant (brand name escapes me -- someone else can supply). Should be good to go.

About the tank. Yes, extend the tee with a nipple (whatever length you need; guessing 12" might do it?) then hang the tank from the elbow. Even better would be to add a ball valve (say in between two 4" nipples) so that you can close the valve and remove the tank without draining the system. These tanks don't last forever, and simple changeout is way easier than draining the system.

Vertical is generally better than horizontal, and easier to support.
 
Reply
Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


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