repeated air problem in hot water baseboard


  #41  
Old 02-18-06, 06:28 PM
Grady's Avatar
Forum Topic Moderator
Join Date: Jan 2005
Location: Delaware, The First State
Posts: 12,682
Received 41 Upvotes on 39 Posts
Sorry

In order for it to suck air, you would have to have a negative pressure condition at that joint. If the joint were at a mono-flo tee, you would experience negative pressure but you should not at an elbow. As long as the pressure inside the pipe is greater than that on the outside water will come out rather than air coming in. With all this being said, that suspious area needs to be repaired.

Who: Your input, please.
 
  #42  
Old 02-18-06, 07:29 PM
Who's Avatar
Who
Who is offline
Member
Join Date: Jan 2006
Location: San Jose
Posts: 2,066
Upvotes: 0
Received 2 Upvotes on 1 Post
The air is probably coming from the feed. Can you isolate that leak from the system until you get it repaired?
 
  #43  
Old 02-19-06, 08:38 AM
X
Member
Thread Starter
Join Date: Nov 2005
Location: USA
Posts: 2,338
Upvotes: 0
Received 0 Upvotes on 0 Posts
sign of delta-P

Duh. You'd think I'd know about which direction the pressure gradient is going; I deal with this kind of stuff daily.... Just shows how discombobulated I am over the whole thing. What's it been, 3+ months now?

Eval and repair of suspect joint early next week.

Who: can't isolate leak as it's early in the series loop for the whole 2nd floor. It's also pretty darn cold here (4F in the backyard this AM) right now with more cold forecast for the week. I think I may, however, turn off the feed tonight and see what happens in a day+; have to go out of town for a week on Tuesday.

Thanks.
 
  #44  
Old 02-19-06, 08:56 AM
Grady's Avatar
Forum Topic Moderator
Join Date: Jan 2005
Location: Delaware, The First State
Posts: 12,682
Received 41 Upvotes on 39 Posts
Discombobulated...LOL

Yep, know just what you mean. The frustration meter has reached about 15 by now on a scale of 0 to 10. If we ever get this thing straightened out, I'm going to save this thread for reference.
 
  #45  
Old 02-20-06, 08:01 AM
X
Member
Thread Starter
Join Date: Nov 2005
Location: USA
Posts: 2,338
Upvotes: 0
Received 0 Upvotes on 0 Posts
induced pressure

Chickened out on the feed shut-off as I'm headed out of town shortly and didn't want it to be the thing I forgot to turn back on before leaving for a week. Did turn it off yesterday afternoon but for only a few hours (maybe 7). No measurable difference in pressure, but I wouldn't expect there to be in just a couple hours if the leak is less than a minor gusher. Nonetheless during the feed-off time the heat cycled several times on all zones and the air is still there. Does that short test eliminate the feed as the source of air? If not, maybe try again when I get back next week.

Question: since we are pumping on the return, could it be that we are inducing negative pressure up at the leak, drawing in air? If the system is at 18#, then we'd need about a 3-4# pressure drop to get below atmospheric. There are experimental approaches (and the ever-popular "theoretical approach") to calculate what the pressure drop is across a circulator, but not sure if I could figure out how that translates back through the system, and I assume it's not constant.

But in any event we'll get the leak addressed. If that doesn't take care of the problem, then back to the drawing board.
 
  #46  
Old 02-20-06, 08:31 AM
Who's Avatar
Who
Who is offline
Member
Join Date: Jan 2006
Location: San Jose
Posts: 2,066
Upvotes: 0
Received 2 Upvotes on 1 Post
Fill a glass of water and watch the entrained air bubbles that are present. If your system is getting makeup water, then it is also getting these air bubbles.

Is your feed located where any incoming air has a good chance of getting eliminated?

I doubt your system is sucking in air through any leaks. Who knows? Nope, I don't. ;-)
 
  #47  
Old 02-20-06, 09:19 AM
X
Member
Thread Starter
Join Date: Nov 2005
Location: USA
Posts: 2,338
Upvotes: 0
Received 0 Upvotes on 0 Posts
The system has a spirovent set up exactly like figure 1 here:

http://www.spirotherm.com/docs/installation/JrIOM-A.pdf

So the answer is yes, the spiro is right where make-up water would be heated and air removed.

(Note we do not have a "spirotop" as illustrated in figure1 [yet...]).

Nor are we "pumping away" from the expansion tank as illustrated in that figure. Which is the way it ought to be; ah the things you learn.... Our circulator is on the return.

Something of an aside. The circulator is presently on the return, hence the question about the pressure drop. But now that I'm moderately educated about heating issues, I'm strongly considering redoing a lot of the near-boiler piping and installing a tekmar 260 outdoor reset control system. Will have to get some estimates as this is way beyond my abilities (and quite possibly way beyond my wallet, but won't know until I get some numbers). That would involve basically creating a separate boiler loop, putting our indirect DHW on its own circulator on the boiler loop, and putting the heating zones on a single circulator fed by a 4-way mixing valve (or via variable speed pump injection) coming off the boiler loop. So if there's a problem with the circulator position or near-boiler piping, this kind of project would sure change things. Unless of course it's a bad feed valve!
 
  #48  
Old 02-20-06, 11:53 AM
Who's Avatar
Who
Who is offline
Member
Join Date: Jan 2006
Location: San Jose
Posts: 2,066
Upvotes: 0
Received 2 Upvotes on 1 Post
Why do you suspect a bad feed valve? If you have a leak and your feed is maintaining the proper cold pressure then it is doing its job.

I'm not so sure that a Tekmar 260 will handle injection pumps. I think you need a 35x/36x series. If the wallet factor picks up look at the tn4 line-up. It looks very impresive and I've been told they are even coming out with an IP interface for it. Dear Santa...

If you are thinking about bringing the near boiler piping up to modern standards, you may just want to consider doing it yourself if you can do the piping. I don't think that combustion related items are part of the realm of us homeowners but piping requires few special tools and truly is a matter of following the manufacturers' instructions.

Boilers aren't common where I live and the people that work on them still tend to be immune from trying to catch up on modern practices like pumping away unless they are doing high end radiant applications and then they couldn't be bothered doing small jobs without charging through the nose. They also tend to want to do everything in copper even when the whole system is black iron. I'm not sure how it is where you are but I'm inclined to work on all non-combustion repairs and enhancements. I've done some system branch changes, got rid of the old expansion tank (amazing how much better that made life - I can't remember the last time I bled the rads but it used to be about 3 times a year). replaced a leaky PRV (that started my curiosity about hydronics), and soon I'll be redoing the piping to a p/s layout that is pumping away as well as tying in the underfloor plates and PEX for the kitchen (and baths in the future). I wish the other aspects of home maintenance were as rewarding as heating.
 
  #49  
Old 02-20-06, 01:04 PM
X
Member
Thread Starter
Join Date: Nov 2005
Location: USA
Posts: 2,338
Upvotes: 0
Received 0 Upvotes on 0 Posts
Misunderstood where you were headed with the feed valve. I get it now.

Still researching the tekmar options. Can't remember off the top of my head whether the 260 does the injection pump. Might be a 4-mixing valve operation. In any event, some boiler loop/system loop isolation is necessary.

I agree on non-combustion home improvement. Have done years of everything in this house (and previous homes). Did some plumbing for sinks and such, but never until now really gave heating much thought. It was a sleeping dog....
 
  #50  
Old 02-20-06, 05:14 PM
Grady's Avatar
Forum Topic Moderator
Join Date: Jan 2005
Location: Delaware, The First State
Posts: 12,682
Received 41 Upvotes on 39 Posts
Originally Posted by xiphias
Question: since we are pumping on the return, could it be that we are inducing negative pressure up at the leak, drawing in air? If the system is at 18#, then we'd need about a 3-4# pressure drop to get below atmospheric. There are experimental approaches (and the ever-popular "theoretical approach") to calculate what the pressure drop is across a circulator, but not sure if I could figure out how that translates back through the system, and I assume it's not constant.
Good thought BUT: Yes atmospheric pressure is aprox. 14psi but that is 14 psia (absolute) & the 18 psi on the boiler is 18 psig (gauge) which is actually 32 psia (absolute). Even if you add in the pressure needed to push water to the second floor, you are only looking at another 6 psig max.
 
  #51  
Old 02-23-06, 05:59 PM
X
Member
Thread Starter
Join Date: Nov 2005
Location: USA
Posts: 2,338
Upvotes: 0
Received 0 Upvotes on 0 Posts
curses, foiled again!

bah humbug! While I'm out, the system is getting a full pressure test. Assuming everything passes, then back to the drawing board.
 
  #52  
Old 02-24-06, 04:09 PM
Grady's Avatar
Forum Topic Moderator
Join Date: Jan 2005
Location: Delaware, The First State
Posts: 12,682
Received 41 Upvotes on 39 Posts
Thumbs up Sense of Humor

You are better than most. By now, most would have long lost their sense of humor. Glad to see you still have yours. Hang in there. When all is said & done I think we will all have learned something.
 
  #53  
Old 03-09-06, 01:34 PM
X
Member
Thread Starter
Join Date: Nov 2005
Location: USA
Posts: 2,338
Upvotes: 0
Received 0 Upvotes on 0 Posts
stump the pros continues....

Well we never did have a full-on pressure test while out of town. Powers that be decided that if there was a leak, then we'd have seen it by now.

Still have not done a more extensive shut-off of the feed valve test per Who's earlier suggestion. Planned for this weekend when I'm around to watch. 48 hr enough time?

Meanwhile, a Taco Hy-Vent was added to the boiler by T-ing off at the pressure relief valve.

Zero effect. Still have same symptoms, etc. Question: why haven't the powers that be considered putting a Hy-Vent or Spirotop at the end of a 2nd floor baseboard run? I think one could fit under the end cap. Pros? Cons? Supply side or return side of a baseboard section? Early in the loop or late? In theory a closed series loop doesn't need one so long as air elimination is done down at the boiler, but in my view theory went out the window a while ago.

The current thinking apparently involves moving the circulator so we're pumping away from the expansion tank. Unfortunately, given the way the system is piped right now, that amounts to fairly major surgery; everything is piped rather compactly and close to the basement ceiling. Not going to be an easy way to move it over and keep the same setup, at least from the way I look at it. And I've been staring at it enough to have at least some clue. Which is why I'm contemplating just changing over the whole system (see other thread). The cost difference might be recoverable in a fairly short time (~5-10 yr, maybe less depending on fuel prices) given the increase in efficiency.

Of course, before going that route, it would be nice to figure out what the heck the air problem is and how to get rid of it!!!
 
  #54  
Old 03-09-06, 02:51 PM
Grady's Avatar
Forum Topic Moderator
Join Date: Jan 2005
Location: Delaware, The First State
Posts: 12,682
Received 41 Upvotes on 39 Posts
Hy-Vent

You can put a Hy-Vent on the upstairs loop but keep in mind the potential for a leak. As far as where to put it, it's kind of a double edged sword. Put it on the begining of the loop & you may not get the air (has not yet come out of solution?). Put it at the end of the loop & the air might not make it that far (trapped horizontally?). As long as you are doing it, why not put one on both ends? You can always screw the cap down tight.

Who: Your thoughts, please.
 
  #55  
Old 03-23-06, 04:53 PM
F
Member
Join Date: Mar 2006
Location: Wet side of Washington state.
Posts: 16,321
Received 38 Upvotes on 30 Posts
It's too bad that you have suffered these many months as your problem is almost certainly caused by the improper location of your circulating pump.

ALWAYS have the circulator pumping away from the expansion tank, ALWAYS! You did not have a problem when the heat only supplied the first floor, your problem arose when you added the second floor heat. This is an absolutely classic case of a misapplied circulator.
 
  #56  
Old 03-24-06, 10:12 AM
X
Member
Thread Starter
Join Date: Nov 2005
Location: USA
Posts: 2,338
Upvotes: 0
Received 0 Upvotes on 0 Posts
indeed! (Warning: Rant alert below!)

Well, it ain't over yet. Getting somebody competent around here, with room in their schedule for a smallish job, turns out to be pretty tough. A proper repipe will be 1-2 people for about a full day, which around here is apparently a small job, at least for the good guys.

But let me give a summary to date, since you've reactivated the thread. (Also since you know about pumping away, I'm not telling you anything new, but for others' benefit, here goes.)

The problem certainly does appear to be an improperly situated circulator, pumping on the return, or more importantly, towards the expansion tank rather than away from it. I ran a bunch of numbers on pressure drop across the circulator, then a hydronics pro did a pro bono calculation for me and came up with the same thing.

The physical problem is this: with the circulator pumping toward the expansion tank, the pressure differential across the pump from suction side to discharge side is achieved almost entirely by creating negative pressure on the suction side. This drop in pressure extends throughout the upstairs loop. The pressure drop is sufficient to cause the entrained, compressed air to come out of solution, causing the air noise. The pressure drop may or may not also be sufficient to suck in air from any valves or fittings. Given that most of the valves in the system are right near the pump intake, this is a definite possibility.

Before the addition, we never had an air problem, period. I presume this is because that while we still had a negative pressure issue, the negative pressure was not high enough to cause the air to come out of solution, or get sucked in through valves, etc. Shorter pre-addition loop = less hydraulic resistance = less head [pressure differential] required to move water -> less pressure change -> air doesn't come out of solution or get sucked in through valves/fittings.

<rant on>

This is probably why there are thousands of hydronic systems that are piped with the circulator pumping toward the expansion tank that never have a problem. You have to cross a threshold before you get unwanted behavior. Most systems don't cross it, and thus there's no incentive for changing practices or getting educated about it (which could explain why this problem was such a mystery to some of those who looked at it). Who would have thought adding ~30-40 ft of piping to a zone would cause this problem? Certainly not me. Four months ago, I was blissfully ignorant about every last bit of just about anything having to do with my heating system. Now I'm hopelessly book-smart about all of it.

Probably the most interesting practical aspect of this, is that had the first plumber in the door said "Hey you're not pumping away and you need to re-pipe the system. Re-piping will be major surgery given the present piping layout and cost XX dollars," I would have flipped out. "How the heck can adding a mere 30-40 ft of pipe upstairs cause me to spend XX dollars in my basement?" I would have said. "We've finished the project, maxed out the budget, now this?" I would not have believed it, thought him a charlatan, and found someone else, who would likely have given the answer I wanted to hear but not fixed the problem. So, painful as it has been, educating my way through this has actually provided some insight and knowledge and made me a better consumer. I'm not at all happy about the mental and physical cost because it has absolutely, 100+% been NOT worth it. I have more important things to do with my time.

</rant off>

When we finally do get things re-piped, will post back with final results.

For further reading, I suggest two things:

1) http://www.contractormag.com/article...chydrotemp.asp which is a short, clear description with figures of pumping away vs. not.

2) "Pumping Away" by Dan Holohan. It can be had for short money at http://www.heatinghelp.com/shopcart/...?category=2-16 and is a quick, fun, informative read.
 
  #57  
Old 03-24-06, 05:50 PM
F
Member
Join Date: Mar 2006
Location: Wet side of Washington state.
Posts: 16,321
Received 38 Upvotes on 30 Posts
You have it exactly!

When "forced" (or pumped) hot-water systems were first being installed, many times just a pump being added to a "gravity" system, the pumps were placed on the system return and pumped into the boiler. This was done because the materials in the pump seals (usually packing rather than mechanical seals) had a shorter life due to the temperature of the supply water. The lower temperature of the return water was easier on the seals. There were several other factors that made the location of the circulator pump much less critical.

Unfortunately, the physics of the pump location in regards to the expansion tank was never fully explained to the masses of designers and mechanics installing these systems and therefore the old practice of installing the pump on the return and connecting the expansion tank somewhere downstream of the pump discharge continues even to this day. As was noted, many (most) residential boiler manufacturers mount the circulating pump on the boiler inlet and then provide a tapping on the boiler for the expansion tank or else show the expansion tank connected to the boiler outlet piping, usually on the bottom of an air eliminator fitting.

Modern pump seals now operate at temperatures in excess of 250 degrees F. with no problem yet the pumps are STILL installed on the return side, either because "that's the way it's always been done" or because the designer or installer simply doesn't know the proper location.

I'm somewhat mystified why changing the "point of connection" of your expansion tank should be such a major repiping job. All you really need to do is have the P.O.C. prior to the pump suction. Perhaps you have a diaphragm tank screwed into the bottom of an air eliminator fitting with a main air vent on the top of the air eliminator. If so all you need to do is remove the diaphragm tank, plug the bottom of the air eliminator and move/reconnect the diaphragm tank prior to the pump suction. Most expansion tanks have a 1/2 inch pipe connection and simply cutting in a tee (and union) prior to the pump suction should not be that difficult. The tank does not need to be connected directly to the main line nor does a diaphragm tank need to be installed with any special orientation.
 
  #58  
Old 03-25-06, 03:41 AM
X
Member
Thread Starter
Join Date: Nov 2005
Location: USA
Posts: 2,338
Upvotes: 0
Received 0 Upvotes on 0 Posts
thanks furd

Understood. That's the basic deal.

Agree that moving the tank should be simple. I recall reading in Holohan's book that there was some caveat about having the tank/pump on the return. Will have to check. Would also have to move the feed valve as it's piped into the base of the spirovent, but as you point out, no big deal.

[Edit: checked Holohan. p. 41-46 address leaving the circ and moving the tank. The caveats are 1) potential problems in taller buildings with 30# relief valves and high required static pressure. Pumping away adds pressure and could pop the relief. Not a problem here. 2) possible incorrect addition of a second pump on the return outside of the main pump/tank return line. Not a problem here, either; looks like he threw that in because it happens during some ill-planned retrofits/additions. So we could do the simple, low-cost alternative. However, if budget allows, I definitely want to do the stuff below.]

So yes, you are likely correct that it could be a simple fix. But I'd like to take advantage of everything I've learned over the past four months that could make this setup a lot more efficient, improving performance and saving money over the life of the system. Namely:

1) boiler is way oversized for the heating load. More for the indirect load (though I think it was unintentional...). Short cycling is an issue.

2) 40gal indirect is on 3/4" pipe and a zone valve. Would like to put it on its own circ and use 1" (tank will accept 1").

3) with the above in mind, use a tekmar 260 to do outdoor reset with DHW priority.

It's a conventional gas boiler, so boiler protection is in order. Thinking of doing that either with a controlled 4-way mixing valve (e.g., use tekmar 360 instead of 260) or possibly thermostatically with an ESBE TV valve (not sure if that would work, though). Could possibly go all out with primary/secondary and injection pump, but for a fin-tube convector hotwater system, that seems like overkill. From what I've read, fin-tube output drops off significantly at supply <140F.
 

Last edited by xiphias; 03-25-06 at 04:53 AM.
  #59  
Old 04-06-06, 06:27 AM
X
Member
Thread Starter
Join Date: Nov 2005
Location: USA
Posts: 2,338
Upvotes: 0
Received 0 Upvotes on 0 Posts
and the answer is...

PUMP AWAY FROM THE EXPANSION TANK CONNECTION. A good purge helps, too.

We left the circulator on the return. We moved the expansion tank connection from the discharge side of the circulator to the inlet side. A MUCH easier job than trying to move everything around on the supply side.

One good purge later, this system is silent.

Pumping away from the expansion tank just completely changes the way pressure operates through the heating system. I reiterate the earlier suggestion to check out the article by Mark Eatherton and the book by Dan Holohan suggested above.

In looking back through this looonnngg thread, I noticed that early on, Chimney Cricket called it when he said to check for "pumping away." Unfortunately, at the time I was basically clueless about what he was talking about and the way he described it didn't sound like that applied to my system. The things you learn....

Also interesting that at least four professionals (three plumbers and one hydronics design guy) stared at and/or worked on this system and nobody said anything about how the circ and tank were set up in relation to each other. One guy eventually did, but he would have moved the circ (major surgery relative to moving just the tank) rather than reposition the tank connection.

Having read all about pumping away and how it changes the pressure characteristics of the system, it is really cool to see that play out in my system. Every single problem we've had has disappeared. There's something to this physics stuff.

FWIW, the Taco 007 is one heck of a pump. The 2nd floor loop has a Total Equivalent Length (TEL) of around 324 ft. That's long. But the pump is pushing 3.95 gpm (probably more) through this loop. It is also fine with both zones (and the indirect as well) open. Right around 4 gpm everywhere. Cool.

Many thanks to everyone for sticking with this. I trust we've all learned something. I sure have!
 
 

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