Expansion Tank Connection to Air Separator Question

  • Active since 1995, Hearth.com is THE place on the internet for free information and advice about wood stoves, pellet stoves and other energy saving equipment.

    We strive to provide opinions, articles, discussions and history related to Hearth Products and in a more general sense, energy issues.

    We promote the EFFICIENT, RESPONSIBLE, CLEAN and SAFE use of all fuels, whether renewable or fossil.
Status
Not open for further replies.

dogwood

Minister of Fire
Mar 22, 2009
825
Western VA
I have two Amtrol Extrol bladder type expansion tanks, an SX160 and an SX30 which I plan to pipe together from their bottom fittings . Once piped together I will tee off a single pipe to connect to either just before, or directly to, the bottom of, a Taco 4900 Series Air Separator. The SX160 accepts 1-1/4" pipe and the SX30 accepts 1" pipe. The bottom of the air separator accepts 1/2" pipe. The piping to and from the air separator from the boiler is 1-1/2".

Should I reduce all the pipe from the expansion tanks to 1/2' and connect directly to the bottom of the air separator? Or should I connect 1-1/4 and 1'' pipes from the expansion tanks to a section of 1-1/2'' pipe teed off from just before the air separator. Does it make a difference? I thought since Amtol had sized the pipe openings on the expansion tanks the way they did, maybe it does. The Tarm Biomass PT3 piping schematic I'm using shows the pipe from the expansion tank connecting to just before, rather than to, the bottom of the air separator. (pp 21, http://www.woodboilers.com/admin/uploads/public/WoodBoilerPlumbingSchematic0111Web.pdf), I'd appreciate any advice.

Mike
 
If you do the arithmetic you'll get a few gallons per hour flow to and from the expansion tanks, so 1/2 inch is plenty big.

I think the large connections on the Amtrol tanks might be because historically expansion tanks that big would be connected to large systems, not tiny residential systems with relatively huge storage tanks.

In a large system you have not only the thermal expansion to deal with, but also pressure spikes caused by trying to overcome the momentum of big columns of water and getting the water moving, so expansion tanks can serve to relieve start-up stress on big pumps, and would therefore require big pipes to absorb the large momentary flows.
 
Thanks Elliot and Huff. I wonder if they even make an 1-1/4 to 1/2' reducer. Maybe a brass bushing. It would have to be brass to go between steel/iron and copper, correct? Here's another question just came to mind. When you go from brass to steel/iron threaded connections, do you need to use a different sort of pipe dope. I bI have a can of Rectorseal #5 Iwas going to use on the black iron to black iron threaded connections, but would that also work on the brass to black iron or steel?

Mike
 
dogwood said:
I wonder if they even make an 1-1/4 to 1/2' reducer.
You bet they do. On a side note, if you get stuck on a Sunday afternoon you might be pleasantly surprised to find Ace hardware carries a lot of oddball stuff.
It would have to be brass to go between steel/iron and copper, correct?
I've consulted professionals, I've googled that one a lot, and I've seen 38 year old copper to iron fittings that look good as new; I have come to the conclusion that it is just not needed in a sealed hydronic system.
Here's another question just came to mind. When you go from brass to steel or iron threaded connections, do you need to use a different sort of pipe dope. I beleieve i have a can of Rectorseal #5 Iwas going to use on the black iron/steel threaded connections but would that also work on the brass to black iron or steel?
Can't imagine that you could go wrong no matter what sealant/lubricant you used with brass fittings.
 
Thanks again Eliot. Just as you said, found a 1-1/4" to 1/2" black iron bushing on PexSupply. I won't worry about brass fittings to transition from black iron to copper, and will go with the Rectorseal #5 on the copper to iron connections too.

Mike
 
dogwood said:
Thanks again Eliot. Just as you said, found a 1-1/4" to 1/2" black iron bushing on PexSupply. I won't worry about brass fittings to transition from black iron to copper, and will go with the Rectorseal #5 on the copper to iron connections too.

And be sure to take note of what Huff is pointing out about having isolation valves between the expansion tanks and the system, with some sort of drain valve between the isolation valves and the tanks, boiler cocks or whatever. You need them to set the tank pre-charge in the future if need be. An isolation valve with a drain cock for each tank would be best probably, although one for both might be adequate.

--ewd
 
Yes,

I would go immediately to 1/2" and then to an isolation valve just like the others are saying.

There is minimal flow as the hot water expands into the tank and cold water out of the tank and thus 1/2" is just fine.

What I have learned after being involved in many installations is that once the water in the storage tanks is hot (at 195F or so and fully expanded) you can let some air out of the expansion tank to reduce its pressure and also that of the entire boiler system.

A few weeks ago I helped a person who was using a 35kw boiler with 1,000 gallons of storage and one XT160 (82 gallons) expansion tank did the job (they have been running now for 2 weeks with no over-pressure situations).

There seems to be many opinions and different charts recommending various size expansion tanks for the same size water tanks and by trial and error we have determined/established one more opinion of this.

Thanks,

Brian
 
With regards to where to tie the expansion tank into the overall system it does not really mater.

Most air separators come with NPT threads on the bottom of the separator and thus the expansion tank could be connected here.

However, I have also connected expansion tanks to other locations in system piping with no problem.

When I was installing my system several years ago I had someone tell me that the expansion tanks should be connected to the piping on the bottom of the storage tanks as this is cooler water than if connected to piping at the top of the tank. The thinking was that cooler water would not effect the rubber bladder material in the expansion tank as much as hot water, thus possibly extending the life of the expansion tank. Thus, this is how I installed my expansion tanks and have never had a problem with them ( I installed a 52 gallon expansion tank for each of my 500 gallon propane tanks).

I have heard this same type of reasoning used for the placement of the circulator pumps (many will say to install the circ. pump on the return side of the heating loop so that it sees cooler water).

However, if you look at the ratings for both hydronic expansion tanks and circ. pumps they both can tolerate water temps of 240-250F.

Hope this helps,

Brian
 
Brian said:
What I have learned after being involved in many installations is that once the water in the storage tanks is hot (at 195F or so and fully expanded) you can let some air out of the expansion tank to reduce its pressure and also that of the entire boiler system.

Help me out here, I got my Plumbing Merit Badge in the '50s, so maybe there's some new stuff I haven't kept up with over the years.

What's the theory here, and in what sense can you claim to have learned something? That the system will never again return to its minimum temperature? That the minimum temperature is not really the minimum temperature? That the maximum design pressure isn't really the maximum design pressure? That the lower the pressure the better?

--ewd
 
EffectaBoilerAgent (USA/Canada) said:
With regards to where to tie the expansion tank into the overall system it does not really mater.

Most air separators come with NPT threads on the bottom of the separator and thus the expansion tank could be connected here.

However, I have also connected expansion tanks to other locations in system piping with no problem.

When I was installing my system several years ago I had someone tell me that the expansion tanks should be connected to the piping on the bottom of the storage tanks as this is cooler water than if connected to piping at the top of the tank. The thinking was that cooler water would not effect the rubber bladder material in the expansion tank as much as hot water, thus possibly extending the life of the expansion tank. Thus, this is how I installed my expansion tanks and have never had a problem with them ( I installed a 52 gallon expansion tank for each of my 500 gallon propane tanks).

I have heard this same type of reasoning used for the placement of the circulator pumps (many will say to install the circ. pump on the return side of the heating loop so that it sees cooler water).

However, if you look at the ratings for both hydronic expansion tanks and circ. pumps they both can tolerate water temps of 240-250F.

Hope this helps,

Brian

It dosn't matter were it's connected to the system ? I thought that it did matter. I guess I need to review some info.
 
Sorry about that.

What I learned is that the pressure in a system that was teetering at the edge of the 30 psi pressure relief valve (and causing it to slightly drip water) was lowered enough after letting some air out of the expansion tank, that it no longer teetered on the edge (at 30 psi). The max system pressure was reduced from 30 psi to approx. 22 psi and saved having to add an additional expansion tank.
.
If one were to never adjust the air in the expansion tank (they come standard with approx. 12 psi) when the water in the storage tanks went from 50F at initial filling to when it got hot at 195F, an additional expansion tank would need to be added to prevent over - pressurization.

Thanks,

Brian
 
How about on the suction side of the circulator ?
 
EffectaBoilerAgent (USA/Canada) said:
What I learned is that the pressure in a system that was teetering at the edge of the 30 psi pressure relief valve (and causing it to slightly drip water) was lowered enough after letting some air out of the expansion tank, that it no longer teetered on the edge (at 30 psi). The max system pressure was reduced from 30 psi to approx. 22 psi and saved having to add an additional expansion tank.

This is great news for those of us needing to save money on expansion tanks! So as the system heats up, all a fellow has to do is keep letting air out until he has whatever pressure he prefers. Sounds like 22 psi is the best pressure, is that right?

--ewd
 
I think I'm reading a disagreement between Effecta Brian and EW.

It may be a somewhat common misunderstanding, as I also have heard from others, that if a pressurized system reaches too high a pressure (25-30 psi+) at maximum temperature, the solution is to let some air out of the expansion tank(s), which will to a point allow an increase in expansion capacity and relieve the over-pressure. Assuming that the initial pressure setting (assume 12 psi) was made at the system minimum temperature point, then the error in this solution is glaringly apparent, and that error will become obvious as the system again cools to its minimum temperature point: likely insufficient pressure to maintain adequate system pressure, a vacuum being created in the system, which will suck air in from some point -- not what is wanted in a pressurized system due to corrosion potential, etc.

If minimum system pressure needed is 12psi at minimum temperature, then sufficient expansion capacity must be available at the system high temperature point. The solution is not to let some air out of the expansion tank; the solution is to add expansion capacity.
 
jebatty said:
I think I'm reading a disagreement between Effecta Brian and EW.

It may be a somewhat common misunderstanding, as I also have heard from others, that if a pressurized system reaches too high a pressure (25-30 psi+) at maximum temperature, the solution is to let some air out of the expansion tank(s), which will to a point allow an increase in expansion capacity and relieve the over-pressure. Assuming that the initial pressure setting (assume 12 psi) was made at the system minimum temperature point, then the error in this solution is glaringly apparent, and that error will become obvious as the system again cools to its minimum temperature point: likely insufficient pressure to maintain adequate system pressure, a vacuum being created in the system, which will suck air in from some point -- not what is wanted in a pressurized system due to corrosion potential, etc.

If minimum system pressure needed is 12psi at minimum temperature, then sufficient expansion capacity must be available at the system high temperature point. The solution is not to let some air out of the expansion tank; the solution is to add expansion capacity.

My thoughts as well. If you start cold at min pressure (12), and heat system to max and let air out - when the system returns to cold it will be underpressured and, if connected to a 12psi regulated feed, will see fresh feed come in to return to 12psi. You have then effectively reduced your expansion area, and when the system gets to max temp again, you will see even higher pressures. I am no pro, but would not recommend at all letting air pressure out of a system when up to temp.
 
If the system has an autofill valve set to a designated minimum pressure, then more water will be added as the system cools to maintain the minimum pressure. Iin general it is not good practice to be adding more water into a closed system due to possible oxygen admission and/or causing a pH imbalance, and perhaps other issues. My system is isolated, no autofill, and my comment as to air being sucked into the system related to a system like mine.
 
Yes - so I think in either case (isolated or constant regulated feed), it is generally not a good idea (IMO at least - sounds like others as well) to let air out of a system at max temp to relieve pressure. You will inevitably be causing an issue at the other (cold) end of the spectrum which may in turn re-introduce or compound the hot end pressure issue, and also introduce others (e.g. fresh oxygen admission either thru fresh air or fresh water). This is not the first time I have seen it suggested to let air out at high temp, but with nothing offered to address, explain or counter the subsequent problem potentials that were raised.
 
What if... in the interest of not buying another or larger expansion tank, but you really need more expansion capacity you just put the expansion tank upstairs in a closet?
Won't that effectively give you more expansion room for your system water?

My reasoning (a generous term this particular morning) is that if you need 10 to 12PSI at the boiler/circulator to prevent cavitation in the circulator, if the expansion tank were 2 floors above them the preset pressure in the expansion tank would only need to be 2 or 3 PSI to give 10 to 12 at the boiler (assuming about 9ft. per floor). In a non bladder tank like all the home-mades being discussed around here, that would give you a lot more expansion room before you get near max pressure down at the boiler. Wouldn't it?

With commercial bladder/diaphragm tanks you still have the issue of not being wise to stretch the rubber beyond the design-volume capacity. So I'm not sure it would be a useful trick for most purchased tanks.

Anybody thought about this angle in detail? Anybody actually done this and compared the results?
 
Letting air out would exacerbate the problem, especially in an auto-fill valve scenario. When the auto-fill adds water to the system during cool down, because air was released, the system pressure would only increase at operating temperature. Now what? Release more air? Those tanks have a bladder and a precharge for a reason me-thinks. Formulas have been developed for properly sizing expansion in a closed system for a reason. Circumventing them usually leads to undesirable results.
 
DaveBP said:
Anybody thought about this angle in detail? Anybody actually done this and compared the results?

You bet. In a tall building it can make a big difference, but to do it right I believe you'd want to run a dedicated run of PEX or whatever all the way up to the elevated tank. And routing of the line to eliminate any possibility of freezing would be important. Cold system pressure of four psig at the highest point in the system is generally accepted.

I worked out the percentage-of-system-volume numbers for various starting and maximum system pressures:

https://www.hearth.com/econtent/index.php/forums/viewthread/56576/
 
ewdudley said:
DaveBP said:
Anybody thought about this angle in detail? Anybody actually done this and compared the results?

You bet. In a tall building it can make a big difference, but to do it right I believe you'd want to run a dedicated run of PEX or whatever all the way up to the elevated tank. And routing of the line to eliminate any possibility of freezing would be important. Cold system pressure of four psig at the highest point in the system is generally accepted.

I worked out the percentage-of-system-volume numbers for various starting and maximum system pressures:

https://www.hearth.com/econtent/index.php/forums/viewthread/56576/


Just curios on what happens if the pressure falls below the 4 psi at the highest point ? Air sucked in somehow ? bad circulation?
 
woodsmaster said:
Just curios on what happens if the pressure falls below the 4 psi at the highest point ? Air sucked in somehow ? bad circulation?
Openings in threaded joints, valve stems, and whatnot that leak very little water-- or none at all-- can easily leak air the other direction due to the difference in viscosity between water and air, and the surface tension of water. But as you suggest the pressure would pretty much have to be less than 0 psig for this to happen.

I think the recommended 4 psig is more than 0 psig for a couple reasons; to err on the high side, perhaps to accommodate momentary dips in pressure when circulators start and stop or when zone valves open and close, to help keep gases dissolved in higher parts of the system, changes in barometric pressure, and to have some extra pressure in case the system get cooler than intended.
 
I stumbled upon this discussion yesterday, and decided to join the group just to comment. I was impressed by several very good and correct comments, and saw a few misconceptions. I'm a 25-year-plus HVAC and plumbing engineer.

The manufacturers can provide info. People behind the counter at local supply houses probably won't have the depth of knowledge you people are looking for - you already know a lot. I'd think the commercial sales reps would be as helpful to the end users as they are to us engineers. Look for your local Taco and Bell & Gossett (B&G) reps online. B&G, as part of ITT, wrote "the bible" on hydronics, air separation, expansion tanks, and pumping. ITT just recently spun off B&G and related businesses as Xylem (named for the part of plants that moves the water up the stem, cute). In New England, Taco rep is Emerson-Swan, and B&G rep is FIA (Fluid Industrial). B&G/ITT has great handbooks that I got long ago in hard copies - I don't find them online though.

Generally, the automatic water makeup, air separator, and expansion tank should all be close to the inlet side of the hot water pump. That's where the water pressure is lowest, so air separation works best there. Also hopefully the pump is downstream of your heat source (boiler etc.) so the water is warmer, helping bubbles come out, not critical in small low-pressure-drop residential systems.

There's nothing wrong with automatic water makeup (with proper reducing valve and backflow prevention) in a plain-water system, but we avoid them in systems with antifreeze to avoid unknowingly diluting the fluid and causing a freeze. Without automatic makeup, you have to somehow pressurize the system - the expansion tank won't do it for you.

I appreciated the comments calling for a shutoff valve and drain in the expansion tank line. That is a must for properly charging the tank air pressure. I've met few plumbers who understand this.

The suggestion to locate the expansion tank upstairs will not help, particularly assuming we're talking about the type of tank with a bladder or diaphragm. Non-bladder types work quite differently, and need the air bubbles from the separator piped to rise into them. With bladder type you relieve the air bubbles into the room, and eventually you've gotten virtually all that corrosive oxygen out.

Also, the idea of bleeding off tank air to adjust the system pressure is wrong, assuming the pressure was set properly offline. As others rightly said, that reduces the amount of expansion the tank can absorb, and will actually lead to higher pressures and more chance of blowing the relief valve.

The makeup water pressure is set with pumps off, and preferably with the system at room temperature. The 4 psig suggestion is correct and typical, as Mr. Dudley said this is to keep bubbles from coming out at high points (which can air-bind the system) and to have just a bit extra. 12 psig is a typical setting, but the way to calculate it is by converting the vertical system height (between fill point and high point) from feet of water to psi, and adding that value to the 4 psi. Tank initial air pressure is the same as the water fill. We say psiG to distinguish Gauge pressure from Absolute (psiA).

The expansion tank's functions are to provide a place for expansion (and avoid blowing the relief valve), and to create a "point of no pressure change" at the pump inlet. You want the pump inlet at a positive pressure to avoid "cavitation" which is damaging to the pump impeller. And you want the makeup pressure reducing valve to always see true system pressure, so it only opens when more water is needed. Again, no biggie in low-pressure residential stuff. But if you have expansion tanks in more than one place (or a big bubble at the top of the system acting like an expansion tank) the pump won't be happy.
 
Phizman, Eliot, anyone, would you mind looking at page 21 the PT3 pumbing schematic from Tarm Biomass, http://www.woodboilers.com/admin/uploads/public/WoodBoilerPlumbingSchematic0111Web.pdf. I would like to know what you think about the location of the expansion tank/air separator/make up water supply relative to the Termovar Loading Unit (which contains the primary circ pump) in this diagram. What confuses me is the pump is on the return rather than the supply side, but the expansion tank/air separator are on the supply side of the boiler. Is this less than optimal, acceptable, or just messed up. Need to know before I plumb it in that way.

Mike
 
Status
Not open for further replies.