Expansion tank question

[cheapsx]

Seems as most diaphragm type tanks accept about 65% of their total volume. If you use a non diaphragm type tank does the same hold true for acceptance volume for that type of tank or is there a rule of thumb for using that type of tank? Also other than height affecting pressure does it matter where the tank is connected to the system? Is it better to do it at the storage or the boiler? In my set up at the boiler would be better so all pumps are pumping away from the expansion tank.

 

[ewdudley]

Seems as most diaphragm type tanks accept about 65% of their total volume. If you use a non diaphragm type tank does the same hold true for acceptance volume for that type of tank or is there a rule of thumb for using that type of tank?.

Most references distinguish between diaphragm tanks and simple closed tanks by assuming that the simple closed tank will start out a 0 psig and the diaphragm tank will start out at some pre-charged pressure equal to the static water column pressure at the elevation of the tank plus at least 4 or 5 psi, which is the minimum system pressure at the elevation of the expansion tank.

So the diaphragm tank won't start accepting any water until there is some heat expansion because it is already pre-charged at the minimum system pressure. But the simple closed tank will accept water until the air inside is compressed to the minimum system pressure, and only then can it begin to accept operational expansion.

If the simple closed tank starts out at 0 psig, the required tank size can be much larger than a membrane tank that is pre-charged, especially if the tank elevation is low in a tall system.

However what the old-timers taught me was that you want to pre-charge a simple closed tank by filling the system and then pumping air into the tank and bleed off water until the water level in the tank is down near the bottom. From then on you just need to keep an eye on it, and the tank needs to be isolated with a valve if opening the system could allow air in the tank to blow back through the system. Thus the minimum size of a pre-charged simple closed tank can approach that of a diaphragm tank.

But this kind of fussing about is from an era where getting the expansion tank charge just right and then keeping an eye on it was just one more task in tending to heating systems that generally required more on-going maintenance, and may not be worth it these days of hands-off trouble-free operation if simply getting a bigger tank is doable.

The only rule of thumb is that the most expansion you can see from 40 or 50 degF up to 210 degF or so is 4% of system volume. Otherwise the size of the tank depends on pre-charge pressure, system height, tank elevation, and maximum pressure at the boiler. The formulas aren't hard, but you need to have all the parameters for a given situation:

(broken link removed to http://www.amtrol.com/enginhandbook.htm)

Also other than height affecting pressure does it matter where the tank is connected to the system? Is it better to do it at the storage or the boiler? In my set up at the boiler would be better so all pumps are pumping away from the expansion tank

For a membrane tank you want the tank connection to be strictly downhill and below the point where it connects to the system such that no air from the system can accumulate below the diaphragm and cause corrosion inside. The tank can still be at any elevation in the system, it's just that the pipe that tees off to the tank needs to run vertically and/or downhill so air bubbles won't travel into the tank.

For a simple closed tank the traditional systems provide a way of insuring that all dissolved gases released by the post-boiler separator are fed to the top of the expansion tank, and therefore the expansion tank is normally above where the line tees off to it.

Notwithstanding all the dire warnings you will encounter as regards 'pumping away', if the storage is adjacent to the boiler, and if the pipes 1.25" or so such that there's only three of four feet of head across the pump, and if the system pressure is 15 psig or so at the boiler, if you ask me it doesn't much matter where the expansion tank is connected in the boiler-to-storage-and-back circuit, but most will tell you it is impossible for this to be correct.

YMMV --ewd

 

[cheapsx]

My thought is this; I can put a 50 gallon non diaphragm tank on the floor next to the boiler so watching it won't be such a pain. My plan was to pipe to the tank from the bottom of the air scoop to the bottom of the tank. Then after filling system allow enough water to tank to cover the bottom inlet and then proceed to charge the tank equal to the fill pressure. I figure after that it's just a matter of watching the water level. I could tap into the boiler return line to get a lower point of entry into the system if that would make it work better. Also I have heard talk of charging the tank with nitrogen to prevent absorption into the water, is it worth the trouble to do so? I suppose that in the future if this were a lot of tending the tank I will put in a diaphragm type but funds are low right now, I have the 50 gallon tank, and that's a big hurdle as for getting the 500 gallons of storage in before winter.

 

[heaterman]

Well said EW. I'll agree with everything but add a precautionary note about pumping away.

The expansion tank is the point in the system at which the circulator cannot change the pressure. (PONPC or, Point of no pressure change)
I have found time after time that locating the system circ(s) in a manner that they move water away from that point, yields far better results for air elimination, less failures of air vents, eliminates problems with fill pressures and a host of other issues. It doesn't really matter where the cirs or the expansion tank are actually located in the piping as long as they are in that sequence.
I looked at a system in a church just today where the caretaker mentioned constant failure and leakage from the float type air vent. Surprise! the expn tank and vent are located very close to the outlet of the circ. I asked if they had problems getting the air out of the system and he said "constantly". As some of the radiation is over 20 feet above the boiler room, they also need pretty good static fill pressure to get water to the top of the system. After they charge it up to 18-20PSI the combination of temperature increase and circulator head has caused the relief valve to relieve itself because everything running is added to the system. If they were pumping away this would be eliminated.
Pumping away is the best practice.

 

[heaterman]

My thought is this; I can put a 50 gallon non diaphragm tank on the floor next to the boiler so watching it won't be such a pain. My plan was to pipe to the tank from the bottom of the air scoop to the bottom of the tank. Then after filling system allow enough water to tank to cover the bottom inlet and then proceed to charge the tank equal to the fill pressure. I figure after that it's just a matter of watching the water level. I could tap into the boiler return line to get a lower point of entry into the system if that would make it work better. Also I have heard talk of charging the tank with nitrogen to prevent absorption into the water, is it worth the trouble to do so? I suppose that in the future if this were a lot of tending the tank I will put in a diaphragm type but funds are low right now, I have the 50 gallon tank, and that's a big hurdle as for getting the 500 gallons of storage in before winter.

My rule of thumb for tank precharge on a diaphragm tank is 2PSI less than static pressure. If you want to run at 15, charge it to 13. etc.

And yes, your idea of the tank on the floor is fine. If you want to post up your boiler volume and length size of connected piping/baseboard I can crunch out a tank size for you.

 

[ewdudley]

Well said EW. I'll agree with everything but add a precautionary note about pumping away.

The expansion tank is the point in the system at which the circulator cannot change the pressure. (PONPC or, Point of no pressure change)

I have found time after time that locating the system circ(s) in a manner that they move water away from that point, yields far better results for air elimination, less failures of air vents, eliminates problems with fill pressures and a host of other issues. It doesn't really matter where the cirs or the expansion tank are actually located in the piping as long as they are in that sequence.

[See above excellent example of what can go wrong when not pumping away.]

Pumping away is the best practice.

Yes we agree that nothing can go wrong if we always pump away, and the need to proceed with caution should be stressed. But I make a very narrow claim about a circumstance where the rule can be ignored.

I claim that for the boiler-to-storage-and-back circuit, where the storage is nearby, and the pipes are fat, and the pressure delta across the pump is tiny, and where there is plenty of system pressure to avoid any possibility of cavitation or flashing, then pumping away is a non-issue.

So pumping away is best practice, except when we don't need to do it.

For instance, referring to the 'Simplest Pressurized Storage System Design' sticky, the '... Before Storage' piping schematic is clearly asking for trouble.

But if we look at the '... With Pressurized Storage' version things are looking good: the load circuit pump is pumping away from the storage tank and there's nothing but low resistance between the PONPC(s) and the load circuit pump. My point is that even though the boiler circs are 'wrong' in the '... With Pressurized Storage' schematic, for as much pressure delta as either boiler circ will pull across the storage tank resistance, it just doesn't matter.

Cheers --ewd