expansion tank (non bladder type) problem

john58

tank becomes filled with water (water logged) in a few hours after draining and then the pressure relief valve begins to open when the boiler fires. the  tank is equipped with a B&G airtrol tank fitting,  should i replace the B&G air trol fitting on the bottom of the tank?

Mike Kusiak_2

Draining tank

Are you just draining the pressurized water from the tank or are you actually letting air in to replace all of the water? If you just connect a hose to the drain valve and let the initial water drain, you probably have not let air in to displace the remaining water. Try hanging a pail under the fully open drain valve and see if water comes out in spurts, as gulps of air enter the tank. Continue until all the remaining water is displaced by air.

Wayne_16

Are you isolating the expansion tank from the system.

Is there a valve you can close, isolating the tank from the rest of the system, when you open the drain valve. 



Minnesota Wayne

cattledog

possible air leak

One thing to check for is a small air leak from the top side of the tank or its fittings which is preventing the tank from developing and holding enough pressure to counter balance the system auto fill pressure.

If you can get a pressure gage on the expansion tank it can help in trouble shooting the problem.

Mark Eatherton

Solutions....

Everyone posted great answers. If in fact there were a leak on the top of the tank, out of sight, you'd have water leaking off the bottom of the tank. Possible, but doubtful.



Based on other answers, you are NOT getting the tank completely drained, and replenishing the air in the system. On the Airtrol fitting, at the very bottom is a small screw. THis is connected to a tube that goes into the top of the tank. While draining the tank, if you will open this screw, it will allow air into the tank, thereby letting the water out of the bottom of the tank, and keeping you from standing around for 2 hours listening to the tank GLUGGG and dribble and GLUGG and dribble.



Once the tank is completely drained, close the vent, close the drain, and turn the water back on. Make certain the system is isolated when you are doing this, or you will be back asking how to purge your heating system :-)



Here is a link to an article I wrote many years ago about expansion tanks.



http://contractormag.com/hydronics/cm_column_75/



Also, if you can see any small automatic air vents on your system, you need to shut them off once initial air has been removed, or it will remove the air charge from the tank again over time.



A common mistake by unknowing technicians of today. If it appears to be an air problem, throw an auto vent on it (WRONG!!)



If it has a bladder style expansion tank it HAS to have an auto vent to expel the air.



If it has the old style tank like you have, you DON'T want any auto vents ANYWHERE on the system.



In 95% of the cases where air is an "issue", it has more to do with the pumps location in relationship to the expansion tank, but that is a topic for a whole nuther thread.



Best of luck!



ME

DanHolohan

Great advice.

I'd also check the gauge glass if there is one. It should be closed all the time because air can leak from the packing on the upper valve. When water reaches the valve, the packing swells and doesn't leak until you drain the tank and give it a chance to dry out.



If there's a second expansion tank (by way of an air-bound radiator or a partially air-bound indirect water heater (such as an Ergomax) the air will migrate between the two "tanks," causing one to waterlog. It also shifts the point of no pressure chance for the pump to somewhere between those two "tanks."

Mike Kusiak_2

Expansion tank location and PONPC

Mark, you bring up some very good points in your article. Your mentioning the expansion tank location and the point of no pressure change brings up a question I have.



 I maintain a large commercial system with high head pumps, pumping away from the boiler and expansion tanks. Although the pressure at the boiler remains constant and equal to the static pressure when pumping, there is a momentary drop in pressure of about 6 PSI when the pumps first start. Conversely, when the pumps shut down, there is a momentary surge in pressure of 6 PSI at the boiler and tanks. During these momentary surges, you can hear water flowing out of and into the tanks. Because the fill valve is located at the boiler, during the momentary pressure drop the valve opens and adds additional water to the system. So, even though the fill valve is set as 12 PSI, the actual pressure is maintained at 18 PSI because the system pressure drops to 12 PSI when the pumps first start.



Is this normal?



I am a bit confused regarding the pressure changes and the flows in and out of the tanks. Since water is practically incompressible, the system volume should not change except for thermal expansion, yet there are large momentary flows from the tanks. The only thing I can think of is that there may be trapped air in other parts of the system, which compresses and expands as the pumps start and stop, allowing a change in water volume in the piping and radiation.



Any ideas?

Jamie Hall

Pressure surge

can't really call it a water hammer (although it is a flavour of water hammer) is quite normal when starting or stopping a large pump.  Years ago in another life I had to deal with this problem (large municipal water supplies) and the only reliable solution which I found -- which is probably not applicable, sadly -- is slow closing valves for the pumps (one for each pump -- these were usually big parallel pumping systems).  The drill was that when you got a system demand, you started the pump against a closed valve, then slowly opened the valve.  Similarly, when the system demand stopped, you closed the valve slowly, then stopped the pump.  Might be able to come up with a way to do that?

Mark Eatherton

I too have heard this....

and I too thought of possibly another point in the system with significant air trappage, but none could be found.



Siegenthaler noted that there is a "droop" factor that occurs in the piping between the PONPC and the pump inlet. It has more to do with the location of the PONPC as it pertains to the eye of the impeller than anything, and the greater the distance, the greater the droop.



When the pumps first start up, you are taking a significant amount of mass from dead still to a state of motion. It takes a second for all things being unequal to equalize, and during this point in time, the droop factor is drastically exaggerated. This is when the PRV is kicking extra water in to the system. One for sure way to over come this would be to replace the old pumps with DCECM technology that has a soft start characteristic, and that is probably not in the budget right now, but if you plant the seeds of change....



I suspect that if there were a way to move the PONPC directly to the inlet of the pumps, your movement of water in and out of the expansion tank would be minimized. The fact that you eventually get to a point of stability (18 PSI) tells me that it is not an air problem, but more of the previously explained "droop factor".



As long as 18 PSI doesn't cause the relief valves to do their thing, I would not worry about it.



As an experiment, if your expansion tank has isolation and a drain cock, and your pumps have drain cocks at their inlets, you COULD (temporarily) move the PONPC from the boiler to the pumps inlets and see what the net effect is....



This is why they make double female washing machine hoses :-)



I once had a homeowner with a mini tube RFH system (BESI) and a high head pump (Gfos 2699) who absolutely did not believe that moving the PONPC would make ANY difference in the operation of his heating system (extremely noisy), and I used the old double female hose trick to move the PONPC from where it was to where it needed to be. It made an IMMEDIATE, SIGNIFICANT difference in the noise, and air elimination of the system. He was sold then...



He had gauges strapped ALL over the system, and you could see the positive pressure results.



Try it if you can, and report back on your findings.



ME

DanHolohan

Your instincts are good, Mike.

Here's an explanation from Gil Carlson. See the section on Multiple Tank Locations. I believe that's what you're seeing on your gages. The pump is confused and searching for the Point of No Pressure Change. 

Mark Eatherton

Egg Zachary....

That is the drawing Siggy has in his Modern Hydronic Heating book that demonstrates the droop factor. You can see it in the blow up porting of the piping between the PONPC and the pump.



That Gil was quite the visionary...



ME

Mark Eatherton

Restrictive orifi ....

I have had situations where I had to move the PONPC to two remote zone pumps, and with a common connection for one PONPC going to two seperately zoned pumps, I made some restricitve orifices to make sure there wasn't a significant amount of water doing the short curcle jurk between the respective loops. I think I took a nickel ($0.05) and drilled a 1/8" hole in it, and soldered it in to the 3/4" copper pipe attached to the pumps inlet and connections to the PONPC.



Worked quite well.



In Mikes case, it would lessen the immediate hydraulic pressure differential that is causing movement in and out of the tanks.



As he said, water is incompressible, and I suspect that the movement in and out of the tanks is the opposite of what's occurring in his minds eye... Only confirmable through the OOhhh AHHhh test :-)



ME

Mike Kusiak_2

Pressure change

Thanks to all for the responses, and to Dan for posting Gil Carlson's article. It really says it all. Gil specifically mentions on the last page the situation of multiple tanks and the flow in and out of them causing pressure changes. He then mentions the possibility of overpressuring the system if the fill valve is located where a dip in pressure occurs, which is exactly what is happening here. Makes one consider that the best place to connect the feed valve may not be at the boiler, but rather at a point after the pump where pressure will not drop below the static level.



In my case, I believe ny situation can be explained by trapped air in the pipes and radiation. For one, I have been involved with this system for several years and have not noticed the large pressure changes until recently. Secondly, over the summer the building maintenance crew drained several sections of piping to replace bad radiator valves and probably did not completely bleed the affected parts of the system. It may be possible that several radiators are airbound and acting as "secondary tanks". Given that we have not had any really cold weather this season, it is not unusual that no one has complained of cold radiators yet, but time will tell....



Since this system involves a multi-loop monoflow system with 200 radiators and 40 apartments in 5 separate buildings, finding the trapped air is not trivial. With the pumping away configuration, small pockets will probably go back into solution and wind up back at the boiler where the auto air vent is located. Larger airbound areas will most likely cause no heat complaints when the weather gets colder and will be addressed at that point.



Thanks again for all the input and new understanding of the situation. I will keep you posted of any further developments.