Interesting Expansion Tank Discovery

[wwhitney]

I believe I answered that in passing in my previous post, if the air tank pressure rises to 80 psi, the water pressure rises to 80 psi, too.

Cheers, Wayne

 

[James Henry]

Drain the tank of all water and charge it with 50psi of air. Then turn the water back on and read the water pressure guage. Now use a tire gauge and read the pressure in the tank. Is it the same?

 

[Sponsor]

 

[go_hercules]

No James, it is not. Like I mentioned above, I did that. The air pressure at that point in time will equalize with the water pressure. In my case, my air pressure is set to 55, when put back into service with a water pressure of 60, the air pressure is now 60. Because the water pressure is greater than the air pressure - they equalize. Until the water is heated ........

 

[James Henry]

No James, it is not. Like I mentioned above, I did that. The air pressure at that point in time will equalize with the water pressure. In my case, my air pressure is set to 55, when put back into service with a water pressure of 60, the air pressure is now 60. Because the water pressure is greater than the air pressure - they equalize. Until the water is heated ........

That's fine with me. I couldn't care less.
May have something to do with temperatures. I'm not a scientist.

 

[wwhitney]

Bottom line: Your water pressure and air pressure gauges will read the same if the following are all true:

1) The system is static (and hence closed).
2) The gauges are accurate and at the same elevation (otherwise correct for elevation difference)
3) There are no intervening components that can maintain a pressure difference between two sides, such as a check valve.

So if you are getting different water and air pressure readings, one of those 3 things must not be true.

Cheers, Wayne

 

[wwhitney]

BTW, is this a diaphragm type expansion tank or a bladder type expansion tank?

The following pairs of simultaneous water/air pressure for the expansion tank are not self-consistent:

1) 0 psi water / 55 psi air (water side of tank open to atmosphere)
2) 60 psi water / 60 psi air (water side of tank pressurized by PRV set to 60 psi)
3) 60 psi water / 80 psi air (air side of tank further pressurized by expanded hot water)

The expected value for (3) would be 80 psi water/ 80 psi air. Are you sure there are no check valves or other devices between the expansion tank and where you are measuring the water pressure?

Cheers, Wayne

 

[hj]

Unless someone has changed the laws of physics, and conservation of energy, the only way #3 can be true is if the air pressure started at 80 psi.

 

[go_hercules]

Okay, you guys really had me second guessing myself on this one. So, what I did this morning was to completely drain the water heater and refill it so that I would be putting in the maximum amount of heat and expansion. Previously what I had seen was virtually no rise in pressure at all after a heating cycle. As some of you pointed out, that seemed impossible. Well, what I found out is that after completely draining and re-heating the pressure did in fact rise, from the initial 60 psi, up to about 65 psi. So, I think in "normal" operation after a shower, I just don't generate enough heat to register. Probably because the water heater is efficient and is re-heating constantly even while showering. So afterwards, only a little "new" heat is put into the system. But with a completely drained tank, I am putting in a lot more heat, enough to register on the gauge. So worst case scenario, I get only a 5 psi rise in pressure. And that's with the pre-charge set at 55 psi, rather than 60 psi.

 

[Reach4]

I applaud your posting and checking further. Your tank draining went above and beyond, but it is good to flush a WH maybe every year or two anyway.

If somebody's water pressure never rises to 90 psi, I think that person is good shape expansion-tank-wise. Your expansion tank may be bigger than the minimum. Nice. The person with an undersized tank with really cold incoming water and a WH set high would be better off to have the precharge at the highest pressure coming in. Because you have a working PRV, the incoming pressure does not change with time of day etc.

 

[go_hercules]

Funny you should mention the 90 psi threshold - in looking at the expansion tank sizing charts, they usually use something like a 40 degree rise in temp from ambient to heated. I would think it is likely usually more. But the more curious thing to me is the charts never tell you what maximum pressure rise they are based on. I, like others I'm sure, thought the tank would just about completely control the rise. Now I know it only limits it to some unknown value.

 

[Reach4]

I picked 90 as a for-sure more-than-safe number. If the pressure does not rise to 90, you have more than enough expansion tank capacity. I don't know that there is an official design max. These things are rated to take 150 psi. Water heaters are rated to take 150. The T+P valves are designed to release water at 150 psi. I saw one document (https://rogergriffith.com/2015/07/31/thermal-expansion-tanks-part-2/) where they set a maximum design target at 150 psi minus 10%, which would be 135 psi, when doing calculations.

 

[go_hercules]

Good info. The funny thing is, before I ever installed an expansion tank, with my PRV set and working at 60, the highest pressure I ever got was 120 after heating. I figured that was way too high, but I have heard of people with T/P valves releasing, toilets running, etc. because of expansion, so I guess my case was marginally "acceptable" to begin with by some standards. Oh well, cheap insurance overall.

 

[wwhitney]

Thanks for the followup! I'm glad physics is still working.

Cheers, Wayne

 

[Jadnashua]

Plumbing is designed (with a margin) for the outlet of the PRV that should never exceed 80psi. So, when sizing the ET, keeping it below that number is what's needed. By having your ET precharged lower than the static pressure, the bladder is stretched all of the time. When precharged to the incoming water pressure, it is at its designed, neutral point most of the time. The more it flexes, the shorter the lifespan. When sized properly, the pressure doesn't go up much, and should be well within the maximum of 80psi (unless you started with 80psi!). The higher the outlet pressure and the greater the temperature rise would all dictate a larger ET, as does the volume of the WH itself.

 

[Dana]

This is where you are mistaken, Figure 3 in the reference is correct.

If you have a closed system with a flexible membrane separating two fluids, the pressure in each fluid will be equal. If not, the higher pressure side would push on the membrane more than the lower pressure side, moving the membrane towards the lower pressure side. This movement would provide more volume for the higher pressure side, reducing its pressure, and reduce the volume for the lower pressure side, increasing its pressure. The movement will only stop once the two pressures are equal.

In other words, the pressures on the two sides of the membrane will be in equilibrium.

Cheers, Wayne

That would be true only if the flexible membrane didn't impart any pressure of it's own. As an elastic membrane stretches there will be several psi difference in the fluid pressures. The diaphram in an expansion tank does not behave like a free-floating piston.

 

[Reach4]

That would be true only if the flexible membrane didn't impart any pressure of it's own. As an elastic membrane stretches there will be several psi difference in the fluid pressures. The diaphram in an expansion tank does not behave like a free-floating piston.

They try to design them so that the diaphragm is not stretched-tight.

 

[Dana]

True, but it doesn't take much "stretch" to add a few psi.

 

[wwhitney]

Yes, I was wondering about this issue. Obviously when the expansion tank is precharged and there is no water pressure on it yet, something is maintaining a rather large pressure difference across the membrane. I believe the bladder type expansion tanks are designed so that the membrane is is pushed up against the steel tank in this configuration, and steel is not very stretchy.

As to the diaphragm type, I checked the specs on one model of diaphragm tank. The 2 gallon nominal model was rated for 150 psig (165 psia), and acceptance volumes were given for 40 psig (55 psia) and 60 psig (75 psia) precharge. The acceptance volumes were as expected very close to 2 gallons * (165-55)/165 = 1.3 gallons and 2 gallons * (165-75)/165 = 1.1 gallons, respectively.

So if the diaphragm can be designed so that at one limp extreme it is against the steel tank (for the air precharge), and at the other limp extreme it has accommodated a a volume change that is a sufficiently large fraction of the tank volume, then this issue of diaphragm elasticity does not arise.

Cheers, Wayne

 

[go_hercules]

That's a good cutaway diagram. I think a lot of people assume the diaphragm is like the head of a drum stretched across the middle. Unless the thing gets completely empty of air, I don't think the diaphragm ever gets stretched much at all. The reason I even care about that is that I am running air pre-charge pressure at 5 psi less than water pressure, so was concerned that the diaphragm would always be stretched a little, but according to that picture I don't think it will be.