Why does air cause pressure stablization problems during hydro test? 2

[johnchrc]

I was having problem with pressure stabilizing during pressure test with test apparatus. Pressure was slowly dropping and then eventually stabilized after ~ 30 minutes.

Particulars:
- Test apparatus contained ~ 9 gallons of water
- Vessel materials are 4130 alloy steel, 110ksi MYS, 30-36 Rc
- Test at RT
- Test Pressure 5,000 psi.
- Air was bled while filling tool but geometry made it impossible to remove all. A .23 gallon (53 cu. in.) pocket probably remained in addition to any entrained air.
- Pressure dropped ~ 1-2 psi/min at 5,000 psi.
- Pressured system to 1,000 psi and bled repeatedly to try and remove all air.

I understand how air can mask a leak because it is compressed and volume will expand without creating the pressure drop normally observed with incompressible fluid like water. What I don't understand is how it's presence could cause the pressure to drop after applying pressure. If there was no leak, I would think you just would require some additional time to pressure system while air is compressed. After reaching pressure, I would not expect it to change and pressure to hold steady.

This stabilizing "phenomenon" is widely accepted as fact in our lab. We routinely blame air in test systems as reason pressure won't stabilize. Of course, we also remove air for safety reasons but the amount I'm discussing is negligible (in regards to safety).

Could someone explain why this occurs and if air is actually the culprit or is something else to blame?

Many Thanks

- CJ

 

[hydtools]

Depending on how fast the pressure is applied, the air temperature could rise due to adiabatic compression. As the air cools stabilizing at the water temperature, its pressure will decrease.

Ted

 

[electricpete]

I thought air gets hot when compressed and cool when allowed to expand. Right?

Even if so, temperature change is worth looking at.

Can the compressed air migrate into solution over time?

=====================================
(2B)+(2B)' ?

 

[MintJulep]

At 5000 psi diffusion of the air into water will occur. As the air diffuses it leaves nothingness behind, creating room for the water to expand into, reducing the pressure.

 

[johnchrc]

test at (RT) room temperature ~ 80F. Pumped water probably a little cooler. There fluid temperature increased ~ 2F so may prevent stabilization but should not cause pressure drop.

 

[hydtools]

Air gets cooler when it is in contact with stuff which is at a lower temperature. Air cooled at constant volume has lower pressure.

Ted

 

[zdas04]

Water at 5000 psig will reduce in volume by nearly 2% so your 9 gallons of space requires 9.18 gallons of water. Once at pressure, if you have ANY void volume at all, then the air in it will be at 5000 psig. Some amount of the air will disolve in the water, and it takes a TINY amount of volume reduction to make a huge difference in pressure.

If your ambient temperature is 80F and your water is at 70F, then if the system was actually liquid full the pressure would increase from the 5000 psig where you stopped the pump to over 6,000 psig when the temperature reached ambient. Since the pressure is going down, you are obviously not liquid full.

If you have 53 in^3 of void space I would expect to be able to disolve that in the water in 6-10 months. If you can't get that air out, then you are simply not ever going to have a successful test.

The alternatives that I would be looking at in your position are (1) pull a deep vacuum on the vessel prior to filling; or (2) test with nitrogen. Many people here will say that a nitrogen test would be irresponsible, and that may be true if you are not knowledgeable in the design requirements for a nitrogen test there is a fair bit of stored energy.


David Simpson, PE
MuleShoe Engineering

http://www.muleshoe-eng.com

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

"It is always a poor idea to ask your Bridge Club for medical advice or a collection of geek engineers for legal advice"

 

[hydtools]

Changing air at 5000psi from 80F to 79F will reduce air pressure to 4991psi.

Can your pressure measuring system really discriminate 1 to 2 psi change in 5000psi?

Ted

 

[johnchrc]

Thanks. Lots of good points.

Yes. we can see 1 psi change. We use pressure transducers and have data sampling around 10 samples/sec.

I did get a successful system test. Pressure eventually stabilized. Just making sure no leaks before beginning testing prototypes.

TI don't care if the air dissolves. I just want it to quickly reach a state of relative equilibrium so I can get a good test. I should note that the trapped air volume I listed was at atm. pressure. I don't know if that was clear but obviously the volume at 5ksi will be substantially less. I also perform multiple pressure bleed cycles to help bleed entrained air.

My question wasn't could I get a successful test with a tiny fraction of air present. We do it daily. I just didn't understand why air would cause pressure to drop and prevent stabilization. Normally air is a system will mask the rate of pressure loss so that cab be a concern. I knew I would get good responses on this web site so I just tossed the question out. I'm satisfied that dissolved air will allow water volume expand slightly causing a small pressure drop. Thanks again all.
-CJ

 

[hacksaw]

Disolved or not, the pressure is increased until it is stable i.e. equilibrium reached, then valved off, checked, and trimmed as necessary.

With an air pocket, you tend to reduce the rate of pressure drop in your test, potentially giving a false sense of pressure being maintained for the duration of the test.


you should be able to workout the time constant for the pressure drop for a given leak rate and volume of trapped air, but it makes more sense to get the air out

 

[moltenmetal]

A hydrotest becomes more dangerous and less sensitive the more air is trapped during filling.

Dissolution of the air at 5,000 psig can be significant and doesn't happen instantaneously. Dissolution of trapped air can generate a pressure loss in a tight system which looks like leakage, but trapped air can also tend to mask undetected leakage of water during the test.

Evacuation of the vessel, even to a fairly coarse vacuum- prior to filling, tends to reduce these problems without adding a lot of effort- provided the vessel is vacuum rated...We find vacuum filling particularly useful for testing piping in place.