Tsted a system with air and it leaked. But with water it passed

[Eng1011]

Hi-
Was wondering if this was possible. We have a water cooled system made up of SS fittings. We tested it in the shop with our pneumatic system using a custom adapter to make the connection. Well it leaks in a few places. The vendor who provided us with the system claims that he tested it, with water, at a much greater pressure and there were no leaks. Is this possible? And Why?

thanks-

 

[Duwe6]

Nope, ASSuming that you actually soaped to find your leaks. "Watching the gauge" is useless and worthless for a pneumatic test.

BTW, good dish soap, like Joy, Palmolive, etc. makes as good a solution as the pricy "Snoop". Save the Snoop bottles and refill them with 'homebrew' solution.

 

[BigInch]

Yes it IS possible to leak air and not water during the test time. Air is "thinner" than water and migrates through leaks easier and faster than water. Leaks can be found at lower pressure. Helium leak test is better than an air test. The thinnest gas, very, very thin and it diffuses faster still under less pressure than air.

From "BigInch's Extremely simple theory of everything."

 

[vpl]

Agree with BigInch -- though I would have said that air molecules are smaller than water molecules and thus can find leakage paths that wouldn't exist for a water test.

Patricia Lougheed

******

Please see FAQ731-376: Eng-Tips.com Forum Policies for tips on how to make the best use of the Eng-Tips Forums.

 

[Duwe6]

I'll agree wity "Yes" if there is a caveat of less than 50 psi. Air, using soap-suds and hydro are equally senisitive "real world" above 75 psig.

Using Helium or Hydrogen for a bubble-test is half-again to 10X more sensitive. And using them for a Tracer-Gas, and chasing the leakage with a gas analyzer is the "Gold Standard" for leak detection -- more than 100X as sensitive as air or hydro.

 

[MiketheEngineer]

I think BigInch is right. Seems as if FREON will leak through Kyrptonite

 

[Duwe6]

Nope - Freon actually goes to prove my argument. A Freon molecule is actually bigger to a LOT bigger that water;
Cl2-F2-C 2-chloro, 2-fluoromethane[R11] Cl3-C-C-F3 trichloro-trifluoroethane[R113a] vs. H2O.

What's happening is that the Freon is a 'tracer gas' and your detector can find VERY small ammounts of it comming out of your lines. Makes for a stringent test.

 

[ione]

I think the point is that the system must not leak (or at least leak to an acceptable amount) the fluid for which it is intended to contain. The same system could leak other different fluids (air in your case), but who cares?

 

[BigInch]

Helieum balloons float at head level a day after you buy them. Its not leaking through the knot. Its passing straight through the rubber. For a helieum trace test, you can even use a low He mixture.

From "BigInch's Extremely simple theory of everything."

 

[rconner]

While some gas constituents of air are indeed very small, and well run (immersed, or carefully “soaped/observed”) air tests are very good for leakage, water is also composed of quite small “molecules”. Was just curious if there was any way you could simply fill one of these “air-leaking” items with water, and THEN over-pressure that enclosure with your own air pressure test fixture? If you see water coming out of some predetermined air leak locations and at fitting wall locations normally exposed to water in service yourself, I suspect you could be justifiably skeptical of the vendors claims or QA. If you do not see water leakage, you might believe that they may very well be telling you the truth.

 

[IRstuff]

"Well it leaks in a few places"

meaning what? how big a leak, how many?

TTFN
faq731-376

 

[SNORGY]

To me...silly me...

It leaked with air, but not with water.

So, the bottom line is, it leaked.

I think you need to find where it leaked the air, and fix the leaks.

Regards,

SNORGY.

 

[stanweld]

Eng1011,
Duration of test, test pressure and surface tension of water can provide clues as to why the hydrotest passed and the air test failed.

 

[racookpe1978]

Been about this thread for a while, but finally noticed the pitfall: Sure, assuming the supplier is right, he might have tested it at a far higher pressure with water.

It might have passed. No real reason to doubt him unless you've had issues before.

But then he took it off the test stand, packed it up and loaded it on the truck, shipped it, unloaded it, unpacked it, re-installed it on the final assembly (or second test stand) at your place.

For small diameter screwed tight fittings? That's going to create many potential leaks. One new leak for each non-welded mechanical fitting.