Measure Air Leakage: Test Sealed Device Without Water

[curiouskk]

I was not sure where to post this so decided to start here. If there is a better section please let me know...

If you put a sealed device on a scale and place in an air tight box then fill with air pressure would the weight change of the device if there was a leak? I am trying to do a test to check watertight seal without water. Thanks in advance.

 

[jim hardy]

Thinking out loud ...

If you put a sealed device on a scale

okay,, the scale will indicate the weight of the device

and place in an air tight box

The scale AND the sealed device being weighed are both inside the airtight box , you say,
okay,,scale will still indicate the weight of the sealed device

then fill with air pressure

Pressurize the airtight box that surrounds both the scale and sealed device , i assume

would the weight change of the device if there was a leak?

Of course the mass of the sealed device will increase as matter moves into it. But what will the scale report?
Hmm. A couple of things are going on at once.
First - is this a spring scale that actually reports force , or is it a balance scale that actually reports mass ?

As you pressurize your sealed box you are surrounding the device with increasingly dense fluid. The bouyancy of that fluid, presumably air, will exert more force upward on whatever it surrounds as increasing pressure raises its density.
So a spring scale will initially indicate less weight because the more buoyant air pushes up on your sealed device , and indeed on the platform of your scale too. Then if there's air inleakage it'll creep back toward heavier as air leaks into your sealed device.

A balance scale will be similarly affected by bouyant forces of air this time acting on both the volume of your sealed device and on the scale's balance weights and moving parts.
Think outside the box ?
Might be more straightforward to just evacuate the sealed device by itself and see of it gets heavier

I am trying to do a test to check watertight seal without water. Thanks in advance.

Ever played with an ultrasonic leak detector? They're really interesting. . I had one that subtracted 20 khz from whatever sound it "heard" and sent it to headphones. It effectively 'translated' ultrasound into normal audio. Made it apparent why a dog can recognize the sound of his master's car engine from a block away. It was great fun at a party...
http://superiorsignal.com/resources/useful-articles/104-successful-leak-detection-using-ultrasonics

 

[curiouskk]

I appreciate your help. From the link you sent it says that a tiny leak can be heard even under vacuum, any experience with this? The leaks that we are testing for are extremely small so not sure how this would work. Curious which you think would have a better chance of showing the leak, the scale or ultrasonic? Could it be as simple as making a small box, pull to a certain vacuum then seal and see if the ultrasonic picks up noise?

 

[jim hardy]

From the link you sent it says that a tiny leak can be heard even under vacuum, any experience with this?

I was always looking for leaks out of a pressurized system, so no I've never personally tried that variation on the theme.

Curious which you think would have a better chance of showing the leak, the scale or ultrasonic?

Air weighs so doggone little,,,,... What is mass of your sealed device and what is its internal volume? What is mass of an equal volume of air?
Ratio of those masses tells you the resolution your scale will need to have.

Call somebody who makes the ultrasonics and ask if their sales guy would bring by or loan you a demonstrator unit?

Here's an old HP article on them.
http://www.hpl.hp.com/hpjournal/pdfs/IssuePDFs/1967-05.pdf

Sorry i can't be more help.

old jim

 

[JBA]

Ultrasonic leak detectors are extremely sensitive, in my limited background using them, the major problem was they were too sensitive. For any pressure differential there is almost no sealed joint, short of a welded container, that will not result in some level of a leakage signal. If you are using some type of a soft O Ring or compressed seal you are almost sure to get a leakage signal of some level. The problem is identifying the allowable signal level; particularly, when using air as a test for liquid tightness.

If you want see how one would work, generally the local distributors of those type of units are very happy to give a demonstration for your application.

 

[jim hardy]

Ultrasonic leak detectors are extremely sensitive,

The ancient HP i had was extremely sensitive. It heard change in somebody's pocket from fifty feet and a car engine's valve train from half a block.
As that article said, noisy machinery will mask the sound of a leak , but in a quiet place they're simply amazing.

Ought to be worth a try.