Hello:

I posted this under ultrasound, but it appears that this forum have more visitors so here it is...I need to know the minimum system pressure required for an ultrasound tool to be able to hear a leak. I have manufacturing compressed air systems operating 10-25psi. Any hel will be greatly appreciated. Thanks,

frankie

UE Systems say their 9000 and 10000 models will enable you to hear a 5 P.S.I. leak through a .005 inch hole 50 feet away. You can purchase optional attachments to greatly increase the sensitivity.

You can buy some special liquids to apply to the areas where leaks can occur, and hear incredibly small leaks with airborne ultrasound tools. I do not know the minimum pressure leaks you can hear, though.

Thank you for your response webcruiser. Since this is for clean room application, the liquids might not be an option. We have the SDT 170...I was thinking about it and since the instrument hears turbulence, i think that the minimum pressure can be calculated as a funtion of the orifice. I will try to reach SDT tech service to see if they have this model or have the data for minimum pressure. If not, I can either do it or I can always experiment and determine the minimum for the SDT for a loose fitting etc. I will post the results here. If anyone has any information on this please let me know.

Frankie,

It is difficult to specify minimimum orifice size, pressure, and distance for detecting a leak. You can certainly get a starting point by what the mfg. says, but there are too many variables to be certain. You really should do some testing of your own to optimize the measurements and to improve your technique. Most real leaks are not nice round holes! There are no uniform standards for rating leak detection sensitivity, so you need to sort out the real from the imaginary (mfg claims).

Walt

Good point, Walt. Hole size and shape, direction, etc. can make huge differencies. Just basic stuff is to get rid of as much background noise as possible. Turn up the sensitivity, and check all the lower and mid frequencies. If possible, test system in static condition, at the highest safe pressure.
Use scanning module to find leaks, and a focusing cone to pinpoint them if needed.

Thank you both for your support. I got a response from Allan Riestra at SDT that is very similar to your post above. Also, he sent me a lab test that I'm attaching to this post for your reference. It seems that with the parabola I can be able to detect the leaks even at lower pressures than 5psi (depending on orifice and distance. I'll test it anyways at mfg. Thank you!

FAQ_170-06__Measure_sensitivity.doc (50 Kb, 11 downloads) Answer and test by Allan Riestra

Frankie,

All very usefull comments made by all, thankyou for posting this question as you can see it has generated some great thoughts from all over the world.(keep them coming!)
I have been asked this question in the past and was not so sure of the correct answer.
My solution so far is to pull out a bicycle tube and put a gauge on it to see if I could simulate the working/system pressure in the enviroment you need to be testing. This test is an easy way of deteriming if you can test while the plant is operational, this can save some time and money.It is also a great way of testing your sensitivity as you know thier is a garanteed leak area.
The lowest pressure leaks I have found are down to 5psi on a air system. I was Using the CSI unit to carry out the test. The plant was not operational at the time.

These are just my thoughts, I hope this can help.

Hooch

I was not too happy when trying to find a leak on a CO2 system using the UE ultraprobe 10000. I could not locate the leak around a valve packing using the gun. I ended up using soapy water and found the leak visually.

The ultraprobe works great for finding bearing faults.