How ultrasound inspection saved hundreds of thousands of dollars.
The service inspector asked to have the system pressurized with compressed air. The test pressure was set at between 25 to 30 psi. The inspector used an ultrasound detection instrument that sensed ultrasound emissions produced by a leak. The instrument then translated these sounds down from the high frequency range. Through headphones supplied with the instrument, the inspector was able to hear the low frequency or “audible” range where the sounds were coming from and see intensity increments via a display panel on the back of the instrument.
While scanning the first section with this ultrasonic instrument, the inspector noted that a 250 lb steam line close by was creating excessively loud ultrasound. Since it was anticipated that the leak sound would have to penetrate through the insulation and would be substantially lower in volume than the steam generated noise, the inspector requested that the steam line be temporarily turned off.
After the steam line was shut off, the inspector used a module to sense airborne ultrasounds, referred to as a scanning module, and located three leaks. The leaks were repaired (welded) within a few hours. The inspector then noted a drop of pressure in the system pressure gauge and determined that there were still more leaks to find. He resumed scanning the pipe system with his ultrasound instrument but was unable to identify any more leaks.
Speaking of pressure, the customer was depending on the inspector to find all the leaks before the evening shift began. He asked for permission to make test points in the insulation by perforating the aluminum skin with a sharp screwdriver. Out of necessity, permission was granted. The inspector made a series of strategic perforations on each side of the 8” piping system.
He switched to a module used to inspect structure borne ultrasounds called a stethoscope or “contact” module. This is a module with a solid metal rod on the end that acts as a wave guide to transfer structure borne ultrasounds produced by the leak to the instrument’s sensor. He carefully inserted the wave-guide into the holes he fashioned and listened for any increases in the sound. It was not until the inspector came up to an elbow in the system that he heard what he believed was the source of the problem. The insulation was removed and to everyone’s relief and the leak was confirmed. The problem became evident.
Although the majority of the system was constructed in 316 stainless steel, the elbow was not. To temporarily fix the leak, soft putty that hardens in less than 10 minutes was fashioned over the leak. It was hypothesized that because the system was pulling vacuum, the putty would be sucked into the leak site just enough to stop the leak. While the system was changed from positive to negative pressure, the inspector made a final scan. Their supposition was correct. All of the leaks had been identified and repaired, and the customer could resume their process.
The inspector has checked back several times and confirmed that the customer was able to run his process for the two weeks necessary to have the part fabricated. For Enercheck Systems, it was just another success story using ultrasound detection equipment to spot leaks. For the customer, it kept order fulfillment on track - potentially saving hundreds of thousands of dollars in downtime.
Article submitted by:
Bruce Gorelick, vice president, Enercheck Systems, Charlotte, N.C., www.enerchecksystems.com
Alan Bandes, vice president marketing, UE Systems, Inc., www.uesystems.com
Comments (1)
The word vacuum stems from the Latin adjective vacuus, which means "empty." It is also the origin of the word, vacuous. Which means having or showing a lack of thought or intelligence.
I am fascinated by the physics behind an ultrasound detector's ability to locate turbulent flow. Most especially in the presence of a vacuum leak.
Discovering an exogenous turbulent flow condition is rather elementary.
However, an endogenous turbulent flow condition is monumentally more difficult to locate. The friction required to excite the piezoelectric sensor is occurring within the vessel and thus shielded by the very container one is surveying, making detection of a vacuum leak far more difficult.
The case study states, "The service inspector asked to have the system pressurized with compressed air. The test pressure was set at between 25 to 30 psi."
When an enclosed system achieves a pressure lower than the pressure outside the vessel, it is defined as a vacuum, actually it is a partial vacuum. But for simplistic purposes we can label lower pressure systems as being a "vacuum" system.
When an enclosed system achieves a pressure greater than the pressure outside the vessel, it by definition is not a vacuum. In actuality it is a pressurized system.
A defect or void in a pressurized system would allow the fluid or gas to escape the high pressure vessel, through a restriction, into an area of lower pressure. Thus resulting in turbulent flow and ultrasonic friction.
I am perplexed how the authors were able to maintain a vacuum state while simultaneously adding 25-30psi of compressed air into the system?
The authors should be applauded for affirming that a contact probe can be used to locate structural ultrasound sources caused by turbulent flow emanating from beneath insulation. Even if the method described appears to be, "induce a compressed air leak" in your piping system.
Regardless, their results are similar to the experiences I had with finding vacuum leaks in an operating system. The methods and results were published in Uptime magazine with my co-author, Mr. Allan Rienstra.
Though unpublished, I have had several more opportunities to validate the concept of detecting vacuum leaks beneath insulation using a contact probe technique at other evaporator units in North America.
Although to be fair, I have not had the need to employ a contact probe to discover compressed air leaks beneath insulation. My ultrasound detector seems to locate positive pressure leaks through the use of an airborne sensor.
1) Posted 12:05 am, 28 January 2012 by Karl Hoffower