A leak you can hear at a coupling is wasting compressor output, and a leak you cannot hear still drops the pressure the cylinders see. U.S. Department of Energy Compressed Air Tip Sheet #3, “Minimize Compressed Air Leaks” (revised August 2004, prepared with the Compressed Air Challenge), says leaks often waste 20 to 30 percent of a compressor’s output. The method and the flow table below are from that sheet. The ultrasound definition is from Robin Marcinkus, UE Systems, in Compressed Air Best Practices on 3 November 2025.
The blank worksheet is the file this title promises: leakage-survey-worksheet.txt. It sits in the compressed air section, next to the purity-class note.
What the walk includes
The sheet says a leak program should include identification and tagging, tracking, repair, verification, and employee involvement. It calls 5 to 10 percent of total system flow a typical target for industrial facilities. After repairs, it says to look at the supply again and reduce compressor run time so it matches the lower demand, with a compressed air specialist adjusting the controls.
Places it names: couplings, hoses, tubes, fittings, pipe joints, quick disconnects, FRLs (filter, regulator and lubricator), condensate traps, valves, flanges, packings, thread sealants, and point-of-use devices. Leakage, it says, rises with supply pressure in an uncontrolled system and is proportional to the square of the orifice diameter.
How the leaks are found
The tip sheet calls an ultrasonic acoustic detector the best way to find leaks, because it recognizes the high-frequency hiss. It also says soapy water applied with a paintbrush is reliable, and that this method is time consuming. Marcinkus defines ultrasound as sound-pressure waves above 20 kHz, beyond human hearing, and describes a survey that tags the location, records the reading, and keeps unrepaired leaks on a list. He attributes a separate pair of figures to U.S. Department of Energy studies of industrial plants: only 50 percent of compressed air produced is used wisely, and approximately 30 percent is lost to leaks. That attribution is his. The tip sheet’s own range for waste is 20 to 30 percent of compressor output.
The same article notes that some proportional valves leak compressed air on purpose, and that this noise can confuse a listening survey. Mark an intentional vent—including twin-tower desiccant purge mufflers, membrane dryer sweep orifices, and timed condensate drains described in compressed air dryer selection: refrigerated vs. desiccant—on the sheet so it is not tagged as a failure. A reject ram’s exhaust is one place that distinction matters, on the reject checklist.
Two cells from the published rate table
The tip sheet table, which it says is used with permission from Compressed Air Challenge training, gives leakage in cubic feet per minute for a supply pressure and an approximate orifice. Two cells from that table: at 90 psig a 1/16 inch orifice is listed as 5.72 cfm, and at 100 psig a 1/4 inch orifice is listed as 100.9 cfm. The sheet says to multiply by 0.97 for a well-rounded orifice and by 0.61 for a sharp one. Copy a flow cell only after the orifice is estimated. The full multi-pressure table (70 to 125 psig across 1/64 in. to 3/8 in.), choked sonic orifice equations, 38–42 kHz heterodyne vs. MEMS acoustic camera comparison, and point-of-use FRL unit diagnostic guide are detailed in compressed air leak detection: ultrasonic survey physics, FRL unit leak points, and orifice ROI.
The blank sheet
| Tag | Place | Pressure | Method | Intentional vent? | Repaired | Checked again |
|---|---|---|---|---|---|---|
| — | Coupling, hose, FRL, trap, valve, other | — | Ultrasonic or soap | Yes or no | — | — |
| — | — | — | — | — | — | — |
The same columns are in leakage-survey-worksheet.txt. Port pressure after a leak is one reason the cylinder page asks for the pressure at the work port, not the compressor nameplate. Purity class is a different record, on the air preparation note.
