Leaks run every hour your compressors do
Compressed air leaks cost money every hour your compressors run, whether the line is producing or not. They are easy to ignore because they make little noise on a busy floor and appear on no invoice. The US Department of Energy’s compressed air sourcebook, written with the Compressed Air Challenge, states that poorly maintained systems can lose 20 to 30 percent of their air capacity and power to leaks, and that a well-maintained system should be below 5 to 10 percent. The fix is a programme, not a one-off: measure the leak rate, find leaks with ultrasonic leak detection, tag them, repair them, then measure again.
This article explains why leaks cost more than they look and how to run a programme that keeps them down. It sits within our industrial compressed air guide.
Why leaks cost more than they look
They never clock off
A leak consumes air whenever the system is pressurised. If compressors run through breaks, nights and weekends to keep a few machines ready, every leak runs too. Off-production demand is often mostly leakage, which is why it is one of the first things a compressed air audit measures.
Small holes grow fast
Flow through a small opening rises with its area, which means with the square of its diameter. Double the diameter of a hole and, at the same pressure, roughly four times as much air escapes. A worn hose connection that starts as a hiss can become a significant flow without anyone noticing a change.
Pressure feeds them
At typical plant pressures, flow through a leak rises roughly in proportion to absolute pressure. So a plant running a higher header pressure than it needs loses more air through every leak, on top of the extra energy needed to compress every cubic metre. The DOE sourcebook notes that raising discharge pressure increases the consumption of every unregulated use, leaks included. Our pressure reduction savings tool shows how setpoint and leakage interact.
They hide other costs
The same DOE guidance notes that leaks also lower system pressure, make tools and machines perform worse, lengthen compressor running hours, add maintenance, and can lead a plant to buy compressor capacity it does not need.
A leak runs three shifts a day, seven days a week, whether or not the line is running.
Measure the leak rate first
Before you hunt for individual leaks, find out how big the problem is. Measure your leak rate with production stopped and all air-using machines idle, but the system pressurised.
Load and unload timing. For a compressor with load/unload control, the DOE sourcebook gives a simple method. Record the average time the compressor runs loaded (T) and unloaded (t) over several cycles, while only leaks are consuming air. Then:
Leakage (%) = (T × 100) ÷ (T + t)
The result is the share of that compressor’s capacity lost to leaks. For example, if it runs loaded for 2 minutes and unloaded for 6, leakage is 2 × 100 ÷ 8 = 25 percent of its capacity.
Pressure decay. For variable-speed compressors or other controls, the same source describes timing how long the system takes to fall from operating pressure to about half that pressure, with the compressors stopped. With an estimate of the total system volume, including receivers and piping, this gives a leakage flow.
Both methods have limits. They can miss air consumed by idle machines that were not fully isolated. But they give a baseline, and a baseline is what lets you prove the programme worked.
Once you have a leak flow, our leak cost calculator turns it into an annual cost using your electricity price and running hours.
Find, tag, fix, verify
A leak programme has four steps. Each one produces a record.
| Step | What to do | Output |
|---|---|---|
| Find | Survey the whole system with an ultrasonic detector, during production if needed | Each leak located and sized |
| Tag | Attach a numbered tag at each leak and log it | Leak register: location, estimated size, priority |
| Fix | Repair in priority order; larger leaks and easy fixes first | Register updated with date and repair method |
| Verify | Repeat the leak-rate test under the same conditions | Measured leak rate before and after |
Ultrasonic detection works because escaping air produces high-frequency sound, most of it above human hearing. The detector converts that sound into an audible signal and a reading, so the surveyor can locate leaks on a working floor. The DOE sourcebook describes it as probably the most versatile leak-detection method.
Where leaks hide
The DOE sourcebook lists the most common problem areas. On a plant floor, that means checking:
- threaded joints and pipe connections, especially where sealant has aged;
- hose connections and drops to machines, where hoses are moved and flexed;
- condensate drains stuck open, or timer drains set to open too often;
- flanges and unions on older, corroded sections of the main;
- machine-side connections inside guards, where nobody looks;
- disused branches that were never isolated or capped.
Make it a programme, not an event
Leaks come back. Joints age, hoses wear, machines are moved and new lines are added. A plant that surveys once, repairs everything and then stops will see the leak rate climb again.
What works is a routine:
A leak programme that lasts
- Measure the leak rate and record it as a baseline
- Survey the full system with an ultrasonic detector
- Tag and log every leak with location, size and priority
- Repair in priority order and record each repair
- Re-measure the leak rate under the same conditions
- Isolate or remove disused branches
- Lower the header pressure to the lowest level that works
- Repeat the survey at an interval set by how fast leakage returns
Give the programme an owner. Leak work slips when it belongs to everyone. Name one person to hold the leak register, schedule surveys and report the leak rate to management. Build repairs into planned maintenance, so parts and labour are ready, and make leak checks part of every machine move or line change. A maintenance contract can carry the survey and re-measurement as scheduled tasks.
Track one number over time: leak flow as a share of total demand, measured the same way each time. If it rises, survey again sooner. If it holds, the interval can stretch.
What fixing leaks is worth
The value depends on your leak rate, electricity price, running hours and how well your compressors reduce power as demand falls. A fixed-speed compressor running more hours unloaded after repairs may save less than expected, which is why control and leak repair are often planned together. Savings are project-specific. We measure them against a baseline and verify them after implementation; we don’t promise a blanket percentage.
Start with the baseline
If you have never measured your leak rate, start there. It takes a non-production window, a watch and a notebook. If the number is high, a structured survey is the next step.
Our compressed air leak detection service covers survey, tagging, a priority register and re-measurement. It fits naturally inside a wider energy optimisation programme. For what an audit covers beyond leaks, read what a compressed air audit is. Then ask an engineer to measure what your leaks cost.