The True Cost of Compressed Air Leaks in Manufacturing

Most compressed air audits find the same thing: 20-30% of the air being produced is leaking out through fittings nobody has touched in five years. On a site running a 55kW compressor continuously, that translates to somewhere between £8,000 and £12,000 a year in wasted electricity, before you factor in the compressor running harder to maintain pressure.

I've done leak surveys on dozens of sites across Yorkshire and the East Midlands. The range is wider than most people expect. A well-maintained food production facility I visited a few years back had leakage below 8%. A fabrication shop I walked the following month was losing over 40%. Both were running similar-age equipment.

Why nobody fixes leaks until something breaks

The invisibility problem is real. Unlike a water leak, a compressed air leak produces nothing you can see and almost nothing you can hear against normal factory background noise. The only symptom is the compressor running more. If you're not tracking specific energy consumption, kilowatts per cubic metre of free air delivered, you'll never notice.

Leaks also develop slowly enough to feel normal. A fitting that sealed fine when installed starts weeping after three or four years of vibration, pressure cycling and temperature swings. By the time it's losing a meaningful volume, it's been there long enough that nobody questions it.

There's a management angle too. Leak hunting doesn't feel urgent when the compressor is still managing to hold pressure. It becomes urgent the day it can't.

Where the air actually goes

In my experience, the biggest culprits are consistent across sites:

Threaded connections are the most common source. BSP fittings assembled without proper thread sealant, or PTFE tape applied in the wrong direction, leak from day one. One turn of tape wound the wrong way and you've got a permanent weep that nobody can hear.

Push-fit fittings on pneumatic circuits are reliable when new. The collet and O-ring wear over time, and on a system that's ten years old, a reasonable proportion of those fittings will be weeping at least slightly.

Flexible hose connections fail at the crimped ferrule, particularly on hoses that get coiled and uncoiled repeatedly. The ferrule fatigues before the hose itself fails.

Condensate drains are the high-volume culprit people miss most often. A float drain that has stuck open dumps 200-400 litres per minute of compressed air continuously. On a busy site with eight or ten drains, finding two or three failed ones is typical.

What the cost actually looks like

The calculation isn't difficult. Take your compressor's power rating in kW, multiply by your electricity unit rate (say 14p/kWh), multiply by annual running hours, and multiply by your estimated leak percentage.

A 75kW compressor running 5,500 hours a year costs roughly £57,750 in electricity. If 25% of that air is being leaked, you're spending over £14,000 a year on air that goes nowhere. A thorough ultrasonic leak survey costs £2,000-4,000 depending on site size. The maths is obvious.

The survey also prioritises repairs. Not every leak is equally expensive to fix, and not every fix is equally quick. A competent engineer with a good ultrasonic detector can survey a medium-sized site in a single day and give you a list ranked by annual loss, biggest savings first.

Fixing the right things, in the right order

Most repairs are fast. Thread sealant on a leaking compression fitting takes ten minutes. Replacing a worn push-fit takes five. The investment isn't in the repair work itself; it's in systematically finding every leak rather than just the obvious ones.

One thing worth doing before anything else: fix leaks before you start talking about downsizing your compressor. Sites that run a proper leak programme often find their existing machine suddenly has capacity it hasn't had in years. That changes the capital investment conversation completely.

What doesn't work is a one-off survey with no follow-up. New leaks develop within 12-18 months. The sites that control leak rates treat it as a recurring maintenance activity, not a project with a start and end date.