An ultrasonic survey at a Tier 1 automotive supplier's plant in Mexico tagged more than 80 compressed air leaks. Priced at the plant's assumed blended electricity rate of USD 0.12 per kWh across 8,000 operating hours a year, that register carried an estimated annual energy cost of more than $26,000. Repairs needed almost no capital, and the estimated payback is under three months. The client is referred to here as INDUSTRIAX, an anonymized designation rather than a company name.

Those figures are estimates built from the survey. Mexico Energy Partners has not published a metered before-and-after consumption comparison for this site, so treat the payback as a projection.

The decision this puts in front of a plant director is small and cheap to get wrong. A leak survey either sits in the maintenance budget as routine work or waits in a capital queue behind equipment replacement. The answer depends on how much of your compressor load is leak load, and on whether the saving reaches your CFE demand charge or only your kWh.

What eighty leaks cost at one plant

The register at INDUSTRIAX averages roughly $325 per leak across the 80-plus tags. That is a useful benchmark, and it is smaller than most published illustrations of a single leak. The reason is size distribution. Most tagged leaks at a working plant are small.

The US Department of Energy publishes the flow through a clean orifice at 100 psig, and the numbers scale sharply with diameter.

Set against that table, the illustrative figures often quoted for individual leaks make sense. A 1.5mm leak, barely audible on a noisy floor, can waste over $700 a year at 8,000 hours and USD 0.12 per kWh. A 3mm leak, about the diameter of a BB pellet, can cost more than $2,800 on the same basis. The ratio between those two figures matches the DOE flow ratio for the same two orifice sizes, which is what you would expect.

How to reproduce these numbers

Two assumptions carry all of the arithmetic above, and neither is stated in most leak reports, including the original version of this one.

The first is system pressure. Leak flow is a function of gauge pressure and orifice geometry, so a cost per leak without a stated psig cannot be reproduced. The figures here use the DOE table at 100 psig.

The second is specific power, the kW the compressor spends per cfm delivered. The $700 and $2,800 illustrations imply about 0.116 kW per cfm, or 11.6 kW per 100 cfm. The Compressed Air Challenge fact sheet on leaks prices a 1/16 inch leak at $523 a year at $0.05 per kWh under constant operation, which works back to about 0.19 kW per cfm. On that basis the same 1.5mm leak at USD 0.12 per kWh and 8,000 hours is closer to $1,150 a year. The illustrations used here are the conservative end of the published range, not the aggressive end.

The same arithmetic reconciles the INDUSTRIAX average. At 0.116 kW per cfm, $325 a year per leak is about 2.9 cfm, which sits between a 0.8mm and a 1.6mm orifice on the DOE table. In other words, the typical tagged leak at that plant was smaller than the 1.5mm illustration, which is why the plant average is well below it.

The USD 0.12 per kWh rate is itself an assumption. It is a blended figure, not a CFE tariff line, and the article does not state the tariff class, the CFE division, the month or the MXN to USD rate behind it. Rebuild it from your own invoices before applying any of this to your site.

Why compressed air is the most expensive utility in the plant

Compressed air is a poor way to move energy around a factory, and the physics is not close. The Department of Energy's sourcebook Improving Compressed Air System Performance puts the wire-to-work efficiency of a compressed air system at around 10%, and states that more than 80% of the electrical energy going to a compressor becomes available heat. Only 10% to 15% of the electrical input reaches the point of use as usable work. The remaining 85% to 90% leaves as heat.

That is the ratio before any leak. Air lost through a leak has already paid the full conversion penalty, so every cfm that escapes is the most expensive cfm in the plant. Compressor systems commonly account for 10% to 30% of a plant's total electricity consumption in the industries where pneumatics do the work, including automotive assembly, metal forming and steel.

For a Mexican plant on CFE's GDMTH tariff there is a second effect. Leak load runs continuously, including through the punta window, so it raises both consumed kWh and coincident demand. Whether repairs reduce the CFE demand charge depends on whether your compressors were loaded during punta and whether the compressor controls translate reduced flow into reduced power. Both need checking against interval data before any demand saving is booked.

What leaks do to pressure and to the compressors

Leaks drag header pressure down. When pressure at the point of use falls below what actuators and pneumatic tools need, cycle times stretch and product quality drifts. The usual response is to raise the compressor setpoint, and that is where the money goes.

Raising the setpoint increases flow through every leak already in the system. The DOE sourcebook states that for every 2-psi increase in discharge pressure, compressor energy consumption rises by approximately 1% at full output flow. Run that in reverse and it is the reason pressure optimization follows leak repair rather than replacing it. Higher pressure also means longer loaded hours for motors, coolers and dryers, so the maintenance cost moves with the energy cost.

Why leaks go unfound

Leaks are inaudible above the ambient noise of a working plant, so walking the floor and listening finds only the worst of them. They are invisible, and they accumulate quietly until they are accepted as background.

Ultrasonic detectors pick up the high-frequency signature of turbulent airflow through a small orifice and work in loud environments. What turns detection into money is the register that follows. Each leak is tagged, given an estimated cfm and an estimated annual cost, and ranked. Maintenance then works the list from the top, which is where most of the value sits, because a handful of large leaks usually carry a disproportionate share of the total. That register is the deliverable from a full industrial energy audit that a maintenance planner can act on.

What to do

• Build the baseline. Run an ultrasonic survey that identifies, tags and prices every leak, and record the system pressure and specific power used to price them. Without those two inputs the register cannot be audited later. This is the starting point for MEP's energy efficiency work on pneumatic systems.
• Make leak detection and repair a standing program. A one-time sweep decays. Survey quarterly or semi-annually, and record the leak reappearance rate between surveys so the program's cost can be judged against what it recovers.
• Lower system pressure after the repairs, not before. Once the large leaks are closed, the setpoint can usually come down to the minimum the end-use equipment requires. At approximately 1% of compressor energy for every 2-psi step, the saving is easy to model and easy to verify.
• Tie every tag to a work order. Push the leak register into the CMMS so each tag becomes a work order with an owner and a close-out date, then re-survey to confirm the repairs held. A leak register that never becomes work orders is a report, not a saving.

One limitation frames the whole program. Leak repair reduces kWh reliably. It reduces billed demand only if the compressors were loaded when the meter set your monthly maximum, and only if the controls actually unload. Test that against the plant's own interval data rather than assuming it.

Get a leak register for your own plant

Tell us your installed compressor capacity in HP or kW, your header setpoint in psig or bar, your annual operating hours, and send one recent CFE invoice. Mexico Energy Partners will run an ultrasonic survey, return a tagged leak register with an estimated cfm and annual cost for each leak, and separate the kWh saving from any change in your demanda facturable. What you get back is a prioritized work-order list your maintenance team can execute, with the pressure and specific power assumptions stated so you can check the arithmetic. We confirm what else we need within two business days. No saving is promised before your own system is surveyed.

Sources

• US Department of Energy, Compressed Air Tip Sheet No. 3, Minimize Compressed Air Leaks, August 2004. Leaks waste as much as 20% to 30% of compressor output, and the leakage rate table at 100 psig by orifice diameter.
• US Department of Energy, Improving Compressed Air System Performance: A Sourcebook for Industry, third edition. Wire-to-work efficiency of around 10%, more than 80% of compressor input energy becoming available heat, and the 1% per 2-psi discharge pressure relationship.
• Compressed Air Challenge, Compressed Air Systems Fact Sheet No. 7, Compressed Air System Leaks. Annual leak cost of $523 for a 1/16 inch orifice at $0.05 per kWh under constant operation, used here to derive specific power.
• Comisión Federal de Electricidad, Gran Demanda en Media Tensión Horaria tariff schedule, as published and in force in August 2025. Structure of the demand charge and the punta, intermedio and base periods.
• Mexico Energy Partners ultrasonic leak survey at the INDUSTRIAX plant, 2025. Source of the 80-plus tag count and the estimated annual cost of more than $26,000.