Compressed-Air Leak Program Under Production Pressure
Your team is asked to approve a compressed-air improvement package at a food-processing plant. The plant operates 7,000 hours per year, five days a week, with two 250-hp fixed-speed rotary screw compressors and one 100-hp backup. At 110 psig header pressure, production flow averages 1,650 cfm and falls to 900 cfm during breaks. Trend data show the lead 250-hp unit loaded 95% of the time and the second unit running at 40% load for most production hours. An ultrasonic survey identified 38 leaks totaling an estimated 420 cfm at 110 psig. Repair quotes total $18,000, but maintenance can only isolate sections during a four-hour Sunday window. Operators raised header pressure from 100 to 110 psig after a dryer pressure drop increased to 8 psi. A vendor proposes a 250-hp VSD compressor for $185,000 and claims 25% energy savings. Electricity costs $0.09/kWh plus $14/kW-month demand. The plant has no automated leak alarms and no permanent flow meter; historical data come from compressor onboard controllers.
1. Estimate the annual energy and demand savings from repairing the 38 leaks. State your assumptions, decide whether the $18,000 repair should be approved now, and identify what should be sequenced with it.
Read the worked answer
Use a marginal specific power of about 18 kW per 100 cfm at 110 psig. That is roughly 0.18 kW/cfm and is consistent with 250-hp units delivering about 1,000 cfm each.
- Leak flow: 420 cfm / 100 = 4.2; 4.2 x 18 kW = 75.6 kW.
- Annual energy: 75.6 kW x 7,000 h = 529,200 kWh; at $0.09/kWh = $47,628/yr.
- Demand: 75.6 kW x $14/kW-month x 12 = $12,701/yr.
- Total approximate savings = $60,329/yr; simple payback = $18,000 / $60,329 = 0.30 yr.
Caveats matter. The demand savings only occur if the leaks are present at the monthly peak and if compressor cycling actually reduces billed demand. The second 250-hp unit runs at 40% load, so removing 420 cfm may shut it off and save a different amount than the linear estimate. Confirm with 15-minute kW data.
Approve the repair, but not as a one-time standalone task. Sequence it with fixing the dryer pressure drop and reducing header pressure from 110 to 100 psig. A 10-psi reduction is roughly 5% compressor power at constant flow; on about 250 kW average load that is about 12.5 kW, or roughly $8,000/yr in energy plus about $2,100/yr in demand. Do not simply add that to the leak savings: leak repair lowers flow and may allow the pressure reduction, so report the combined measured result. Verify leak repairs with ultrasonic survey and a pressure-decay or flow test, then re-measure compressor kW.
2. The VSD compressor vendor claims 25% savings. Evaluate the claim and recommend how to sequence the $185,000 purchase against leak repair and pressure reduction.
Read the worked answer
The 25% claim is not credible for the post-repair load without a part-load model. Vendor claims often compare a VSD against fixed-speed compressors running unloaded or against a leaky system that should be fixed first.
After repairing 420 cfm of leaks, average production flow may fall from 1,650 cfm to about 1,230 cfm. A 250-hp VSD rated near 1,250 cfm would then run near full speed for most production hours, where the VSD has little or no efficiency advantage and may be slightly worse. The VSD does not eliminate leak waste; it only changes how the compressor matches flow.
Sequence the work as follows:
- Repair leaks and fix the dryer pressure drop.
- Reset header pressure to the lowest stable value, likely near 100 psig.
- Install a temporary flow meter and power logger for at least 30 days.
- Re-baseline the flow profile, including breaks, weekends, and seasonal production.
- Evaluate a VSD sized for the actual trim load, not the pre-repair load. A smaller VSD plus fixed-speed trim may be better than one large VSD.
Do not sign the $185,000 purchase before re-baselining. If a VSD is still justified, require a performance guarantee at specified flow, pressure, and kW, and verify it after installation.
3. Design a 12-month M&V plan for the leak program and pressure reset. Identify the IPMVP option, measurement points, baseline adjustments, and how to avoid double-counting.
Read the worked answer
Use IPMVP Option B, retrofit isolation with continuous measurement, with the boundary at the compressor discharge to the plant. Measure compressor kW, header flow, header pressure, run hours, and dryer pressure drop. Use existing controller logs for a 12-month baseline plus temporary power and flow meters where permanent instrumentation is absent.
Adjust the baseline for production volume, operating hours, ambient conditions, and product mix. A practical model regresses compressor kW on flow and pressure, because flow is the main driver. If production data are unavailable, use operating hours and a representative flow profile.
Avoid double-counting by measuring the system after each phase. If leak repair and pressure reset are implemented together, report a combined savings estimate from the measured before-and-after kW, not the sum of two engineering estimates. Attribute savings using the flow and pressure model, and state the uncertainty.
For demand savings, use 15-minute interval data and check whether the reduced load coincides with the monthly billing peak. Verify leak repairs with ultrasonic tagging and a flow or pressure-decay test. Build persistence into the plan: repeat the leak survey at 6 and 12 months, because leaks recur, and report savings net of any new leaks.