Nitrogen Pressure Test Calculator

Yes. A sealed nitrogen pressure test loses pressure whenever the gas cools, with no leak at all, because pressure follows absolute temperature (Gay-Lussac's law). A 300 psig test that cools from 75 F to 50 F should read 285.3 psig in the morning, a 14.7 psi drop that is entirely temperature. The way to tell a real leak from temperature is to correct for the temperature change, which is what this calculator does: enter both pressures and both temperatures and it tells you how much of the drop is unexplained.

It depends on the test pressure. At 200 psig it changes about 4.05 psi for every 10 F of gas-temperature change, but at 500 psig it changes about 9.72 psi per 10 F. The change scales with absolute pressure: dP per degree F = (gauge pressure + 14.696) divided by (temperature F + 459.67). This is why the flat "5 psi per 10 degrees" rule of thumb is only right near the middle of the usual test range: below about 250 psig it overstates the change and above it, it understates it badly. Use the pressure-specific value from the table, not the rule of thumb.

Correct for temperature first. Compute the pressure a leak-free system should read at the ending temperature (starting absolute pressure times ending absolute temperature divided by starting absolute temperature), then subtract it from your measured pressure. If the leftover residual is within your gauge tolerance, the change was temperature and no leak is indicated. If the measured pressure is well below the temperature-corrected expected pressure, that gap is a real loss, so soap the joints and recheck the fittings before charging.

Test at or below the pressure rating of the lowest-rated component in the sealed section, and never above what the equipment manufacturer specifies. Many residential split systems are pressure-tested in the few-hundred-psi range, but the exact test pressure is set by the manufacturer and your local authority having jurisdiction, not by a rule of thumb. Always use dry nitrogen, never oxygen or a flammable gas, and never test a system past a component's rated pressure. This calculator does not set or certify the test pressure; it only separates temperature from leakage.

Hold time is set by the manufacturer and the AHJ, not by this tool. As field practice, residential jobs are often held long enough to see a stable reading and VRF and larger commercial systems are held much longer, sometimes a full day. The longer the hold, the more the ambient temperature swings, which is exactly why you correct the reading for temperature before calling it pass or fail. A two-reading test (pressure and temperature at the start, again at the end) is what makes the correction possible.

The gas warmed. Pressure rises with absolute temperature just as it falls, so a test left in the sun or one that started cold and warmed through the day will read higher than it did when you pressurized it. If this calculator says your reading is higher than temperature predicts, recheck the entered start and end temperatures, because a rising reading with no temperature change would be physically impossible on a sealed system.