We get so much slack when we push Hydrostatic Pressure/Leak Testing vs. Pneumatic testing. So many think that pneumatic is easier and cheaper, when in reality it is NOT. In fact, the B31 series makes it clear that HYDROSTATIC is the first choice for pressure and leak testing. The code makes it clear that pneumatic is for when the piping pressure and leak test can NOT be done by the hydrostatic method due to the weight of the test fluid (i.e., water), the quality concerns with the test medium not being fully removed from the piping, and/or the safety and health concerns if the test medium is not being fully removed from the piping (water reactivity). It is this last exception that got used recently on a Hot Oil system. The contractor claimed that, if any water was in the system when the oil began to heat up to 485°F, the piping would be at risk of damage due to water expansion. I bought off on that concern and allowed them to use the pneumatic method. But here’s the catch… they used a standard air compressor that delivers compressed air with a dew point between 35°F to 50°F. Here is the issue…
If we were concerned about a single drop of water being present at start-up, one would think the air used in the pressure/leak testing would be VERY DRY, like a dew point of -40°F, not 35°F to 50°F. With that kind of moisture content in the testing air, we can certainly have moisture dropping out of the gas, meaning we now have liquid inside our piping that we originally claimed was a hazard, and hence why we are doing pneumatic testing!
While the Code doesn’t prohibit a 35°F dew point, it is generally not recommended for hot oil boiler piping. Thermal fluid systems are extremely sensitive to water contamination for several reasons:
- Steam Flashover: When you fill a system with hot oil (>212°F), any residual liquid water or condensed moisture will flash into steam. This expansion can lead to pump cavitation, pressure spikes, and potential relief valve operation or rupture disc failure.
- Oil Degradation: Small amounts of water can cause the thermal oil to emulsify, leading to premature oxidation, foaming, and reduced thermal efficiency.
- Corrosion: While the system is hot during operation, the period between testing and startup (the “lay-up” phase) allows standing water to cause internal corrosion, which can then circulate through the heater and foul heat transfer surfaces.
So the bottom line:
if we claim moisture is a hazard and such that pneumatic testing is necessary, then we better be using DRY AIR/GAS. I also specified a dew point of -40°F, which is the dew point of N2!
Dew point is defined as
the temperature to which air must be cooled to become saturated with water vapor.
Because this is a saturation property, the behavior depends heavily on how you manipulate the pressure.
If we compress the air, we are effectively increasing the water vapor concentration (relative to the volume). Consequently, the dew point temperature rises as pressure increases. If we take air with a 50°F pressure-dew-point at atmospheric pressure and compress it to, for example, 100 psig, that air will have a much higher “pressure dew point” (likely exceeding 100°F). This means the air will drop out moisture at any temperature below that new, higher dew point.
Since we are dealing with critical piping systems like hot oil boilers:
If we compress 50°F dew-point air for a pneumatic test, moisture will condense in the receiver tank, piping, and regulator during compression—even if the ambient temperature is well above 50°F. At any given pressure-temperature combination where the temperature is below the calculated pressure dew point, you will have liquid water forming in our piping system.
If you use 50°F dew point air for a test in an area where ambient temperatures drop (like an outdoor site at night), we are virtually guaranteed to accumulate liquid water in the piping, which presents the exact “flash-to-steam” risks discussed previously.