Pressure test with air?

pretty much all the components with the exception of the boiler can handle 40-60 psi

Hydronic work done during freezing weather is often air tested first

Tiny air leaks can be tricky to find, leak detection bubble solutions work well

you need to cap off any auto air vents on the system, the purger at the boiler for one.

A test block like this may be the simplest, unscrew the expansion tank and install it there. If you can’t isolate the boiler get a 30 psi version of this and stay below that 30 psi relief setting.

" Take a cap and drill it. Put a valve stem in it. "

Interesting idea. 29/64" or 0.453" or 11.5 mm.

There is this style too, more sturdy used for LT tires / wheels

Also schrader valve to 1/8" NPT, at NAPA Autoparts stores and maybe also Harbor Freight.

1/8" NPT

The discharge side of a pressure relief valve is not air tight.

I have heard of an engineering rule of thumb. The rule of three. If your system is supposed to operate safely at 30 PSI then you test the system to 90 PSI. Three times the safe limit of operation. This is a safety test, not a leak test. If all you are looking for is leak before you add water, then 20 PSI or less should be fine. I agree that sometimes on a large volume system, finding a leak in the boiler room after filling the last radiator on the third floor can be a real PIA. But if you valve the install properly, you can

1. isolate sections as you fill the boiler / system
2. repair those leaks before filling more of the system
3. repair future leaks without emptying the entire system

I rarely used air pressure to test for leaks. When I did use air pressure, I made it easy on myself and did sections, not the entire system.

@Jamie Hall … You? Wrong? Saints preserve us!

I heard this while working in the theater, referring to support and safety cables for lamps, catwalks, and temporary set pieces that actors needed to come in contact with. Theater was a hobby, and I always thought that was a good engineering rule of thumb.

It seems that the boiler industry is a little less demanding, with a 1.5-times rule for boilers. And they don’t tell you to use air pressure. They recommend testing with water, as most of us here recommend.

I haven’t heard “hemp houses” in some time. I remember working with some old hemp-rope backdrop flies that were counterweighted with sandbags to make them easier to operate at my kids’ high school. If we ran out of sand for the sandbags, we would use a random freshman as the counterweight.

With my limited experience in theater, I was voluntold to help out with the school play. That lasted for seven years while my children attended—even during the years when they weren’t in the play.

Our community theater used an old trolley barn, and it had no equipment that we didn’t supply ourselves. When we went on the road, though, we performed in some pretty nice performing arts centers that were a little more modern, with motorized overhead flies and lighting bars that could be lowered to the floor for servicing.

But this has absolutely nothing to do with pressure-testing a heating system.

I wonder how many “Off Topic” icons we’re going to get for this?

I air test for leaks a lot and the safety concern is real. A tiny pinhole will hiss on air where water might only weep, so air actually finds the small stuff faster. Keep the pressure low and let the gauge settle before you call it, since a temperature swing will move the needle and fool you.

Speaking as the colleague in question 👆🏻 ( Hi Alan!)

Here’s a pressure-testing puzzle that stumped quite a few engineers, contractors, and other problem solvers I’ve shared it with.

Back in the early 2000s, California inspectors I worked with would accept a 100 psi air test held for at least 30 minutes on hydronic PEX tubing before concealment—whether in a slab or behind drywall.

I’d connect my compressor, bring the tubing up to 100 psi, and isolate it. If there were no leaks, the gauge would stay at 100.

Later, the inspectors began requiring a hydrostatic test. That’s when things got interesting.

I’d fill the tubing with water, pressurize it to 100 psi, and isolate it. An hour later, the gauge would read around 50 psi. I’d bring it back to 100, then return an hour later to find it around 75. After another top-up to 100, it might drop to 85.

That pattern would continue, with progressively smaller drops, until it finally held at 100 psi. Assuming there were no leaks, that’s where it would stay.

Same tubing. Same test pressure. Different behavior with air versus water.

I posed this question to plenty of people over the years without getting a satisfactory explanation, until I eventually figured it out.

What do you think was happening?

Temperature? I usually spend 5 minutes or so topping off the N₂ when I pressure test my linesets. I'm pretty sure it's just the gas settling in to the temp of the copper.