What happens when an air compressor tank rusts?

[Michael Handrinos]

Why not disconnect the air line and cap off the fittings? Pressurize the tank, record the reading and come back maybe a hour or so. If it drops then I'd say it's time for a new one. Repairs to the old tank would probably be close to buying new.

Michael

[Dan Hintz]

Why not disconnect the air line and cap off the fittings? Pressurize the tank, record the reading and come back maybe a hour or so. If it drops then I'd say it's time for a new one. Repairs to the old tank would probably be close to buying new.

Such a test would only show if there is currently a leak, not if there was a weak point in the tank that will eventually rupture.

[Daniel O'Neill]

This makes me want to go home and open the pressure relief valve on my compressor.....

[John Donhowe]

Okay, I can appreciate that an exploding tank is dangerous. Is it clear, though, that the incidents mentioned were actually due to rusting of the tank, or were they due to some other cause, such as a bad pressure relief valve or cut off switch? By analogy, most water heaters need replacement because they leak from rusting out, but I've only heard of ones exploding as a result of lacking a properly function TPR valve, not from rust. So, any rust through in a compressor would result in a (slow) leak before reaching the level of a catastrophic explosion. Just sayin' ....

[Roger Feeley]

I never could remember to bleed the tank and I hate getting down on my knees to do it. So I got one of those automatic tank bleeders at harbor freight. It opens the bleed valve for just a moment every time, the unloader pipe pressurizes or depressurizes. So it fires twice every time the compressor comes on.

[John K Jordan]

certifications by the American Society of Nondestructive Testing in MT, RT, UT and PT

This caught my eye - sounds like me in the 70's! I worked for PTL and did NDT and in-shop and field structural steel inspection. Spent some time at Quad Cities nuclear plant unit 2, mostly RT but also dye and ultrasonics. Then to NDT on research reactor element manufacturing in Oak Ridge, then weld inspection. Wow, that was a lifetime ago. I haven't even thought to remember it for years.

In Oak Ridge we also inspected and tested pressure vessels, mostly hydro but occasional air pressure testing in an underground bunker. My thought on this thread was also get it tested or replace it. I've also seen the aftermath of failure and it was not pretty. Going to the shop now to drain the compressor tank again. Just for good measure.

JKJ, happily retired, playing and raising llamas in TN

[Glen Blanchard]

Lots of good information here. I took this opportunity to upgrade my air compressor. Installed and operational. With an easy to access drain valve!

Thanks.

[Steve Milito]

Okay, I can appreciate that an exploding tank is dangerous. Is it clear, though, that the incidents mentioned were actually due to rusting of the tank, or were they due to some other cause, such as a bad pressure relief valve or cut off switch? By analogy, most water heaters need replacement because they leak from rusting out, but I've only heard of ones exploding as a result of lacking a properly function TPR valve, not from rust. So, any rust through in a compressor would result in a (slow) leak before reaching the level of a catastrophic explosion. Just sayin' ....

Water is non-compressible fluid while air is a compressible gas. The dynamics (physics) of a breach of a container containing a pressurized fluid is completely different than that of a highly compressed gas which will undergo a huge volume expansion at failure.

[Robert Engel]

Oy.

Common sense tells me it all depends on how thick the wall of the tank is.

Come to think of it, I can't even remember when I drained mine last.....

[Keith Outten]

In almost every case of a compressor failure the tank will rupture before it leaks simply because it will fail before the thickness of the material is deteriorated to the point that it can leak before rupture. The minimum wall thickness required to maintain an operating pressure is much higher than the thickness necessary to prevent a leak. This is why an ultrasonic test is the only sure way to verify that a tank is safe to operate because it will determine the actual wall thickness of the tank. Thickness tests taken from the upper portion of the tank will provide you with the original wall thickness so you can determine how much damage has been done.

Steve is right, a hydrostatic test using water is a much safer test. When a vessel full of water ruptures there is rarely any danger BUT both pressure tests if successful will only account for the current pressure capability of the tank it won't prove how much damage has already been done from corrosion at the bottom of the tank. After a successful test the tank could fail a week, six months or one year later if the corrosion continues. Pressure testing is used to prove new vessels most of the time it is rarely used as an inservice testing method.

[Keith Outten]

This caught my eye - sounds like me in the 70's! I worked for PTL and did NDT and in-shop and field structural steel inspection. Spent some time at Quad Cities nuclear plant unit 2, mostly RT but also dye and ultrasonics. Then to NDT on research reactor element manufacturing in Oak Ridge, then weld inspection. Wow, that was a lifetime ago. I haven't even thought to remember it for years.

In Oak Ridge we also inspected and tested pressure vessels, mostly hydro but occasional air pressure testing in an underground bunker. My thought on this thread was also get it tested or replace it. I've also seen the aftermath of failure and it was not pretty. Going to the shop now to drain the compressor tank again. Just for good measure.

JKJ, happily retired, playing and raising llamas in TN

A pressure test that fails will leave a lasting impression. Even when you are investigating a failed test after the fact you rarely forget how much damage is possible. Basically a high pressure tank is a bomb without a fuse.
I got started in the inspection field in the 70's as you did and I enjoyed being in the business and the fact that we saved lives and property. I remember dozens of situations at NASA Langley where we were inspecting steam pipe installed in the 1930's. The pipe ran in the ceiling of office spaces and we often found pipe that was so thin it was ready to burst and the line was right over top of a person's desk. When we obtained authorization to shut down the system for repair the facility manager would complain big time so we would remove the pipe and cut it in half and drop it on his desk. After seeing the condition of the pipe he would always be happy to see us in his facility from then on.

Jim Creech who is the Moderator of our Boat Building Forum here was the Nondestructive Testing Supervisor for Advex Machine when I worked at NASA Langley . Advex was a sub-contractor to the company I worked for at NASA so Jim and I could write a very interesting book on our experiences at NASA Langley. Imagine three ten inch diameter pipes almost touching each other that had 10,000 psi air, liquid oxygen and methane all connected to one test chamber used to test engines for the National Aerospace Plane that Ronald Regan wanted to build.

[John K Jordan]

So, any rust through in a compressor would result in a (slow) leak before reaching the level of a catastrophic explosion. Just sayin' ....

It is possible for rust to slowly degrade and weaken the steel wall or weld of a pressure vessel then catastrophically fail. As one even tiny section fails, the pressure is still high and can rip the tank apart to cause larger areas to fail with an explosive release. This is the reason pressure testing with compressed air always done in an underground bunker. Normally, vessels are hydro tested which means filling completely with water under high pressure. Since water is not compressible, even a tiny leak will "instantly" reduce the pressure in the entire tank and all you get is a water leak. Hydro testing is done out in the open air.

[John Donhowe]

Thanks, all for the info- I stand, gratefully, corrected!

[John Sanford]

An observation, if I may.

Scaring people with stories of 10,000psi tanks failing makes for a good tale, but some readers are going to look at their 125-200psi compressors and go "order of magnitudes less, not to worry."

This video shows the explosion of a normal home garage compressor.

[Dan Hintz]

An observation, if I may.

Scaring people with stories of 10,000psi tanks failing makes for a good tale, but some readers are going to look at their 125-200psi compressors and go "order of magnitudes less, not to worry."

This video shows the explosion of a normal home garage compressor.

Skip to 1:25 to see what happens (watch behind the car)... all video before that is irrelevant.

Looks like the weld blew out on that tank. A weld, if done properly, should be stronger than the surrounding metal. I question the quality of that weld.