Air Compressor Warning

Has anyone found approved reliable methods of testing, and not the U tube videos that proliferate the net ?

Is there any reason why the methods appropriate to small steam boilers should not apply to larger vessels?

Edited By not done it yet on 24/01/2021 19:29:07

Fair comment, but I had a number of large compressors on building sites throughout my working life as well as workshop compressors & i am only repeating a comment from one of the testers, when we asked him about the small cheepo ones in the joinery shop. They were kept in cages. More to prevent people damaging the motors rather than prevent explosive damage. I was worried about how often they should be renewed.. I was quite careful with the ones used in the working environment.

As for the video, i might still be unsure about the safety valve. I would imagine a bit of dirt could easily make it stick & a bang like that blow the grit free. That is probably why model steam engines have to have 2 safety valves.

Hydro testing is a perfectly acceptable method of air receiver testing.

Air lines, as in Neil's link, are another thing entirely. Works both ways, I've been in a power station where a gate valve was cracked open in an auxiliary cooling water system that had not been filled. Valve and flange left the 18" pipe when the slug of water from the pumps hit it. That was only two bar, but the water was traveling pretty fast when it hit the valve...

Hydro testing as per boiler code is totally appropriate in a club or home capacity.

My opinion only!

As to the original post. I also suspect that the safety valve and the pressure switch failed, rather than the tank failing. Metallurgical examination would be required to show how it could have failed that way at normal operating pressure even with a 20% loss of wall thickness.

Edited By Mark Rand on 24/01/2021 20:27:57

Commercial air receivers in Australia have to be registered (AS 43434:2014) and externally inspected every 2 years and internally every 4 years. Home compressors do not have to comply with this design code but recommended for your own safety.

Opening the drain valve will remove the water but not the moisture inside of the receiver. The moisture will condense and will cause rusting. This is why on a steam engine, the boiler is emptied (blown down when the pressure is about 5-10 psi)) while it is still hot. The manhole and inspection bungs are removed and the residual heat in the boiler evaporates the remaining moisture. This will not happen in a air compressor receiver because it is cold and a sealed vessel.

Paul

(PS. When did you check your compressor tank ??? )

Agreed a metallurgical exam is needed to prove it, but the way fatigue cracks propagate is well known.

Looking at a cross-section through the tank, it can be seen once started a fatigue crack crowbars itself into the metal every time the tank flexes. Expansion due to pressure opens the crack and contraction due to emptying closes it. Both actions concentrate force at the point. At first the crack gets deeper very slowly, but, as the crowbar gets longer, the rate of growth accelerates. When the crack is deep enough, the metal can't resist the pressure and tears like a sheet of paper along the flaw. Nothing obvious until the tank rips open with a bang.

Fatigue cracks can be started with with a ding, scratch, or corrosion. When an item is subject to fatigue cracking, the best way to avoid catastrophic failure is to take the item out of service at the end of it's design life, i.e before a slow moving crack has had enough time to become dangerously deep.

Highly stressed components like aircraft wheel struts are often polished to increase their strength by removing the tiny surface imperfections that act as stress raisers. It's not just for show!

Dave

You could fit an adv as they do on trucks!

Dave, your representation of a defect in a round tank is the wrong way round; the wide part of the V should be on the inside. Which would make the pressure pushing outwards even more effective at enlarging it.

Not at all NDIY - indeed that's exactly how decompression chambers were tested in my day - wouldn't think it would be any different today. Most were about 4'6" dia and around 12' long, the doors each end were shut and the entire chamber filled with water and the inspector would couple up his pump just as an inspector would for a boiler. The vent would be capped off once the water came out of the vent on top. Believe it or not given the size difference the hand pump was really not that much bigger than we would use. Pressure would be taken up and held and the circumference measured before and after pressurising.

As said a decompression chamber is just a big volume tank with doors on - I would imagine that that is still how it's done.

Tug

That is probably why model steam engines have to have 2 safety valves.

Sam,

So do compressors - one electrical switch (probably not too applicable to steam engines🙂 ) and the mechanical over pressure valve. They don’t blow up more than once! It was why the compressor in the workshop (I mentioned earlier) was caged - not to prevent damage to the unit by the workers.

I took a careful look at the tank, which still looks new, and found that the date of manufacture was 1992. I've ordered a new compressor.

Since I have a 150psi, 50 litre compressor under my bench, this post gave me a bit of a fright. Mine's an Airpower similar to that shown in one of the later videos. At 3mm wall thickness, the hoop stress in the tank at 10 bar is 44MPa, giving a 5x safety factor against the 2% proof stress value for MS. And way below the fatigue limit for MS. Even at 2mm thickness, the SF is still about 3.5.

In the OP's video, he thinks the failure initiated at a seam, not where the worst rust from condensate was, and I saw that the delivery pipe from the pump seemed to have blown off the unloader valve. So I wonder if the failure was due to overpressure.

I drain my tank every time I use the compressor, only takes a few minutes, but surprising how much water comes out..

Was stunned to learn that the Chemical firm at the bottom of my garden were carrying out a pressurised process (which failed) resulting in neighbours in close proximity being compensated for their washing hung out to dry and having cars parked close to the "occurrence" being professionally cleaned.

As an Ingineer I was horrified to learn the ONLY safety feature on the process was a burst disc.

Regards Ian.

ndiy. I asked the same question a few years ago and never got a satisfactory answer but then the manufacturers must have a resun for this layout, Slill windering why?

Bursting discs are not proof against damage where explosions are concerned.

They used to be a feature of main propulsion marine engines., but the varnished brown paper variety were found not to be safe where crankcase explosions were concerned.

Photographs have been published showing two flights of steel stairs, from cylinder head to bedplate, on the engine severely twisted; but the disc was still intact. The rate of pressure rise is such that the damage is done before the disc can rupture.

Many years ago, Blackstones experimented with "explosion proof" crankcase doors. A colleague told me of a door weighing over 50 Kg being thrown across the shop!

A colleague was beside an engine which suffered a crankcase explosion. The die cast sump fell onto the bedplate, leaving the flange still securely bolted to the crankcase.

I witnessed a crankcase explosion, in a car engine, from across a car park. Quite frightening and damaging!

A colleague lost the ends of fingers when an air receiver exploded, because the oil carried over, vapourised and mixed with the air, exploded.

For air receivers, frequent draining of condensate HAS be a safety measure.

Howard

Pretty sure it's to stop liquid collecting in the drain cock and corroding it.

I don't tilt my compressor to drain it. Rather I leave just enough pressure in the tank to blow the contents out when the cock is opened As water is removed by the outgoing air blast, not by gravity, the cock can be positioned above the water line. For the same reason it probably doesn't matter if the outlet pipe protrudes a little inside the tank.

Dave

I was not the initiator of that comment. I only replied. My reason for this was quite clear in my post, I thought?

I have an Ingersoll Rand (No relation ) automatic float operated drain system for my compressor. That would really need the drain to be at the bottom of the tank. The other problem with it at the moment, is that it's sitting on a shelf and not yet attached to said compressor!

My new compressor arrived yesterday and obviously I want it to last a long time. It will seldom be used continuously for any length of time - mostly for just a few minutes, barely long enough for the pump to start up.

The instructions say to release the pressure and drain it every day after use, and this got me thinking. Suppose the tank is empty and I fill it. The pressure stretches the tank metal and water in the air enters the tank. Suppose 1cc of water is included for every tankful of air. I use, maybe, 1/4 of the tank contents, so 0.25cc more water gets in. Suppose now, that no more air is used for a week, say. What's best? To release the pressure and drain the water, then wait a week and refill with air, or to leave the tank full and empty it every 4 weeks, say?

If I drain each time, then 4cc of water passes through the tank, but if I drain every 4 weeks, then only 2cc of water gets in. Each time the tank is emptied and refilled, it goes through a stretch/release cycle, which can't be beneficial, and just draining the tank doesn't dry the surfaces - which remain damp. So what's best?

.

A clear demonstration of the fact that it is helpful to give attribution when you quote another.

I have never understood why some just pluck a few words, without that courtesy.

MichaelG.