Compressor safety, or how do I make sure an old compressor is safe?

[S. Heslop]

A neighbor across the road who restores old cars gave me a spare compressor he had in his shed. I'm pretty excited about it!



But being fairly old (about 30 years), and having sat in a shed for a while, i'm nervous about it possibly being dangerous. It runs fine and I got it up to 40psi before deciding that was far enough. There's a large plug on the back with a valve for getting the water out that I could possibly remove to inspect the inside of the tank, but I thought i'd probably best not remove it until asking if it was necessary.



The guy also said that the output regulator is shot and leaks air, although it isn't currently it probably will start doing it again if I try fiddle with it. I can't find any information on this compressor online and i'd have no idea how to go about potentially repairing it.

Anyways any advise is appreciated.

[John Rudd]

A new regulator is the least of your issues......you can pick one up from eBay or machine mart....

In order if importance.....

A hydraulic test of the receiver if you have doubts about it......
The relief valve, needs testing to make sure it goes at the correct pressure, which should be below the MWP of the receiver, around 120 psi ....
The pressure switch, should stop the pump at its set pressure....the setting needs to be below the relief valve setting....100psi.....would be ball park.....the differential on the switch doesn't want to be too wide or too narrow....if the switch is set to trip at 100, then 75-80 psi should be a decent value for resetting...
None of these figures are cast in stone, mainly dependant on the pump...but ought to be a start for you...obviously if you aren't going to use the compressor for anything serious and just for a bit of spraying, then the settings could be lowered if you have any doubts....

I'm sure mine will not be the only opinion expressed......


If you are up sure about anything shout up.....don't take risks....compressed air is just like any other compressed gas and can be just as dangerous, but you already know this....


If you have access to a megger, you could quickly zap the motor for a quick insulation test.....

[David Jupp]

Main potential hazards are probably, receiver corroded and weakened, relief valve jammed or restricted, loose pipe or fittings coming adrift under pressure.  Oh, and not forgetting electrical hazards

I'd suggest, inspect receiver via one of the plugs/fittings (the difficult bit is to decide how bad things are if you do find corrosion).  Strip and clean the relief valve to make sure it operates freely and that there are no restrictions (that should protect things even if the pressure switch isn't working.  Make sure everything is securely attached.  Small Jubilee type clips are often frowned on for securing flexible air hoses (but realistically they are still widely used) - use something better if you can find it.

Check condition of plug/cable, make sure earth is connected. 

Replace that clear reinforced hose - it has probably gone very stiff/brittle.

[appletree]

The Air receiver is the more difficult bit to check, these type of compressors tend have rather thin receiver walls. Obviously corrosion, particularly pitting is the big issue, if you can borrow a borescope (Lidl do them) you could inspect through any of the connection ports and soon decide the condition. Other options are do away with the receiver altogether, operate at lower pressure, however there is still a lot of force available at 40 psi. There is no need to have the compressor near you as long as you keep pipework large enough to avoid pressure drop. Visual inspection in some form is the order of the day   

[vtsteam]

I think inspection on practically any tank through a borescope (and I have one) won't tell much except to show that yes, in most tanks unless right off the showroom floor, there will be some rust. But how much, how bad? You can't really tell that by sight.

If it was mine, I'd fill the tank with water and hydrotest it to about double what I'd want to run it at, and hold it for a minimum of 5 minutes checking the gauge. If it's filled with water, it can't explode, just leak.

If you look elsewhere on this forum you'll find a boiler feed and hydrotesting pump I made from scratch. You just need that and a decent liquid filled pressure gauge in the range you want to check (maybe $10 on ebay).

One of the handiest things I've made here, and definitely something that can provide peace of mind.

[Kjelle]

That is a quite good compressor, sold by LUNA, a Swedish company... I doubt it's as old as your neighbour says, I'd say maybe 20 years.... LUNA is still in business, do a google, and you'll find them. I think they might stock whatever parts you need, and there might even be a spare parts list and sketch on their website...

It was touted as better than the Chicom stuff other were selling at the time it was new, I think they were made (partly) in Italy... Don't quote me on that!

Kjelle

[S. Heslop]

I think inspection on practically any tank through a borescope (and I have one) won't tell much except to show that yes, in most tanks unless right off the showroom floor, there will be some rust. But how much, how bad? You can't really tell that by sight.

If it was mine, I'd fill the tank with water and hydrotest it to about double what I'd want to run it at, and hold it for a minimum of 5 minutes checking the gauge. If it's filled with water, it can't explode, just leak.

If you look elsewhere on this forum you'll find a boiler feed and hydrotesting pump I made from scratch. You just need that and a decent liquid filled pressure gauge in the range you want to check (maybe $10 on ebay).

One of the handiest things I've made here, and definitely something that can provide peace of mind.

Found the thread where you made that pump.

http://madmodder.net/index.php/topic,9920.50.html

So I figure the way to do things would be to take the thing apart and plug all the holes, test it to about 200 or more psi, then reassemble it and keep an eye on the gauges as it fills to make sure the motor cuts off at the right points. Is it at all likely that the gauges will be out of calibration? I suppose I could test the relief valve too while hydrotesting it.

Also there's a few plates on the tank that say that it was made in 1982. In Swedish and Italian. It's Italian made it seems.


It seems my favourite image host has gone down. Hopefully not for good, but good image hosts never seem to last for too long. I guess they're only losing money unless they're pulling some bullshit like photobucket or imgur.

[awemawson]

[quote author=S. Heslop


It seems my favourite image host has gone down. Hopefully not for good, but good image hosts never seem to last for too long. I guess they're only losing money unless they're pulling some bullshit like photobucket or imgur.
[/quote]

That's why I always upload pictures to the forum rather than use a third party. That way the images are always there so long as the forum survives. Nothing worse than reading an interesting old thread where the pictures have all gone 

[appletree]

Most receivers only have small inspection openings, so even the official insurance inspector can only assess using his eyes and aids such as mirrors etc. At work we had air receivers little larger than yours the Inspector only had a 2 inch diameter hole to look through admitted they were heavily constructed but they never went away for pressure testing. Have you considered contacting the manufacturer for advice, or a pressure vessel tester for their opinion on the way forward for a receiver of this class.

[appletree]

Just had another look at your pictures, I think the vessel is better quality than suspected, the wall thickness of the inspection plug boss is of a good section and I would say the hole large enough to inspect through, after all that is what it is there for rather than a small pipe connection.
Any rust is likely to be at the lowest point a dentists type mirror would allow you to see, even if condensate has rested in the bottom it still needs oxygen for rusting to occur.

[S. Heslop]

That's why I always upload pictures to the forum rather than use a third party. That way the images are always there so long as the forum survives. Nothing worse than reading an interesting old thread where the pictures have all gone 

I wish i'd done that sooner. I trusted iforce.nz since its been up and stable for a good few years now. I figured you wouldn't be able to embed attached pictures but it seems to be working fine.  If iforce doesn't come back i'll probably go through my old threads and attach the pictures, i've kept them sort-of organized in folders.

[S. Heslop]

Just had another look at your pictures, I think the vessel is better quality than suspected, the wall thickness of the inspection plug boss is of a good section and I would say the hole large enough to inspect through, after all that is what it is there for rather than a small pipe connection.
Any rust is likely to be at the lowest point a dentists type mirror would allow you to see, even if condensate has rested in the bottom it still needs oxygen for rusting to occur.

I imagine in a compressor there's plenty of oxygen about. Corrosion might be a concern but i'm also worried about fatigue. Hydraulically testing it would put me more at ease than just checking for corrosion.

[appletree]

Obviously when in use there is plenty of oxygen about in use but if the vessel is unused and the valves shut the oxygen will become depleted. I would inspect either way, going back to my earlier post as to what happens in industry, i do not see why the vessel would be fatigued, the vessel is designed in the first place even if the safety valve is not working which you can check, the maximum pressure achievable is determined by the compression ratio of the compressor. 

[vtsteam]

There's been plenty of oxygen and moisture condensation in most any compressor tank that's been used. That's why they put drains on the bottom. And likely rust, at least surface rust.

If your eyes can determine rust thickness under scale over every square millimeter of the inside of a tank through a borescope, and you know the thickness and composition of the metal and can then calculate the allowable stress left after deductions for the rust locations you've found, then that's the way to go.

Otherwise a pump you can make yourself and a reasonably priced gauge will actually tell you the strength of what you have to the degree you are interested in.

[David Jupp]

In an industrial setting PSSR would apply - which requires a scheme of examination to be in place.  Such schemes would be pretty similar for most compressors around this size - manufacturers often supply one with the machine (so may be worth asking).

Test pressure should be marked somewhere.  For smaller vessels, the original plan could have been no routine test, but with a 'use by' date - though the label shown looks like it has space to record routine tests.

[vtsteam]

One other suggestion if you are worried or if on examination or test you don't like what you see..... buy one of those refillable portable air tanks and replace your present one. They're pretty reasonably priced.

[Lew_Merrick_PE]

Your local welding gas company should be able to provide you with a pressure test & certificate for a reasonably low cost.  ???

[Henning]

Does it say anything more than Luna compact on the side of the tank? There should also be a sign somewhere...
If it does, i may be able to help with schematics and/or parts lists and maybe even a user manual.

[mexican jon]

I really do have to commend everybody on being so safety conscientious  but re-hydro testing a small compressor that probably has had little usage is probably going to cost more than what a new one could be purchased for  .  The big difference between a compressor and a boiler (I know they are both pressure vessels) is compressors have a set pressure control and a pressure relief valve and they charge to the set pressure at a given rate. Boilers whilst having a pressure relief valve basically get to pressure as quickly as you can heat the water and therefore are more susceptible to stress damage  .

I’m not for 1 minute saying that compressors don’t go bang because they do but generally a compressor that has been used in a stable inside environment will very rarely have a tank that won’t outlast the pump or other components  .

But as I said at the start I really do have to commend everybody on being so safety conscientious 

My main compressor in the workshop was built in 1948 and is still going strong  .

[mattinker]

Simon,

Plug all the holes, fill the tank up with water and inflate with a high pressure bicycle pump! I use the metal valve holder from an inner tube silver soldered into a screw plug to connect the pump. Commercial vehicle tyre fitters will give you old inner tubes. Pump up to twice the working pressure to test. I have a 500 litre tank that I need to test, half a ton of water to fill it up! An other alternative high pressure pump is a pressure washer.

Regards, Matthew.

[mexican jon]

We normally test 1.5 times max working pressure for the initial test and certification then periodic testing would be to max working pressure  .  And just so you know our normal working pressure is 15000 PSI (yep fifteen thousand)

This is for receivers rather than boilers (don’t know if boilers are 2 times max design working pressure)

[edward]

if you have a Scuba shop near you they will have the tools to look inside, and an ultrasound thickness measuring device if there are any dodgy spots. it might cost, dependant on how nice they are feeling.

[edward]

Forgot to add: not all Scuba shops test in house, so it would need to be one that does.

Scuba cylinders are typically 232bar WP and 348bar TP made of steel and used in salt water, so divers tend to be quite anal about testing.

[David Jupp]

Stick to the test pressure on the label - (I don't read Swedish, but it looks like 8 bar test with 7 bar working ?).   Certainly don't try 2x working - you may cause damage.

[awemawson]

Simon,

I have a pressure test pump like the one in the picture below and would willingly test your tank were distances and geography not such a challenge. Though I must confess I've never hydro statically tested my compressor tank just regularly drained condensation and given it a visual check over (it's in an otherwise unoccupied steel Portakabin - if the worst happened it would be contained)