Air compressor size recommendation

[Cerebus]

The end plates didn't deform under normal pressure. They deformed at tens of bar! How much pressure does it take to deform a 10mm round steel plate with an approx 30cm diameter? Right now there is a blow-off valve besides the automatic switch on/off.

It deformed, that's enough isn't it? How much does it have to deform before you realise quite how potentially dangerous an object you've got there?

I knew you wouldn't take any notice, I've seen enough of your comments to know that, but please for once stop and think. There is a reason there are laws about pressure vessels. If you are strictly the only person at risk from operating this I'll happily wave you on and chalk it down as "evolution in action". But if there are any other family members, friends, co-workers or neighbours within the potential blast and shrapnel radius then I really do whole heartedly urge you to think about this again. If the simple "for safety's sake, your own and others" argument doesn't persuade you, consider your liability at law. I don't know what the penalties are in the Netherlands, but here in Blighty you can get a £5000 fine and/or 3 months in jail for operating a pressure vessel lacking the correct approvals - which clearly an old, home made, failed and repaired vessel doesn't have or you would have said so by now.

[Kjelt]

The Darwin award goes to......

But seriously I was the day before yesterday with my compressor dealer.
I wanted the autovalve but that was too expensive so I bought an extension pipe with manual bulletvalve so I have no excuse to drain the water each week.
I noticed that there was quite some rust when I removed the old watervalve, so I asked him if I should let the tank be inspected and what could happen when it ruptures.

He said that at 10bar when it rusts through there is a slight opening you will notice immediately, worst case it will rupture.
In the 25 years he worked there he never heard of an exploding tank with max 10 bar of pressure.

So are you serious with exploding schrapnell at 10bars or are you talking about gas tanks with 300 bar pressure, there I can imagine that it would give some blast.

I for instance would like to know how much bar was on this vessel before it exploded

[drummerdimitri]

So I think I will go for the 50L model as it is the same price as the 100L and much more suitable for home use IMO.

Found belt driven models from three brands: Newco (385$), Fiac (395$ )and Fima (425$). They are all Italian bur is one better than the
others? Spec wise they are all the same so I'm not sure which one to get?

[Mechatrommer]

The end plates didn't deform under normal pressure.

the failure point usually not on the metal surface area unless they are badly corroded, watch out at the welded lines, poor welding technique will make those spots as potential failure points.. for diy tank, preferably i will stress test it in controlled area at say 150% of normal operating point...

[SeanB]

Probably the FIAC, but they probably all are alike anyway.

As to pressure testing tanks the standard is to use water and a hydraulic pump, and this will be typically to 1.25 times the rated maximum pressure of the tank. The only exceptions to this are for oxygen systems, where the test medium is pure methanol or ethanol, so as not to contaminate the inner tank wall with corrosion, as this will act as a catalyst when pressurised with oxygen. Iron burns quite well in pure oxygen, you really do not want to start this though inside a tank with high pressure. one reason a lot of oxygen equipment is made in stainless steeld, pure nickel or aluminium, as all of these form a hard oxide layer on the surface, passivating it.

Of course if you want a nasty gas try fluorine, as literally everything burns in it, including things that you would normally consider non combustible, like asbestos, PTFE,  sapphire, concrete, brick, rock. The only reason you can store it and use it in process units is that slowly introducing it into a system, which is only made from pure nickel, allows the formation of a thin coat of nickel fluoride on all exposed surfaces, which is able to withstand further fluoridation and protect the underlying metal. If this wears away too fast you get a reaction that is best viewed from a distance, and it finishes typically with a large very hot hole in the ground when the fluorine finally all has reacted with what used to be your plant and property.

[nctnico]

The end plates didn't deform under normal pressure. They deformed at tens of bar! How much pressure does it take to deform a 10mm round steel plate with an approx 30cm diameter? Right now there is a blow-off valve besides the automatic switch on/off.

It deformed, that's enough isn't it? How much does it have to deform before you realise quite how potentially dangerous an object you've got there?

I knew you wouldn't take any notice, I've seen enough of your comments to know that, but please for once stop and think.

I'm thinking this: why replace a tank with 10mm thick walls with a new tank with walls which are a few mm (or less) thick? Only because someone (who can weld very well) welded it together a couple of decades ago and (unintentionally) tested it at many times the operational pressure? It got stress tested the sh*t out of it and it survived. It just doesn't make sense to replace it. As Kjelt noted it is not working at several hundred bar but at 8 bar and it is a small tank too so the amount of energy available to do damage is very limited.

The video linked by Kjelt above shows a commercial tank which exploded but without knowing why it exploded it is impossible to draw any conclusion. Was the tank bad or was the pressure cranked up way too high (due to a malfunction)?

BTW here is someone with a certified tank which blew open:

[CatalinaWOW]

Kjelts video shows the danger potential of these tanks.  But contrary to the commentary on the video it is not obvious that rust is the culprit.  While there is rust on the tank it doesn't appear to be enough to be the primary cause an of an explosion, though it may have been the last link in the chain.  If rust was the only failure in this tank I would say that there is clearly not enough initial safety margin.

Whatever the cause, the danger is real.  In most of the US periodic re-certification of these tanks is not required by law.  But world over it is required by good sense.  The bigger the tank the more the need.   A good check would not only verify the strength of the tank, but also confirm the auto shutoff and safety relief valve(s) are working properly.  Plural on safety relief because this is one place redundancy makes a fair amount of sense.  That same corrosion that is endemic on the inside of these tanks has an enormous tendency to freeze up relief valves.