Building an Air Tank

[greg]

Hi All,

I have an Endless Air fitted to my car at the moment, and have been trying to fit an air tank for a while now but the area that i have is a bit small / cramped / and not really ideal for fitting an off the shelf type tank of any decent size.

I was playing around with the idea of building a tank to fit into the space.

Does anyone know what the guidelines are for safely making an air tank? I am running a 90psi pressure valve, and would definately have a pressure release valve on the setup.

How do you pressure test your tank for the first time?

i was thinking of using - for example - some lengths of 100x100 2mm or 3mm box with ends welded on.

Is this a good idea? or should i just accept that a small off the shelf tank is the right solution?

Thanks,
Greg

[MissDrew]

The tray on my hilux is an air tank.
Its made of a mix of 3mm and 5mm side wall RHS.
When making it we would just weld it all up then plug the compressor on to it and let it build up pressure. Then while it was still filling up we would go around every weld with a spray bottle of soapy water. Once we had mark a few leaks we would disconnect the compressor, let the air out and then grind and re-weld. Then reconnect the compressor and do it all over again.
8 years later and its still going sweet and never leaked different story about fittings and hoses tho

[ISUZUROVER]

I made sliders that double as air tanks. I run them at 120psi.

75 mm OD x 4 mm wall. 6mm ends, and 10 mm flatbar welded on to attach the fittings.

Welded it up, put bungs in all the holes, and pressurised it. Soapy water to check for leaks.

[nicbeer]

been thinking this myself

at what size does the tank loose its effectivness?

like for running airtools (ratchet, grinder)

i am also running a/c endless air on my vit engine.

[ISUZUROVER]

been thinking this myself

at what size does the tank loose its effectivness?

like for running airtools (ratchet, grinder)

i am also running a/c endless air on my vit engine.

The bigger the better, but anything is better than nothing.

If you are running air tools continuously (like grinders/sanders), then the tank makes bugger all difference.

[mrw82]

I have endless air (home made jobbie) on my zook, whenever I use it I lock the engine at 2000rpm.
at first I had my roll bar plumbed up as a tank, worked it out to be about 4 litres capacity. found the compressor was cycling a bit too much, so I put on an elcheapo bunnings tank (19 litre capacity) and found the compressor was struggling to keep up, running for a long time to pressure up.
so I ditched both Ideas and got a 9 litre aluminium tank. perfect, it cycles on and off nicely while inflating tyres.

I would recommend around a 10 litre tank for any endless air setup, definately no larger then 15 litres. bigger tank capacity is not always better.

[Tojo]

no it is not a good idea and yes you would be better off purchasing an off the shelf tank. A compressed air receiver is a pressure vessel. Pressure vessels have very rigid standards for design, verification of design, manufacture, pressure testing and in service inspection. They have these standards because pressure vessels are potentially very dangerous. The size of the pressure vessel you are considering making is quite small and thus the potential risk is also smaller than for a larger pressure vessel.

If you really want to make it yourself use thick wall pipe / tubing in preference to RHS. Make sure all welds are strong. Make sure you have a pressure relief valve with the correct setting. Pressure test it by pumping it up with compressed air or dry nitrogen and immersing in a big tank of water.

[toaddog]

If you are going to test it yourself do not use air or gas, fill it with water and pump that up. If you use a gas/air of some sort and it lets go it may explode. If you are using water then it just springs a leak.

[Dyna Beast]

A old bbq gas bottle makes a great receiver.Is a pressure vessel to start with and was originally made for far greater pressure than a air receiver.Lot of home made compressors use them.
Cheers

[dogbreath_48]

If you are going to test it yourself do not use air or gas, fill it with water and pump that up. If you use a gas/air of some sort and it lets go it may explode. If you are using water then it just springs a leak.

That's right.

I wouldn't be too worried, 90psi isn't much. If you're really concerned about safety definitely go with the mechanical pressure relief valve as well as the electric switch

[jet-6]

Is it just me or what, im forever chasing tiny leaks in my system, drops froj 130 psi to 0 in 48 hours, i cant find a leak that small!

[Struth]

90 psi is a lot.

2 psi blew an autoclave door over 6 feet in diameter across a factory and through a brick wall at Dunlop.

The volume of compressed air plays a large part in how catostrophic any failure will be because the actual amount of air escaping from the receiver is very relevant to how much force is created. Or put correctly how much force is escaping from the failed tank is relevant to how much compressed air is in the tank and at what pressure.

Use pipe not RHS for a stronger tank, RHS will concentrate the Psi in the corners creating stresses and likely failure points.

If you want a large receiver then buy one.

Cheers

[nicbeer]

A mate has one in his crusier made out of 100x100 4-6mm wall.

damn strong and end caps to suit. looks good and works great.

[greg]

Thanks for the comments guys... It's interesting to see that i am somewhere in the middle as far as safety vs building your own goes.

I did some more measuring under the car last night - it's looking like i might have to run multiple little tanks to get any sort of volume at all.

MrW82 - i found a site yesterday that suggested that the ideal tank size is equal to the CFM output of the pump. This is apparently the most efficient setup for pump cycling etc. That said - at 6cfm for my smaller model Endless Air = (30x30x30x6) = 55.8L, i'm never going to get near that under the car.

[Tomo_89]

i used a fire extinguisher for my air tank.. run 100 psi in it no worries.. but if you do make your own be sure to do any pressure testing with water

[PacMan]

hi

i use a fire extinguisher too.

there are very handy because there are painted from inside - so no rust.

greetings
chris

[nastytroll]

be careful using fire extinguishers for receivers, some are only tested to 60-70psi.

A gas cylinder is tested to something like 4300-4600psi, test preasure will be marked on them in Mpas. Someone always has one of these out of date kicking around, and they use a more common thread.

The downside to old lpg cylinders is the stench, it takes ages to get the stink out.

[BEU77y]

I've used a 4.5kg fire ext. Works out to about 5-6L capacity. I'm pretty sure the guage that was on it read up to 150psi or so. Different sorts of extinguishers might have different ratings.

The other handy thing about them is the size. 140mm dia. My chassis rail is 150mm. Sits alongside the chassis rail real nice.

[r0ck_m0nkey]

be careful using fire extinguishers for receivers, some are only tested to 60-70psi.

What extinguisher are they?

Some general working pressures for Extinguishers are 100psi for water, 175 - 195psi for a Dry Chemical and right up to about 800psi for CO2. Even a Water Extinguisher which is the lowest pressure is pressure tested to 2.5MPa (290psi) They are a pressure vessel no different to anything else and are rated, looking at the guage (usually in kPa) is the quickest way to know roughly what the working pressure is

As others have said, to pressure test properly, it should be done with water. Fill the vessel with water then pressurise it SLOWLY. The vessel should be in a securely caged area and then you simply watch for water leaks. Typically if it's going to leak it will do it at fairly low pressures. once at test pressure, let it hold pressure and watch for any leaks. Don't kid yourself if you think 100psi isn't much, i've witnessed the head of a fire extinguisher being pressurised come off at low pressure, it went about 3m into the air, hit the tin roof of the shed and went through it. That's enough force to kill.

The only technical problem with using an extinguisher (except CO2), is wall thickness. It can dent easily and if it is dented with the metal surface scratched it is defectable and should not be used. So if going underneath, should have some guard protecting it from damage.

there are very handy because there are painted from inside - so no rust.

No they're not. Obviously the stainless steel and smaller alloy ones won't rust, but the steel ones certainly do. Typically they will rust, fatigue and crack around the neck of the extinguisher where they are welded.

[nastytroll]

be careful using fire extinguishers for receivers, some are only tested to 60-70psi.

What extinguisher are they?

Some general working pressures for Extinguishers are 100psi for water, 175 - 195psi for a Dry Chemical and right up to about 800psi for CO2. Even a Water Extinguisher which is the lowest pressure is pressure tested to 2.5MPa (290psi) They are a pressure vessel no different to anything else and are rated, looking at the guage (usually in kPa) is the quickest way to know roughly what the working pressure is
quote]

Thanks for the tech, I don't recall the brand of extinguisher but was suprised at the low rating myself.

[ISUZUROVER]

While I agree, pressure vessels can be very dangerous, anyone who would have sufficient trust in their welding abilities to weld structural components like chassis and suspension bits should be able to weld up an air tank that won't fail.

However - if you use smaller diameter pipe for the main part of the air tank, the force on the end caps (and the chances of Shit Happening if anything goes wrong) will be smaller.

e.g: Lets say you are building a tank for 90psi as mentioned.

You have the option of 3" pipe and 6" pipe.

An end cap on a 3" pipe will be approx 7" square in area. An end cap on a 6" pipe will be around 28" square in area.

That means that at 90psi, the force on the 3" end cap will be around 290kg. But the force on the 6" end cap will be around 1.16 TONNES.

[80's_delirious]

I have an alloy CO2 cylinder sitting in the shed ready to use. As rock_monkey said they are tested to far more than the working pressure, and Co2 cylinders are typically drawn alloy, so no rust and no welds.
they are also long and fairly small diameter so easier to tuck up out of harms way

[Slayer]

just modeled up a simple 100x100 3mm, 10 mm end plates rhs tank using strand7, the fea program we use here at tafe in mechainical engineering. with an internal pressure of 100psi the stress concentration in the corners is marginal because of the radius, anyways its not an issue. maximum fibre stress was 43Kpa.. anyone who knows anything about mild steel knows it wont yield untill around 210MPa.. therefor this design has a safety factor of almost 5 thousand lol.. as an engineering rule of thumb we use a weld strength factor of .9, so your still fine
anyone who thinks it will explode, go out the shed and rig your self up some fittings and pump a coke bottle up to 100psi, it will easily hold the pressue.. why wouldnt a 3 mm steel section.

just to put things into perspective, a 1000psi pressure still only stresses the material to .3 of a Mpa..

just some more professional advise using the software thats avalible to me.. take it as u wish..

http://www.strand7.com/

[nicbeer]

just modeled up a simple 100x100 3mm, 10 mm end plates rhs tank using strand7, the fea program we use here at tafe in mechainical engineering. with an internal pressure of 100psi the stress concentration in the corners is marginal because of the radius, anyways its not an issue. maximum fibre stress was 43Kpa.. anyone who knows anything about mild steel knows it wont yield untill around 210MPa.. therefor this design has a safety factor of almost 5 thousand lol.. as an engineering rule of thumb we use a weld strength factor of .9, so your still fine
anyone who thinks it will explode, go out the shed and rig your self up some fittings and pump a coke bottle up to 100psi, it will easily hold the pressue.. why wouldnt a 3 mm steel section.

just to put things into perspective, a 1000psi pressure still only stresses the material to .3 of a Mpa..

just some more professional advise using the software thats avalible to me.. take it as u wish..

http://www.strand7.com/

that sounds gd.

what difference to the output does it make using less wall in the end caps?

as this is basically exactly what i am planning to do for the zook.

[fester2au]

The other thing to think about is in the end caps as mentioned. Notice almost all pressure vessels talked about above have curved end caps or ends as this spreads the pressure better than a flat cap. I have seen people use tapered ends for this reason if they can't use domed ends. So cutting the box or tube for that matter on an angle and capping that is supposed to reduce the pressure across the flat surface. read that somewhere but can't recall where so anyone shed further light on this. The assumption was the angled end was a safer design for the home handy man to fabricatre than a flat ended design.