Square Gas Tank

[Brian GParticipant @briang]

Hi

My son and I are returning a 25-30 year-old Steamlines "Rheidol" 16mm locomotive to use and decided to start with hydraulic testing, under which the butane tank sprang a leak at just over 100 psi.

The tank is a square 20 gauge brass tube 1.25" x 1.25" x 1.625" long. A 6BA stay connects the two 18 gauge brass ends and is nutted externally. We would like to make a new tank that resembles this closely as it is visible in the cab of the locomotive.

The planned replacement uses 18 gauge tube and 1/8" ends, with a single 1/8" bronze stay. This should both increase the stiffness of the tank and more than double the area of the silver soldered joint.

Should we also add cross stays to the tube or just build and test this design, (Or is it better to exactly copy the failed tank as it is an established design?)

[Brian GParticipant @briang]

For 16mm Rheidol

[JasonBModerator @jasonb]

Looks to me like you are staying the wrong plates, the 18swg (16 on drawing) are a larger area and less than half the thickness of the ends so will tend to flex outwards a lot more than the end plates.

[Brian GParticipant @briang]

Hi Jason, the tube is 16 gauge, my drawing is right (or at least matches the materials I have bought), my posting is wrong. On the original tank the ends are also thicker than the tube but this is the only stay fitted.

Brian

[vintagengineerParticipant @vintagengineer]

Why not fit a round tank inside the square tank?

[Paul KempParticipant @paulkemp46892]

You say you were pressure testing it and it leaked at 100psi – what pressure were you intending to go to? Pure Butane in its liquid phase will be 2.6 bar at 38 degrees C I believe. How hot are you expecting the tank to get? 90 psi as 2 X WP should be adequate for a pressure test surely?

Paul.

[JasonBModerator @jasonb]

Gas tanks need to be tested to around 300psi, I'll check the exact figure later

 

EDIT for 70/30 gas mix the tank needs to be tested to 333.5psi

Edited By JasonB on 24/02/2017 18:35:19

[Neil WyattModerator @neilwyatt]

Posted by Paul Kemp on 24/02/2017 17:42:40:

You say you were pressure testing it and it leaked at 100psi – what pressure were you intending to go to? Pure Butane in its liquid phase will be 2.6 bar at 38 degrees C I believe. How hot are you expecting the tank to get? 90 psi as 2 X WP should be adequate for a pressure test surely?

The pressure rises exponentially with temperature and there's a good chance of the tank getting hot (e.g. if boiling water or steam is directed at it).

Neil

[Brian GParticipant @briang]

As we do not possess a test certificate for the tank or boiler, we decided to test both before presenting the boiler for testing. An earlier thread carried a link to this test schedule, so our intention was to test to 185 psi and to use only pure butane.

**LINK**

The tank is in a well ventilated position right under the roof of the open backed-cab and in operation is unlikely to be affected much by the boiler, but it could become heated in the sun as the model doesn't have the double roof of the prototype.

I had wondered about a round tank concealed in a box, but didn't want to make any change to the gas capacity until we know the run time of the loco. As the tank/valve/jet assembly is easy to remove, one option might be to do this as a display tank, with an alternative longer cylinder (perhaps rated for mixed gas) for extended running, although the off-axis location of the filler (dictated by the existing hole in the cab roof) could be an issue.

Brian

[Author]

Posts