STEPNEY 5"g terrier gordon smith safety valve

[jma1009]

one of the questions i asked back in june 2011 when i started building my 5"g terrier STEPNEY was whether anyone had a suitable 'squat' gordon smith safety valve design. anthony kendall very kindly supplied me with all the gordon smith articles from EIM, and both jim scott and george ray have also provided much help and advice, so the above pics are my version of same. the internals are as per the POLLY drawing SV#2E, and i ordered the spring, ball, and drawing. the externals are to suit the BOXHILL type boiler.

anyway yesterday evening i had ago at making same. as the external shape of the safety valve isnt critical and is hidden by the outer dome casing, i fitted a 1/2" x 26 tpi adjuster to the top which avoids the rather fussy top housing for the spring as there is sufficient clearance with the large adjuster to simply counterbore the adjuster for the spring as per a gordon smith drawing george ray very kindly sent me.

anyway this evening i tested same on my compressor and the results were very disappointing. i took everything apart and found the dimensions again spot on to the drawings.

anyway i very carefully checked through the articles supplied by anthony kendall and noted that the distance between the bottom of the shroud and the valve seat with the ball in the valve should be 25 thou. my POLLY drawing didnt show this dimension but showed the 3/16" ball having a 60 thou distance from the shroud. this was spot on upon remeasuring same. however upon making up a bit of brass with a 5/32" dia reamed hole same as the seat in the valve body i found the distance between shroud and valve seat to be only 15 thou. i machined off 10 thou from the shroud, tested the safety valve again and it worked perfectly! i was quite sure that my original machining of all the parts was within 1/2 thou.

anyway anyone making a gordon smith safety valve with a 5/32" seat and 3/16" ball may like to note that a careful check of the distance between shroud and valve seat is required and if as in my drawings the dimension is shown as 60 thou between shroud and bottom of the ball it should be 50 thou, or a 25 thou gap between shroud and valve seat.

the above valve is only 9/16" tall including the washer . i have subsequently added a small groove to the top of the rod sticking out of the top to aid releasing same just in case the valve sticks.

i will test same with steam as opposed to air over the weekend hopefully on one of my other locos.

incidentally the POLLY 3/16" stainless ball supplied with the drawing and spring is magnetic. it will get thrown away as magnetic stainless balls are liable to corrode in steam locomotive boilers. i will replace it with a non-magnetic stainless ball.

cheers,
julian

[Deleted]

I made a pair of pop valves for my LMS dock tank, using my own drawings but following Gordon Smith's EIM articles for the critical parameters. They worked perfectly first time, with almost no adjustment needed. I've just re-checked my drawings and the valve seat is 5/32" dia, the ball 3/16" dia, and the distance from the bottom of the shroud to the face of the valve seat is 0.020". Working pressure is 80 psi and they go off with soft 'pop' as intended. Gordon Smith certainly knows his stuff!

[jma1009]

thanks superseven for your comments which confirm my own views that as expected small changes in the dimensions can make a BIG difference. having made numerous 'scale' fiddly boiler fittings over the years and ventured into the wonderful world of injectors this is something ive got used to.

when i stated 'very disappointing' the valve opened like a normal 'pop' valve with quite a sudden opening at 90 psi, and would not shut without gently tapping the brass guide with my finger.

when i stated 'worked perfectly' the valve opened far less suddenly and there was no pressure increase with my rather large ex WD compressor running, and the valve shut off after a drop of 2-3 psi. so needless to say i was then very pleased with the result. i will convert all my locos to Gordon Smith designed safety valves in due course. gordon's contribution to the miniature steam aspect of model engineering must be the single most important development in the last 5 years?

too wet here to do a steam test this weekend.

cheers,
julian

[Deleted]

I'm a very long way from making safety valves and haven't researched yet as to how close Don's are to the prototype but reading this gives me encouragement for making my own as close as possible to the outline of the real thing while keeping a reliable miniature valve.

Great work as always Julian...

Pete

[Jim]

I made Boadicea's safety valves to Gordon Smith's articles and they work just as Julian describes. One valve a makes a 'buzzing' sound when blowing off due to the hole that the valve spindle passes through not being a close fit so I'll be remaking that. As Julian says 'small changes makes big differences. I also made my own valve springs as I couldn't get any here that would fit the design.

Jim

[jma1009]

hi jim, hi pete,
thanks for your comments.

pete, don's valves are 'traditional' and so i can recommend in due course using the nearest of the POLLY drawings that will be suitable. however, don was quite capable of working out spring loads and dimensions when it suited him (which wasnt very often as its quite a bit of complex maths though in his drawing office days it was bread and butter stuff).

jim, gosh im very impressed you made your own springs! im afraid calculating spring loads and doing the maths is beyond me. if GS had said use such and such SWG stainless wire wound so many turn to the inch to a certain ID i could have coped but alas this sort of info is lacking. as pete knows i did wind the springs for STEPNEY's buffers from quite large SWG wire. ive got a large tin of different SWG stainless and spring wire in very long coils. usually in my case its winding very small stainless or phos bronze springs for steam brake cylinder relief valves and whistle valves. at least i know what SWG and turns per inch STEPNEY's springs are so can copy same with some degree of accuracy, though the minimal cost from POLLY or GS's supplier hardly makes it worthwhile for parts that are so crucial. the Lee catalogue can be downloaded on line for the springs GS used.

btw i will still make 'proper' pop valves as per the traditional methods for locos that in fullsize had them - though in my current stable of locos this should be nil! GS didnt dismiss traditional pop valves - only that the accuracy required for the stepped recess he considered difficult for the amateur. i worked out my own easy way of achieving same many years ago, but that's another story!

incidentally quite a bit of my stock of piano wire has been used over the years to line up fullsize traction engine cylinders with the crankshaft etc!

cheers,
julian

[Jim]

Oct 21, 2013 22:14:30 GMT jma1009 said:

jim, gosh im very impressed you made your own springs! im afraid calculating spring loads and doing the maths is beyond me.
julian

I must own up Julian and reluctantly admit my ability at maths is limited to what I can work out on my fingers, I was the despair of my poor maths teacher who went bald during the time I was in his classes.

In truth I couldn't get the stock springs recommended by Gordon Smith but like you I had some of the correct SWG SS wire and after some trial runs was able to wind a pair of working springs similar to the ones specified.

Sorry for misleading you there Julian but my mind went blank when I saw the formula needed to design the spring for the job.

Jim