How to select the correct Gordon Smith safety valve

[spindizzy]

can anyone tell me how to select the correct ball diameter/seat size I need for a Gordon Smith design soft pop safety valve suitable for a 4" scale Burrell traction engine. I would like to disguise two soft pop safeties in the chimneys of standard design safety valves. A friend has a similar set up on a 4 1/2" and he uses Ross Pop type valves with a 1/2" ball.

[Jim]

According to the Australian Boiler code the size of a safety valve's orifice which is then used to determine the ball diameter is arrived at by the working pressure and area of the heating surfaces set out in a Nomogram.

The Nomogram was first compiled by Henry Greenly and is in his hand book on model locomotive construction. The only change to Greenly's work in our Copper Boiler code is the conversion from Greenly's Imperial units to metric which is used out here.

Jim

[jma1009]

Hi Ian,

Polly Eng do the Gordon Smith safety valves. The drawings balls and springs are very reasonably priced.

There are not many in the list (see online catalogue p.55) for big boilers.

My own personal view is that the larger the safety valve (such as your's will require), the less problems with a conventional design or copying fullsize so far as the accumulation test is concerned.

If you want to design your own Gordon Smith safety valves then you will need the original EIM articles and be prepared to do a lot of reading and study and research. There are also a number of well known errors in the original articles (and even some errors on the drawings sold by Polly).

Cheers,
Julian

[mbrown]

Thanks Julian,

Maybe you could tell us what the errors are. I made two Gordon Smith pop safety valves for my BMR loco, working scrupulously to the dimensions in EIM. They work perfectly well as conventional safety valves, but "pop" (soft or otherwise) they aren't.

It would be good to know how to make them work as intended.

Malcolm

[4fbuilder]

Hello Malcolm,

I've built several pairs of the "Gordon Smith" design soft pop valves, only one pair didn't work properly, goodness knows what I did but when I carefully examined the measurements they were nothing like the drawing! I think I must have made it up as I went along. Seriously though, the pairs I have built do exactly what it says on the tin, a gentle soft lift with about a 5lb differential with an almost snap shut action. On each occasion I've always used the correct springs as supplied by Polly Engineering. The valves are that good they will rise and shut in unison.

I can only speak as I find.

Regards,

Bob

[mbrown]

Thanks Bob,

I made mine before Polly started stocking the springs so I got mine from Lee Springs to the spec Gordon set out in the EIM articles. As I say, they work OK, opening and shutting cleanly so I guess it's possible that I am just expecting a more defined pop action than they are meant to give. But if Julian (whose work on injectors I have good reason to respect) thinks there are errors on the drawings, I'd be interested to know the whys and wherefores.

Malcolm

[Deleted]

Does anyone have the articles that they wish to share?... I fancy doing my own when the time comes so I can get the exterior as close to scale as possible...Hearing that Polly do the drawings and the correct springs has whetted my appetite...

Pete

[mbrown]

Jul 27, 2017 17:55:04 GMT @greenglade said:

Does anyone have the articles that they wish to share?... I fancy doing my own when the time comes so I can get the exterior as close to scale as possible...Hearing that Polly do the drawings and the correct springs has whetted my appetite...

Pete

I can dig them out at the weekend and if you send me a PM with your email address I can send you scanned copies.

Malcolm

[Deleted]

Thanks Malcolm....I'll send PM now....

Pete

[jma1009]

I would agree with Bob and Malcolm that the Gordon Smith safety valves work very well indeed and I would not these days make anything else.

I described on here a few years back the 'squat' valve from Polly of Gordon's design and the error on that drawing which was found by testing the valve then going back to the EIM articles and working through Gordon's design formulae.

There is a list of errors in respect of the original EIM articles. A forum member very kindly supplied me with downloads/scans of the original articles plus the list of errors via email. It would be wrong of me to name the forum member without his consent and he may not appreciate being inundated with requests!

Pete I will send you an email or pm.

Cheers,
Julian

[steamer5]

Hi Julian,
As I've got a couple of different safeties to make & Gordons design is the chosen one, so I'd be very interested in what errors were found. Hopefully the member who sent you the error information will kindly see this thread & provide the info or give you permission to post the same.

Cheers Kerrin

[Deleted]

A big thanks to Malcolm for sending me the relevant articles..I very much look forward to learning from them..

Kind regards

Pete

[spindizzy]

Thank you for your replies gentlemen.

[Tony K]

Jul 27, 2017 9:15:16 GMT jma1009 said:

Hi Ian,
Polly Eng do the Gordon Smith safety valves. The drawings balls and springs are very reasonably priced.
There are not many in the list (see online catalogue p.55) for big boilers.
My own personal view is that the larger the safety valve (such as yours will require), the less problems with a conventional design or copying full-size so far as the accumulation test is concerned.
If you want to design your own Gordon Smith safety valves then you will need the original EIM articles and be prepared to do a lot of reading and study and research. There are also a number of well known errors in the original articles (and even some errors on the drawings sold by Polly).
Cheers, Julian

Regret I am late - bin on me yollidizz but might have been me.

The articles are in my webspace as a pdf here.

I will leave them there for about a week for those interested - might be copyright issues?

You mentioned errors - the errors I know of are....

There is a measurement ∅A in bottom LH diagram on page 300, which is not the smaller diameter as illustrated, but the larger diameter

Page 301 - middle drawing of body - 11/16 measurement should be 11/32 I think.

I think the spring references for Lee Spring have changed e.g. LC-026C-3-SS has become LC 026C 03 S or something.

Hope it helps - 'every addition to the knowledge of man is an expansion of the power of mankind'.

All that said, they do work well, the main advantage to me is they close again at a pressure nearer to blow-off and the minimum steam is wasted i.e. the hysteresis is less. My experience is, with Gordon's design set to 100 psi, it will close at above 95, whereas simple designs often close much lower e.g. 90 at best and even then may need a clout.

[steamer5]

Hi Tony,
Many thanks for pointing out the errors!
Page 300 The drawings on pg 301 when compared with the table on pg 300 confirm this.....why didn't I find this
Page 301. Given the outside diameter is 1/2" boring out to 11/16" might make the wall thickness a little negative! Once again why didn't I see this! So your 11/32" is a much better size!
Thanks for the heads up on the spring references numbers

Cheers Kerrin

[spindizzy]

I've made a pair of valves based on the internal dimensions of the tall no. 5 design and I am a bit disappointed with the shut off performance. I purchased the correct springs and balls from Polly, after an unsuccessful use of a 1/4 reamer I bored out the seat diameter to 6.40 to true it up. The ball was shutting off on air ok but not completely sealing. The letting by was much worse on steam. Also when I adjusted it to get exactly 100psi the valves started accumulating past the +10%, backing off the blow off point to 95psi the valve worked and did not accumulate past 100psi. This is ok but the valves were not shutting until about 85 psi. I'm pretty confident that the dimensions and surface finish are to drawing. I did wonder if the ball was to tight in the shroud, I went for a light press fit, should there be a few thou clearance between the ball and shroud? Also is the 0.093 dimension between the bottom of the ball and the edge of the shroud correct? Thanks

[builder01]

I have not made the specific safety valve you are dealing with. But, I have made the Gordon Smith safety valve designed for the Super Simplex. It was easily adjusted to open at about 100 psi and closes at about 95. On the drawing, it looks like the cavity in the shroud, for the ball, is made with an ordinary twist drill. This suggests a fit to the ball that is not tight. No where does the drawing specify any sort of interference fit for the ball to the shroud.

The distance from the end of the ball to the shoulder on the Simplex safety, is shown specifically as .075" to .080". So, I assume your distance of .093" is probably quite important. the bore that the ball covers is pretty important, if it is too large, the shroud will tend to be much closer to the bottom of the bore, it could even touch if you get carried away "seating" the ball. Be gentle.

David

[jma1009]

Hi Ian,

I am going to have to dig out my notes but I had a similar problem on the Gordon Smith squat safety valve I made for Stepney when first tested, and the correction was very simple and quick but at the time required a good deal of study of Gordon's original EIM articles before going into the workshop.

I will have a look tomorrow as day off work and report back.

Cheers,
Julian

[steamer5]

Hi Spindizzy,
At the risk of getting shot down in flames & having not yet made my replacement safety's to Gordon's design here's my 2 cents worth......

Reading Gordon's article by making the hole that little bit bigger will I think do 2 things..
1, allow the ball to sit a little lower & expose a bigger ball area to the steam pressure thus requiring a different spring rate, on page 252 Fig's 6 & 7 show the difference of a larger seat has, the graph on page 252 shows an increase in lift of 0.001" has quite a big difference in pressure change.
2 change the escape angle of the steam, thus change the action of the steam on the shroud & thus the action of the safety valve

My suggestion is to make a new body, or make a 2 part body, as per page 289 ie the seat end(allows you to recover the top part) the only caveat is to make sure that NO thread is above the ball seat. & it looks like a new shroud / ball holder, as David pointed out & Gordon's article (pg 299) shown the ball sits in a center drilled seat, such that the ball fits without being a press fit. That way IF you make the hole too deep you can skim a bit off to get the depth correct, easier than trying to get it spot on! ( measure length of shroud with ball inserted, then measure the shroud without ball gives the amount of ball sticking out, not trying to teach egg sucking here but suggest an easy way to measure the ball protrusion!) from the article it looks like you MAY have to much ball sticking out, & with it a press fit adjusting it isn't going to be easy!

I'm sure if the above is a load of rubbish somebody will correct me & point you in the right direction!

Cheers Kerrin

[robmort]

For those interested a set of drawings of various GS safety valves is available here