Safety-Valves and Steam Accumulation Test Queriess

[Nigel Graham 2Participant @nigelgraham2]

All of us making and using steam engines know that the safety-valves should hold the boiler pressure steady even with the blower hard on, so at maximum steaming rate.

What though, of this problem?

 

Preparing a small and fairly old 7¼” g. locomotive for its certifying after a long lay-up in good storage conditions, I was applying the proper, current MELG tests as a preliminary to the society boiler-tester’s official ministrations.*

On the Steam Accumulation Test (SAT), both safety-valves were open fully at the correct pressures, yet with the blower-valve only partially open the pressure still rose above the specified 88psi (110% W.P. of 80psi), and though this slowed at 90 it was still creeping up above 95. I closed the blower before reaching 100psi.

From the previous certificates either this had not happened before, or they pre-dated the SAT requirement and the testers then were satisfied with the valves opening on and slightly over the line. These certificates and the boiler itself pre-date the “Wussy” (W.S.E.) requirement so that is possible.

 

Hence……

Why should safety-valves that presumably coped with such a test in the past, did not do so today, yet appear to work properly otherwise? (Yes, I had de-scaled them, in citric acid, but without dismantling.)

Or might these valves be slightly undersize for the boiler, so is there in fact any real guide to the passage size for safety-valves, perhaps based on boiler capacity and working-pressure?

 

Or too fierce a blower – would slightly larger nozzles (just drilled holes in a flat disc on this one) soften the draw?

 

What of the coal? This loco was formerly a club portable track workhorse, our members’ engines have become used to rough old rock of perhaps lower calorific and higher impurity values; while some people’s idea of optimum lump size is firehole diameter. Also, the boiler had always worked with tap-water, hard in our area. (My tests used rain-water.) So the boiler might never have been operating at its best, other than when still quite new.

The coal I am using for this was originally obtained by a friend from, I think, an independent Forest of Dean mine quite some years ago, and is a fairly hard grade, perhaps nearer anthracite than God’s Own (Welsh Steam Coal, of course). It burns remarkably cleanly even during steam-raising with a somewhat unscientific electric fan on the chimney. It also created little or no clinker, although my tests were of short duration and driving limited to a few up-and-downs of a test-track all of five yards each way.

 

Or is a combination of a cleaned boiler, rather fierce blower and particularly good coal of a sensible size for the fire, likely to just push the SAT over the edge?

 

I have driven miniature locomotives over many years, I have helped a good few tests, but I have not previously encountered this difficulty with boilers fitted with their designed safety-valves. If anything, safety-valves like to defeat maintaining full pressure!

As well as the immediate problem with this locomotive, it makes me question if I have under-estimated the valves I need for my steam-wagon boiler, as I could only try to compare that with a few published locomotive designs with vaguely similarly sized boilers.

….

*Though the MELG instruction-book might be thought only for engine designers and Club Boiler Testers, I think it should be on all engine-owners’ bookshelves, or at least available in the club-room for members’ benefit.

[duncan webster 1Participant @duncanwebster1]

Crude solution that will work, flatten the blower feed pipe to limit the maximum feed to the blower. Don’t be too enthusiastic at first, it’s a one way street.

If you want to be less brutal, you want smaller jets, not bigger.

[noel shelleyParticipant @noelshelley55608]

The first part of the test should be to check/calibrate the gauge. Is the gauge your using accurate, or has it become tired ? You know the rules ! The club inspector will be an intelligent man and will know what is safe, if slightly out and judge accordingly.  If the problem bothers you then use safety valves with bigger passages or use cheap coal for the test ? Reduce the effectiveness of the blower in some way ? The boiler may well have a design safety factor of 8 or 10 so the risk of a catastrophic shell failure is unlikely and the hydraulic test will go to 1.5 WP, so there is still a margin. Good luck. Noel.

[Nigel Graham 2Participant @nigelgraham2]

Thankyou!

Duncan –

I don’t really want to be that brutal! A restrictor in the line, perhaps in the valve outlet, would be better and can be changed by experiment.

The blower ring does not have discrete jets. It is a cylindrical annulus with tiny holes in its flat top.

 

Noel –

Actually I have just become a club boiler inspector but obviously can’t certify my own boilers!

I have though given it preliminary tests before taking the loco to the club this morning.

Oh woe and thrice woe. Not only did we have leaky clack-valves to contend with, but also the dome flange joint decided to start leaking, and one is just beginning to appear on the superheater’s wet header joint – which with water coming from the drain-cocks also shows the regulator is leaking .

This all by hydraulic test. There was no point trying to steam the engine, nor in writing a certificate with “FAIL” across it.

So a lot of work to do.

The pressure-gauge is slightly out but probably not drastically. Opinion was that enlarging the safety-valve passages is worth trying, including extra holes through the cap-nut, which has only four. Also countersinking their inlet ends.

.

Oh, and the petticoat-pipe fell off. Again. It clearly needs better than a little clip.

These things are sent to try us….

[noel shelleyParticipant @noelshelley55608]

Sorry to hear of your woes Nigel. Good Luck. Noel.

[Nigel Graham 2Participant @nigelgraham2]

Thankyou Noel.

The behaviour I found really worrying was the safety-valves apparently not able to pass sufficient steam.

I have though, not come across much guidance on their proportions, just designs for specific engines.

[noel shelleyParticipant @noelshelley55608]

Within reason open up the body ID, so long as the ball or valve will drop back on to it’s seat , then open up the holes in the cap or increase the number of holes. I’m sure there must be calculations somewhere ? but how do you know the weight or volume of steam you wish to vent. The only thing you know is the pressure ! Noel.

[Martin Johnson 1Participant @martinjohnson1]

To improve the safeties, you need to consult Gordon Smith’s work.  It needs adequate ports and passages but also the right spring in terms of load AND rate or stiffness.

I think a lot of models have blowers that are way too fierce, so a restrictor or re making with smaller nozzles is the way to go.  It will be kinder to the boiler aanyway.

Not all clubs pay attention to the SAT.  One club I was in reasoned that once the valve lifted at the correct pressure an engineman worthy of the term would shut off the blower.  The engine should never be unattended when in steam anyway.  I was actually very happy with the testing undertaken by that club.

Martin

[duncan webster 1Participant @duncanwebster1]

I’d put more holes in the cap first, and make sure the spring isn’t obscuring them at the bottom end. Calculating capacity of a safety valve is not to be undertaken lightly. There was an excellent series in EIM some years ago, and even further back an article by Bill Hall in ME. I have copies of both, if you want them send your email by PM.

[Paul KempParticipant @paulkemp46892]

Nigel,

A few thoughts;

As far as I recollect the Steam Accumulation Test (SAT) has been part of the test code as long as I can remember and certainly before the review to make the code broadly compliant with the PSSR 2000.  The only really significant changes to the code as far as I remember with the addition of the requirement for a Written Scheme of Examination (WSE) was clarification on the annual “steam test” to include a cold examination and requirements for captive valve spindles (later clarified further).  Small boilers and gas tanks are irrelevant to this discussion.  One thing to note on the WSE is the form refers to sections of the code in multiple places and only has proper meaning when read in conjunction with the code.  Therefore, every WSE issued should really be accompanied by a copy of the code as the code is part of the WSE.

Given your boiler has been previously certificated it should at that time have had and passed a SAT.  In which case something has changed.  If you have done no work on the boiler that has changed anything connected with steam production like drafting, grate, smokebox arrangement, details or components for blower or safety valve then it should pass now.  But it doesn’t!  I have come across this situation once with a 5” gauge Maxitrak “Dixie”.  The fix was as you / Duncan discussed, introducing a restrictor in the steam line from the blower valve.  However, there was much procrastination as to why this had become necessary.  Turned out between boiler tests the ashpan had been replaced.  The original was a pressed tin, the thickness of a mouses ear which had rusted through at the retaining tube.  The replacement was more substantial and as the ashpan supports the grate and there was a small difference between the original and the replacement, the grate ended up slightly lower in the firebox than before and the area of the vent allowing air in under the fire was slightly larger.  That was the only explanation we could come up with.  So, have you unwittingly made any small changes in resurrecting the loco that may have had an effect and improved its steaming capability?  New grate perhaps, slightly wider fire bar spacing?  Improved the seal of the smokebox door where previously there was a small air leak?

On the safety valves, if they were previously adequate and nothing has been changed they should still be so, unless they were just on the borderline previously and something else has changed. Is there any possibility the springs have been replaced with a slightly different rate spring since the last SAT?  It doesn’t actually need a very large “hole” in a boiler to let the white stuff out but if the hole is restricted because the valve doesn’t lift much off the seat…..  The other thing to check is when they lift, how far they lift.  That will determine the area of the annulus through which steam can escape.  The relief holes in the top of the body must have a total area larger than the annular area across the valve / seat.  As soon as the steam passes the seat it is expanding so there will be a larger volume at lower pressure to exit the top of the valve.  Spring rate and the annular area above the seat between the valve and the body will determine the hysteresis of the valve, ie lifting to closing boiler pressure.  A lower rate spring will give a larger lift but also give you a lower closing pressure so replacing the springs with a lower rate spring will relieve more steam (assuming the holes in the body can cope) but the valve may not reseat until the pressure has dropped to 50lb in the boiler – not particularly helpful.  So be careful what you do to the safety valves.

Good luck.

Paul.

[Nigel Graham 2Participant @nigelgraham2]

NB: MY PC IS COLLAPSING AROUND ME..

So I may vanish like the Cheshire Cat, unexpectedly. Which might relieve the “mouse”…)

It will need professional attention. It had blocked access to one of my most important spreadsheets (before I could back up its latest edition), and now to ‘Word’ and Excel’ generally.  It also freezes for the Hell of it.

 

To business.

As far as we can determine from its documents, my Steam Accumulation Tests on this boiler were the first.

There is NO mention of them on any of its previous, SFMES, certificates. The boiler was made non-commercially in 2000, by a retired commercial builder of miniature boilers, to replace the original steel one that had expired. He was a Society member so built it as a favour to the Society. So we don’t have any manufacturer’s certificates either: it was tested as being “home-made” by a member, though I am sure he would have carefully tested it to at least the standards of the time before it was installed.

So it is possible the boiler testers were more concerned with the ability of the safety-valves to lift at the right pressure, than with maintaining it as the SAT investigates.

The initial suggestions to countersink the inlet ends of the passages, and add more holes to the cap nuts, would be first to try as they do not affect the mechanical behaviour. They simply give a freer flow to the steam.

…

I have not made any engineering changes that could affect the boiler’s steaming rate – but washing the boiler out and using a particularly good coal might have some bearing on the subject.

The grate and ashpan (which never had a damper) are as fitted in that rebuild; so in theory nothing has changed since the locomotive was last used some eight or ten years ago. Though idle it has spent at least some of those years indoors, snuggled up to a radiator in Winter. (Lucky thing – comfier than its owner…)

””’

Another approach to the blower problem is to limit the control valve opening, e.g. by a spacer on the spindle (if the design allows) or an external stop that acts against the handle. This one has a lever-type handle, not a hand-wheel. These modifications would be simple to make and to change.

As far as I know the two manifold valves do not have captive spindles. (I’ve not tried unscrewing them!)  I don’t think retrofitting them, although wise and perhaps policy in some societies, is called for anywhere. It is an advisory not stipulation, but I could verify that from the orange-on-white book; and obviously to consider if I replace these ones anyway.  Such a valve can readily be given restrictor-spacers to experiment.

I should add that irrespective of locomotive, it has always been my practice to close the blower a little at a time as pressure rises, and at blowing-off to leave it open only just enough to make the fumes go the right way.

On this loco the blower was effective in my tests last week, from only about 20psi on the gauge!

 

…..

So essentially. combing advice on here and from my club’s boiler-testers and other engine-owners, first approach is to smooth the flow of steam through the safety-valves without changing the basic design, and some form of choke or opening-restriction on the blower control.

”””’

Another problem I have to sort is the reluctant injector, and one suggestion passed on by a fellow club-member, is to fit a larger steam-valve and pipe. The theory being an injector wants plenty of steam, not just “strong steam”. Volume as well as pressure, or more likely, minimising pipe losses. I must admit that’s new to me but there could be something in it.

…..

Regarding a club boiler tester thinking you don’t need a SAT because any decent engine-man would close the blower (and damper if fitted)…

That is all very well and of course he should, but it ignores human fallibility such as the driver distracted.

Let’s just say my own Societies (I am in two) would not countenance that sort of nonsense. An incomplete test and record would be wilful negligence potentially voiding the club’s insurance, as well as by which a host club of an event like a rally, could bar track access to a visiting locomotive.

 

[Paul KempParticipant @paulkemp46892]

Nigel,

If your previous certificates are pre the PSSR 2000 review then you won’t final any mention from memory of the SAT on the certificate I don’t have any old certs immediately to hand but I am pretty sure it only said a steam test had been carried out.  The new certs have a line saying the system is safe for a WP and the pressure was not observed to rise more than WP + 10% so they do have a kind or reference but only by referring to the code do you get the method – max firing rate, full blower.

If you can find a copy of the last code prior to the review and the new style certs (I think it was the green book from memory but could well be wrong) I am almost 100% sure you will find reference to checking the safety valves at max firing, rate full blower (or similar text).  So while the SAT was not directly mentioned on the certificate it would still have been done.  It’s a bit like the current certificates not mentioning the water gauge has been tested and shown to be functional, pressure gauge checked for accuracy etc.  That is why the code book underpins the WSE.

You can do a crude check of you safety valve by lifting the valve further once it lifts (carefully with a pair of pliers from the side) if the vent holes are restricting it you will not see / hear a noticeable increase in steam flow.

On captive spindles the first version of the code post review stated “shall” be fitted.  Most locos did not historically comply and at the next review this was changed to “should” giving the inspector some leeway.  You are correct, you do not have to retrofit to captive spindles, you would have been required to do so by the old version though.

Yes injectors do need volume of steam as well as pressure and they also need a free flow of “cold” water.  One problem with tank engines is the water in the tanks gets too warm.  Try feeding the injector from an independent “cold” water tank before you get too carried away, you may find it works fine.  If so, have an independent tank in your driving trolley that feeds the injector.

Best regards,

Paul.

[Nigel Graham 2Participant @nigelgraham2]

This loco has no WSE either, but I think that is a recent development, after the last time it was tested.

It’s no problem to prepare one though, and using a sample form I have written the required details in the correct order to facilitate the boiler tester transcribing it to a real form.

I don’t have copies of the old MELG code to examine. There may be ones in our club-room, though it’s possible they were thrown away to ensure no-one tried using them.

.

With the loco back home, I partially dismantled the two manifold valves to ascertain their types. They need overhauling anyway.

That feeding the injector has a captive spindle created by a flange with its operating thread on its periphery – the same 3/8″ X 32tpi as the rest so very delicate-looking with only two or three full thread turns. Curious way to do it, but it works, albeit with a slower action than I’d like.

The blower-valve, of smaller diameter, is non-captive but has a long thread so needs a lot of unwinding to release. It has a 3/16″ BSW or perhaps 2BA thread, so very small lead, but when I tested the boiler less than half a turn over-drew the fire as I’d observed. A simple restrictor would be an 1/8″ tube between it and the hollow-stay connector.

This captive / non-captive difference is a bit odd, as the turret and valves were clearly made at the same time by the same person.

…

The tanks on this engine have an air-gap from the boiler of no less than 1/4″ at narrowest, open all round (no closure plate between the tanks and boiler on their ends or along the top). So I would not think the water would become very warm. Indeed the injector was side-tank fed on the engine’s original form. As it was a portable-track hauler, used at Summer events, the Sun was likely the tanks’ main heater, not the boiler.

Even so the injector on there now had been plumbed for a driving-trolley tank, and this is how I used it.

 

I know they deal with vastly greater volumes of water, but how do full-size tank locomotives manage, especially the pannier and saddle versions? I have, briefly, driven a full-size narrow gauge saddle-tank locomotive whose injector and feed check-valve are directly on the back-head, and have seen similarly on a vertical-boiler NG engine. On both the injector is expected to lift the water into itself as well.  One of my inherited engineering text-books describes the injector with its example having a single control with very quick-acting thread for both water and steam, boiler-mounted and apparently capable of lifting water from a tank at floor level.

I know injectors do not like water that is too hot, but that is not the only reason for them refusing to inject at times. Yet an injector that needs also pull the water up before pushing it into the boiler will have steam running through it for a very short but appreciable time before the water arrives.

We “cannot scale Nature”, says Received Wisdom – very true, but I can’t help thinking sometimes RW leads to missing the real problem. It’s an intriguing point and I have not seen definitive answers to it.

I am tempted as what was a service turns into restoration to fit this locomotive so its injector (1 now, perhaps 2 in future) has/have the choice of water source, and see what happens.

 

It’s quite a large injector – 40oz/min – intended to be connected with 1/4″ pipes, but there’d be no harm in fitting a larger, and faster-acting, steam-valve and 5/16″ or 8mm pipes, with 1/4″ for the last little bit into the unions. Apart from aesthetics; but although it should look reasonably “right” this loco is a rather freelance version of a freelance design, not a fine-scale rendition of a specific prototype.

[Paul KempParticipant @paulkemp46892]

Nigel,

Do some reading on “lifting injectors”.  Full size injectors are slightly different beasts often with moving or split (along their length) cones to facilitate easy starting or “automatic restarting”.

Paul.

[Nigel Graham 2Participant @nigelgraham2]

Paul –

I am well aware that “full-size” injectors are more complicated than their miniature versions, and good luck to anyone who tries to make a working exhaust-steam injector for a miniature BR Standard-class locomotive! (I doubt very much it would be feasible and I don’t seriously suggest trying. A “simple” injector of near-prototypical external appearance is a different, and laudable, aim.)

However, they still work in the same way! The split combining-cone is cone and starting valve in one component. That would be very hard to follow reliably in miniature, hence the ball-valve instead.

Some injectors on “full-size” plant had single-piece cones and ball or poppet valves; at least in the smaller, simpler versions. They were probably simpler hence cheaper to manufacture and to service.

A miniature injector should lift water, to a limited extent, but is better with gravity feed as was full-size railway and traction-engine practice anyway, with very few exceptions.

 

I had raised this to ask if using the tanks to feed the injector on a miniature tank locomotive is quite the Wrong Thing as we are told.

As built, my own loco, a doubled “Juliet”, was used for much of its earlier career with a combined driving / passenger-truck having no water-tank. Although its axle-driven pump would have been the main feeder on busy portable-track events, I do not recall its side-tank fed injector having problems definitely caused by water temperature. Scale was the main difficulty, in our hard water area.

The chapter on problems in D.A.G. Brown’s book on making these devices, is comprehensive but does not mention feed-water being too hot. It covers dimension errors in the device (sometimes, in commercially made ones), scale or dirt (obviously), and air-leaks. These last can be very devious and hard to trace, such as, he explains, flux residue in a soldered joint eventually dissolving to leave a minute air-passage through the joint. Though the more likely course of air-leaks is on flexible tender to locomotive feed-bags consisting of simple rubber or plastic tube pushed onto the pipe ends.

He also details the prototypical Giffard-pattern injector he made for his Manning-Wardle saddle-tank locomotive. I am confident Mr. Brown would have lagged its boiler very well indeed, but such an installation would still be an eye-brow lifter for many.

 

What I have also picked up from his book is that the 40oz/min injector it sports might be too large (its axle-pump definitely so, for continuous, steady feed). It may repay trying a smaller injector, on the same 1/4″ pipes apart from needing short 3/16″ pipes in the smaller unions. Mr. Brown points to inadequate water, not steam, feed as the main difficulty. In principle it should be possible to use injectors sized to keep a reasonably steady water-level without hogging the steam output.

 

So while obviously I hope the tanks are not a “free” feed-water heater, just what is the truth about tank v. tender feeding injectors on miniature tank locomotives?  Or does it just potentially swap temperature with air-leaks?

[TomfileryParticipant @tomfilery]

Nigel,

Your spreadsheet problem might be caused by Microsoft.

See the link below for more info – though bear in mind that, as it is a welding forum, suggested solutions may not be meticulously detailed.  However, it may give a clue as to what the problem is and help you find a solution.

Regards Tom

https://www.mig-welding.co.uk/forum/threads/more-microsoft-shenanigans.140209/

[Nigel Graham 2Participant @nigelgraham2]

Well, not only not detailed but apparently about Outlook, mainly, but Microsoft’s own none-too-helpful help site revealed other users having problems with Excel after a recent “up-date”.

My PC reached a point that neither Excel nor Word would open at all, even in the proffered “Safe Mode”. Nor did two third-party programmes, though TurboCAD and Alibre still functioned.

Luckily I found a full re-installation repair, and used that. It took over an hour!

This not only unlocked everything. It let me find some lost photos – MS had moved them to OneDrive without telling me. It also allowed me to open the damaged spreadsheet in Safe Mode, re-save it under a new name and copy it to external storage. The message said it was severely damaged so formulas [sic] might not work, but it is a pure index, without any formulae written by me in it. Another spreadsheet that does contain some simple formulae and graphs, still seems to operate properly.

 

======

 

Regarding safety-valves, closely examining of of mine showed the exit holes in the top are partially eclipsed by the spring, and I wonder if that is a common problem with these.

[Author]

Posts