Stephensons Valve Gear

[Phil H1Participant @philh196021]

I am currently building a 3 1/2" locomotive with Stephenson's valve gear and I am just a bit curious about the weighshaft and more specifically its bearings.

The bearings seem to be very well engineered (bronze bearings fitted into the 1/8" steel frames etc). I can understand the need to keep the mechanism nice and snug but do these bearings really see much wear?

Phil H

[Phil H1Participant @philh196021]

[Nick Clarke 3Participant @nickclarke3]

Posted by Phil H1 on 26/09/2019 17:00:44:

I am currently building a 3 1/2" locomotive with Stephenson's valve gear and I am just a bit curious about the weighshaft and more specifically its bearings.

The bearings seem to be very well engineered (bronze bearings fitted into the 1/8" steel frames etc). I can understand the need to keep the mechanism nice and snug but do these bearings really see much wear?

Phil H

Yes – not in the rotational sense so much as fore and aft with every stroke.

[Phil H1Participant @philh196021]

Thanks Nick,

The rest of the motion (pivots on the eccentric rods etc) recommends steel to steel contact with case hardened ends and simple steel pivots. How does that hold up? For example, a bit of watch making would be required but small bearings could be fitted.

Phil H

[Nigel Graham 2Participant @nigelgraham2]

There can be a small but oscillating load on the weigh-shaft and its bearings, so the shaft does need those bearings.

If the shaft is simply held in plain holes in the frame, giving a very narrow bearing surface, it will in time wear the holes oval and cut an eccentric groove in the shaft, worsening the situation and probably losing valve travel and timing accuracy.

You could add small gun-metal bushes in places like the eccentric rods.

I may be wrong but I have an idea phosphor-bronze is inappropriate for unhardened mild-steel pins or shafts: the steel will wear more than the bronze.

[julian atkinsParticipant @julianatkins58923]

Hello Phil H,

The weighshaft bearings do wear. In Stephensons valve gear they get quite a 'kick', which can be transmitted back to the reverser, and all the bearings etc need to be jolly good fits and with good alignment

I 'beef up' the weighshaft bearings.

Cheers,

Julian

[Paul MParticipant @paulm98238]

I am building a 3 1/2" loco to a LBSC design. As a novice I have to do a lot of research to find out appropriate materials for constructing the loco ,as the drawings supplied by Reeves, have no reference to materials.

I am about to start work on the Stevensons valve gear and would appreciate some guidance regarding materials selection or the best reference books to help.

[SillyOldDufferModerator @sillyoldduffer]

Posted by Paul M on 28/10/2019 08:43:43:

I am building a 3 1/2" loco to a LBSC design. As a novice I have to do a lot of research to find out appropriate materials for constructing the loco ,as the drawings supplied by Reeves, have no reference to materials.

I am about to start work on the Stevensons valve gear and would appreciate some guidance regarding materials selection or the best reference books to help.

Not an engine builder myself, and am prepared to have my views questioned. But they might attract better answers!

Generally speaking:

• Boiler and Pipe work to be made of Copper, not Brass. This is because steam and hot-water slowly dissolve the Zinc in Brass leaving it weak and porous. Steel boilers are difficult to make safely and to the satisfaction of a boiler inspector.
• Most fittings will be Brass – injectors, whistle, safety valve, etc.
• Mild-steel for all structural work. EN2, 5 or 8.
  • Bright Mild Steel is strongest and comes with pre-finished surfaces but it tends to warp when machined and is more expensive. (Not a major problem in my experience, but it does happen.)
  • Black Mild-steel is slightly weaker, unfinished, covered in mill-scale, and cheap. It doesn't warp when machined. May need to be pickled.
  • Both Bright and Black can be had as 'free-cutting' alloys – (EN1A / EN1B) to make them easier to machine. Bit weaker and more expensive, but saves time and bother when turning and milling. Worthwhile.
• Anything made from a casting will be cast-iron. Cast-iron isn't particularly strong in tension but it has the property of being slightly self-lubricating, making it attractive for steam-cylinders and pump-bodies. Used for complex shapes easily reproduced in a mould, often saving masses of time compared with fabricating the same object from metal stock.
• Bronze is best for bearings, but brass, or cast-iron can be used. Bronze is preferred for bearings because it's tougher and lasts longer, but it's more difficult to machine than brass, and costly. I tend to use Brass for lightly loaded bearings and if it's easy to change the bearing. Heavy loadings and being difficult to replace make Bronze worth the extra effort.
• Aluminium – probably not much use on a steam loco. It's strong and light, but reducing the weight of a loco tends to reduce track adhesion causing the wheels to slip.
• High Tensile Steels like EN16 are stronger than Mild-Steel, and can heat treated to harden or toughen the metal. Good for strong and hard-wearing parts like axles, bolts and gear-wheels. Somewhat harder to machine than mild-steel, and a few are downright unsuitable. I use them only when mild-steel won't do a good job.
• Alloy Steels are available for a wide range different purposes, for example Stainless Steels resist corrosion. Not sure any are useful on an LBSC loco.

For a Stephenson Gear, it might be worth using EN8 (a strong mild-steel) rather than EN1 to resist buckling. Or a heat-treatable steel to reduce joint wear by hardening the metal. Not sure it's worth the trouble myself. EN1 isn't exactly weedy, and can be case-hardened if need be. As the strength of a mild-steel member can be increased simply by beefing up its dimensions, I feel the need for a high-tensile steel like EN16 only comes into play when it's desirable to keep the weight down for performance reasons, or because a part has to fit within a limited space, or because a hefty lump of metal spoils the look of the model. My guess is LBSC mostly used mild-steel, but don't be surprised if an engine builder says I'm wrong!

I recommend getting a book! There are thousands of different steels, dozens of different Brasses, perhaps 30 different cast-irons, and 40 odd Bronzes. Tubal Cain's 'Model Engineer's Handbook' helps focus on useful facts and materials.

Dave

[Phil H1Participant @philh196021]

Dave,

Great input and very helpful but I must issue a warning about distortion on steel.

Black mild steel very definitely distorts when machined. It might be less likely to distort than BMS but it sure does suffer depending on the geometry. However, I have never found an issue with restraightening black or bright mild steel.

Phil H

[Anonymous]

Posted by Phil H1 on 28/10/2019 14:24:05:

Black mild steel very definitely distorts when machined.

That's interesting; I've machined a lot of hot rolled steel and never noticed a problem with distortion. How did the problem manifest itself?

Andrew

[Phil H1Participant @philh196021]

Andrew,

Yes I am also puzzled because I see the same information about black mild steel being so stable on this site over and over again but it is not my experience.

Typically, it has happened when the edge of a rectangular bar has been skimmed (along its edge) or the face of a plate has been skimmed – almost exactly what you might expect of BMS. Another case was when I was trying to get 'clean' bright steel by filing a face of a 1" wide by 1/8" thick piece of black steel about 6" long. I achieved a nice clean shiny surface but it was banana shaped by about 1/16".

My point is that distortion can occur – it is not completely stable.

Phil H

[duncan webster 1Participant @duncanwebster1]

Posted by SillyOldDuffer on 28/10/2019 12:16:40:

Not an engine builder myself, and am prepared to have my views questioned. But they might attract better answers!

………

For a Stephenson Gear, it might be worth using EN8 (a strong mild-steel) rather than EN1 to resist buckling.

Dave

buckling is governed by section shape/size and young's modulus, not by yield strength, and just to be pedantic (who me) I would call EN8 a carbon steel

[Anonymous]

I tend to use much larger sections, so may be they're less prone to distortion. When I made some press tooling for the flypress I skimmed the surfaces of hot rolled steel to form the blanks:

After the curved surfaces were CNC milled the tooling fitted together very well; certainly less than a thou or two gap:

But these blocks are about 5" x 2.5" x 3/4".

Andrew

[Nigel Graham 2Participant @nigelgraham2]

Paul M:-

I would not become too hung up on steel grades for the well-established designs most of us work to. By all means use the best you can and indeed the more common grades the professional designer might pick; but it is highly unlikely the majority of the many locos built to designs by LBSC< Evans, etc. have many "special" steels in them, (quite possibly including the designers' original example!); and I have not seen many published model-engineering drawings that actually specify materials to that degree anyway.

Low-carbon (mild) Steel:

EN1A is free-cutting, and suitable for most lightly-stressed components unless having to be hardened. The components in a 3.5"g loco's motion work are small but the forces in them are not all that large either.

Unlikely on a miniature loco but if you fabricate anything by welding, do not use EN1A for it. It will seem to weld all right but the joint is likely to be brittle.

EN3A is a bit stronger but still plain mild-steel – I'd be happy making motion parts from it. And it is weldable.

EN8 is of the medium tensile strength grades – ideal for axles, studs carrying (relatively) heavy loads, etc.

BTW I believe officially the standards have changed from EN-number, but I don't have the equivalents to hand.

Silver-steel –

– (if you can be sure it actually is the precision-ground oil-hardening steel it's meant to be) is for pins etc. that need to be hardened and tempered – and as a tip, lead kept just molten is about right for the tempering colour of blue-purple.

Also used for making small special tools like D-bits for cutting valve seats.

Gauge-plate, or oil-hardening ground flat stock,

– can be thought of as the flat bar version of silver steel and is often recommended for expansion-links as it is hard-wearing. It's often used for making form-tools, especially for turning fancy bits of brass-work like globe-valve bodies.

To get the best out of these alloys they really need to be heat-treated as per the manufacturers' instructions but I am sure most of us get perfectly serviceable results without precision muffle-furnaces and molten-salt baths!

Stainless-steels –

I would respectfully disagree with SillyOldDuffer. There are plenty of applications on an LBSC design for stainless-steel of appropriate type. LBSC, Martin Evans et al tended to call it "rustless steel".

The generic name is of a bewildering range but "our" stockists generally sell the free-cutting (with due care on tooling) varieties, and you use this where rusting would otherwise be a problem: piston and valve rods, screw-down valve spindles, etc.

Chosen with care, it has a higher tensile strength than mild steel, which combined with its corrosion resistance makes it ideal for mounting-studs on boiler fittings and the cylinder and valve-chest covers.

It can be awkward to machine (depending on grade) but if approached with due care can give beautiful results. The golden rule is the tool must cut not rub: abrasion with a mis-set or blunt tool will work-harden the surface of some stainless-steels very rapidly.

Precision-ground stainless-steel is good for bearing pins though it can't be hardened – but as with silver-steel or precision-ground mild-steel, the advantage is lost if you go and turn the stock diameter down to the bearing diameter, as on so many of these old designs!

Brass / Bronze:

For bushes carrying non-hardened steel pins & shafts: gunmetal or leaded bronze (or in fact cast-iron). Them or Phosphor-bronze for hardened ditto.

Boiler bushes must be of gun-metal or phosphor-bronze NOT brass, but brass is OK for fittings like clack-valve bodies. You can use the two bronzes for the fittings too, of course, but plenty of builders do use brass for those, without undue problems.

Boilers:

Again with due respect to Dave, I cannot let his " Steel boilers are difficult to make safely and to the satisfaction of a boiler inspector." go unchallenged, or rather, un-amplified!

It is safe to say we cannot now build steel boilers without being coded welders and using certified materials, with all the entailed rigmarole and costs. We don't want dissatisfied boiler inspectors!

We can build copper boilers – but do NOT use brass even for fittings bushes. Those must be phosphor-bronze or gun-metal. Your club's boiler admirer may wish to view progress to ensure soundness of what will be hidden parts, and ought be able to advise you, too.

NB: if you do build the boiler you must put a lot of heat into the copper, to raise the metal to silver-soldering temperature as quickly as possible, and to complete the joint quickly and smoothly without rushing it. Poor and over-long heating risks ruining the flux and even dissociating the silver solder. The boiler inspector won't be happy – nor will you! (After seeing some forlorn examples, I had cold feet rather than hot metal and contracted my project to Western Steam!)

As SOD (Dave) says, there is plenty of literature available to help you. He cites a few books, also peruse the TEE Publishing catalogue – they sell by post and electronic mail order, and usually have a stand at the major exhibitions. Most of my own model-engineering library is from them.

I hope all that helps – we look forwards to progress reports!

[Paul MParticipant @paulm98238]

Many thanks to all who have posted such informative and helpful responses on this topic.

It has certainly given me plenty of help and steered me in the right direction.

[SillyOldDufferModerator @sillyoldduffer]

Posted by Paul M on 29/10/2019 08:55:30:

Many thanks to all who have posted such informative and helpful responses on this topic.

It has certainly given me plenty of help and steered me in the right direction.

Me too! This forum is most educational, an excellent source of expertise and clarification.

Thanks to everyone for clearing up my misunderstandings. It all helps.

Cheers,

Dave

[Nick Clarke 3Participant @nickclarke3]

Posted by Nigel Graham 2 on 28/10/2019 23:34:08:

It is safe to say we cannot now build steel boilers without being coded welders and using certified materials, with all the entailed rigmarole and costs. We don't want dissatisfied boiler inspectors!

Like Duncan above, never let it be said that I am at all pedantic, but however without differing at a practical level, the boiler code does not quite say this.

Para 6.2 says that "If welding is undertaken by anyone who is not a coded welder the inspector shall require that weld samples be made available for inspection and testing prior to the commencement of the welding of the boiler, or that the welder should have proof of test pieces being satisfactorily tested within the twelve months prior to the jointing being undertaken. Weld samples shall be tested by appropriate testing laboratories."

In practical terms it would probably be beyond the reach of most of us to meet the conditions required, but also note that steel is not mentioned and with TIG machines available on the High Street, complying with the code would imply that this applies to TIG welded copper too.

 

Edited By Nick Clarke 3 on 29/10/2019 12:18:19

[Nigel Graham 2Participant @nigelgraham2]

Yes, you are right, Nick, but my point was as you say by "in practical terms". Basically the whole lot is out of hand for anyone but a full-time welder with the appropriate certificates and experience. And what does it cost to have the samples tested? Possibly not much – if you are the Accounts Manager for a big company that can share such overheads around its invoices.

Interesting point about welded copper boilers. I know there are one or two professional builders of welded-copper miniature boilers. I don't know if the copper has to be traceable – possibly not for an amateur but quite likely for the professional, especially if his boiler-making is part of a wider range of trade work anyway. It does have to be de-oxidised grade for welding, which may mean traceability and conformity being de facto.

I don't think the MELG Code has anything on this.

Incidentally the original Pressure Equipment Regulations – the document by and for lawyers for whom a 'file' is a ream or two of printed paper – did mention two, and only two materials: stainless-steel and aluminium-alloy. I am not sure which of these would give our lot the more kittens if suggested as boiler materials! They also specify a lower minimum test factor than we use: a rather odd 1.4 WP, if I remember correctly. (Nothing to stop anyone specifying higher, of course.)

[Author]

Posts