Society Gauge Widening Tool

[grovenor-2685]

Les I have thought the same. I will try to summarize the thread on one side of A4 for here and/or the Snooze.
But the salient points are briefly:
1.It is a question - why does the Society Triangular Gauge give only half the gauge widening on curves that the prototype would use?

Because the gauge designer thought that was enough, and it has proven to be so for many years now.
Note: Models are not the same as the prototype in this respect, first the P4 standards have compromises in wheelset back to back and checkgauge that increase clearances to allow for curves of less than scale radius, secondly we can build sideplay into our loco chassis to a greater extent than the prototype, these factors mean that strictly prototypical gauge widening is not essential. Modellers using the more exact S4 variant of the standards or indeed any modellers may wish to use the prototypical figures of course.

2. How long would a gauge be that gave prototypical Gauge Widening
3. Alan Turner explained how gauge widening is a stepped process on the prototype, which a triangular tool cannot replicate. He gave several options presenting them in graph form, showing how little the Society gauge gives, and that a tool giving full widening at 5.5 chains will then give you too much at lower radii.
4. My son produced a simplified graph and all the numbers represented by that graph.

Points 3 and 4 answer 2.

5. I realised that the question needed to be reframed more precisely as 'why does the Standard say 528mm is the radius at which maximum permitted widening occurs.'

That can only really be answered by Russ Elliott who seems to have introduced it into the standard when he revised it. It was not part of the original standard and is not really relevant.

6. I asked this question as a riddle in a new topic and FINALLY began to get answers that made sense from Martin Wynne, whose absence from the thread had surprised me.
7. Martin established that the 528 came from BRSMB standards 60+ years old.

I don't think Martin established any causal link between these two references to 528mm, it may just be co-incidence.

8. He established in the early hours of this morning that a 47mm length tool could be both practical and prototypical (I think!)

As also covered in points 3 and 4 and in the new series from Will.

9. The rest of the thread is taken by very many various misunderstandings, clarifications, perhaps some obfuscation, some resistance to questions being asked, and now denial of the obvious, that the standard needs updating.

I don't think it at all obvious that the standard needs updating. Following the standard works just fine and I have seen no report of any specific problem in following the standard. It might be better to remove the reference to 528mm radius as it is really irrerevant but as that does not affect use of the standard its not of itself enough to justify an update.

10. However, a lot of the thread is taken up by what remains a perennial question, whether the prototype stepped function can ever be adequately replicated by a triangular tool that gives a continuously varying function, and the difficulty of teasing out this issue in the written word.

If 'perennial' means that Julian keeps on about it Surely it is self evident that if you want to model the prototype stepped GW then you copy it by using stepped gauges. Such devotion implies that equal accuracy will be demanded fir all the rest of the wheel/track interface.
Regards

[grovenor-2685]

As Richard Chown said earlier. 1432 wasn't a success and it would be fascinating and delightfully irrelevant to the thread to know why. He says there that bogie hunting increased.

I have a paper somewhere on that exact topic, only in printed form though, I'll try and scan it one of these days.
Regards

[grovenor-2685]

The thread has highlighted that the triangular gauge works well for ply and rivet, for which it was designed, but if used incorrectly on a curve can cause the track to be under-gauge. Used correctly, it gives a small degree of gauge widening on curves, but if used with moulded chairs, it can again, result in under-gauge track.

from what I have heard there are triangular gauges and triangular gauges! Mine, which are orinal Studiolith versions only have slots deep enough to hold the rail head to gauge, they do not give any problems with moulded chairs as the rail can inclne as required. I have heard of, but not seen, gauges where the slots are so deep that the foot of the rail is held preventing use on inclined rail. This would cause the mentioned effect with moulded chairs.

Rather than demand a "correct-length-for-gauge-widening tool", would in not make more sense to look for a new design which accommodates inclined moulded chairs? It could be made at a length which creates appropriate widening for (say) 1.1 metre radius which could be a reasonable compromise. Anyone wishing greater verisimilitude could, as now, use individual gauges; in steps, per prototype.

If your gauge has the deep slots and you want to use it with moulded chairs just file it down until the slots are of an appropriate depth, it would only take a few minutes and would not affect the accuracy as none of the filing would be on a gauging face.
Regards

[Julian Roberts]

I agree that the standards document may need an update, primarily to correct the 528 anomaly, but a major re-write is surely OTT?

Hi Les

I've had no time since Tuesday to reply, that the 528 is the only figure in all the standards that I was/am suggesting might be updated. Obviously that would depend on others' agreement which is (equally obviously) in short supply...err somewhat less than that....

[grovenor-2685]

As Richard Chown said earlier. 1432 wasn't a success and it would be fascinating and delightfully irrelevant to the thread to know why. He says there that bogie hunting increased.

I have a paper somewhere on that exact topic, only in printed form though, I'll try and scan it one of these days.
Regards

Paper found and scanned. I can email to anyone who wants it. File comes out to 3.5Mb.
Paper was written for the ICE by Richard Harvey of AEA Technology and came to me via HMRI in the course of work.
Regards

[Russ Elliott]

Oh, gawd. I think I might have to read this thread.

[grovenor-2685]

Oh, gawd. I think I might have to read this thread.

Yes indeed, we have been missing you
Regards

[Philip Hall]

Russ, We will be missing you for quite a bit longer if you read all this from the beginning...

Philip

[Julian Roberts]

May I apologize that there were several posts I did not reply to over the last few pages. This was due to lack of time. I was unable to get my head round some of Will's, to reply properly, I hope he did not take any playfulness of mine as an insult. Summarizing the thread properly in my own words will take more time than I have just now. The very brief synopsis quoted by Keith above neglected to acknowledge the help he has given.

That can only really be answered by Russ Elliott who seems to have introduced it into the standard when he revised it. It was not part of the original standard and is not really relevant.

Keith, This is covered fairly broadly in this thread while you were on holiday - the original Protofour document in August 1966 had no mention of from what radius the GW standard (a maximum of 0.22mm, obviously the prototype minus 12%) applies. The idea of a P4 minimum radius equivalent to EM appears in the second instalment of that document in September '66, and by the Jan '67 article the "at 528mm" has been introduced, albeit as a footnote. Thanks to your website I noticed this. At 1250 a.m hopefully I am remembering these facts correctly, but they are all on this thread:

528mm is the radius at which a 31mm P4 triangular gauge gives 0.22mm widening. The EM gauge triangular tool is 31mm long, but on EM gauge that gives a 0.2mm GW at 620mm, which is the EMGS GW standard.

I don't think Martin established any causal link between these two references to 528mm, it may just be co-incidence.

Yes, I think that was all that Martin was implying - but isn't the mystery why the MRSG went along with it, and the above coincidences? They had thrown out all the other standards and then breathed new life into this one, and here we are 50 years later still talking about it. Maybe Russ knows...

[Martin Wynne]

They had thrown out all the other standards and then breathed new life into this one, and here we are 50 years later still talking about it. Maybe Russ knows...

Hi Julian,

Before deciding that 528mm is a silly minimum radius, it would be worth deciding what "minimum radius" actually means?

On the Great Western main line?

On a South Wales colliery branch?

On the sidings round the back of the gasworks?

If the latter, 528mm might not be so silly. i.e. the absolute minimum for any successful use of P4.

regards,

Martin.

[Russ Elliott]

Maybe Russ knows...

I sent you an e-mail a couple of day's ago, Julian.

[Julian Roberts]

Hi Martin

Thanks for your post just now. Keith wrote as below about a fortnight ago replying to stuff of mine

I am asking if more of us less good modellers may be able to get good results more easily if we junk the trainset minimum radius that is -?- if I am right? - an integral part of the gauge widening standard....? Isn't a minimum of 3' 3" small enough nowadays?

I don't recall anyone proposing train set radii for P4! General recommendation is usually 4ft, with a bit of care you can manage with less, I have a few bits down to 3'6". Tin Venton on Clutton, I believe uses 3ft.

Hi Russ

I have just replied to the email thanking you

[Martin Wynne]

Martin
Keith wrote the above about a fortnight ago replying to stuff of mine

Yes, but for what rolling stock? Just saying this person uses this radius or that doesn't mean anything unless you know what rolling stock is running on it.

Here is some data for industrial turnouts, used by short 4-wheel wagons and 0-4-0 shunters:



You can see that the smallest turnout has a radius of just 12.192 metres radius. That's a daft metric conversion, it is actually 40 feet radius.

At 4mm/ft that is just 160mm radius -- a long way below 528mm.

regards,

Martin.

[Russ Elliott]

Crikey, what a thread!

The origin of the '528mm' is the first 'mature' P4 standard, as expressed in January 1967 MRC. I was a gullible teenager when I first immersed myself in the mysteries of the 'P4 standards', and believed everything (well, almost) the MRSG gurus and gospel issued. This 528mm value was merely repeated when issue 2 of the P4 Track and Wheel Standards appeared in 1998 (although it had been put to bed in committee many years earlier, '84 I think). I didn't question it, and no one batted an eyelid about GW at that time. As Martin has pointed out, the '528mm' seems to have been established in folklore sometime before 1967. I was utterly unexcited about the '528mm' at the time, and still am. There were more important political issues on the standard at the time.

But let's roll the clock back a bit.

The late Malcolm Cross was obsessed with curves, and he did some pioneering work in introducing the modelling public to the subject. I can imagine Malcolm's dream was maybe to present Cyril 'Mr Doubting Thomas Number One' Freezer with an 8' x 6' tabletop layout in P4, just for the hell of it, to prove that it could be done. 6-wheel coaches to boot.

But it is also evident there must have been considerable ambivalence within the MRSG over the subject - with a BB on the conservative side (as we now generally regard it), and a CF larger than the prototype equivalent, would there need to be much stress put on the implementation of GW? There were commercial arguments as well: producing a triangular gauge 50-or-so mm long was always going to be a lot more expensive than a 30mm one. Joe Brook Smith was not obsessed with popular evangelism like Malcolm was. And the commercial reality of the nascent P4 era dictated that multiple or different types of constructional TG gauge would not be economic, and might be perceived as too confusing to the potential P4 punter.

The result was a compromise length. It was a compromise arrived at by fallible individuals with divergent visions of what P4 could and should be. The compromise gauge length was a balance between, on one hand, giving some GW on curves, and on the other hand, constraining the amount of GW at crossing work, viz the Digest's hallowed words of GW "should not be applied to any sections of pointwork where CG, CF or BC dimensions are specified". I think the key word there is 'should', and not 'must', because the same gauge was being promoted for constructional use in the vicinity of crossing work. Accordingly, to compound the compromised mess, Joe knew, and had known for a long time, it was necessary to introduce a physical gauge for CG to get the check rail in the right place! Everyone was equally offended.

We live with the baggage of this messy history. Yes, everyone genuinely wanted a horse, but ended up designing a camel.

[Paul Townsend]

We live with the baggage of this messy history. Yes, everyone genuinely wanted a horse, but ended up designing a camel.

I am glad you have got some mojo back, we missed you.

Quote from " Brief history of camels":
Camels have been domestic animals for thousands of years.

I suspect they have served mankind for longer than horses so thats OK then!

[DavidM]

Very eloquently said, Russ.

I too am glad to see you back!

David

[Martin Wynne]

it was necessary to introduce the notion of CG (completely alien to the prototype, btw)

Hi Russ,

Welcome back.

I don't want to disagree with you when you have only been back 5 minutes, but I am going to anyway.

Your above statement is just plain wrong.

Check Gauge is a well-known and fundamental concept on the prototype. Here's an extract from BRT3:

check_gauge_brt3.png

regards,

Martin.

[Russ Elliott]

Your above statement is just plain wrong.
Check Gauge is a well-known and fundamental concept on the prototype. Here's an extract from BRT3:

Yes, Martin. Sorry, I didn't make myself clear. What I meant was to compare the GW-inducing triangular artefact setting TG and the physical artefact (a constructional gauge) that set CG, which to my knowledge the prototype has never used, which preferred (where GW was felt necessary) to use special chairs. That was the contradiction I was trying to point out. I'll edit my post later. It was a bit late at night!!

[grovenor-2685]

The origin of the '528mm' is the first 'mature' P4 standard, as expressed in January 1967 MRC.

Ah, missed that one, sorry for attributing it to Russ then. I do note that it did not get a mentuion in the official P4 Manual published in 1970.
I suspect that Julian is right in that it was related to the triangular gauge dimensional compromise as explained by Russ. All surmise now as we can't ask Joe Brook Smith anymore.

I would note that GW does not feature in any of the 6 governing equations given in the standard, it is essentially just an arbitrary figure without explanation. Perhaps an equation would help establish the standard without reference to a particular design of gauge.
GWmax = (EFmin+BBmin+TWmin)-TGmin-OL where OL is the minimum acceptable width of tread remaining on the rail (another arbitrary number!)
Hence GWmax = (0.35+17.67+1.85)-18.83-OL = 1.04-OL.
To arrive at GWmax = 0.22 then OL would need to be 0.82 which is not unreasonable given a railhead width of 0.92
Choosing an OL of 0.79 would give a GWmax of 0.25 and still has a reasonable amount of tread on the rail.

All comes down to judgement!
Regards

[Russ Elliott]

Choosing an OL of 0.79 would give a GWmax of 0.25 and still has a reasonable amount of tread on the rail.

I think one man's OL is another man's 'oh hell'. Here's an Exactoscale skinny wheel (TW 1.67mm, and yet to be made an honest woman in our standards) sitting on some 19.05mm TG:

twmin.png

(Btw, has the forum search facility disappeared? And can it detect attachment filenames? I'm trying to find the thread I first posted the above sketch.)

[Julian Roberts]

Russ, Martin

You had the same conversation on 5 April 2015! referring back to 2012! But obviously a lot has happened in the 12" to the foot world since then.

viewtopic.php?f=5&t=2021&hilit=gauge+widening+at+the+crossing&start=25


Martin's post on the same page on 6 April illustrates the "Gap D", the checkrail flangeway, as :

D is the check rail gap. The width of this gap doesn't matter a damn, providing it is wider than the wheel flanges. It's whatever you end up with after setting A and C correctly. But where the check rail is combined with a wing rail in complex formations (i.e. in parallel-wing V-crossings) it must be the same as B. This means that it is not possible to have gauge widening through such formations, such as a tandem turnout.

In this context I have never understood the Scalefour Digest bit that you quote Russ, but as you say these are hallowed words I ask with trepidation

constraining the amount of GW at crossing work, viz the Digest's hallowed words of GW "should not be applied to any sections of pointwork where CG, CF or BC dimensions are specified".

Surely those values are constant but unaffected were GW to be introduced?

I waded/skimmed through the entire Network Rail Track Design Handbook document Martin kindly posted here last week (Aug 23) on track.

In all 208 pages there are just 9 lines of text given to the subject of GW, and the interesting one says

"Gauge Widening must not be applied to S&C without permission of the Head of Track Engineering at Network Rail HQ"

- in other words (presumably), something done in very exceptional circumstances. But maybe an everyday event in the model world where tight radii are the norm?

[John McAleely]

(Btw, has the forum search facility disappeared? And can it detect attachment filenames? I'm trying to find the thread I first posted the above sketch.)

Nope, top right, most forum pages (as part of the header graphic on most screens).

Typing in twmin as a term popped out:

viewtopic.php?p=3073#p3073

[Russ Elliott]

Thank, John. Silly me! (My brain still isn't working yet.)

[grovenor-2685]

I think one man's OL is another man's 'oh hell'. Here's an Exactoscale skinny wheel (TW 1.67mm, and yet to be made an honest woman in our standards) sitting on some 19.05mm TG:

I almost added a comment to be careful if using wheels that are non-compliant on the skinny side! But left it out as the discussion is now about the standards* rather than the effect of non-compliance.
Clearly, if you reduce the TWmin without changing other values a sensible value for OL (OverLap) will result in a reduced GWmax.
This does demonstrate that 'prototype' GW values really imply 'prototype' BB/CG values as I have said several times.
Regards

*Despite what the topic title says.

[Martin Wynne]

"Gauge Widening must not be applied to S&C without permission of the Head of Track Engineering at Network Rail HQ"

- in other words (presumably), something done in very exceptional circumstances. But maybe an everyday event in the model world where tight radii are the norm?

And also maybe in the steam-era world of bullhead track and local p.w. gangs?

In those days there really were sidings round the back of the gasworks. Does the "Head of Track Engineering at Network Rail HQ" know what a gasworks is?

The modern railway is a very different world from the one most of us model.

Martin.