Society Gauge Widening Tool

Discuss the prototype and how to model it.
User avatar
Julian Roberts
Posts: 1388
Joined: Sat Jan 09, 2010 11:33 pm

Re: Society Gauge Widening Tool

Postby Julian Roberts » Thu Apr 28, 2016 6:51 am

Very many thanks Keith. Plenty to digest here.

I wasn't quite clear. While different companies at different times may have done things in different ways in detail and maybe at different rates, was there in general, i.e. as a matter of principle, anything optional about gauge widening on curves, or was it something that was always done?

Terry Bendall
Forum Team
Posts: 2416
Joined: Sun Jul 27, 2008 7:46 am

Re: Society Gauge Widening Tool

Postby Terry Bendall » Thu Apr 28, 2016 7:12 am

John Palmer wrote:he set out to show that a mint gauge capable of gauging the 4mm equivalent of 0.75” maximum GW would need to be 54mm long, but that such a gauge would yield excessive GW in all other cases.I don’t doubt that this is so, but surely the graphic proof of such an argument is somewhat academic since there has been no proposal to produce a mint gauge that is 54mm in length, the only such gauges so far actually produced being either 35mm or 40mm in length (my new one is 40mm long – I’m neither bragging nor complaining).


It has already been pointed out that the mint gauge was not designed or intended to be used as a means of setting gauge widening.

Terry Bendall

andrewnummelin
Posts: 374
Joined: Mon Jul 21, 2008 9:43 am

Re: Society Gauge Widening Tool

Postby andrewnummelin » Thu Apr 28, 2016 7:44 am

grovenor-2685 wrote:
...
Gauge_widening.pdf

Regards

Nice numbers here!
Note the plain track gauge - if you model modern image and use P4 standards you already have some gauge widening.
(I remember learning over half a century ago that London Underground and British Railways used different gauges. I've often wondered why/when there was a reduction of 1/8" - it certainly does not correspond to a rounding of metric dimensions.)
The tables imply step changes in gauge but no indication is given over what distances these are made? Was it always from one sleeper to the next?
Regards,

Andrew Nummelin

User avatar
LesGros
Posts: 546
Joined: Thu Oct 08, 2009 10:05 pm

Re: Society Gauge Widening Tool

Postby LesGros » Thu Apr 28, 2016 8:52 am

Andrew nummelin wrote:
...The tables imply step changes in gauge but no indication is given over what distances these are made? Was it always from one sleeper to the next?

Highly unlikely I would have thought; Track is laid to smooth the way, hence transition curves from the straight. Gauge widening, if applied over the distance between sleepers, would probably feel more like a wobble than a transition.
LesG

The man who never made a mistake
never made anything useful

User avatar
grovenor-2685
Forum Team
Posts: 3917
Joined: Sun Jun 29, 2008 8:02 pm

Re: Society Gauge Widening Tool

Postby grovenor-2685 » Thu Apr 28, 2016 11:32 am

Bear in mind that, for bullhead track you use different chairs for each step in the gauge widening, there can be no intermediate values.
Regards
Regards
Keith
Grovenor Sidings

John Palmer
Posts: 825
Joined: Fri Jul 15, 2011 11:09 pm

Re: Society Gauge Widening Tool

Postby John Palmer » Thu Apr 28, 2016 1:33 pm

Terry Bendall wrote:
John Palmer wrote:he set out to show that a mint gauge capable of gauging the 4mm equivalent of 0.75” maximum GW would need to be 54mm long, but that such a gauge would yield excessive GW in all other cases.I don’t doubt that this is so, but surely the graphic proof of such an argument is somewhat academic since there has been no proposal to produce a mint gauge that is 54mm in length, the only such gauges so far actually produced being either 35mm or 40mm in length (my new one is 40mm long – I’m neither bragging nor complaining).


It has already been pointed out that the mint gauge was not designed or intended to be used as a means of setting gauge widening.

Terry Bendall

Terry, I didn't say that I envisaged using the mint gauge to set gauge widening at the construction stage. My intention was to use it as a diagnostic tool in the way described by Paul Townsend. What I am trying to establish is whether Alan Turner regards it as being unfit for such a purpose, and if so, why.

User avatar
LesGros
Posts: 546
Joined: Thu Oct 08, 2009 10:05 pm

Re: Society Gauge Widening Tool

Postby LesGros » Thu Apr 28, 2016 2:30 pm

grovenor-2685 wrote:Bear in mind that, for bullhead track you use different chairs for each step in the gauge widening, there can be no intermediate values.


Ah, that makes perfect sense. A specific chair for each 3mm step; something new learned every day.
Thanks Keith
LesG

The man who never made a mistake
never made anything useful

User avatar
grovenor-2685
Forum Team
Posts: 3917
Joined: Sun Jun 29, 2008 8:02 pm

Re: Society Gauge Widening Tool

Postby grovenor-2685 » Thu Apr 28, 2016 2:59 pm

With bullhead the steps are 6mm (1/4")
The bottom table on my pdf.
The different chairs are needed as they are check chairs and set the flangeway.
Regards
Regards
Keith
Grovenor Sidings

User avatar
Will L
Posts: 2516
Joined: Sun Jul 20, 2008 3:54 pm

Re: Society Gauge Widening Tool

Postby Will L » Thu Apr 28, 2016 6:14 pm

John Palmer wrote:...Terry, I didn't say that I envisaged using the mint gauge to set gauge widening at the construction stage. My intention was to use it as a diagnostic tool in the way described by Paul Townsend. What I am trying to establish is whether Alan Turner regards it as being unfit for such a purpose, and if so, why.


While it may well make sense to use a Mint gauge to check the results of track construction with three point gauge of the same effective length because there results will match for the same curvature of track. To perform similarly relevant Mint style gauge check on construction done by any other method would require both a knowledge of what gauge widening was intended, what the radius of the curve was, a whole selection of mint gauges of different lengths and Alan's graph to help you chose which one to use.

Do we see this as a sensible proposal?

User avatar
LesGros
Posts: 546
Joined: Thu Oct 08, 2009 10:05 pm

Re: Society Gauge Widening Tool

Postby LesGros » Thu Apr 28, 2016 9:42 pm

grovenor-2685 wrote:With bullhead the steps are 6mm (1/4")
The bottom table on my pdf.
The different chairs are needed as they are check chairs and set the flangeway.
Regards

:oops: Oops! ... I should have scrolled down.
LesG

The man who never made a mistake
never made anything useful

User avatar
Julian Roberts
Posts: 1388
Joined: Sat Jan 09, 2010 11:33 pm

Re: Society Gauge Widening Tool

Postby Julian Roberts » Thu Apr 28, 2016 10:01 pm

Will L wrote:
Do we see this as a sensible proposal?


No way!

I wonder if I am looking at this in a totally different way to everyone else.

I hope he will forgive me if I am getting this wrong, but Alan is gently sending up the whole idea when he produces all those different colours. He originally thought I had this all wrong, on April 21st. His original graph, before he edited it to correct a "minor error", showed my 54mm length as being way out, totally off the wall. But (he may regret it) after the correction he has ended up corroborating what I am saying, that the Society gauge gives only around half prototypical widening. (And that's not what I am saying, it is what Russ Elliott said, but no one noticed it seems.)

Here is the graph again:

[img]
Alan Turner's Gauge Widening graph Gauge Mint 5.png

[/img]

There are only three curves that matter, the Society curve (the bottom one in green), the one in purple that gives minimum prototypical widening, i.e. just meets the prototype criteria for 5.5 chains, and the one at the top in black that gives the prototype's maximum gauge widening at 5.5 chains. This last one (54mm) literally meets the criteria but that is to misunderstand (as I did) the function of the prototype "steps". This function, it looks to me and from the fact that Alan suggested it, is met by the purple 45mm one.

There is only one proposal I am making, and only as a debating point in the first place, and that is whether we might add a 45mm gauge as another (fourth) approach to gauge widening, using a separate tool ( a track gauge set to 18.83 + 0.20 - 25) for curves below 3ft 3ins. This would be without altering anything else - i.e. I am not suggesting any heresies like overscale flanges, overscale flangeways (at least, no more overscale than P4 flangeways are already), etc, so any visual difference could only be seen by those with very exceptional eyesight.

To have a separate gauge to make track less than 1000mm curve is hardly very complicated compared with Will's springy beams...

Why? To be faithful to the prototype. Not because of a slavish or pedantic fetish to model exactly what is there even when invisible, but because of the numbers. We are using up between half and all of the extra tolerance that we get in P4 with the narrower B2B by making our track undergauge on curves.

Assuming Alan's graph is correct, and as far as I can judge the actual numbers to be from this graph, between 1000mm and 1450, we are between 0.14 and 0.08 narrow to gauge. As the P4 B2B criteria are between 0.12 and 0.22 narrower than the prototype we have lost between half and all of the extra tolerance P4 gives us for those curves. And that's if we have gauged our wheels correctly. As I wrote on Sunday (and no one has made any comment so far):

If we can accurately locate the wheels as per the Scalefour standard, they are between 17.67 and 17.75 apart. The prototype “window” is between 17.87 and 17.89. Thus while our wheels may be as much as 0.22 narrower, they may be set only 0.12 narrower than the prototype.


1000mm curve, we are undergauge by 0.14. Wheels at 17.75 may be only 0.12 narrower than prototype. So the tolerance has disappeared and is 0.02 negative to the real thing.
1450mm curve, we are undergauge by 0.08. Wheels at 17.75 only have 0.04 tolerance compared with the real thing.

IMHO. Please correct me if I am wrong.
You do not have the required permissions to view the files attached to this post.

User avatar
Julian Roberts
Posts: 1388
Joined: Sat Jan 09, 2010 11:33 pm

Re: Society Gauge Widening Tool

Postby Julian Roberts » Thu Apr 28, 2016 11:13 pm

burning the midnight oil now..

To me there are two questions that arise.

First is from Alan's graph, why all the (banana shaped) curves slightly miss the prototype curve steps. None of them meet all the steps. So it looks as though, subtly, something else is going on.

Second, from Alan's graph one can see that the less the gauge widening, the sharper the curve can be before the track goes "too wide", i.e. wider than the scaled down prototype maximum of 0.75", 0.25mm. More gauge widening, less minimum radius, if using one tool. Why did the P4 founding fathers choose such a small minimum radius at the expense of being closer to the gauge widening? I have yet to read the archive, but doubt I am going to get an answer to that.

It is as though the criteria was to have one track tool to keep things simple. But why to such a sharp curve if the idea was to bring true scale modelling? I can only think that an A4 point is 520mm radius so maybe that was the criteria for a minimum.

It seems to me on reflection that the 80% correct Mark 2 gauge widening method is nearer to the prototype in function where it matters, giving 0.2 out of the correct 0.25mm, and it really doesn't matter that it is overgauge from 1450.

After all this whole thing is about a space (0.2mm) about the thickness of a human hair probably, or a pinhead. A good title would be "how many railway modellers can dance on a pinhead?" But in a way this is a definition of civilization, that we can have a debate like this, and not be the citizens of Aleppo for example in terror for our lives.

User avatar
Will L
Posts: 2516
Joined: Sun Jul 20, 2008 3:54 pm

Re: Society Gauge Widening Tool

Postby Will L » Thu Apr 28, 2016 11:48 pm

Julian Roberts wrote:I wonder if I am looking at this in a totally different way to everyone else.

Probably
I hope he will forgive me if I am getting this wrong, but Alan is gently sending up the whole idea when he produces all those different colours.

Probably
...To have a separate gauge to make track less than 1000mm curve is hardly very complicated...

what you would be talking about here would be a new three point gauge, with a matching Mint gauge as a go/nogo checker
...compared with Will's springy beams...
now that is sacrilege. CSB are not complicated, but if I'd just said they were a good idea without the supporting info as to why and how, the other half of the fraternity would have had a wonderful time telling us, on the basis of no evidence, why they didn't work. CSB only seem complicated if you can't be bothered to try and follow the argument. On current performance Julian I would have said you should have not trouble with CSB at all.
Why? To be faithful to the prototype....

But the prototype is clearly a distinctly arbitrary system more to do with the practicality of building the track than any theoretical correct degree of track widening. So to be prototypical our track/gauge widening should be equally arbitrary.

Alan Turner
Posts: 643
Joined: Sun Jul 20, 2008 4:24 pm

Re: Society Gauge Widening Tool

Postby Alan Turner » Fri Apr 29, 2016 8:08 am

My first response is that, as has been repeatedly said, the "Mint" gauge was not produced for setting gauge widening.

Second you are looking at this issue from the wrong end of the telescope. The gauge widening shown in my graphs is the minimum required to enable the gauge to traverse the curved track. The arbitrary length of the gauge therefore affects the gauge widening to enable the gauge to traverse. What this says about the performance of actual rolling stock I have no idea.

regards

Alan

User avatar
Julian Roberts
Posts: 1388
Joined: Sat Jan 09, 2010 11:33 pm

Re: Society Gauge Widening Tool

Postby Julian Roberts » Fri Apr 29, 2016 8:16 am

I have edited this, [] show edits. To clarify my meaning, as I had to dash out earlier.

Will L wrote:
Julian Roberts wrote:...To have a separate gauge to make track less than 1000mm curve is hardly very complicated...

what you would be talking about here would be a new three point gauge, with a matching Mint gauge as a go/nogo checker .


[I think] you've got me wrong there Will. Look at the graph. Below 1000mm or 3ft 3ins the green line goes above the red line. So to stay within the maximum widening of 0.22 (P4) to 0.25 (prototype) you would have a simple 18.83 + 0.22-5 track gauge, i.e. [pretty much the same as] the Protofour Mk 2 gauge with GW (the 'gauge widening factor'). So that would be for curves below 1000mm, [above there you'd have] the 45mm rectangle, triangle, mint, whatever it is, it's the length that matters. So you would need just two gauges. Whether it is rectangular or triangular is immaterial. Any Mint gauge [for checking the track] ought to have the same length in case it is used for curves, which it perfectly well [can] do, as Paul has demonstrated.

I wrote

Why? To be faithful to the prototype....


Will wrote

But the prototype is clearly a distinctly arbitrary system more to do with the practicality of building the track than any theoretical correct degree of track widening. So to be prototypical our track/gauge widening should be equally arbitrary


Oh dear, this is a bit confusing. and difficult to explain exactly. What I mean is, that the important thing is to have somewhere nearer the prototype gauge widening, I think you understand what I mean there. But I mean, the intent of the prototype. The curving banana shaped lines on Alan's graph surely represent what really happens, as a curve gradually sharpens the wheel and flanges [of any vehicle] take up a gradually [and at an increasing rate] more offset angle. [Because we can use a triangulation tool, unlike the prototype] our infinitely variable means of fixing the track, whether by solder or glue, means [that] we follow this curve [on Alan's graph], but the prototype can't do that and has to work in steps. So I don't suggest slavishly following that [stepped] aspect of the prototype. It is just a question, what length of triangulation most closely does what the real thing would do if it also worked in an infinitely variable way.

(It is a bit like my mother's first car bought in 1970, a Daf 33. It had infinitely variable "Variomatic" transmission instead of the usual gears that work in steps. A much better concept, but not so practical as it relied on belts that give more friction and can break. But Ford continued and improved the concept in some of their Fiestas, and she bought one of those later.)

And as far as I can see Alan's purple 45mm option most closely does that, but it could be argued about, which represents best what the prototype would do if it could work like that.

The dunce in the class of Alan's graphs is the Society gauge length. All the others at least touch the red lines once, but our gauge doesn't even manage the lowest level of the red lines. But it only matters [below] around 5.5 chains, as above there the P4 extra tolerance is sufficient to pretty much make up the difference.

Of course all this is predicated on Alan's graph being accurate...

User avatar
Julian Roberts
Posts: 1388
Joined: Sat Jan 09, 2010 11:33 pm

Re: Society Gauge Widening Tool

Postby Julian Roberts » Thu May 05, 2016 9:44 pm

I think most has been said that there is to be said on this thread, but to tie up some loose ends, first:

Paul Townsend wrote:I am not knocking your analysis, as an engineer I appreciate this sort of theoretical approach, but as you are against the mint please design a gadget that does what many of us use the mint for and satisfies your analysis too.


Paul I know you did not ask me.

I don't own one, but apart from the question of how long it should be, the only problem I see with a mint gauge is that it can't be used on the crossing area of the diverging road of a turnout. But if the "rails" of the present tool were replaced with three pegs of similar depth, one at each end of one rail, and one half way along the other, thus making a three pegged triangle that sits inside the (model railway) rails, it could be used to slide through and check the diverging crossing area. Only the gauge at the crossing flangeway gap would not be precisely checkable. The pegs might have to be in a harder material than brass perhaps.

Why not make the tool two sided? One side like it is, and the top, at present just flat, as I have just described. Then it could carry out the function it already has, plus the only bit it can't do.

I have no idea how practical such a tool would be to make.
Last edited by Julian Roberts on Fri May 06, 2016 8:04 am, edited 1 time in total.

User avatar
Julian Roberts
Posts: 1388
Joined: Sat Jan 09, 2010 11:33 pm

Re: Society Gauge Widening Tool

Postby Julian Roberts » Thu May 05, 2016 9:55 pm

I forgot to say -

The pegs would be the same diameter as the thickness of the "rails" on the present tool

User avatar
Julian Roberts
Posts: 1388
Joined: Sat Jan 09, 2010 11:33 pm

Re: Society Gauge Widening Tool

Postby Julian Roberts » Wed Aug 03, 2016 8:59 am

The questions behind this thread are:
1 whether the Society Track Gauge tools that automatically gauge widen on curves give less gauge widening than the prototype
2 what is prototypical gauge widening
3 why is the Society Gauge Widening less than the prototype, by how much, and does it matter
4 would running in P4 be improved if prototypical gauge widening was used, and would there be any downside?

Very helpfully, Gauge Widening has been represented as a series of graphs by Alan Turner. The only question of the above that his graphs have shown fairly definitively is that, yes the Society Tools give less widening than the prototype.

Keith has from the beginning of the thread clarified that in P4, because the B2B is less than the prototype, less gauge widening is needed. However, in view of the equally widely accepted nostrum that in P4 narrow-to-gauge track gives problems, I don't quite understand this logic, and it is to investigate this issue that I am still pursuing this thread, with the idea that maybe using prototypical gauge widening might further increase reliability of running. And/or, the more generous prototypical gauge widening might make good running more easily achieved by those of us who are less expert at track laying than the professionals.

My civil engineer son Matthew, who currently lives in New Zealand, has kindly drawn out what seemed to me the most relevant synthesis of Alan's graphs. It shows some additional features. He has also sent me the spreadsheet of the numbers that the graph represents.

So here is the graph. In addition to showing the "maximum" gauge widening that I suggested earlier, and showing Alan's "minimum" gauge widening, and the prototype gauge widening in the form of steps, and the Society 29.75mm tool gauge widening: it shows the P4 Mark 2 Track Gauge with 0.2 mm gauge widening washer inserted, and the Scalefour Digest 1.2 gauge widening standard. This says

"In P4.......gauge widening should not be necessary unless using long-wheelbase stock around sharp curves"

As "long" and "sharp" are not defined, it is difficult to know what this means, which is why the Standard of zero gauge widening is in a dashed line, indicating uncertainty.

Similarly, the P4 Mark 2 (with 0.2mm washer inserted) tool gauge widening becomes a dashed line, indicating uncertainty, beyond the point where a curve is so shallow as to not merit the use of the gauge widening washer. The problem with this is defining in ordinary use where that point is. Clear on the graph is the fact that it gives widening much nearer to the prototype. The answer to my fourth question above might be known by people who have used both this tool and the Mark 1 tool.

Gauge Widening Graph.pdf


Edit - the Webmaster has explained about the attachments - thanks John



Here, next, is the spreadsheet. On this the steps of the prototype gauge widening (5.5, 7, and 10 chains) are in red, in imperial feet and inches as well as millimetres, in 4mm scale. There is a number for every 3 inches of increase in radius.

Additionally, I asked Matthew to work out what length tool gives the Scalefour Digest 1.2 maximum gauge widening of 0.22mm at 528mm radius. (Just under 21 inches, or 1' 9".) This appears on the spreadsheet - it is 31mm, slightly longer than my rectangular gauge, though someone earlier on this thread gave the length of his Studiolith one, and I think it may have been this dimension.



Gauge widening Version 2.xlsx


What is apparent from the spreadsheet is that the 45mm tool is incorrect, as the minimum gauge widening of the prototype in 4mm scale is 0.1666mm at 5.5 chains, not 0.1721mm. So the correct tool length will be somewhat less.

The correct figures and line on the graph will hopefully appear soon from New Zealand - or maybe some maths genius will come up with them here first. What I surmize is that the 45mm curve that (looking carefully) is just above the minimum step at 5.5 chains, once adjusted (to say 43mm) to meet that, will also be much nearer or coincide exactly with the first step at 10 chains, and be half way up the step at 7 chains. Thus I reckon it will show "true" prototypical gauge widening, or something as near to that as is useful to know.
You do not have the required permissions to view the files attached to this post.
Last edited by Julian Roberts on Thu Aug 04, 2016 12:12 am, edited 1 time in total.

User avatar
grovenor-2685
Forum Team
Posts: 3917
Joined: Sun Jun 29, 2008 8:02 pm

Re: Society Gauge Widening Tool

Postby grovenor-2685 » Wed Aug 03, 2016 9:27 am

I thought we had done this to death!
Use the Triangular gauge and it all works.
Have you read this?
Unfortunately, when Rob revamped the website he left out some of the illustrations as he felt the scan quality was not good enough. I'll see if I can find the originals for you.
Regards
Regards
Keith
Grovenor Sidings

User avatar
grovenor-2685
Forum Team
Posts: 3917
Joined: Sun Jun 29, 2008 8:02 pm

Re: Society Gauge Widening Tool

Postby grovenor-2685 » Wed Aug 03, 2016 6:07 pm

Here are the missing images from the March 1967 MRC article.
fig3.gif

fig4.gif

There was another article on this topic in Model Railways for March 1973 which used to be on the site but seems to have been dropped completely. When I get time I'll put it up on my website.
Regards
You do not have the required permissions to view the files attached to this post.
Regards
Keith
Grovenor Sidings

User avatar
Julian Roberts
Posts: 1388
Joined: Sat Jan 09, 2010 11:33 pm

Re: Society Gauge Widening Tool

Postby Julian Roberts » Thu Aug 04, 2016 12:03 am

Thanks Keith.
I don't know if we are at cross-purposes. I had seen the Malcolm Cross article on curves, though not the graphs. I don't understand them so far, they will repay further study.

I am not talking about very sharp (in model railway terms) curves - e.g. less than 4 feet radius - nor long wheelbase 6 wheel stock. Sideplay on centre axles can mostly pretty easily be made greater than on the prototype, that is not the issue either. Incidentally I was asking people last night at the club if they knew how sharp are the sharpest curves near here - the line at Carstairs that takes trains to the Edinburgh line when coming from Glasgow, and the curve that takes the train from Queen Street towards the West Highland Line at the top of Cowlairs bank, opposite Eastfield depot. It would be interesting to know the radius of what feels like a very sharp curve, taken at a very low speed, say 10mph. But I guess even such curves are not much sharper than a 4 foot radius in 4mm scale...and it is the gauge widening needed for the sharpest prototypical curves that interests me, as in the model railway context these are not as rare as on the real thing.

My thinking is simply from what I read in Iain Rice's "An Approach to Building Finescale Track". I hope it will be OK by Wild Swan to show the excerpt here:

2016-08-03 20.53.38.jpg

2016-08-03 20.54.49.jpg


Well I can't get the second image to come out vertically, apologies. Anyway I am sure you don't need to read it.

"an ounce of practice is worth a ton of theory" - here is a 4ft radius curve, with the gauge written by each "sleeper". It is gauged with my Society Triangular Gauge, which incidentally is not exactly triangular, having 4 "claws", and is 32mm long. 18.93 - 6 is the range, so that's 0.1+ widening. Looking at the spreadsheet for 4 feet, a 29.75mm tool would give 0.9 widening, and the prototypical widening, converted to 4mm scale, would be the 0.25mm maximum.

2016-08-02 13.11.12.jpg
You do not have the required permissions to view the files attached to this post.

User avatar
Allan Goodwillie
Posts: 915
Joined: Sun Jul 20, 2008 7:00 pm

Re: Society Gauge Widening Tool

Postby Allan Goodwillie » Thu Aug 04, 2016 9:47 am

Hi Julian,

As far as I know the tightest curve on the Scottish region used by mainline engines was that at Haymarket Central on the Gorgie curve where the line from the sub came in to the main line. I made a check with Richard Chown as it was his job to know these things. Unfortunately Richard could not remember the radius of the curve, but thought it was a good specimen.

Richard mentioned a number of things. One was that the outer rail of the pair was made up from 60ft lengths as normal, but the inner rail was 59' 9" with one bolt hole sawn off to keep the rail ends opposite one another. :shock: The second thing was that check rails were for track under 10 chains for passenger traffic, but not necessarily "required" at that for freight. Where check rails were used it was normal practice to make sure that the joints in the check rails were in the middle of the lengths for track so that the joints did not coincide with both. Now check rails are not used so much even on fairly tight curves like the one at Bathgate heading west - which is quite exciting to go around. :o A rail is placed in the middle of the four foot to help maintain alignment. Although this is a modern development - I must go and take a photo sometime!

One last comment made by Richard to do with the current discussion is that, in the present Track "Bible" it is suggested that where track is of tight curvature and longer wheelbase vehicles involved there is a recommendation (but only a recommendation) that track widening can be considered. It is not laid down with a hard and fast rule although there are a range of figures for consideration. Richard's comment was that nothing was ever hard or fast in his day either as there was much to consider that was track related and the local geography and geology that could require different solutions. There is even an anomaly outside his kitchen window to do with the chairing of the track on the Glasgow -Edinburgh main line at the moment.

In the days of colliery lines branching off the main line, there was a particularly tight one branching away at, I think, Cardenden - it may be that only colliery engines used it so, perhaps, less remarkable if that was the case - I only remember seeing colliery locos on it. As you know I am trying out a few practical experiments myself, but too early to come to any conclusions yet.

Richard, being a S7 modeller has also thoughts on this - I can see a long conversation coming on as we drive down to Wigan with the layouts!

It may be an idea to have a look at Mike Gilgannon's points from his fiddle yard which we still have in the club, They have not been discarded, but some of them seemed impossibly tight curves for S4. I will get my S4 gauges out to see what the gauge widening was on them, but all his locos and stock ran through them perfectly behind the scenes. Mind you he never had anything longer than his beautiful 4-4-0 or his Cardean. If you think about it, most railways don't have to cope with anything more than an 0-6-0 coupled wheelbase.

On Grayrigg I did have to install a check rail on one of the inner mainline curves to cope with the bigger engines and as you know I can get my A4 around the inner circuit. Mike's Cardean did not have any outside cylinders to foul the bogies. I had to do a bit of grinding, not seen from most angles, of the back of the cylinders just to make it happen.

I remember one time looking forward to seeing Ian Middleditch's Duchess run on Grayrigg. Ian is better known for his beautiful GSWR locos, but he had put together a beautiful kit version of the Duchess just to have as a display. You could even see through the centre of the middle axle as per prototype! It ran beautifully along the straight, but disappointingly would not go around the curves, even on the outer circle, unless the front bogie was removed. It was intended as a glass case model anyway.

My good friend Bob Hetherington who has also brought up his Stockton and Darlington locos from time to time also found that they would not go around the corners on Grayrigg without trying to climb the rail. Again this is no condemnation of Bob's wonderful locomotives, it is just that they were built to greater exactitude than mine with less sideways travel built into the chassis than I have in my 2-8-0's and 4-6-0s. I think Bob's minimum point is a 10 on his own layout. In fact I think he is busy replacing them with a couple of 12s at the moment.

As you know we have had discussions about the new C&L track gauges and their choice of gauge widening and why they maybe felt they should introduce them and replace the previous type they supplied - all conjecture on our part and probably not for discussion on the forum. I have been experimenting with them on the new points.

Another area this discussion effects, although no one has mentioned it, is if we are using tighter than prototypical curves with double track mainline stock we may encounter other problems. On Grayrigg for example on the Northern approach to the station I have to be careful as to what passes there. For example, Dave has a Bachman class 40 which we use on the "Royal Scot" (Which is about to be replaced by 10000 and 10001, now we have got them running). The class 40 clips other coaching stock going the other direction due to its overhang on the curve.

Richard when on the phone commented about the Glasgow O Gauge layout with 4 track mainline had the same problem, but only after the introduction of an LMS twelve wheeler coach and some 70 ft GWR stock which when they met on the curves would batter one another. So not just a problem on S4 layouts.

I built the track with the correct spacing recommended by the late Derek Genzel which you will find in the S4 Digest sheets,I even have a set of profiles for such things including the station platform one, which my WD's fail on, so a certain amount of modification has been required on Grayriggs platform. On Gilgannon Junction I have deliberately given a bit more space between platform and train. Not noticeable from viewing angles it will allow enough space for most of the bigger engines I am building over the next few years.

Prototype spacing between tracks can be just a fantasy on a model railway where tighter than prototype curves are to be found. Derek was a great theorist and could be heard at S4 dinners from several tables away. I had the pleasure of sitting with him on a couple of occasions and realised that he really enjoyed theory, I never did see his layout, but he did encourage testing of ideas and I believe that he had a hand in the S4 test rigg which was in use for many years at Scalefourum. In fact the East Group including myself had the pleasure of operating the system for a few years at Scalefourum. All very interesting it was too, there are many tales there, but I diverge. Theory put into practice with sometimes startling results.

Back to the class 40 - it is correctly dimensioned with correctly tapered nose to deal with the same problem in real life (Unlike the Joueff product of many years ago.) Longer steam locos have the same feature. The fault lay with the 6' between the tracks being set according to the diagrams supplied and my unquestioning attitude to given wisdom.

I have been following this thread as you know with much interest as various figures have been bandied about, however I have no experience of the triangular gauge that the Society uses so have no comment to add to what has been said up to now, except that the person who never questions things will never discover anything new and recent comments on QI which I do enjoy, pointed out that within 10 years 2/3 of all known facts will have changed to some degree! As an aside I also enjoy that delightful Ms.Coren's lateral thinking quiz, must be turning into a :ugeek:

One thing I might add is just a comment on the perception of curves. Grayrigg can only be seen from the inside of the layout and curves seen from the inside never seem so bad just due to the straight on angle you are looking at them from. The impression the layout gives is that of a proper main line. It is now more than twenty years old and has lots of experimental features - some of which have been very successful, others less so, but what a lot I have learned from building it. I am putting all of this into practice on my new layout/s so we will see what comes of that in the next few years, if I am spared. I have been heavily involved with both modelling groups for years as you know and feel that I can take more of a back seat as there is plenty of talent around in both groups and I may get more time to do my own thing.

:) I have, as you know, had a preview of some of your ideas on compensation and think it will make a great article, again it questions the wisdom handed down from the days of Mike Sharman, a really good practical modeller and charming man, always happy to discuss his work and techniques. I know he wouldn't mind.

I am definitely going to have a go at altering the next bogie loco I build - perhaps I may alter the Jubilee that I am on at the moment, it is still at the early stage just now although it may act as a distraction for the starters group. My bogies stay on the track doing what I already do, but it may be interesting just to do one differently, plus from time to time I have had the odd tender engine misbehave in the rearwards direction, which I have always put down to there not being enough weight above the rear driving axle - which does seem to tie in with your own conclusions. I will be fitting the Jubilee with about a 40:1 gear ratio gearbox to go with the motor I have for it to allow for express speeds on Grayrigg, although I know your thoughts about your own locomotives and that will fit fine on Gilgannon Junction as there will be only a few through expresses. It is always exciting to see a train coming down from Shap around the banked curve at speed on a Glasgow - Manchester. :idea: I have a Patriot to do as well and I might do it conventionally to compare the two.

There should be a thread saying things to definitely avoid or experiments that did not work, but it would probably be only myself or one or two others who would contribute as most probably don't like to admit that they can get it wrong as it may include a sense of loosing face in front of our peers. I have just completed an experimental chassis that was started 30years ago by my long lost friend Jim Pugh which has a combination of compensated and sprung suspensions. It has been on the back burner so long. It works OK but I do not think it works better than either convention on its own, It does have one or two behavioural quirks. One for the tried it,interesting, but maybe a dead end just the same.

Enjoy the fireworks Julian! :D

User avatar
Allan Goodwillie
Posts: 915
Joined: Sun Jul 20, 2008 7:00 pm

Re: Society Gauge Widening Tool

Postby Allan Goodwillie » Thu Aug 04, 2016 9:51 am

I have taken your image Julian and tried to make it clearer, easier to read for everyone if they click for magnification. Ian's comment at the foot of the strip is interesting.

file.jpg
You do not have the required permissions to view the files attached to this post.

User avatar
Noel
Posts: 1972
Joined: Wed Jun 23, 2010 1:04 pm

Re: Society Gauge Widening Tool

Postby Noel » Thu Aug 04, 2016 3:33 pm

For those who don't have the book, the final paragraph is a lead in to the importance of the check gauge, which is the b-to-b + 1xflange thickness, Iain's contention being that this must be constant, but the b-to-b may need to vary if the wheel profile or flange thickness varies varies. He says that this is not an issue for EM and P4, with consistent standards, but is for OO.

However, this begs the question of whether all P4 wheel manufacturers actually use the same profile; I seem to remember the suggestion elsewhere in this Forum that this was not entirely true.
Regards
Noel

User avatar
Julian Roberts
Posts: 1388
Joined: Sat Jan 09, 2010 11:33 pm

Re: Society Gauge Widening Tool

Postby Julian Roberts » Thu Aug 04, 2016 4:18 pm

Allan - Thank you for re-arranging that image. And your previous comments. The article on compensation you refer to does not specifically say, and I was wondering whether and how to, that I have not made any bogie locos except 4-4-0's and 0-4-4's: so I am not sure whether that stuff will apply to bigger locos.

I was in correspondence with Richard about all this gauge widening business and will ask him if I can post his reply here.

grovenor-2685 wrote:
Keith - can one read the archive? Is it on the Society website?

See http://www.scalefour.org/history/history.html
and http://www.scalefour.org/members/protofoursociety/
Regards


I was interested to read the archive to try to understand why the gauge widening is less in P4 than the prototype.
It was fascinating to skim through the issues of "Prototype" and "Precision". Back to the days of my youth - long hair, manual typewriters, a convention held at Heathrow Airport of all places, difficult to get to at the time!
It made me appreciate just how good we have it in this hobby with plentiful supplies of wheels, mainframes, rail - all these basic things were in scarce supply. The steam loco mainframe programme (I assume that is what we now can get from Alan Gibson/Colin) was a major labour of love.
The "bumpy" (quoting Iain Rice in his Approach to Building FS Track) politics are only hinted at, very much underplayed, and I am not much the wiser as to what went on.
The aim was to make true scale modelling practical and feasible to the kitchen table modeller, using devices to make sophisticated measuring to 2 decimal places unnecessary - all very new at the time but what we take for granted.
This is where the Track Gauge Tool comes in to this thread. It is in the same mould - you don't have to know the curve radius to put in the correct gauge widening, the tool does it for you. You don't have to know anything about the subject at all. If you want to be pedantically correct you can measure your curve and get a gauge to the "prototype" widening - but it is unnecessary. Except, the gauge doesn't give the full amount that the prototype does, not that this is said anywhere - at least not until the Mark 2 version came out.
In one of the issues - 12 - there is reference to "political issues" regarding adopting standards that are not an absolute scaling down of the prototype. In Issue 13 of the "Prototype" the latest iteration of the P4 standards is given, and a promise is made that in the next issue there will be a full explanation of them. But explanation is there none, nor in any subsequent issue. What does it mean, a minimum radius of 528mm with maximum gauge widening of 0.22mm? How does that fit into the newly ever-increasing faithfulness to the prototype?
So my hunch is this. All this "white heat of technology" development of the Protofour Standards was in a quite different era to ours, and a particular aspect that was so different is that model railways had sharp curves. Today we accept that a model Duchess won't go round an A4 point and don't try to make it do so, even in RTR 00. Back then, that may not have been such a welcome reality. I reckon that the other side of the prototypical faithfulness coin, realistic minimum radii curves, may have seemed a less desirable development in that it might make the new standard less popular or attractive. Thus in Issue 15 there are compliments to a layout demonstrating the practicality in P4 of a mainline with 2' 9" radius curves. It's as though how small you could get the radius in the new standard was a definition of machismo - regardless of how prototypical it was in itself. Particularly, I reckon people didn't want EM to be able to better P4 in this regard.
Our own Mike Gilgannon appears in Issue 7, confessing his widely shared scepticism that the new standard will work for curves.
One of the saddest pieces by Bernard Weller is in Issue 41 where he says that "for reasons we would rather forget" Studiolith jigs are unavailable following the closure of that company. I get the impression of a life totally dedicated to this extraordinary quantum leap forward and excellence in model railways, I don't know how accurate that may be.
If one looks at any of the graphs drawn to show the gauge widening tool, it is obvious that the gauge widens gradually to a point where it exceeds the prototype's 0.25mm maximum. All that varies is when it hits that point, which would imply the minimum radius possible. The P4 curve on the graph hits its arbitrary maximum of 0.22 at 528mm, which would have been I expect the same or similar to EM. 0.22mm I assume to reflect the narrower BB of P4. Personally I don't see what the problem would be with choosing something nearer to the real thing, and, if one really must make very sharp curves, widening more than 0.22 or 0.25 for that matter - after all there is about 1.5mm to play with before the wheels fall between the rails. But that would be unprototypical...

A prototypical widening tool would be a little less than 45mm, and would hit the prototype maximum widening at about 3' 3", more or less the same as Will L's chosen minimum radius. If I built a bit of track for the Calderside extension using such a tool, there would be no practical difference visually or to the running of anyone's stock, except perhaps a greater reliability of road-holding.

There was another article on this topic in Model Railways for March 1973 which used to be on the site but seems to have been dropped completely. When I get time I'll put it up on my website.
Regards

That would be great Keith - can you alert me/us when you have done so?


Return to “Track and Turnouts”

Who is online

Users browsing this forum: ClaudeBot and 1 guest