Do we need 1:20 Rail Inclination

Discussion of model and prototype wheel/rail interaction.
User avatar
Will L
Posts: 1528
Joined: Sun Jul 20, 2008 3:54 pm

Do we need 1:20 Rail Inclination

Postby Will L » Tue Aug 07, 2018 4:05 pm

Seems we have had a lot of comment recently about the problems we get into trying to model the prototypical 1:20 inward inclination of track which appears to be leading to gauge narrowing.

In the end you do have to ask, why are we bothering.

I know it has been said that without the rail inclination the self steering effect of a coned wheel is lost. But I'm sure that isn't true. As far as I can work out, inclining the rail so it matched the conning on the wheels is all to do with evening the wear across he rail head. (Otherwise it would be concentrated on the inner edge and presumably(?) contribute to the gauge corner cracking problem). The natural tendency for coned wheel to self centre on the straits and steer round corners isn't materially affected, and in case anybody is thinking otherwise, works for us as least as well as for the full size railway.

Visually I really doubt if anybody can tell whether rail is inclined correctly, or not, without significant instrumental assistance.

So, given that rail wear is hardly likely to bother us, would we be better off suggesting the next generation of functional chairs leave out the cant?

Written as a fully signed up member of the "getting it all right" tendency
Edited to remove the tenancy tendency
Last edited by Will L on Wed Aug 08, 2018 6:40 pm, edited 1 time in total.

Phil O
Posts: 181
Joined: Sun May 05, 2013 5:23 pm

Re: Do we need 1:20 Rail Inclination

Postby Phil O » Tue Aug 07, 2018 5:26 pm

As the head of the rail should be slightly curved, I very much doubt that in our scale it makes much if any difference one way or the other.

Phil

User avatar
David Thorpe
Posts: 717
Joined: Tue Oct 21, 2008 11:13 pm

Re: Do we need 1:20 Rail Inclination

Postby David Thorpe » Tue Aug 07, 2018 5:57 pm

Will L wrote:So, given that rail wear is hardly likely to bother us, would we be better off suggesting the next generation of functional chairs leave out the cant?


Yes.

DT

Philip Hall
Posts: 1087
Joined: Mon Aug 10, 2009 7:49 pm

Re: Do we need 1:20 Rail Inclination

Postby Philip Hall » Tue Aug 07, 2018 6:04 pm

I'm with Will on this. Peco have already done it on their new track and it looks OK, you can't tell by looking at it and I doubt whether it makes any operational difference on a model. The inward lean also gives very minor problems of alignment when joined up to ply and rivet trackwork.

Interestingly, I have recently made up some test lengths of Exactoscale FastTrack to see how the rail sections that I have in stock actually fit the mouldings, and in many cases the rail has been inclined inwards, some were almost vertical, and one panel had the rail leaning outwards! These also varied from a loose fit to a fairly tight one. But in all the cases the gauge was within reasonable limits limits and the stock ran OK.

Philip

Andrew Ullyott
Posts: 131
Joined: Fri Aug 08, 2008 2:31 pm

Re: Do we need 1:20 Rail Inclination

Postby Andrew Ullyott » Tue Aug 07, 2018 7:55 pm

I've built quite a bit of 'light railway' track in 4mm, code 55 and 60 flat bottomed rail soldered to coper clad. There is no cant and it is entirely prototypical. I also have built quite a bit of conventionally chaired track. I can't say I've ever noticed any difference in the way the trains move comparing the two.

But then I've also been known to not model things I can't see from 3 feet...

Alan Turner
Posts: 498
Joined: Sun Jul 20, 2008 4:24 pm

Re: Do we need 1:20 Rail Inclination

Postby Alan Turner » Tue Aug 07, 2018 8:00 pm

For a time BR laid crossings with vertical rail.

Also Swiss rail doesn't have any inclination and North American rail has a 1:40 inclination.

regards

Alan

User avatar
jim s-w
Posts: 1603
Joined: Wed Jul 30, 2008 5:56 pm

Re: Do we need 1:20 Rail Inclination

Postby jim s-w » Tue Aug 07, 2018 8:10 pm

Andrew Ullyott wrote:
But then I've also been known to not model things I can't see from 3 feet...


12mm is pretty close! :D

User avatar
dcockling
Posts: 572
Joined: Mon Jun 30, 2008 10:11 am

Re: Do we need 1:20 Rail Inclination

Postby dcockling » Tue Aug 07, 2018 9:53 pm

Will L wrote:"getting it all right" tenancy


Tenancy? :shock:

perhaps you're wearing your 'hare' shirt Will :D

Hardwicke
Posts: 530
Joined: Sun Sep 14, 2008 10:25 pm

Re: Do we need 1:20 Rail Inclination

Postby Hardwicke » Wed Aug 08, 2018 7:32 am

If you let a pair of P4 wheels roll freely along track that has the correct 1 in 20 angle and then try it with wheels and track that don't (00 or P4) you will see the difference it makes.

The answer is, yes we need it. Otherwise why bother in P4?
(I have a booklet from the Research Dept at Derby from 1950 on the behavior of 4 wheeled vehicles on track. Explains a lot. You need the inclination on curves more to prevent derailments).

It improves the wheel to rail contact too; essential in P4 for electrical continuity.

User avatar
jon price
Posts: 427
Joined: Mon Jun 07, 2010 2:34 pm

Re: Do we need 1:20 Rail Inclination

Postby jon price » Wed Aug 08, 2018 8:58 am

Although with all the other factors a single axle is not guaranteed to behave properly, which is why so much effort goes into the suspension for pony trucks. Two axles held parallel seem more legit as a measure of practical track configuration.

User avatar
Guy Rixon
Posts: 506
Joined: Sun Mar 27, 2011 6:40 pm

Re: Do we need 1:20 Rail Inclination

Postby Guy Rixon » Wed Aug 08, 2018 9:41 am

Hardwicke wrote:It improves the wheel to rail contact too; essential in P4 for electrical continuity.


Have you tested this? I'd be interested to see the results of a proper study.

Hardwicke
Posts: 530
Joined: Sun Sep 14, 2008 10:25 pm

Re: Do we need 1:20 Rail Inclination

Postby Hardwicke » Wed Aug 08, 2018 9:50 am

I don't need to. It is logic. A wheel with a 1 in 20 cone will touch a vertical rail on a knife edge. If both have the same inclination the contacted area is larger.

Hardwicke
Posts: 530
Joined: Sun Sep 14, 2008 10:25 pm

Re: Do we need 1:20 Rail Inclination

Postby Hardwicke » Wed Aug 08, 2018 9:58 am

It might be in Russian but shows eloquently why you need the cone AND inclination. Unless you think physics don't apply.
https://youtu.be/agd8B-31bjE

martin goodall
Posts: 864
Joined: Fri Oct 02, 2009 6:20 pm

Re: Do we need 1:20 Rail Inclination

Postby martin goodall » Wed Aug 08, 2018 10:56 am

I have never attempted to reproduce the 'tilt' of the rail when building and laying track.

Lets face it - it's only ONE SEVENTY-SIXTH of full size, so it just doesn't matter.

(And I agree that omitting this tilt in the rail makes it easier to get the track gauge right, including necessary gauge widening on curves.)

"Getting it all right" has its practical limits (and I always thought it was a rather silly slogan anyway). I am satisfied with "artistic realism", subject only to getting everything to run smoothly - hence my use of EM wheels on P4 track.

Hardwicke
Posts: 530
Joined: Sun Sep 14, 2008 10:25 pm

Re: Do we need 1:20 Rail Inclination

Postby Hardwicke » Wed Aug 08, 2018 10:59 am

To upset the applecart, the concerns about coning and inclination seem to be based on opinion, not fact or evidence. When challenged with facts, people throw their arms in the air and make up excuses not to believe. Are we regressing as a group, just as society seems to be generally?

User avatar
jon price
Posts: 427
Joined: Mon Jun 07, 2010 2:34 pm

Re: Do we need 1:20 Rail Inclination

Postby jon price » Wed Aug 08, 2018 11:28 am

So my question is: What is the contact area of a P4 coned wheel on vertical track, and what is the contact area on inclined track. In THEORY the contact between a circle and a straight line is almost zero, so the width of a molecule; and as the circle turns the molecule in contact is replaced by another. So the contact point is infinitessimal, but it is surrounded by a variable zone of arcing, whose size is dependent on the current drawn. If the contact point is extended to a line then this would also be a molecule wide, with a variable zone of arcing. Clearly a very inefficient way of transmitting electricity, so the difference between contact on vertical or inclined, whilst impacting on such things as wheel wear, is not likely to make much of a difference to the inefficiency of the contact. Looking at real world railways that draw their power from the track, the live rail is always vertical, the top horizontal, and the collector a wide shoe equipped with a way of pressing the shoe onto the rail.

Philip Hall
Posts: 1087
Joined: Mon Aug 10, 2009 7:49 pm

Re: Do we need 1:20 Rail Inclination

Postby Philip Hall » Wed Aug 08, 2018 11:32 am

Another point, possibly more important, is how practical it is to produce a moulded chair, with the correct inclination, that fits the rail section accurately without strain or slop. Unless the fit is good, the rail can be held at various angles, which my tests have proved happens. There are various rail sections around and there has been much comment about this (see the Forum a while ago) and I cannot be the only one with large stocks of rail from various sources which are not going to be thrown away. Moulds wear, as we have seen, and rail web thicknesses vary with the consequence of opening up the jaws of a chair. There are too many variables in practice.

In the Ward-Platt article in the Snooze recently, it was interesting to read again that he got the smoothest running from parallel treads with no coning. This after a great deal of exhaustive testing. I’m not suggesting we should go down that route, but one day I intend to have a go and see how it pans out. That gentlemen carried out tests the like of which I have never heard of before or since, with very good results it seems. A shame he is no longer around to give us his thoughts on this debate.

Much as we would like to have near perfection, it’s not going to be practical for a result that to my mind will not have a lot of benefit or visible improvement. As for electrical conductivity, there may well be a slight difference in pickup, but it will be so small between canted and vertical rail. Far better from my experience, when the prototype permits, to have as many collectors as you can.


Philip

Armchair Modeller
Posts: 1047
Joined: Thu Jan 06, 2011 7:34 pm

Re: Do we need 1:20 Rail Inclination

Postby Armchair Modeller » Wed Aug 08, 2018 11:46 am

I built my track on both versions of Neversay with copperclad sleepers and Masokits chairs. It turned out vertical without me even thinking about it. I have been able to run trains on it very successfully - including propelling a wagon with sprung suspension at something in the region of a scale 300mph. The video is around on here somewhere to prove it.

Weight and momentum don't scale very well. What works in practice on our models is relevant, not necessarily what has to be done on the real thing. If the inclination of the rails is very difficult, if not impossible to see then surely, pragmatism takes priority.

EDIT

From previous comments in other topics we seem unable to guarantee that the wheel and rail profiles we use in 4mm scale are identical to the prototype anyway.
Last edited by Armchair Modeller on Wed Aug 08, 2018 12:51 pm, edited 1 time in total.

User avatar
Le Corbusier
Posts: 942
Joined: Wed Feb 11, 2015 3:39 pm

Re: Do we need 1:20 Rail Inclination

Postby Le Corbusier » Wed Aug 08, 2018 12:32 pm

Speaking purely from a practical standpoint, I am not sure how much I care :? I use exactoscale chairs and exactoscale gauges. By so doing the railhead is gauged correctly and the stock appears to run smoothly. I have built copperclad using society gauges and I assume the track held vertically ... again the stock ran smoothly.Gauge narrowing has been a problem on the chairs ... but I have always assumed that was down to me rather than the system - in one case by using the society gauges which don't allow for the chair incline, and in the other due to insufficient allowance for expansion due to heat (I think). hey ho ;)
Tim Lee

Hardwicke
Posts: 530
Joined: Sun Sep 14, 2008 10:25 pm

Re: Do we need 1:20 Rail Inclination

Postby Hardwicke » Wed Aug 08, 2018 9:12 pm

I've used K&L, C&L, Exactoscale and P4 Track Co, all essentially the same. This was after seeing lots of problems with riveted track that needed cosmetic chairs anyway. Since then I've seen beautiful riveted track and I have made occasionally appalling P4 Track Co stuff, so it depends on the builder. I've always had built in rail inclination. With sprung stock I have to watch the gradients as the wagons roll as easily as the real thing. I still stand by the need to the 1 in 20. If not I might as well play with the Hornby tinplate.

User avatar
Martin Wynne
Posts: 743
Joined: Mon May 14, 2012 4:27 pm

Re: Do we need 1:20 Rail Inclination

Postby Martin Wynne » Wed Aug 08, 2018 9:47 pm

Hardwicke wrote:I don't need to. It is logic. A wheel with a 1 in 20 cone will touch a vertical rail on a knife edge. If both have the same inclination the contacted area is larger.

Which will reduce the contact pressure and make electrical pick-up actually worse. The minimal knife edge contact has a very high contact pressure, which will cut through dirt and crud. And inclining the rail to increase the contact area works only if the inclination angle is exactly the same as the coning angle on the wheel. How likely is that? Otherwise you are back to a knife-edge contact. Wheels differ, rails differ.

But the major reason for not inclining model rails is that it makes it all but impossible make set bends and knuckle bends in pointwork, without the proper conical press tools. Even the prototype has difficulty holding inclined wing rails down flat. That was one reason for the change to vertical rails in pointwork for about 30 years from 1970 to 2000. It was the arrival of cast crossings which allowed a return to inclined rail.

regards,

Martin.
39 years developing Templot. And counting ...

Hardwicke
Posts: 530
Joined: Sun Sep 14, 2008 10:25 pm

Re: Do we need 1:20 Rail Inclination

Postby Hardwicke » Wed Aug 08, 2018 10:24 pm

Point taken Martin,
I'll admit you know far more about track than I do. Actually I realised after I wrote it that the wheel still only has a minimum point of contact. I still believe that it reduces derailments though. And yes, crossings were vertical. I'll find that booklet I mentioned.

User avatar
Le Corbusier
Posts: 942
Joined: Wed Feb 11, 2015 3:39 pm

Re: Do we need 1:20 Rail Inclination

Postby Le Corbusier » Wed Aug 08, 2018 10:28 pm

Martin Wynne wrote:But the major reason for not inclining model rails is that it makes it all but impossible make set bends and knuckle bends in pointwork, without the proper conical press tools.


Really not sure what I'm doing ... but following your advice my set bends and knuckle bends are definitely working and I am definitely using Exactoscale chairs and definitely not using 'the proper conical press tools' :o .... :D
Tim Lee

Philip Hall
Posts: 1087
Joined: Mon Aug 10, 2009 7:49 pm

Re: Do we need 1:20 Rail Inclination

Postby Philip Hall » Wed Aug 08, 2018 10:59 pm

Martin raises an interesting point about contact pressure, which was very important in the days when I still used a HiFi turntable and vinyl records. A stereo cartridge stylus can often have a tracking pressure of just one gram, but the contact pressure at the stylus point is measured in kilos or even greater, which is why a stylus can track complicated music on a vinyl pressing with minimal wear. I could also say that to me it still sounds better, but then I’m showing my age.

I have for some time arranged pickups as a wire bearing on a flange edge, which is a tiny point of contact, and experience has shown that they stay cleaner longer than scrapers on the back of a wheel. I also think that a very smooth tyre surface is better than a rough one, but here I am straying a way from the original topic.

In that vein, the modern rail sections we seem to have, with much sharper edges to the head, coupled with a vertical rail, could render better pickup! And if the prototype at one time used vertical rails and we didn’t notice when we were at at the lineside...

Philip

User avatar
Will L
Posts: 1528
Joined: Sun Jul 20, 2008 3:54 pm

Re: Do we need 1:20 Rail Inclination

Postby Will L » Wed Aug 08, 2018 11:16 pm

Hardwicke wrote:If you let a pair of P4 wheels roll freely along track that has the correct 1 in 20 angle and then try it with wheels and track that don't (00 or P4) you will see the difference it makes.

Will you? I wasn't at all sure, but I agree experimentation is far better than speculation so I thought I'd give it a go. Out came my little test track, built of copperclad sleepers and vertical rails and including a 3'6" reverse curve, a pair of Exactoscale coach wheel and a camera. The result was reliable, easy to reproduce and is as follows.

That I think you will agree is pretty faultless. I was going to repeat the experiment with a bit of ready to run P4 track which I believed was set at 1:20, but as it turns out I coudn't put my hands on the required bit of track, but given the performance above, no I din't think you will notice the difference.
Hardwicke - in a second post wrote:It might be in Russian but shows eloquently why you need the cone AND inclination. Unless you think physics don't apply.

I've used this one myself to try and convince the non believers and heretics that the self steering effect of conical wheels works as well for us as it does on the real railway, but this clip doesn't show the inclination is equal important to the tracking effects. What it does suggest (at about 1:05) is that without the rail inclination it is possible that the weight of the train applied through the coned wheels might be enough to spread the track. That may be true on the prototype, but again this is a problem we wont be troubled by, as, unlike the self steering effect, the weight/material strength relationships doesn't scale.

Hardwicke - back to the first post wrote:The answer is, yes we need it. Otherwise why bother in P4?

So sorry can't agree we need it. The second point is harder, as I tried to say before, I do believe that modelling in P4 is about getting as close as practical to the original. You wont catch me using EM flanges, but if it isn't visible either physically or operationally and including it is causing problems, there has to come a moment when your model will deviate from the original.

I have a booklet from the Research Dept at Derby from 1950 on the behavior of 4 wheeled vehicles on track. Explains a lot. You need the inclination on curves more to prevent derailments).
I'd be interested in seeing what it does have to say.


Return to “Vehicle dynamics and trackholding”

Who is online

Users browsing this forum: No registered users and 0 guests