Better point geometry for OO gauge layouts

[martin_wynne]

1 hour ago, Pacific231G said:

We've had this discussion before and we're not going to agree but I am in any case more interested in the early origins of these terms.

 

Hi David,

 

There is no harm in disagreeing.

 

It's true that the term points (in the plural) has evolved to have different meanings for different people (including the writers of RAIB reports), but not I think for many P.W. engineers.

 

But there can't be much argument that a point (singular) means a single switch blade, and in some cases only the actual tip of it. The early origin of the term being clearly a reference to a pointed object.

 

Hence point-locks, point-rodding, point-levers, clipping the points, etc., all of which make sense only when applied to a movable object.

 

cheers,

 

Martin.

[Trog]

An exception to not using the P word, is catch and trap points, but even then if more detail was required you would describe them as being made up of a single or pair of switches.  For larger formations I would tend to say single lead, crossover, S&C.  Only describing the S&C more fully if the person I was talking to needed to know it was a double junction etc.

[Denbridge]

On 05/08/2019 at 18:36, Foden said:

But are templates available for OO like they are for EM and P4 modellers to work from?

DCC concepts and C&L produce oo templates 

[Pacific231G]

17 hours ago, martin_wynne said:

 

Hi David,

 

There is no harm in disagreeing.

 

It's true that the term points (in the plural) has evolved to have different meanings for different people (including the writers of RAIB reports), but not I think for many P.W. engineers.

 

But there can't be much argument that a point (singular) means a single switch blade, and in some cases only the actual tip of it. The early origin of the term being clearly a reference to a pointed object.

 

Hence point-locks, point-rodding, point-levers, clipping the points, etc., all of which make sense only when applied to a movable object.

 

cheers,

 

Martin.

The term I really don't like is switch and crossing work simply because it's clumsy English. I far prefer pointwork and we don't after all feel the need for point and signal boxes.

I've found it quite informative to compare certain French terms as, unlike the Americans and Germans, they tended to use literal translations of the words used by the British engineers who worked on their early railways, so giving an idea of what those English terms actually were. So you have aiguilles- needles or points, croisement meaning the whole crossing section but coeur de croisement (literally heart of the crossing) for common crossing. Aiguillage  has tended to be the common word  equivalent to points but also meaning pointwork more widely,  More formally, branchement is the equivalent of turnout used by PW and appearing in internal catalogues etc. 

 

Curiously frog meaning a crossing doesn't appear in British Railway Track but it is very often used in the War Office's contemporaneous 1940 manual Notes on Military Railway Engineering pt 2. Curious because, apart from preferring FB to BH rail and, where possible, limiting the range of frog angles, using straight switch rails and un-handed frogs/crossings  to mimimise the range of supplies needing to be held, this immensely practical manual generally follows  British civilian P.W. practice.

We may grumble about Peco's use of a single divergence angle for Streamline but the British military standardised on only three frog angles No.6,No 8 and no 12 but with just No. 8 prefered for most general work apart from congested docks etc with small shunters (No.6) and lines where high speeds were expected (No.12) 

Edited September 22, 2019 by Pacific231G

[St Enodoc]

10 hours ago, martin_wynne said:

 

Except for UK railway modellers, the general public and everyone else using British English (including the OED) correctly refer to points meaning the moving rails which deflect trains from one line to another. i.e. a set of pointed blade rails, and the fittings and working parts which go with them. Hence station announcements such as "train delayed because of frozen points".  The general public and everyone else using British English (including the OED) do not concern themselves with how those working parts are fitted into the rest of the rails.

 

Where they are fitted into the rails to create a junction where one line of track separates into two lines of track, p.w. engineers create a formation called a turnout. Hardly anyone outside the p.w. world needs to know that term. Even railway operating staff barely need to know it because most of the time they are concerned only with the working points at the business end of it. Even signallers for example have levers with labels such as "Points No. 3" with no indication about whether they are part of a turnout, a slip, or any other formation.

 

But railway modellers do need to know the term turnout because model track is usually supplied in such units. In most of the world modellers do correctly use the term turnout for such an item. That's why Peco with a big export market use it. But for some reason, in the UK modellers tend to refer to a turnout as a point despite the fact that a point means something else entirely, i.e. a single point blade.

 

Does it matter? Not really. But it's a shame to use the wrong term when you could just as easily use the right one -- especially if the object of the exercise is to replicate the prototype.

 

cheers,

 

Martin.

That's true enough from a permanent way engineering perspective but others, such as operators and S&T engineers that I have worked with, use the term "points" to describe the whole shebang.

[Dungrange]

3 minutes ago, St Enodoc said:

That's true enough from a permanent way engineering perspective but others, such as operators and S&T engineers that I have worked with, use the term "points" to describe the whole shebang.

 

But, that's because all that matters to the S&T engineers is the moving part - the point blades  - and what position these are.  Like the general public, they have no need to differentiate between the point blades and the common crossing or any of the other S&C components that make up the turnout.  For anyone with no need to distinguish between the different parts of a turnout, then the generic use of 'points' to mean all S&C components is acceptable if imprecise.

[Mallard60022]

On 05/08/2019 at 20:41, Tiptonian said:

 

There is Marcway, still here after all these years. Study their website thoroughly. If you can't find what you want, they custom build what you need.

 

http://www.marcway.net/point.php

http://www.marcway.net/list2.php?col=head&name=Marcway+00+%26+EM+Pointwork

….and they are very good quality and not that expensive at all.

P

 

[Mallard60022]

May have been mentioned earlier but ICBA to check due to the weather so I'll delete this if someone kindly points (sorry) out. Red Leader of Grantham fame has done an article in RM about 'adjusting' Peco Streamline Point geometry. Works a treat, but the timbers are still rubbish positions.

P

[Pacific231G]

On 22/09/2019 at 10:21, Mallard60022 said:

May have been mentioned earlier but ICBA to check due to the weather so I'll delete this if someone kindly points (sorry) out. Red Leader of Grantham fame has done an article in RM about 'adjusting' Peco Streamline Point geometry. Works a treat, but the timbers are still rubbish positions.

P

There was a fairly detailed article in Loco-Revue in Jan 2014 about modifying Peco turnouts to get a realistic spacing between tracks on crossovers. Obviously they don't have a problem with Streamline being to H0 scale and the sleeper spacing and dimensions are closer to French mainline practice than any other RTL track (Tillig and Roco are a bit Germanic in design) Timbering is probably a bit off though since crossovers are different from single turnouts.

I can't post it (copyright) but I can scan and PM it if  anyone wants it. Peco's large radius points have a shallower crossing angle than small and medium but are then curved to get the same exit angle. I think the crossing angle is about 10-11 degrees so it should be possible to lose that curve and get close to a no. 6 turnout. 

Edited September 25, 2019 by Pacific231G
clarity

[Compound2632]

I have posted these elsewhere but they illustate the point @Pacific231G has made about the curve through the crossing on the large radius points. Trimming to get a crossover with 45 mm tack centres, medium radius:

 

 

With the large radius points one ends up with a bit of a joggle rather than a short bit of straight between the two curves, resulting in a less flowing bit of trackwork. It may not be so bad mating a large radius point with a diamond crossing or slip.

 

and large radius (using the bullhead templates):

 

 

 

With the large radius points one ends up with a bit of a joggle rather than a short length of straight between the curves, resulting in a less flowing piece of trackwork. It may not be so bad with a large radius point and a diamond crossing or slip.

 

Edited September 23, 2019 by Compound2632

[Buhar]

HI Stephen

 

Which thread were these posted in previously and, more pertinently, was there any useful ensuing discussion?

 

Alan

[Compound2632]

@Buhar, it was a layout design topic - here and here; I had suggested trimming the points to achieve more prototypical centres, the benefits being outlined here, viz.

• it's (near enough) the prototype dimension
• it gains you 0.5 cm baseboard width
• it gains you a few cm in the length of a crossover - could even be enough to squeeze in large radius in lieu of medium!*
• in trimming the points, you loose the hideous angled sleeper and can replicate the through-timbering often used on the prototype.
• any over-bridges or other structures can be of the correct prototype proportions.

The topic owner was bold enough, or foolish enough, to take up my suggestion and his xuron to his points. 

*Second thoughts on that as discussed above.

Edited September 23, 2019 by Compound2632

[Buhar]

Thank you, Stephen.

 

I've done that in the past with FB Code 75, but without noticing the curve/kink issue and without really getting the timbers well lined up.  Stock ran OK but I'd love to see a way of straightening out the diverging rails.  I wonder if Red Leader's activities on Grantham hold a clue.  I'll have a poke around there.

 

Alan

[Nick C]

The Bluebell's PW newsletter tends to use "points" or "set of points" to describe the whole assembly, though I have spotted "turnout" in one or two places...

[RedgateModels]

On 21/09/2019 at 13:35, martin_wynne said:

 

Except for UK railway modellers, the general public and everyone else using British English (including the OED) correctly refer to points meaning the moving rails which deflect trains from one line to another. i.e. a set of pointed blade rails, and the fittings and working parts which go with them. Hence station announcements such as "train delayed because of frozen points".  The general public and everyone else using British English (including the OED) do not concern themselves with how those working parts are fitted into the rest of the rails.

 

Where they are fitted into the rails to create a junction where one line of track separates into two lines of track, p.w. engineers create a formation called a turnout. Hardly anyone outside the p.w. world needs to know that term. Even railway operating staff barely need to know it because most of the time they are concerned only with the working points at the business end of it. Even signallers for example have levers with labels such as "Points No. 3" with no indication about whether they are part of a turnout, a slip, or any other formation.

 

But railway modellers do need to know the term turnout because model track is usually supplied in such units. In most of the world modellers do correctly use the term turnout for such an item. That's why Peco with a big export market use it. But for some reason, in the UK modellers tend to refer to a turnout as a point despite the fact that a point means something else entirely, i.e. a single point blade.

 

Does it matter? Not really. But it's a shame to use the wrong term when you could just as easily use the right one -- especially if the object of the exercise is to replicate the prototype.

 

cheers,

 

Martin.

 

Of course the Americans have to use different terms, it's Switch and Crossing rather than Points and Crossing

[DCB]

16 hours ago, Compound2632 said:

I have posted these elsewhere but they illustate the point @Pacific231G has made about the curve through the crossing on the large radius points. Trimming to get a crossover with 45 mm tack centres, medium radius:

Don't forget trains only just missed each other on full size railways, several people lost body parts, like heads, by poking them out of carriage windows, and an open door between trains would do a lot of damage. Today some lines have negative clearance as trees brush passing trains. I get away with sub 44 mm on layouts with few issues.

 

 

With the large radius points one ends up with a bit of a joggle rather than a short bit of straight between the two curves, resulting in a less flowing bit of trackwork. It may not be so bad mating a large radius point with a diamond crossing or slip.

 

and large radius (using the bullhead templates):

 

 

 

With the large radius points one ends up with a bit of a joggle rather than a short length of straight between the curves, resulting in a less flowing piece of trackwork. It may not be so bad with a large radius point and a diamond crossing or slip.

 

If you cut the webs between the sleepers on the large radius "Points" you can tweak the curve out of the rails leaving the frog by gripping the rail very firmly with End cutters or nippers.  I reckon this narrowing the gap between adjacent main line tracks  is about the most worthwhile mod you  can do to improve layout appearance, and also some space saving on small layouts. I just wish I had done it 30 odd years ago!

Edited September 24, 2019 by DavidCBroad

[Pandora]

On 09/08/2019 at 22:19, The Johnster said:

They were called turnouts, never points, when I worked on the railway in the 70s.  The other term, which was used for more or less any sort of junction, was ‘shunt’, but this may have been local to the WR.  

 

 

Wherever I work in the country north or south, we  call the 6 foot part of the crossing the shunt . we also use phrases long timbers or long bearers when not describing a typical  sleeper.

Being cruel,  model railway points are little more than representations of  5 mph hand points in yards or sidings. All the important bits and pieces to make them safe and strong for main line speeds are not attempted,for plain line track   150 metres is the tightest curve I have seen, which scales to about  5 to 6 feet in 4mm,  200 metre curves are more common but subject to 10 over 20 mph speeds  and such curves have continuous checkrail on the inner rail.

Edited September 24, 2019 by Pandora

[Pandora]

13 hours ago, Compound2632 said:

@Buhar, it was a layout design topic - here and here; I had suggested trimming the points to achieve more prototypical centres, the benefits being outlined here, viz.

• it's (near enough) the prototype dimension
• it gains you 0.5 cm baseboard width
• it gains you a few cm in the length of a crossover - could even be enough to squeeze in large radius in lieu of medium!*
• in trimming the points, you loose the hideous angled sleeper and can replicate the through-timbering often used on the prototype.
• any over-bridges or other structures can be of the correct prototype proportions.

The topic owner was bold enough, or foolish enough, to take up my suggestion and his xuron to his points. 

*Second thoughts on that as discussed above.

At the points of interference , where trimming  of ends is suggested,  and under the "frog"  ,in 12 inches to the foot  it would probably be one long span of timber known as a long bearer

[34theletterbetweenB&D]

14 minutes ago, Pandora said:

... 200 metre curves are more common but subject to 10 over 20 mph speeds  and such curves have continuous checkrail on the inner rail.

I did once raise this with a once well known P4 group, very aggressively pushing the superiority of their 'pursuit of perfection' , that the absence of a check rail on a somewhat under 3m radius curve with a passenger train operating over it at a scale 40mph did tend to undermine their argument. This was not received with gladness...

[Trog]

2 minutes ago, Pandora said:

At the points of interference , where trimming  of ends is suggested,  and under the "frog"  ,in 12 inches to the foot  it would probably be one long span of timber known as a long bearer

 

Would not use the word long myself S&C timber or bearer, using the word long suggests a longitudinal timber, what the Southern call a wheel timber. But these things all vary from place to place.

[Compound2632]

1 hour ago, DavidCBroad said:

If you cut the webs between the sleepers on the large radius "Points" you can tweak the curve out of the rails leaving the frog by gripping the rail very firmly with End cutters or nippers.  I reckon this narrowing the gap between adjacent main line tracks  is about the most worthwhile mod you  can do to improve layout appearance, and also some space saving on small layouts. I just wish I had done it 30 odd years ago!

 

That would work for a crossing but not for the combination of point and diamond crossing or slip, as the difference in crossing angles comes into play. A quick mock-up showing how far towards the crossing of the slip the curve of the long point extends, for 45 mm track centres:

 

 

More generally, to model a railway, almost everyone has to make compromises on length and curvature. It's the nature of the beast. I don't think folk should be criticised for continuing to make those compromises while endeavouring to work to finer standards.

Edited September 24, 2019 by Compound2632

[martin_wynne]

31 minutes ago, 34theletterbetweenB&D said:

I did once raise this with a once well known P4 group, very aggressively pushing the superiority of their 'pursuit of perfection' , that the absence of a check rail on a somewhat under 3m radius curve with a passenger train operating over it at a scale 40mph did tend to undermine their argument. This was not received with gladness...

 

The problem there is to decide whether it is a model of say a 5 chains prototype curve (in which case it needs a continuous check rail), or a model of a much gentler prototype curve (which doesn't) or even a straight, but which has had to be bent to a sharper radius simply to fit the available model layout space.

 

If you take the former view, it would mean that almost all curves on all model railways would need a continuous check rail -- which would look very odd and not at all prototypical. If you are modelling a typical prototype, continuously checked curves are quite rare in running lines, so ought to be similarly rare in models to look right. Perhaps a token one on the sharpest curve on the layout, and not on other curves.

 

cheers,

 

Martin.   

[The Johnster]

We are getting down to the fundamental root issues of compromise in railway modelling.  It is to all intents and purposes impossible to model a railway to scale in terms of curvature if it is to incorporate trains running at scale speeds above about 15mph in the spaces we have available, and this is before we even begin to consider the core problem of 00, that the track gauge is miles (all right, 2.85 mm) out.  Manufacturers of RTP track must perforce adopt an unprototypical fixed geometry for it, and even where flexible plain track is available, the points and crossings must be to a fixed and rigid geometry.  Only hand made track can overcome this, irrespective of scale or standard.

 

P4 is an attempt to overcome the incorrect track gauge of 00, ti 'get it right' and generally requires handbuilt turnouts and crossings, but the space restrictions are the same, and curvature is still much too sharp in scale terms for running at even moderate scale speeds.  The use of scale couplings and sprung scale buffers imposes a lower restriction on the curvature; when I laid track for my layout I was still using scale couplings and imposed a restriction on myself of Peco medium except for the small of the loco release crossover; my 'reversion' to t/ls came upon my realisation that I was no longer steady handed or had the eye-hand co-ordination to manage the scale couplings, and have been able to save space with setrack turnouts and curves in my fiddle yard..

 

I submit that this sort of compromise is typical of what most modellers do; if a situation is forced upon you it is reasonable to exploit any advantages (in my case, 3 extra fiddle yard roads) that the new situation affords.  The train set curves are kept off the 'scenic' part of the layout, a more or standard 00 effort using largely RTR stock.  i do not claim it to be scale, but try to 'get it right' to the greatest possible extent within the standards and specifications I have adopted.  Apart from the points, there is no dead straight track anywhere on the 'scenic' area of the layout, but no curvature sharp enough to require check rails.  Everything follows a very gentle curve of about 12' (I've never bothered measuring it) radius, which creates an optical illusion of greater length and gives me the 'look' I want.  I've decided on a line speed of 40mph for the branch, but trains on the scenic area never get up to or down from as fast as that!  

 

A P4 layout has reasonable pretensions to being 'scale', which means that curves of a certain scale sharpness should have check rails, but it is not unreasonable to leave them off if you want to represent a 'faster' curve.  

[Pandora]

One type of S&C is the switched diamond,  how many modellers  have ever noticed one, it resembles a fixed diamond but has two sets of motorised switch rails which move in correspondence to set the route through the unit,  there are plenty of them to be seen in the South

Edited September 25, 2019 by Pandora

[Clive Mortimore]

2 hours ago, Pandora said:

One type of S&C is the switched diamond,  how many modellers  have ever noticed one, it resembles a fixed diamond but has two sets of motorised witch rails which move in correspondence to set the route through the unit,  there are plenty of them to be seen in the South

Before the 1977 refurbishment of the throat at Kings Cross there was a curved double slip with a switched diamond where the locos left Passenger Loco before entering the tunnel and then backing on to their trains. I did attempt to draw it, for a now abandoned model of Passenger Loco. It had a 1 in 5 crossing at the station end and a 1 in 8 at the tunnel end. The K crossings were 1 in 6.

 

I never got as far as putting the timbers in, or the check rails etc.

 

I am sure Martin or one of the others with more knowledge than me will correct me. Diamond crossings of a greater crossing angle than 1 in 8 have switched diamonds.  Not often modelled as 1 in 7 crossing angles create longer points/turnouts than most modellers seem to be able to fit on their train sets.

 

Says me with my two layouts with Peco track