16.5mm traditional OO gauge. Classic steam era pointwork.

[gr.king]

All of these templates generated by rules relating to switch geometry and crossing angle make an interesting comparison with the dimensions arrived at mathematically, purely by considering what happens when a true 3 foot radius curve diverges from a straight line.

[jim.snowdon]

* I had thought that Peco's use of a 600mm sleeper spacing in H0 scale represented only the busiest modern main lines so  was a bit oversleepered for most steam era lines but I've just found an article in a 1949 edition of Loco-Revue on sleeper spacing that gives 600mm as already then the standard SNCF spacing for main lines (category I & II) with track panels 18 and 24 metres long using 46kg/metre and 56kg/metre FB rail. The interesting comment in the context of this topic is that for Category III lines- branch and freight only lines effectively- there is a great deal of regional variation on sleeper spacing and that the majority of this rail is chaired bullhead (double champignon). I do sometimes wonder whether Peco Streamline was originally modelled in H0 based on French prototypes, perhaps with an eye on European exports.

 

By British standards, 600mm between sleepers is tight, equating to 30 sleepers per 60' (18.188m) length.  In the mid 1960's, only continuously welded track, using 113lb/yd flatbottom rail, came close, at 28 per 60', or 650mm spacing. Jointed track had 24 - 26 sleepers per 60', giving spacings of between 2' 7" and 2' 4" (785 - 710mm) after allowing for the more closely spaced sleepers either side of the fishplates.

 

Full size sleeper lengths in the UK are 8' 6" for plain track, however if 00 gauge track is to be given naything like scale appearance, the sleeper length needs to be shortened so as to keep the overhang outside of the rails proportinately correct, otherwise the track will look visibly narrow gauge.

 

Jim

[martin_wynne]

All of these templates generated by rules relating to switch geometry and crossing angle make an interesting comparison with the dimensions arrived at mathematically, purely by considering what happens when a true 3 foot radius curve diverges from a straight line.

 

That is the way toy set-track is designed, but not the prototype.

 

Easily done in Templot as a gaunt turnout having zero gaunt offset. Here is one such, 36" radius, 1:5.21 RAM curviform, 16.5mm:

 

 

 

No such prototype exists or could exist, but don't let that stop you building one.

 

Martin.

[MikeTrice]

Out of interest has anyone questioned the quoted radius of SMP plastic based OO turnouts? Although quoted at 36" radius my measurements suggest they are closer to 33" radius.

[AndyID]

Hi Mike,

 

If using Templot to create commercial track comparable to Peco it is better to use all the available adjustments.

 

Using A and B switches does not give the easiest radius with crossings shorter than 1:7, especially in the 00 gauges. Size A-5 in particular is an unpleasant size which is not suitable to represent anything other than itself -- in other words a turnout intended for cramped goods yards and industrial sidings.

 

Try changing to the 9ft straight switch in Templot, and you will find not only easier radii, but also save a bit of length. You can ease the radius further by changing from a "regular" type V-crossing to a "generic" type of crossing.

 

You can ease the radius even further by using the non-prototypical "short model switches" in Templot. Which would be entirely justified if comparing with Peco.

 

To match toy set-track in Templot, use a "gaunt" turnout and set the gaunt offset to zero.

 

I suggest templates 1 and 2 below as suitable for Peco-type commercial track:

 

 

Martin.

 

 

Hi Martin,

 

Why follow American practice with "square on" timbers? Equalized timbers look much better because they draw less attention to the large crossings angles.

 

Cheers!

 

Andy

[Grovenor]

 

Why follow American practice with "square on" timbers?

Because it is also British practice for most post group and BR designs.

Regards

[martin_wynne]

Why follow American practice with "square on" timbers? Equalized timbers look much better because they draw less attention to the large crossings angles.

 

Hi Andy,

 

It's not following American practice. Both timbering styles are common UK practice and available in Templot. There are prototypes and reasons for both styles, which I have written about at length on several occasions, so not again, please.

 

I used square-on here because this and similar topics are preoccupied with the timber sizes and spacing, which are easier to see and compare when square-on. I agree that if modelling these short turnouts I would used an equalized (skewed) timbering layout. But no-one here is modelling track -- they are trying to represent running-line turnouts symbolically with much shorter turnouts. It's not an exercise in replicating the prototype.

 

Martin.

[Joseph_Pestell]

I think I can cope with people calling the components, and the whole, anything they like, so long as the overall appearance is markedly better that the current leading ready made offering.

 

I keep thinking about practical possibilities for minimum effort home-built points/turnouts, as well as the "we want a better ready made product" campaign. Anybody got any winning strategies to suggest on that front?

 

Yes.

[jim.snowdon]

That is the way toy set-track is designed, but not the prototype.

 

Easily done in Templot as a gaunt turnout having zero gaunt offset. Here is one such, 36" radius, 1:5.21 RAM curviform, 16.5mm:

 

 

 

No such prototype exists or could exist, but don't let that stop you building one.

 

Martin.

Martin,

 

Such turnouts do exist in the prototype - I had a yard full of them on Croydon Tramlink when I was the Engineer for that system. That said, they were nominally 25m radius, with a crossing angle of a little over 1:2, fully curved through the turnout leg.

 

Even without going to such extremes, with curved switches, there is a natural switch/crossing combination that will result in a constant curvature from switch tip to crossing. These are provided for in the design handbooks, although by no means always used.

 

Jim

[martin_wynne]

Even without going to such extremes, with curved switches, there is a natural switch/crossing combination that will result in a constant curvature from switch tip to crossing. These are provided for in the design handbooks, although by no means always used.

 

Hi Jim,

 

Indeed. But not tangential with the main at the switch tips (points). The points must always be blunted back from the tangency position by some means or geometry. And the turnout curve is often non-tangential with the main. I do have the design handbooks.

 

Gaunt turnouts do not have a switch of any kind. There is intended to be a gaunt offset to match the gauntlet track. Setting the offset to zero produces the toy set-track result shown.

 

Martin.

[gr.king]

There will inevitably be those who disagree, but I think Mike's P4 A5 re-scaled to 16.5mm gauge, or in other words a true HO scale throughout, turns out to be a more helpful illustration than I might ever have hoped. The effective plain sleeper length comes out around 30.5mm, mid-way between current Peco and SMP / BRMSB, while the average spacing of the 3.5mm wide timbers is close to the 8mm figure I was suggesting for the hypothetical bronze / silver specifications for improved ready-to-use points. I believe currently that the illustration below, arrived at simply by blacking in the timbers on Mike's original, gives good reason to think that such "HO British steam era" points could be visually acceptable alongside either current Peco or SMP Scaleway plain track.

 

Proportions within the point are correct, the slightly greater number of timbers compared to true 4mm scale helps to avoid emphasis on the stubbiness of the turnout, and if the only way to get something better were to be via reliance on Peco then the production of the "bronze" specification ought to appear to them to be a relatively painless modest advance on what they've been making for a long time.

 

[jim.snowdon]

Hi Jim,

 

Indeed. But not tangential with the main at the switch tips (points). The points must always be blunted back from the tangency position by some means or geometry. And the turnout curve is often non-tangential with the main. I do have the design handbooks.

 

Gaunt turnouts do not have a switch of any kind. There is intended to be a gaunt offset to match the gauntlet track. Setting the offset to zero produces the toy set-track result shown.

 

Martin.

Martin,

 

So do I, and not just the PWI Manuals. Some very fine entry angles, sufficiently close to tangential to be regarded as thus for our purposes, can be obtained using undercut switches. However, since that type of switch geometry is way beyond the practical limitations of all but the most well equipped model engineers, I do not propose going any further. Suffice it to say that the variation from tangential are sufficiently close to zero to be become less significant than the variations incurred by wheel wear and wheelsets running off-centre.

 

The reference to "gaunt turnouts" was new to me, although now you have put it in the context of gauntlet (or interlaced) track it is clearer. It isn't a term that I came across in full size trackwork, particularly in tramways, where interlaced turnouts are far from uncommon.

 

Jim

[AndyID]

Hi Andy,

 

It's not following American practice. Both timbering styles are common UK practice and available in Templot. There are prototypes and reasons for both styles, which I have written about at length on several occasions, so not again, please.

 

I used square-on here because this and similar topics are preoccupied with the timber sizes and spacing, which are easier to see and compare when square-on. I agree that if modelling these short turnouts I would used an equalized (skewed) timbering layout. But no-one here is modelling track -- they are trying to represent running-line turnouts symbolically with much shorter turnouts. It's not an exercise in replicating the prototype.

 

Martin.

 

Hi Martin and Keith,

 

I know square-on was common in the UK too - I was being a bit naughty by suggesting it was only us 'mericans that do it that way

 

I wanted to draw attention to the fact that equalized timbering is another option that might help to improve turnout appearance, and that Templot supports it. Just because Peco makes them square on does not mean they have to be that way.

 

Cheers!

Andy

[martin_wynne]

So do I, and not just the PWI Manuals. Some very fine entry angles, sufficiently close to tangential to be regarded as thus for our purposes, can be obtained using undercut switches. However, since that type of switch geometry is way beyond the practical limitations of all but the most well equipped model engineers, I do not propose going any further.

 

Thank goodness for that.

 

The whole matter of modelling such switches has recently been discussed at length on the Scalefour forum.

 

I also have the Railtrack and Network Rail design manuals, in addition to the PWI books, tram track designs and a lot of historical data. So I don't feel that I'm lacking in any information about such matters, thanks.

 

But very little of all that is relevant in this topic. Or even welcomed by some here, who don't regard building track as part of modelmaking, but simply as a means to run trains reliably, in the same vein as building the baseboards.

 

Martin.

[Clive Mortimore]

 

But very little of all that is relevant in this topic. Or even welcomed by some here, who don't regard building track as part of modelmaking, but simply as a means to run trains reliably, in the same vein as building the baseboards.

 

Martin.

Now I got told off when I suggested some on this thread think we are not real modellers because we don't build our own track.

 

I cannot wait to see your scratchbuilt 1/76th scale figures.

[martin_wynne]

I cannot wait to see your scratchbuilt 1/76th scale figures.

 

You have lost me there Clive. I don't see the relevance of that to track?

 

Several posters here have said that they use Peco track because it is cheap and reliable, and the actual appearance barely comes into it. Which is fair enough for their railway.

 

Martin.

[Clive Mortimore]

You have lost me there Clive. I don't see the relevance of that to track?

 

Several posters here have said that they use Peco track because it is cheap and reliable, and the actual appearance barely comes into it. Which is fair enough for their railway.

 

Martin.

If you go back a few pages, I stated that there were people who posted on this thread who said that those who do not make their own track are not model makers. To which there was a big NO, sorry but "Or even welcomed by some here, who don't regard building track as part of modelmaking, but simply as a means to run trains reliably, " is a put down on those who don't make their own track.

 

I build my own locomotives, rolling stock, road vehicles, buildings, and scale figures. Even the chips on the hot plate in the canteen at Pig Lane Depot are scratchbuilt. I don't build my own track anymore but use Peco as it it more reliable than what I can build. Does that make me less of a modeller? Hence my challenge to see your scratchbuilt scale figures.  Model making is more than track building.

[Pacific231G]

 

* I had thought that Peco's use of a 600mm sleeper spacing in H0 scale represented only the busiest modern main lines so  was a bit oversleepered for most steam era lines but I've just found an article in a 1949 edition of Loco-Revue on sleeper spacing that gives 600mm as already then the standard SNCF spacing for main lines (category I & II) with track panels 18 and 24 metres long using 46kg/metre and 56kg/metre FB rail. The interesting comment in the context of this topic is that for Category III lines- branch and freight only lines effectively- there is a great deal of regional variation on sleeper spacing and that the majority of this rail is chaired bullhead (double champignon). I do sometimes wonder whether Peco Streamline was originally modelled in H0 based on French prototypes, perhaps with an eye on European exports.

 

By British standards, 600mm between sleepers is tight, equating to 30 sleepers per 60' (18.188m) length.  In the mid 1960's, only continuously welded track, using 113lb/yd flatbottom rail, came close, at 28 per 60', or 650mm spacing. Jointed track had 24 - 26 sleepers per 60', giving spacings of between 2' 7" and 2' 4" (785 - 710mm) after allowing for the more closely spaced sleepers either side of the fishplates.

 

Full size sleeper lengths in the UK are 8' 6" for plain track, however if 00 gauge track is to be given naything like scale appearance, the sleeper length needs to be shortened so as to keep the overhang outside of the rails proportinately correct, otherwise the track will look visibly narrow gauge.

 

Jim

 

600mm spacing of sleepers is indeed tight by British standards which is why I think Peco's current "OO/H0" (i.e. H0)  offering may effectively be an H0 scale model of  European rather than British track. For the commonest wooden sleeper length of 2.6m (8ft 6in)* Peco's 29mm sleeper length is actually very slightly short even in H0 scale (30mm would be closer) and much shorter in OO scale. 

As you suggest the sleeper length is a difficult area. If it's scale length for OO then the greater proportion of the sleeper outside the rail willemphasise the narrow gauge. On the other hand, if it's in proportion to the gauge but spaced to scale then it would be 30mm long corresponding to 8ft 6ins in H0 scale, but slightly wider at 3.3mm than an H0 scale sleeper. I think this is bound to be an aesthetic judgement of what looks most like the full size railways to someone looking at the model rather than a mathematical calculation. I think it's misleading to talk of a "correct" size of British sleeper in OO but the original BRMSB dimensions of 32mm x 3.5mm for "standard" OO (equivalent to 8ft by 10 ¹/₂  inches)  still seems a reasonable compromise.    

 

*The standard dimensions of  a standard gauge wooden sleeper seem to be the same across Europe including Britain

[AndyID]

Here you go. This is an REA semi-curved A + V-5 with a curviform V-crossing and equalized (skewed) timbers. It's constructed with flat- bottom rail soldered to copperclad timbers.

 

Edit: Substitution radius is 31.01 inches

 

 

 

[mightbe]

If the goal is to produce a general-purpose turnout, suitable for mainline down to goods yard, can we use geometries outside the REA logic? A switches are not known for being particularly graceful..

 

A5s etc look best in a cramped goods yard or similar, as the combination of switch and crossing angle results in a severe curve. Martin's suggestion to use the geometry of 9' switches or GWR loose-heeled switches (without being modelled as such) is sound. Because the switch length is shorter, more of the turnout's length can be devoted to a larger and/or smoother curve, which is not to say that it should be a single radius AT ALL. Tangential curves would be a step in the wrong direction. Many turnouts only begin to have a radius after the planing of the switch rails (I would hazard that this is the norm, at least in N. America). 

 

The question of curviform crossings is a tricky one. I would be inclined to simply use regular crossings, as they are probably the most useful to a RTP modeller--they produce good crossovers and can be trimmed without changing the trajectory. Curved crossings can come later in curved turnouts. 

 

Perhaps a successful launch could consist of 4 turnouts--L/R with 36" minimum radius and 1:6 divergence, and L/R 60" with 1:8. Arranged in crossovers, both would give 50 mm spacing*. Matching diamonds, single slips, and wyes can be the first expansion.

 

*NEW IDEA: Crossovers would give 50 mm spacing, but ladder arrangements would produce a larger spacing, mimicking the 10' way used in yards, etc where a man would need to be able to get in between wagons. The effect could be lessened for other uses by trimming, but is a novel feature which could be potentially attractive to those seeking more realistic turnouts. This thinking would do away with Peco's "substitution radius" lark which has little basis in prototype practice.

 

Quentin

[MikeTrice]

There will inevitably be those who disagree, but I think Mike's P4 A5 re-scaled to 16.5mm gauge, or in other words a true HO scale throughout, turns out to be a more helpful illustration than I might ever have hoped. The effective plain sleeper length comes out around 30.5mm, mid-way between current Peco and SMP / BRMSB, while the average spacing of the 3.5mm wide timbers is close to the 8mm figure I was suggesting for the hypothetical bronze / silver specifications for improved ready-to-use points. I believe currently that the illustration below, arrived at simply by blacking in the timbers on Mike's original, gives good reason to think that such "HO British steam era" points could be visually acceptable alongside either current Peco or SMP Scaleway plain track.

 

Proportions within the point are correct, the slightly greater number of timbers compared to true 4mm scale helps to avoid emphasis on the stubbiness of the turnout, and if the only way to get something better were to be via reliance on Peco then the production of the "bronze" specification ought to appear to them to be a relatively painless modest advance on what they've been making for a long time.

It might be worth considering a few other factors here.

 

The scaled down P4 turnout is effectively "HO". This might look OK from the top but when viewed side on, the sleepers would be too close in relation to Locomotives etc sitting on the track. There have been  many examples showing that increasing the spacing of sleepers on Peco track improves its appearance. If you are hopefull of persuading Peco to provide revised moulded timbering for their turnouts it would be just as viable to persuade them to do a "British" spaced version of their flexi track.

 

Going back to what the eye actually sees we know the track gauge at 16.5mm is too narrow. Having scale 8'6" sleepers and just narrowing the gauge would result in too much overhang which is why when reducing the gauge SMP and C&L reduce the sleeper length to 32mm. The sleeper overhang is now approximately correct. When comparing the prototype with this revised OO track the only time the eye will be able to notice differences is when looking down in plan view. It is very rare in the real world that you get to do this, which is why the C&L and SMP track is so popular.

 

If you take my earlier comment regarding Peco doing new mouldings for their flexitrack then it would be possible to persuade them to adopt 32mm length sleepers with more prototypical spacing. The turnouts would then need to be revised to suit. In that respect the A5 - OO/HO Universal timbering might be more appropriate. Time permitting I will try and mock something up later to illustrate.

[PatB]

As you suggest the sleeper length is a difficult area. If it's scale length for OO then the greater proportion of the sleeper outside the rail willemphasise the narrow gauge. On the other hand, if it's in proportion to the gauge but spaced to scale then it would be 30mm long corresponding to 8ft 6ins in H0 scale, but slightly wider at 3.3mm than an H0 scale sleeper. I think this is bound to be an aesthetic judgement of what looks most like the full size railways to someone looking at the model rather than a mathematical calculation. I think it's misleading to talk of a "correct" size of British sleeper in OO but the original BRMSB dimensions of 32mm x 3.5mm for "standard" OO (equivalent to 8ft by 10 ¹/₂  inches)  still seems a reasonable compromise.    

 

*The standard dimensions of  a standard gauge wooden sleeper seem to be the same across Europe including Britain

 

Given that I've seen claims of people mistaking old Wrenn Universal for EM, and given that an awful lot of modellers, even quite advanced ones, are quite happy with the appearance of Peco and, in at least some cases, possibly don't even know it's wrong, there's clearly quite a bit of leeway in such an aesthetic judgement. I agree that the BRMSB compromise is as good as any other.

[MikeTrice]

Try this:

[MikeTrice]

Another drawing showing variations on Peco Code 75 sleeper spacings.

 

7mm as supplied compared with spacing them out to 7.5mm and 8mm (dimensions are to centre line of sleepers):

[Pacific231G]

Going back to what the eye actually sees we know the track gauge at 16.5mm is too narrow. Having scale 8'6" sleepers and just narrowing the gauge would result in too much overhang which is why when reducing the gauge SMP and C&L reduce the sleeper length to 32mm. The sleeper overhang is now approximately correct. When comparing the prototype with this revised OO track the only time the eye will be able to notice differences is when looking down in plan view. It is very rare in the real world that you get to do this, which is why the C&L and SMP track is so popular.

 

If you take my earlier comment regarding Peco doing new mouldings for their flexitrack then it would be possible to persuade them to adopt 32mm length sleepers with more prototypical spacing. The turnouts would then need to be revised to suit. In that respect the A5 - OO/HO Universal timbering might be more appropriate. Time permitting I will try and mock something up later to illustrate.

This sounds like persuading Peco to "see the error of their ways" and return to the OO/BRMSB/British steam era fold when it's time to renew the moulds. That's simply not going to happen because a very large part of the market for Peco's Streamline track is the global H0 world and such a change would immediately alienate those customers. The decision made by Sidney Pritchard many years ago to avoid the OO compromise by working to H0 scale has proved to be very astute which is why such a large chunk of the worldwide market for model railway track is supplied by a factory in Devon (and not China!!)

 

I think the best you could hope for from Peco would be a "traditional steam era bullhead" range (the BH distinguishing it from Streamline for a reason other than scale) but, even assuming their mass production track building technology could handle BH, it would only have sales potential among a probably quite small section of those using OO with almost no export potential* and it's probably also a gently shrinking market as the steam era passes further into history.

 

Such a range would be different from the 83 line range to NMRA HO sptandards which was probably aimed at defending and expanding US sales(where Peco is regarded as a premium brand)  but with a good global sales potential as well because modelling N. American railroads is a popular branch of the hobby outside of America in a way that modelling Britain's railways is generally not.

 

I could be wrong but I think 16.5mm traditonal OO gauge points are a niche product- albeit a potentially quite large niche- so it probably needs a niche manufacturer to produce them. Is Marcway's current £25 for a simple OO turnout an out of reach price compared with the £10-12 for a Peco Streamline  equivalent ?(I've not used Marcway so have no idea of their quality or ease of use) 

 

Coachman, you posted while I was composing and I agree with most of what you say apart from having to go to EM/P4 for a more British look.  I've seen enough layouts including recent ones such as Bradfield Gloucester Square and some of Chris Nevard's as well as many others in the past to prove that, despite the 'orrible compromise, a OO layout can look very good indeed (so long as you don't look at the stock end on) even though if we were starting from scratch we wouldn't start from here!! I'm not sure which of Cyril Freezer's quotes you're thinking of but the one that comes to my mind is that it's not worth getting too concerned about the narrow gauge of OO since that's probably closer to scale than almost anything else from the radius of points to the length of platforms. (and both he and Sidney Pritchard had been EM gauge modellers) 

 

 

*There might be a small market in France where BH was in widespread use and survived later than in Britain but it's a tiny minority of the more fine scale modellers who are probably already adequately catered for by SMP/Marcway/C&L.