3-foot radius curves in P4

Discuss the prototype and how to model it.
Andrew GW
Posts: 14
Joined: Mon Apr 14, 2014 5:19 pm

3-foot radius curves in P4

Postby Andrew GW » Fri Jan 01, 2021 5:51 pm

Happy New Year to all.

Whilst I fully appreciate that the minimum recommended radius for P4 is 4 feet, I'm also aware that a number of P4 layouts successfully use curves down to at least 3 feet, even with main-line rolling stock.

My own layout plan requires a minimum radius of exactly 3'0", 915mm, mainly on hidden sections. I'd appreciate advice please on whether gauge widening is necessary at this radius; and if so, how much. I'll be using 1920's GWR rolling stock, including tender locos and 60' corridor coaches - ensuring sufficient sideplay, bogie movement and coupling freedom.

The section of layout I've built so far uses straight 1:5 turnouts (with GWR loose-heeled switches) with a design radius of 943mm on the turnout roads. The short-wheelbase stock I've completed to date seems to negotiate these OK (I suspect that what derailments do occur are down to faults in my track construction), but at present I have nothing larger to test. I laid these turnouts with a Society triangular track gauge, giving measured gauge widening of up to 0.2mm. However, I'm considering other construction methods for future trackwork, which will require me to define the gauge widening at the design stage.

Many thanks,

Andrew

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

Re: 3-foot radius curves in P4

Postby Will L » Fri Jan 01, 2021 8:47 pm

Andrew GW wrote:Happy New Year to all.

Whilst I fully appreciate that the minimum recommended radius for P4 is 4 feet, I'm also aware that a number of P4 layouts successfully use curves down to at least 3 feet, even with main-line rolling stock.

My own layout plan requires a minimum radius of exactly 3'0", 915mm, mainly on hidden sections. I'd appreciate advice please on whether gauge widening is necessary at this radius; and if so, how much. I'll be using 1920's GWR rolling stock, including tender locos and 60' corridor coaches - ensuring sufficient sideplay, bogie movement and coupling freedom.

The section of layout I've built so far uses straight 1:5 turnouts (with GWR loose-heeled switches) with a design radius of 943mm on the turnout roads. The short-wheelbase stock I've completed to date seems to negotiate these OK (I suspect that what derailments do occur are down to faults in my track construction), but at present I have nothing larger to test. I laid these turnouts with a Society triangular track gauge, giving measured gauge widening of up to 0.2mm. However, I'm considering other construction methods for future trackwork, which will require me to define the gauge widening at the design stage.

Many thanks,

Andrew


We have disused this at some length in the past. See this thread
viewtopic.php?f=96&t=5030
Read far enough and you will get to plots of graphs that should answer your question. What is achievable depends on the fixed wheelbase of you locos and whether or not you can get sideplay into the centre axles.

davebradwell
Posts: 1174
Joined: Fri Jan 18, 2019 3:48 pm

Re: 3-foot radius curves in P4

Postby davebradwell » Fri Jan 01, 2021 9:01 pm

You're pushing your luck with mainline stock on 3ft curves and especially 1 n 5 points which worry me more than the curve and just don't look realistic. It isn't really to do with whether stock can be made to go round them but the side effects - like how much of the model you have to cut away. If you're shunting across 1 in 5s the angles are getting large and chances of just pushing stuff off the track increase. You'll have a job keeping coach gangways together, the extra swing on loco bogies can mean bogie wheels short on brake shoes of leading drivers and so it goes on. Suggest all sprung buffers to avoid lengthening couplings. Gauge widening is part of the P4 spec

I just about get mainline stuff around metre curves but I do have a 3ft rad semi-circle round which 0-6-0s and 0-8-0s pull 21T hopper wagons - no propelling. The final catch is how to lay a true radius with no bits even tighter.

It's P4, Jim, but not as we know it.

DaveB

User avatar
Julian Roberts
Posts: 1389
Joined: Sat Jan 09, 2010 11:33 pm

Re: 3-foot radius curves in P4

Postby Julian Roberts » Fri Jan 01, 2021 9:29 pm

Hi I think this is an overview of the subject - https://85a.co.uk/forum/view_topic.php? ... forum_id=1 - scroll down to my post on 7th November. Basically the triangular tool gives you all you need, except you can't use it easily in parts of a turnout.

The prototype maximum was 3/4", in 4mm scale 0.25mm. The gauges you can get are in increments of 0.1, so you'll get 0.3 max. Steve Hall tried more than this and it was counterproductive.

You'll need every trick in the book to make 3ft radius work (other than SWB stock) I'd have thought as Dave B says - I have no actual experience beyond a 2ft radius test track which works fine, with a dummy turnout (i.e. no operational switch). Gauge widening through the switch of the turnout is one http://www.clag.org.uk/switch-traverse.html, and there's no harm in gauge widening through the crossing though the check rail remains 18.15. The Digest says you shouldn't, but Russ explained to me why that is there. In my opinion the tricky area is gauging the diverging stock rail after the stockgauge.

Terry Bendall
Forum Team
Posts: 2420
Joined: Sun Jul 27, 2008 7:46 am

Re: 3-foot radius curves in P4

Postby Terry Bendall » Sat Jan 02, 2021 9:37 am

Andrew GW wrote:Whilst I fully appreciate that the minimum recommended radius for P4 is 4 feet, I'm also aware that a number of P4 layouts successfully use curves down to at least 3 feet, even with main-line rolling stock


Some people have made this work, others have come to grief. It is not just the radius of the curve but other factors such as the gauge widening and the design of the locos and the stock which others better qualified than me have mentioned in their replies and in the other posts mentioned.

I think it is always better to keep to 4 feet as the minimum radius and if necessary modify the track plan. I appreciate that doing that may mean that what you would like to do may not work at all. There have been quite a lot of instances on here and elsewhere where people have used tighter curves and have found that they cannot be made to work successfully and have then had to undo things or have a major re-think. This of course wastes time, effort and probably some money as well. Just because we can do something doesn't mean we should. :)

Terry Bendall

User avatar
steamraiser
Posts: 558
Joined: Thu Dec 31, 2009 4:49 pm

Re: 3-foot radius curves in P4

Postby steamraiser » Sat Jan 02, 2021 11:08 am

Tim Venton's layout Clutton has run successfully with 3ft radius circles at each end of the layout, using tender and tank locos, a diesel, bogie coaches and a full range of freight stock apart from any six wheelers.
On the locos you will need a smidge more side play on the one or two axles.
My test track has 42" radius curves which does not cause any problem.

My approach in this case would be to draw a curve of slightly less than 3ft. Mark the axle spacing on a piece of straight card. Place the card with for an 0-6-0 the first and last axles on the curve, then measure how far off the curve the middle axle is. That is the amount of side play you will need.

You may also find that a continuous check rail may be needed as per the prototype did on tight curves.

I suggest you lay a piece of test track to a 3ft radius to play with first.

Gordon A

User avatar
grovenor-2685
Forum Team
Posts: 3918
Joined: Sun Jun 29, 2008 8:02 pm

Re: 3-foot radius curves in P4

Postby grovenor-2685 » Sat Jan 02, 2021 11:15 am

It also makes a big difference whether these 3ft radius curves are properly transitioned at the ends, without transitions you will get into problems with couplers and buffers, especially pushing.
See http://www.norgrove.me.uk/GWRtracknotes/R1774A.pdf
Regards
Keith
Grovenor Sidings

User avatar
Tim V
Posts: 2867
Joined: Tue Jul 29, 2008 4:40 pm

Re: 3-foot radius curves in P4

Postby Tim V » Sat Jan 02, 2021 12:26 pm

Having said all that, if you set out to use 3' radius curves, you can get it to work.

Coaches - I used the Keen couplings - these kept the buffers in contact whatever the radius. I believe they are good down to 15". Clutton got to four coach trains, but could easily have gone longer.

8 coupled locos? I've got those down to 2'6", though frankly they weren't very reliable.

The only caveat is - as often happened at shows, some exquisite loco would appear. It would get placed on the track for a photo opportunity, quick run up and down and fall off on the corners. Don't expect visiting stock to perform on those corners. Your own stock, built for those corners will cruise everyone else's layouts!
Tim V
(Not all railways in Somerset went to Dorset)

Enigma
Posts: 533
Joined: Tue Aug 13, 2013 2:49 pm

Re: 3-foot radius curves in P4

Postby Enigma » Sat Jan 02, 2021 2:07 pm

Years ago I had a 'roundy' layout at home 12' x 10' with 3' curves and everything of mine ran OK including RTR coaches with just wheel changes. Mind you, I ran mainly pannier tanks and a small prairie (modified Lima with scratch built chassis) but visiting locos usually went straight on at the curves. Most of these were only built and tested on straight test tracks although I know that now, some of them work fine on continuous run layouts with much larger radius curves.

My current 'plank' has pointwork based on Peco Setrack small radius points and works fine with 0-4-0 and 0-6-0 industrials.

I doubt that a Duchess would get round though....................

Terry Bendall
Forum Team
Posts: 2420
Joined: Sun Jul 27, 2008 7:46 am

Re: 3-foot radius curves in P4

Postby Terry Bendall » Sat Jan 02, 2021 4:46 pm

steamraiser wrote:My approach in this case would be to draw a curve of slightly less than 3ft. Mark the axle spacing on a piece of straight card. Place the card with for an 0-6-0 the first and last axles on the curve, then measure how far off the curve the middle axle is. That is the amount of side play you will need.


What a good idea and very simple to do. Learn something new everyday.

Tim V wrote:Having said all that, if you set out to use 3' radius curves, you can get it to work.


Just goes to show that the "rules" don't have to be followed all the time. :D

Terry Bendall

Philip Hall
Posts: 1947
Joined: Mon Aug 10, 2009 7:49 pm

Re: 3-foot radius curves in P4

Postby Philip Hall » Sat Jan 02, 2021 5:04 pm

Until quite recently, I have arranged most engines to be able to go around 3’0” curves or thereabouts. Most engines will do this, unless (like Guy Williams once said) you want to go around the edge of a soup plate. However, I wouldn’t like to have a big 4-6-0 or the like charging at 60 plus around such a bend.

Since I found myself in the fortunate position of being able to have a minimum of 4’6” radius, many benefits have come about.. Closer tender spacing for one, along with the ability to have engines and passenger stock properly close coupled instead of looking like a loose coupled goods train. And I can contemplate pacifics and long trains at some speed without them looking like Hornby on the carpet in front of the Christmas tree.

But don’t be put off, as Tim says, if you design the requirements in from the start, it can be done subject to some limitations.

Philip

Stephan.wintner
Posts: 109
Joined: Sun Mar 15, 2020 11:04 pm

Re: 3-foot radius curves in P4

Postby Stephan.wintner » Sat Jan 02, 2021 5:22 pm

A perhaps foolish question - north american railroads, both 12"=1' and 3.5mm=1', commonly use spiral easements, particularly where a tight radius is needed. Yet Keith's PDF makes no mention of such. Was that not done in the UK?

Stephan

User avatar
grovenor-2685
Forum Team
Posts: 3918
Joined: Sun Jun 29, 2008 8:02 pm

Re: 3-foot radius curves in P4

Postby grovenor-2685 » Sat Jan 02, 2021 6:01 pm

Spiral Easements = Transition Curves, very much used in the UK but critical conditions for buffer locking generally occur when they are not provided. Hence my mention of transitions in that post. Turnouts and crossovers are common situations where transitions are not used.
NB the latest designs of high speed turnouts in the UK do use transition curves but they are of much larger radius anyway.
Regards
Keith
Grovenor Sidings

Stephan.wintner
Posts: 109
Joined: Sun Mar 15, 2020 11:04 pm

Re: 3-foot radius curves in P4

Postby Stephan.wintner » Sun Jan 03, 2021 12:10 am

Cheers Kieth, I follow your logic. I also found some discussion in scale four digest 21.1.

Stephan

Andrew GW
Posts: 14
Joined: Mon Apr 14, 2014 5:19 pm

Re: 3-foot radius curves in P4

Postby Andrew GW » Thu Jan 07, 2021 12:27 am

Thanks everyone for the replies – all taken on board and all much appreciated. I may well follow Gordon A's suggestion and produce a 3' radius test track first.

It was Tim V’s “Clutton”, a section in Iain Rice’s "Creating Cameo Layouts" (p44-45 "Engineering for Curved Trackage") and passing comments in various Forum discussions that persuaded me to consider curves below 4’. Accepting a 3’/915mm minimum radius, and a maximum train length of 3 x 60’ coaches plus a 40’ brake, lets me condense my overall plan from a U-shape that would obstruct a desk and window (and require extra storage for portable boards) into an L-shape that could be permanently set up without affecting the other uses of the room.

I think 3’ curves do sound feasible, subject to the cautions you’ve mentioned; I am building all stock to suit (and thanks for the extra tips on that). All of the sharp curves/turnouts will also have continuous check rails.

All the 1 in 5 turnouts (which will only be traversed by short stock) and all but one of the 3’ curves are on a section of layout depicting a Weymouth-inspired harbour tramway. I’m reasonably happy about using 3’ curves here, a scale 228’ 7”, since the real Weymouth line went down to 223’ radius. Admittedly passenger stock required special couplings and gangway disconnection to traverse that, but hopefully (noting Dave B and Tim V’s advice) I can avoid this on the model.

My main worry was the remaining 3’ curve, the only one to take “main line” locos (say up to 17’ 6” fixed wheelbase on some 2-4-0’s). This is on a non-scenic “train hoist” in the corner of the “L”, giving access to the storage sidings. I have since managed to ease this to 1080mm, by following grovenor-2685's advice and adding a transition curve on the previously almost-straight scenic section.

The reason for my original query, the value of gauge widening needed, is that I’m considering various CAD/CAM methods to produce both the harbour and non-scenic trackage with maximum accuracy – so as mentioned above, I'll need to include gauge widening at the design stage.

Will and Julian, many thanks for the links. I did search both the Scalefour and Templot Forums before starting a new thread, but somehow missed these. From Will’s plots, it looks as if gauge widening won’t strictly be necessary unless I have less than 0.8mm side play on the longest locos – but that applying protoypical values of gauge widening (up to the 0.22mm max. in the P4 standard) would reduce reliance on side play and might give me more margin.

I’d appreciate it if anyone can confirm the correct length of GWR loose-heeled switch to use in a straight 1 in 5 turnout (235’ turnout road radius). The GWR "Table of Leads Etc. Old Type Switches" (https://www.scalefour.org/downloads/gwr ... /R1755.pdf) shows 9’ switches. However, p45 of David Smith’s "GWR Switch and Crossing Practice" describes how special “9ft/10ft” switches were used where turnout curves went below about 300’; a 10’ switch pair with the turnout-side tongue planed as for a 9’ switch, which modified the stock rail alignment to provide 7/16” gauge widening through the turnout. Were these special switches a later innovation (the table above is dated 1937), or did the decision of whether to apply gauge widening – and hence which switches to use – depend on circumstances, e.g. the stock using the turnout?

Thanks again.

User avatar
grovenor-2685
Forum Team
Posts: 3918
Joined: Sun Jun 29, 2008 8:02 pm

Re: 3-foot radius curves in P4

Postby grovenor-2685 » Thu Jan 07, 2021 11:13 am

Andrew GW wrote:The reason for my original query, the value of gauge widening needed, is that I’m considering various CAD/CAM methods to produce both the harbour and non-scenic trackage with maximum accuracy – so as mentioned above, I'll need to include gauge widening at the design stage.

I suspect ypou are being to finicky here, how are you going to install the rails with CAM? Does this imply you are going to 3D print track bases with the specified gauge widening? Using traditional track laying methods, soldering or glueing then use of a triangular gauge as in all the literature provides the required widening with no effort beyond making sure you use the gauge the right way round. Using pre-made sleeper bases, eg the P4 track co fasttrack then the only option is to use the gauge widened bases.
Regards
Keith
Grovenor Sidings

User avatar
Guy Rixon
Posts: 909
Joined: Sun Mar 27, 2011 6:40 pm

Re: 3-foot radius curves in P4

Postby Guy Rixon » Thu Jan 07, 2021 11:23 am

If I had a CNC mill, I'd consider milling bases for inset track from single-sided PCB, cutting grooves for the running and check rails and gapping the base between the rails at the same time. Assuming --- and it's a big assumption --- that this gives automatically-accurate track when the rails are soldered into the grooves, it would be a big win for me, trading off manual skills which I lack for programming skills which I have. Others might find this a poor trade-off.

Andrew GW
Posts: 14
Joined: Mon Apr 14, 2014 5:19 pm

Re: 3-foot radius curves in P4

Postby Andrew GW » Fri Jan 08, 2021 12:00 am

Keith, you may well be right - I'll only know for certain once I have a test piece complete. I'm thinking along broadly similar lines to Guy (bearing in mind this is for inset track), but slightly different materials and methods. I need a number of identical turnouts and since I'm pushing curve radii to the limit, I'm wary of the issue Dave B raised - with hand building, it' can be difficult to ensure a true radius throughout.

User avatar
Julian Roberts
Posts: 1389
Joined: Sat Jan 09, 2010 11:33 pm

Re: 3-foot radius curves in P4

Postby Julian Roberts » Fri Jan 08, 2021 2:17 pm

Hi Andrew
In case this helps - I couldn't find this till now. This is from viewtopic.php?t=2021&start=25




Martin Wynne wrote:Here's some info which I have posted many times on other forums, for 00 modellers new to trackbuilding, but it applies equally for P4.

Image

A is the check gauge. It is the most critical dimension in pointwork. If this dimension is too small, wheels running from left to right can hit the nose of the vee and very likely derail, or at least bump. If this dimension is too large, the wheel backs will bind or jam on the check rail. To make sure it's correct, the check rail is set using check gauge tools. For 00-SF and 00-BF this dimension should be 15.2mm. You can use the same check gauge tools for both these standards (they are both running the same wheels).

B is the crossing flangeway gap. It's also important. If this dimension is too small, the wheel backs will bind or jam on the wing rail. If this dimension is too large, the gap in front of the nose of the vee will be too wide, and the wheels may drop into it with a bump. This gap is set using a small piece of metal shim called a crossing flangeway gauge shim. For 00-SF it should be 1.0mm thick. For 00-BF it should be 1.3mm thick.

C is the track gauge. It shouldn't be less than the specified dimension, but it can be wider. It is often widened on sharply curved track to ease the running of long-wheelbase vehicles. The track gauge is normally set using roller gauge tools, or alternatively using a 3-point gauge tool, which automatically widens the track gauge on sharp curves. For 00-SF this dimension shouldn't be less than 16.2mm. For 00-BF it is normally 16.5mm.

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

regards,

Martin.


So gauge widening only alters C and gap D which becomes wider as required - the flangeways B do not change nor does the check gauge A. In P4 the flangeway B is 0.68

So if you have GW you need to use the checkgauge tool, not the shim flangeway gauge tool, to set the checkrail the correct distance from the opposite running rails and V

User avatar
stephenfreeman
Posts: 153
Joined: Wed Oct 08, 2008 9:13 am

Re: 3-foot radius curves in P4

Postby stephenfreeman » Wed Jan 13, 2021 8:04 am

Not that I suspect most here are that bothered (it's not P4 or S4) but it's nearly 5 years since Martin wrote it, nobody has questioned it. I know what he means but it is not what says. 00sf is as he has said many times EM-2 therefore A should also be the equivalent of EM-2. A should definitely not be 15.2mm in EM. The check rail gap would be substantial.

davebradwell
Posts: 1174
Joined: Fri Jan 18, 2019 3:48 pm

Re: 3-foot radius curves in P4

Postby davebradwell » Wed Jan 13, 2021 9:11 am

You lost me there - what is EM-2? I looked on the EMGS website but they continue to keep their standards a closely guarded secret. I hope we do better. It can't be the same as 00 or it wouldn't be called EM.....would it?

Returning to 3 foot or other sharp curves, might I suggest you get a curved template machined to make laying a true curve more likely - some of the mechanised baseboard suppliers can do this in mdf. Handy for checking that larger curves don't contain trouble, too.

DaveB

User avatar
johndarch
Posts: 162
Joined: Tue Jan 29, 2019 11:24 am

Re: 3-foot radius curves in P4

Postby johndarch » Wed Jan 13, 2021 9:20 am

Good point Dave. Consistent radii and transition curves will go a long way to curing problems and not just on small radii.

Paul Cram
Posts: 75
Joined: Sun Jun 09, 2019 10:17 am

Re: 3-foot radius curves in P4

Postby Paul Cram » Wed Jan 13, 2021 10:21 pm

davebradwell wrote:You lost me there - what is EM-2? I looked on the EMGS website but they continue to keep their standards a closely guarded secret. I hope we do better. It can't be the same as 00 or it wouldn't be called EM.....would it?

Returning to 3 foot or other sharp curves, might I suggest you get a curved template machined to make laying a true curve more likely - some of the mechanised baseboard suppliers can do this in mdf. Handy for checking that larger curves don't contain trouble, too.

DaveB

It is OO with the gauge reduced to 16.2mm through the crossing giving a flangeway gap the same as EM i.e 1mm hence EM -2.

User avatar
Martin Wynne
Posts: 1172
Joined: Mon May 14, 2012 4:27 pm

Re: 3-foot radius curves in P4

Postby Martin Wynne » Wed Jan 13, 2021 10:40 pm

davebradwell wrote:You lost me there - what is EM-2?


https://85a.uk/00-sf/history.php

https://85a.uk/00-sf/

Martin.
40+ years developing Templot. Enjoy using Templot? Join Templot Club. Be a Templot supporter.

davebradwell
Posts: 1174
Joined: Fri Jan 18, 2019 3:48 pm

Re: 3-foot radius curves in P4

Postby davebradwell » Thu Jan 14, 2021 9:25 am

Get it, the dash is a minus, not a dash. Obvious really. We do tie ourselves in knots with wheels standards, though and it's some sort of national characteristic. Thank you. Long may we remain free of such mutations in our society.

DaveB


Return to “Track and Turnouts”

Who is online

Users browsing this forum: ClaudeBot and 1 guest