Minimum Curve/Radius- Any Truth To It?

[ProjectRedcar]

Been wondering since my layout is planned to use a lot of Radius 1 curved track pieces (I'd say 90% of curves on it are Radius 1). However, when I look at what the manufacturers say, they state a minimum curve, usually 2, for the models I have my eye on.

 

Is it a matter of covering themselves in case a more extreme radius ends up breaking the model? Or have they actually tested it in-house?

 

Also, if I have a Class 37 from Bachmann and then one from, say, Accurascale, both OO gauge and both the same scale, is it reasonable to assume both of them will make it around the same curves?

 

Any help would be appreciated. If it ends up that I have to change all my track to be Radius 2 or higher, I will. 

 

And for reference, on my layout, models will run at low speeds (about 10-20 MPH in reality, tops) around those Radius 1 curves.

 

Cheers

[McC]

Most 4 wheel wagons will handle r1 and some stock with bogies but locos with longer wheelbases will simply struggle or derail at r1 (as well as looking very unrealistic). Generally r2 is a minimum for all but small shunters. 

[Michael Hodgson]

Most recent locos require radius 2 or better (the odd one requires even better).

You will usually get away with radius 1 with some short wheelbase locos and 4-wheel wagons.

And no, you can't assume that the same loco by different firms will be equally tolerant

 

oops McC beat me to it!

Edited October 8, 2022 by Michael Hodgson

[ProjectRedcar]

3 minutes ago, Michael Hodgson said:

Most recent locos require radius 2 or better (the odd one requires even better).

You will usually get away with radius 1 with some short wheelbase locos and 4-wheel wagons.

And no, you can't assume that the same loco by different firms will be equally tolerant

 

oops McC beat me to it!

 

4 minutes ago, McC said:

Most 4 wheel wagons will handle r1 and some stock with bogies but locos with longer wheelbases will simply struggle or derail at r1 (as well as looking very unrealistic). Generally r2 is a minimum for all but small shunters. 

Thanks!

 

I'll update them all to r2 at a minimum

[The Johnster]

The RTR manufacturers’ minimum radius spec assumes perfectly laid and level track and their own locos/stock, which will have been tested during development on their in-house test track layouts, see various TV output concerning Hornby.  This TTBOMK includes running through reverse curves and pointwork of the specified radius in both directions at a medium speed, but not AFAIK hauling or propelling stock unless the model is a multiple unit, happy to be proven wrong on that point.  
 

Of course, such testing is representational and cannot take account of production/assembly variations or QC issues on individual models; back to backs on brand new models vary considerably in my experience, and I check all mine first thing out of the box.  They can stilldrift over time, presumably a function of temperature-related expansion and contraction as well.  
 

So you should be ok hauling trains on reasonably well laid track of the specified radius tolerance of the stock, but I would be less confident about propelling, especially at speed through reverse curves, and my recommendation would be ‘minimum radius +1’ for this.  Best practice is to use the largest radii the space will allow in the interests both of reliable running and of realism, but of course there’s never enough space, and the inevitable compromise has to be between space, realism, and what will run reliably. 
 

And there is little more frustrating than a layout that doesn’t run reliably…

Edited October 8, 2022 by The Johnster

[Huw Griffiths]

I don't think it helps that a number of "toy train sets" include significant amounts of 1st radius curves.

 

OK - I suspect that, even if the locos / stock included in the sets are of some length, they might well be older models (perhaps "set up" to go round tight curves) - but you might not always find too many "full fat" versions of long locos, rolling stock etc in the more affordable train sets.

 

For example, it's not exactly unknown to find "local pick-up-goods" themed sets, with a very short loco (whose design might sometimes be of distinctly dubious origin) and a handful of short, 4 wheel, wagons. There have also been passenger themed equivalents, featuring the "celebrated"* short 4 wheel coaches. I haven't noticed too many sets featuring the 1960 "Evening Star", or multiple units of classes 156, 158 or 159.

 

(*OK - different people have their own opinions regarding the short "traditional" Hornby 4 wheel coaches. As I'm a member of DEMU, I've got other reasons for not buying them in bulk - let's face it, they don't exactly reflect the "postwar" railway scene ... !)

 

[The Johnster]

These are the Hornby ‘Smokey Joe’ family, and are of less dubious origins than some of us like to think, though admittedly not well modelled.  ‘Smokey Joe’ is a Neilson pug, as operated by the Caledonian and the North British, largely as dock/harbour shunting engines, and as used by industrial owners.  The full Caley livery is a bit fanciful, I doubt such lowly beasts were given this treatment even when they were new, but the unlined LMS is ok. 
 

GWR 101 has a prototype as well, an experimental proof-of -concept oil-fired loco for the Wrington Valley Light Railway, but which never left the confines of Swindon Works.  The G W R initials livery is far too late for it, and TTBOMK it carried a plain unlined green livery.  The Dowlais 0-4-0 was real as well, designed and built in-house at Dowlais Steel Works in South Wales.  The other 0-4-0 steam locos are fictitious, but the diesels are based on the BR North British TOPS Class 06 and the Bagnall on an industrial type from that company.  
 

But none of these are good models.  The driving wheels are too large in diameter cf. the prototypes, and the Neilson is too long.  The outside cylinder locos have dreadfully crude and toylike representations of the slide bars and crossheads, and the diesels should be jackshaft driven.  The train sets they feature in are highly unlikely; few of them had vacuum brakes to work passenger stock, even the 4-wheelers (based loosely on a Somerset & Dorset joint prototype, apparently) and the freight wagons they are paired with are nothing like the types they worked with typically in reality, being chosen for colourfulness and play value.  
 

That, of course, is their real purpose, toy train sets with no real intention of even attempting to represent any sort of prototype reality in model form.  No.1 curves are used to fit the circuit of track on the dining room table and everything is put away for tea to be served.  0-4-0s are suitable for this because they are easy to put on the track for small children, and we are not the target market.  Imaginary liveries or tie-ups with biscuit or soft drink companies provide colourful locos and colourful branded vans and wagons for them to pull.  Fidelity to prototype is irrelevant. 
 

Which is a pity, because, basic and crude though these offerings are, if they were closer to scale size and had more appropriately sized driving wheels (which they could have been designed to be whilst still ticking the ‘single piece body tooling/cheap to produce’ box), there would have been a possibility of working them up into acceptable models, with details added by the modeller.  
 

This market had been addressed in recent years with industrial 0-4-0s and now 0-6-0s to ‘full fat’ specifications, and a price consistent with that approach, and since I have been critical of Hornby’s starter set 0-4-0s here it is only fair that I point out that they were the first RTR company to produce an 0-4-0 to ‘full fat’ scale and detail spec and high running standards, the W4 Peckett.  Dapol had a fair shot with the L & Y pug, but were unable to keep the motor out of the cab and the running was only ‘as good as you’d expect’ from a model of that size and era.  But there are now high quality locos capable of being used on no.1 curves!

 

Now, if only someone would produce an RTR full fat spec jackshaft drive 4-wheeled diesel…

 

 

[DCB]

The Hornby Dublo models from 1938 used 15" Radius or or very nearly 1st Radius,  Triang used 13.5 " Radius and even the Triang Hornby 9F gets round 13,5" It explains the yawning gap between their engines and tenders. Playcraft  I also used  sub 1st  Radius. My current project uses 12" radius but I am modifying just about every bit of rolling stock.

Most modern RTR can't get round 1st Radius, most can get round 2nd radius plain curves though some is marginal especially when new, but some recent RTR is very iffy on 2nd radius set track points, My Heljan 67 and Bachmann 64XX derailed constantly when I tried them on Hornby set track points which older stuff sails over.  For modern RTR I would use 2nd Radius as a minimum curve and 2ft as minimum point radius to be sure. 

 

Edited October 8, 2022 by DCB
Change a lot to Some

[Phil Himsworth]

I have a Heljan class 27 that didn't like the inside line of a curved point in a second radius corner; the bogies could not turn enough due to details hanging down from the loco body.

 

I have fixed it by committing the crime of trimming those hanging details slightly so it worked.

 

(Then I removed the offending point completely, mind, as I had another loco that wouldn't go round it at all either)

[DK123GWR]

3 hours ago, DCB said:

a lot of RTR is very iffy on 2nd radius set track points

Usually I would agree, but somebody (I think it was actually you) once suggested shortening set track points by cutting the rails on the two diverging routes in order to save move the tracks closer together. The few modern locos* I have tested on an Inglenook which uses points modified in this way have usually not had problems with th radius (and the track is very poorly laid) so this setup may be suitable for a depot layout. I would guess that minimising the length of the curve (when it's already shorter than the loco) reduces the amount that bogies need to pivot, which may help.

 

*Vi Trains 47, Bachmann 08, new tooling Hornby King, Castle, and 61xx

Edited October 8, 2022 by DK123GWR

[Harlequin]

@ProjectRedcar You might like to share yhour proposed trackplan in the "Layout & Track Design" area to get some feedback before you commit too far because it might be possible to ease the radii even wider to improve both the look and running qualities.

 

Edited October 8, 2022 by Harlequin

[Jol Wilkinson]

Quite simply, use the biggest radius you can, it looks better.

[RobinofLoxley]

In case there is any doubt about what is happening think about a 3 axle item rolling down straight track towards a curve. Each axle has a certain amount of sideways movement, and as they enter the curve they are displaced by the curvature of the track. The sharper the curvature the more displacement occurs so that by axle 3 the first one may be so far displaced that axle 3 can't follow and so one or other pop out of the track. It can be worse on turnouts as the track isn't completely parallel through them. The locos are checked to be in spec for axle movement and wheel spacing but really it's done at the design stage using the track radius as a limiting parameter. 

 

This is why axle separation is important and why simpler linkages have more chance of getting round. 

 

My layout would be perfect for 2-8-0's and 2-10-0's but I wouldn't rely on them getting round my second radius half circles.

[Steamport Southport]

Funnily enough I was looking at Hornby train sets and apart from those with the small 0-4-0Ts they are all either 2nd or 3rd radius.

 

Mainly because they sent me an email about this and was pondering whether we should get one for my nephew.

 

https://uk.Hornby.com/products/tornado-express-train-set-r1225m

 

This is third radius.

 

https://uk.Hornby.com/products/flying-scotsman-train-set-r1255m

 

 

TLDR: If you want big locos don't use first radius.

 

 

Definitely not if you want an 18000 gas turbine, ISTR that struggles to get around third radius!

 

 

 

Jason

[ikcdab]

Nothing else impacts a layout design or track plan as much as the minimum radius of the curves.

R1 is 371mm which scales up to 28m radius in real life. 

The minimum radius on London underground is 60m and (I think) 200m radius on mainlines. So you can see how super unrealistic R1 really is.  

Ian

 

[The Johnster]

The biggger the better when it comes to curvature radius, though I appreciate that the space issue is contradictory to this.  Looks more realistic and runs better.  A no.1 setrack curve scales out to somthing like a chain and a half, since real curves are traditionally measured in chains.  There are 22 yards in a chain.  Such a prototype curve may exist in industrial or dockyard sidings, but would only be used at very low speeds by very short fixed wheelbase locos with very big buffers.  No.4 setrack radius, 'express' would in prototype use be very severely speed restricted, we are probably talking about less than 15mph.  

 

It was quite early in my modelling life, my mid teens in fact, that I decided that I couldn't be doing with any of this malarky because it looked unrealistic and the horrible huge gap between buffers upset me; this was the 60s and that was what RTR was like in those days, and there were only 2 radii of which turnouts were only available in the smaller one.  It was about the same time that magazines introduced me to the 'layout on a shelf', fy-terminus or fy-station-fy, which released me from roundyrounds; I accept that this does not suit everyone.  I 'graduated' to using flexible track mainly in order to lay larger radius curves for better running and realism; I wanted to get into kits, which were mostly incapable of running on setrack anyway.  2' radius was the recommended minimum, and experience soon taught that this is very close to the limit of what flexi will tolerate before you start pulling out of gauge or tearing the rail out of the chairs.

 

There is a danger with flexi that one can lay a curve to an average radius without being aware that it can be very much sharper in the central sections, and I would advise anyone laying it for the first time to be aware of this potential issue.  The geometry is different as well, with rnning lines closer together.  When you lay the track, before you fix it in it's final postition and certainly before you ballast it, take your longest vehicle with the largest overhang at the ends and underhang in the centre, usually a bogie coach, and tape felt tip pens to the corners and centre position so that the tips are resting on the baseboard.  Then push it gently around the layout so that the pens leave traces on the boards; where they intersect will be where stock will foul on opposite lines or at the ends of sidingns, and will show the clearance you need on curves to avoid collisions plus a safety margin of half the width of the pen ot accommodate wobble at speed and any small details that stick out like door handles.  The pen marks can also be used to determine clearance for platforms and lineside detail.

[AndrueC]

I model in N but I can say that my 4-6-2 Queen Elizabeth was a bit uncomfortable on 2nd radius curves and struggled on Settrack turnouts which in N are first radius. I also have a Hall which although rated at 2nd radius or higher was unable to negotiate my 2nd radius curves. I also had some JNLA wagons which despite being double bogied couldn't handle my turnouts.

 

So for N at least 2nd radius should be considered the absolute minimum. My new layout uses Streamline throughout and I've taken care to ensure that the tightest point of the tightest curve is 3rd radius.

 

As for ensuring the radius of curves I designed a plan using software that warned me if part of a curve was too tight. Then I printed out my track plan 1:1 and glued it to the baseboard. Then I glued my track to that.

Edited October 8, 2022 by AndrueC

[The Johnster]

27 minutes ago, ikcdab said:

Nothing else impacts a layout design or track plan as much as the minimum radius of the curves.

R1 is 371mm which scales up to 28m radius in real life. 

The minimum radius on London underground is 60m and (I think) 200m radius on mainlines. So you can see how super unrealistic R1 really is.  

Ian

 

 

Quite.  This is why buffer locking becomes a problem with scale couplings, and isn't when scale dimensions are used as in P4.  The restricting factors when you are drawing up your trackplan are firstly, as you say, mimimum curve radius, and secondly, but not by much, is the length of your longest locomotive for run around loco release headshunt purposes.  Thirdly, and again not by much, is clearance for your longest rolling stock, and everything else is just cosmetics... 

 

General advice for good running; flat and level baseboards, good alignment at baseboard joins, care in track laying, avoid reverse curves without a section of plain track between them as much as you can (check out the famous 'Minories' track plan to see how this can be done with turnouts), avoid changes of gradient close to turnouts especially facing turnouts close to summits, avoid curvature on gradients.  If you have to use sharp curavture, use setrack rather than stressing flexi, and use RTR gradient blocks which will avoid the corkscrew effect problem.  Do not, unless your woodworking is of cabinet making standards, attempt to make a helix for setrack radii, it won't work, shell out for a preofessionally built one despite the price, it's worth it.  Ensure that you can get to all track easily for cleaning purposes, and don't rely on track cleaner wagons in inaccessible tunnels; in fact, don't have inaccessible tunnels...

 

Check the back to backs on all locos and stock when you acquire them, don't rely on factory settings.  Replace any plastic wheels with metal; your locos will haul more and much less crud will be spread around the layout.  Ensure that railheads, wheel rims and pickup surfaces, and pickups are clean, and if you are using setrack turnouts or insulfrog Peco, be aware that short wheelbase locos may not be able to cope with the longer dead frogs of large radius or curved turnouts.  Ensure that all pickups are properly adjusted across the full range of that axle's sideplay, and if there are any sprung axles (Bachmann panniers, 56xx, Hornby generic Jinty chassis) that the spring is adjusted so that all the pickup wheels sit properly on the railheads on level track.  Avoid traction tyres, they are Satan's expectorant.

 

If you are using tension lock couplers, try as far as possible to standardise on one make for all stock across the layout to avoid any issues such as overriding caused by different bar profiles and buffer lock.  Try to ballast vehicles to a consistent axle load, about 25 g per axle is generally recommended.  You will find this difficult on wagons with low floors such as Lomwacs, Weltrols, Bogie Wells and so on, and on modern intermodals with very open and visible underframe detail.

[RJS1977]

Although most models are marketed as "recommended minimum 2nd radius curves", some of those will manage tighter curves - though it may put more wear on the mechanism. For example, I have seen a Bachmann Class 70 negotiating 1R curves on one of the layouts at Horsham Model Worlds.

 

In terms of "bigger curves look better" (talking about layouts here, not anything else...!), it's not uncommon for some layouts to have tighter curves in non-scenic areas, particularly where reversing loops are involved, in which case appearance doesn't matter.

 

That said, my only real regret in redesigning my father's layout when he moved house in the late 90s was the use of 1R curves on the reversing loops - just before "minimum 2R" became the default for new models. That said, we haven't bought much in the way of new locos for that layout since then anyway!

[DCB]

3 hours ago, RJS1977 said:

I have seen a Bachmann Class 70 negotiating 1R curves on one of the layouts at Horsham Model Worlds.

 

I saw a 00 Hornby class 47 and 3 Mk2 coaches gong round a 12" radius at Bekra Models Newton Abbott.  The lady behind the counter admitted it had been heavily modified...

[John Tomlinson]

On 08/10/2022 at 01:08, ProjectRedcar said:

Been wondering since my layout is planned to use a lot of Radius 1 curved track pieces (I'd say 90% of curves on it are Radius 1). However, when I look at what the manufacturers say, they state a minimum curve, usually 2, for the models I have my eye on.

 

Is it a matter of covering themselves in case a more extreme radius ends up breaking the model? Or have they actually tested it in-house?

 

Also, if I have a Class 37 from Bachmann and then one from, say, Accurascale, both OO gauge and both the same scale, is it reasonable to assume both of them will make it around the same curves?

 

Any help would be appreciated. If it ends up that I have to change all my track to be Radius 2 or higher, I will. 

 

And for reference, on my layout, models will run at low speeds (about 10-20 MPH in reality, tops) around those Radius 1 curves.

 

Cheers

 

I'd look at this question a different way.

 

All our layouts, pretty well without exception, have curves far tighter than prototype. But the bigger the radius, the less wrong it looks. So if you can get a layout design acceptable to you in the space available, by using radius 2, or bigger, I'd go for it.

 

John.

[The Johnster]

Tbh, I doubt if I would ever be happy with a setrack layout; we all compromise but settle at different points on the compromise spectrum, and I have no wish to criicise others who opt for more RTR solutions lest I come in for criticism myself from proper scale modellers...

 

There is a possible exception in my case though; perhaps a dock or industrial layout with very sharp curves and exclusively short wheelbase stock.  It could be ultra-portable, say 4'x1', and self contained, so I could sit in front of the tele with it or take it out on to the patio in summer.  Stop it, Johnster, you've already got a layout, and that's not finished yet!

[Il Grifone]

The general minimum radius for locomotives seems to be 5 chains (roughly 4' 4" in 4mm scale*). The drawings state 3 chains  for Mk 1 coaches and 1 chain for the humble mineral wagon, but do not specify how, or even if, they are coupled. Presumably these radii intend 'dead slow'.

 

* It's better to err on the right side and say 4' 6". 1 chain = 22 yards (approx. 20 metres, which I believe is the unit in the metric age) or 264mm in 4mm scale. This equates to 10 inches plus a centimetre, but the round 10 inches is probably near enough.

Edited October 10, 2022 by Il Grifone

[meatloaf]

Ive test run the following over first radius on my layout. Only Bachmann 37 wont play nice on 1st radius

 

Class 08

Class 20 - both old and new tool

Class 24 - new tool

Class 25

Class 37

Class 45

Class 47 - both old and new tool

Class 66

 

 

[newbryford]

On 08/10/2022 at 04:38, DCB said:

My Heljan 67 and Bachmann 64XX derailed constantly when I tried them on Hornby set track points which older stuff sails over. 

 

 

Heljan 67??

 

Have you considered that it might also be a mismatch between newer wheel and older track standards, rather than just the sharp radius?