Practical experience of a minimum radius layout
Article written by Harold Jones

With the death of my Father some 30 years ago, I had to completely revise my layout. From two sheds connected by a garden line, with a station in each, I had to fit a new layout into one shed and move my workshop into the smaller shed.

The introduction to the featured layout in the No. 0 issue of the Model Railway Journal suggests that we “ignore the tenets and protocols of conventional finescale modelling”. I heartily agree…

This started my “minimum radius quest”. I wanted a double track main line, with two stations, believable lineside industries and townships. In a 16 x 8 foot shed. As a retirement present, I treated myself to a new shed which is some 19’ 6” by 7’ 7” internal dimensions. I already had a reasonably sized loco stud, so the first problem was to find out what was the maximum radius that I could fit in my shed and then how to get my stock round this.

As I have said many times, after much experimentation, trial and error, I found that with a level one-piece baseboard, new track and the use of laser cut track templates, it was achievable. I also found that gauge widening and superelevation was not necessary. As my stations are both at the same end and the platforms are only partially modelled with the remainder being under cover, the lack of superelevation and the tight radius is not noticeable.

My existing stock only required minimal alteration, mainly adjustment in the cylinder front bogie/pony truck area and some thinning of wheel bearings.

Emboldened by this success, I bought a BR 9F 2-10-0 chassis kit from Premier Components and experimented. My reasoning being that if I could get this monster round a 3’ radius curve, then the sky was the limit.

I laid a 3’ radius curve on a level piece of ply that I had in my “come in useful” stock and away I went. With some thinning of bearings, I succeeded. I ordered a DJH 9F and spent many happy hours building it.

Anyone want my redundant 9F chassis frames? Please apply.

My latest challenge is the LNER P2 2-8-2 which belts round my bends, as with a lot of my stock, it is still awaiting completion. A further benefit of minimum radius is that that of geometry. Taking an oval track and forgetting transition curves for a moment, for a space 6’ x 8’ with a 3’ radius curve you have a 2’ straight length either side. Now with a 10’ x 20’ space you would have a 14’ straight length either side, as opposed to a mere 10’ if you used a “traditional” 5’ radius curve. You would also be able to have more scenery/buildings on the outside of the track.

I use Peco track and points, including their Set Track points in the loco depot. I do not have any problems with them, the only work is to alter the wiring to suit the more “picky” DCC.

The following describes how I go about getting a large loco round my tight radius curves. Initially, all my locos were built with rigid frames, though with experience I have used compensated and sprung frames. The actual process is quite straight forward. Start with the chassis. Fit all the wheels and try it on the 3’ radius test track. Just push it up and down, quite often I have been surprised when it has moved easily. If not, observe closely to see where the problem(s) are.

I find it beneficial if the front bogie is on a rigid, fixed pivot, with a limited degree of side movement and some downward springing on the bogie to give weight to the bogie and provide some guidance to the loco on entering a curve. If bogie/pony wheels are fouling the cut out in the frames, then file until it is clear. If the driving wheels are being forced off the rail, then check which wheel is affected and thin the bearings. The aim is to have a wheeled chassis which, when on a 3’ radius curve, will still have a little bit of sideways movement. The prototype made use of tapered frames and “cut outs” to cylinders to aid cornering, so should we. The MR 4-4-0 compound springs to mind. This is the basis of a minimum radius loco. Obviously, such items as cylinders, steps and splasher clearances will all have to be allowed for. But you are on the right track.

When making adjustments (not butchery) to the loco, remember that the aim is to have a chassis which will run freely on a 3’ radius curve. You can then fit the body, at this point I do not bother with full valve gear or a detailed body. The requirement is to confirm that the loco can still negotiate a 3’ radius curve.

If you have fouling on steps or cylinders, dependant on the degree, you can ease them out slightly, file a little of the cylinders or you can experiment by removing wheel bearings to see if you can change the position of thinned bearings to alter the pivot point of the chassis relevant to the curve, to avoid fouling. On a loco with a leading or trailing bogie, restricting its lateral movement will have an affect on the amount of swing at the other end. I have found that by taking these steps slowly and testing each option I can reach my aim. I have NEVER had to resort to thinning or non-prototypical removing of flanges (the 9F prototype had flangeless centre drivers) When thinning bearings, it is important not to introduce too much sideways slop as this can be a cause of buffer locking. With regard to pony trucks, I have tried springing, but found that the position of the pivot point has a greater affect on track holding, so the main point is for it to avoid fouling anything. The position of the motor and gearbox will also have a bearing on options open to you to achieve a free running chassis.

A crucial point, which I can’t repeat enough, is that of baseboard and track. A true, level one piece baseboard with a smooth kink free well laid curve. It helps if you arrange for the rail joints to be on the 3’ 6” radius parts of the curve, thus avoiding a rail joint on the tightest 3’ radius part. Too often you read about people experiencing derailments and blaming the stock. Check your track first…and second and then again. When your loco can successfully negotiate your 3’ test track, try it on your layout before you finish building it. The dynamics are different between it being pushed round and propelling itself.

I have read of using a video taken on a mobile telephone, played back at slow speed, to view just what was causing derailments. Don’t give up, keep trying and ignore accepted wisdom.

In all things in railway modelling, before we start devising our own solutions, we should look to the prototype for an answer. The prototype also had to avoid buffer locking, the simplest solution is to ban the use of affected vehicles where it occurs, but this may not be practical, so practical solutions must be found. These include transition curves, larger diameter or oval buffer heads and a swivelling coupling hook such as on the BR class 4 2-6-4T.

I have never had to use extra links in the couplings, rigid or solid couplings, bars across the buffers or any other “solutions”, My main aim is to operate as per the prototype, this requires that coaching stock is propelled over crossovers, reverse curves and tight radius curves in the carriage sidings. I also add/remove coaches and tail traffic during normal operation using the train engine or a dedicated shunting loco.

Photograph 1 shows two coaches on my 3’ radius test track. As this shows, the buffers are in contact and would still be on a much tighter radius curve. My use of 3’ as my minimum radius is dictated by my loco stock and available baseboard width. Buffer locking occurs at the beginning of a curve, when the angle between two vehicles, being propelled, is greatest, so if we introduce a transition curve, of sufficient length to allow the second vehicle to gradually enter the curve until the danger of buffer locking is over, then we prevent buffer locking. My article in the May 2021 Gazette describes my method of laying track to help achieve this.

The track in the following photographs is replicating the inside curve of my double track continuous main line in my shed.

Photograph 2 shows two vehicles about to enter the transition curve, the buffers are in contact and in line, the screw link coupling is connected.

Photograph 3 shows the second vehicle about to enter the transition curve, the buffers are still in contact and the coupling is tight but not under stress.

Photograph 4 shows both vehicles fully on the 3’ radius part of the curve. To demonstrate how effective a transition curve is, the first vehicle is a GWR dreadnought which is one of the longest of coaches.

Photograph 5 shows my LNER P2 2-8-2, on the 3’ curve, coupled to the dreadnought coach.

Photograph 6 shows just how tight a radius can be successfully negotiated.

Further experience shows that you must pay attention to the configuration of the vehicles – an 0-6-0T may have a greater overhang at one end, possibly resulting in buffer locking in one direction, some coaches have the bogies set further inboard than normal, again this must be accounted for i.e., not all vehicles of the same length can be treated equally.

Tight radius curves impose their own rules, buffers must work smoothly and gently, file off the little pip left from machining the buffer head, it helps if the coupling hook spring allows a little movement and for bogie stock, there is nothing to prevent the easy swing of the bogies. Another important point is that of weight, it helps the vehicle resist the forces trying to force it to take a straight line. It also follows that the buffer beam must be of sound construction.

I have not mentioned goods stock, mainly, because if a coach can go round a 3’ radius curve, then so can a goods wagon, though some attention to reduce excess lateral slop in the axles and the wheel bearing may be beneficial for long wheel-based stock.

My layout appeared in the November 2020 Gazette, in which I mentioned that I was now using the old Tri-ang tension lock couplings on my coaching stock.

This has nothing to do with buffer locking but in order to make prototypical operation possible, as it is the only easy way in which I can remotely couple/uncouple stock which is out of sight under the buildings which are hiding my tight curves from view.

For those who like to see this theory put into action I have a short video showing a rake of coaches being shunted over a cross over and then round my tight radius curve into the platform. My video for the Guild Spring Show which demonstrates track laying for tight radius curves, is still available on the Guild’s You tube site.

I have also posted videos and information on the Guilds Forum – search under “Jones5669”

Happy Modelling.