Is there any need for side control springs on loco bogies and pony trucks?

[Will L]

There has been quite a bit of traffic on the forum over the Christmas/New Year period suggesting many of us haven’t got anything more pressing to do, so perhaps it’s a good time to indulge in a little theoretical, possibly contentious, discussion.

There is a common assumption that side control springs on bogies and pony trucks exist primarily to help guide our locos round corners. While it is certainly true that these springs may have some effect, steering is primarily done by the coned wheel effect. It is certainly not why these springs were provided on the prototype. It is true that the physics behind the coned wheel effect doesn't scale, but in this case the effects of scale actually work in our favour, which becomes obvious when you considering the speed with which model trains will go round sharp corners.

On the prototype, the primary role of bogies and pony trucks was to carry loco weight which the civil engineer would not allow to be carried on the driving wheels. Given that track curvature requires the rigid driving wheel base to be fairly short, big loco's had short driven wheelbases and big overhangs. Under these it was convenient to place the carrying wheels the civil engineer was insisting upon. Given the nature of the heavy reciprocating machinery that drives a steam loco, there was always a fairly strong tendency for a loco to "box" or oscillate from side to side in time with the pistons. Long overhangs front and back just amplify this tendency. Therefore, side control springing on bogies and pony trucks were used to dampen down this tendency to oscillate, particularly when travelling fast in a straight line. Modern stock bogies which also tend to oscillate when running at high speed use shock absorbers rather than springs to do the same thing.

So what are the implications for us model builders. If your purpose is to build an "as close to the prototype as possible" model, then you want to replicate quite a lot of stuff which serves no operational purpose on the model, but, even then, how many of you have built a loco with a full set of boiler tubes? For the rest of us, who are trying to produce something that looks as much like the real thing as is practicable but would also like it to run nicely too, we might want to consider what makes it a good running vehicle in our world.

We are not concerned that our locos weight might destroy out track so, the primary purpose of carrying wheels goes away. So we don't necessarily need to carry any body weight on bogies and pony trucks at all, so long as they are, in turn, heavy enough in themselves to run without causing trouble.

We don't have significant reciprocating masses and piston hammer blows to worry about, and so most of our locos show no tendency to oscillate. They run smoothly enough to be guided down straits and round corners purely by the coned wheel effect.

That said, there are some wheel arrangements, particularly 4-4-0s where the short driven wheel base and the long overhang mean that a bogie does have a role in helping a loco run realistically and well.

From all this we can conclude
a. Many pony trucks and some bogies are best not carrying any weight other than there own. Attempting to do so only complicates building the model, even more so if you try and include side control springing. The result is as likely to be the cause of derailments than a cure.
b. Even when a weight carrying bogie is required, simple friction between the bogie pivot and the main frames will be enough to stop any tenancy for the loco to oscillate, and the coned wheel effect will successfully guide it round corners.

So, despite knowing there will always be the occasional exception, it is best to save yourself work and potential problems by forgetting side control springs on bogies and pony trucks.

[Philip Hall]

Will, I do agree. I've built/repaired one or two 0-4-4T engines over the years, one compensated and one rigid, and they didn't have side control on the bogies. I've also converted lots of M7s, T9s and Dukedogs (getting on for 20 I guess) and in all of those cases, although the chassis was rigid, in every case the bogie was fully supporting the rear of the engine (M7) or in the case of the T9 and Dukedog, held down by a well adjusted spring. Any side control was just friction as you have outlined. And none of them waggled from side to side and they all took corners well. Waggling in my experience is usually down to wobbly driving wheels.

Weighted pony trucks or ones sprung down are another matter. I've done both and they all work so I'm not sure I have a preference.

Philip

[petermeyer]

Will

Interesting and timely as I am just finishing the chassis on an LNWR Experiment 4-6-0. In this kit, the whole of the front chassis is articulated. There is no provision for side control and I am minded to leave it off. In the minimal testing I have done on the layout, it does run.

Peter

[Jol Wilkinson]

Will

Interesting and timely as I am just finishing the chassis on an LNWR Experiment 4-6-0. In this kit, the whole of the front chassis is articulated. There is no provision for side control and I am minded to leave it off. In the minimal testing I have done on the layout, it does run.

Peter

Peter,

presumably this is the Brassmaster Experiment kit. I have one to build and thought I would probably follow the same approach I used when I built a LRM Precursor Tank kit, which were originally supplied with the Brassmaster chassis. BM never got around to doing the "body" kit so LRM produced that and after initially using the BM chassis now have their own dedicated etched frames kit with a sprung bogie.

With the BM chassis, on which the bogie is pivoted longitudinally and compensated, I attached a wire spring the the underside of the body which ran forwards and then located in a hole in the top of the pivoted front frames. This provides some vertical load on the bogie wheels and some lateral control, probably more important on the 4-4-2T than on the 4-6-0.

As you have found the 4-6-0 works okay, I may therefore just stick with the BM chassis as designed, possibly with a bit of weight added to the front frames

Jol

[ted.stephens]

As you say a key purpose of the bogie is to keep the load on the driving wheels within the limits of the permanent way and structures.

Another key reason to have a bogie is that it limits the potential for yaw, side to side movement of the front and rear caused by the vehicle rotating about its centre. If this is to great then there is the potential for the buffers to move sideways to such an extent that buffer lock occurs or sticky out parts of the loco such as cylinders foul infrastructure, in particular platforms and through girder bridges.

Whilst for modelling purposes we may choose to allow a little extra clearance to platforms, etc. it is not possible to avoid the problem of buffer lock.

To illustrate this here is drawing based on an Adams Radial chassis, renowned for its ability to go round small curves. The frame and buffers are in green, cylinders - red, wheels - grey, rail running edge - blue.

Buffer01.jpg

The centre line of the buffers is 6” (2mm) outside the rail running edge, both sides.

Take a typical turnout, B8, radius 186.8m (2451mm) which most locos can run over is used as an example. This radius is show below in red.

Buffer02.jpg

In close up

Buffer03.jpg

The buffers of an adjacent vehicle are shown on the curve. This demonstrates that there is no particular problem with potential buffer lock due to curvature.

However when you take into account loco yaw which can happen due to clearances between the wheel flange and the rail running edge and between inside of wheel and axle box face things look a little different. The frame and wheels have been turned to place the leading wheel flange tight against the outer rail running edge and the trailing wheel flange tight against the inner rail running edge.

Buffer04.jpg

Buffer05.jpg

Next add in the movement between inside of wheel and frame. As before the frame has been turned to be tight against the inside of the leading wheel on the outer radius and the inside of the trailing wheel on the inner radius.

Buffer06.jpg

Buffer07.jpg

Now you are clearly at risk of buffer lock. It should be noted that this illustration does not show yaw on the adjacent vehicle or take into account any track fault which could make the situation worse.

Remember - this illustration is based on 2500mm radius curve. Imagine what it is like on the type of curvature that many of us use on our layouts.

If you limit the sideways bogie movement you can limit the yaw and thus avoid buffer lock. However a hard stop limit would jolt the bogie and potentially cause a derailment. Including a spring in the mechanism will pull the front end of the loco round to follow the bogie in a progressive manner to avoid the sudden shock.

So “Do you need side springs” - Yes - because it avoids buffer lock.

An interesting footnote - I came across this diagram for the leading truck of an SECR N Class loco in “A Guide to Locomtive Building” by Mike Sharman.

Buffer08.jpg

As far as I can tell the sideways movement is controlled by opposing inclined planes rather than springs. I have highlighted these in red and black. Any thoughts on this?

Buffer09.jpg

[davebradwell]

Your pony truck appears to use a similar principle as employed in the LNER Cartazzi truck - as the front swings, the springs are compressed because the sliding action is uphill. It also embodies the fundamental requirement I put in the Fell loco thread that the side movement must have a definite centre as the 2 wedges act like a notch. The swing links appear to be just holding it all together rather than on Gresley's swing link pony that embodies the same lifting action and notch as the wedges.

If you get buffer locking I suppose you just have to fit effective side-control and add enough weight to keep it on the track. It can be done but when you pull the front in the rear will stick out more unless you do both ends. I suspect keeping the front in will require more force than just keeping all in-line on straight track. I suspect the prototype used a force that would relieve the pressure on the leading flange rather than eliminate it entirely which is what you are trying to do here.

DaveB

[zebedeesknees]

There has been quite a bit of traffic on the forum over the Christmas/New Year period suggesting many of us haven’t got anything more pressing to do, so perhaps it’s a good time to indulge in a little theoretical, possibly contentious, discussion.
So, despite knowing there will always be the occasional exception, it is best to save yourself work and potential problems by forgetting side control springs on bogies and pony trucks.

Nah, while it's on the bench, if in doubt - do it!

Saves a lot more work later..

Ted.

[Stephan.wintner]

I believe I've seen similar wedge designs on other leading trucks (north american) and understood them to apply lateral and vertical force to the frame, which in turn shifts on it's equalization and heels the loco into the turn. I've always understood the intent of a lead truck to be guiding the loco into the curve, not carrying weight (or, not solely carrying weight).

Whether that's needed on a model I'll leave to the experts (and look forward to more discussion).

Stephan

[petermeyer]

Jol Wilkinson wrote:

Peter,

presumably this is the Brassmaster Experiment kit. I have one to build and thought I would probably follow the same approach I used when I built a LRM Precursor Tank kit, which were originally supplied with the Brassmaster chassis. BM never got around to doing the "body" kit so LRM produced that and after initially using the BM chassis now have their own dedicated etched frames kit with a sprung bogie.

With the BM chassis, on which the bogie is pivoted longitudinally and compensated, I attached a wire spring the the underside of the body which ran forwards and then located in a hole in the top of the pivoted front frames. This provides some vertical load on the bogie wheels and some lateral control, probably more important on the 4-4-2T than on the 4-6-0.

As you have found the 4-6-0 works okay, I may therefore just stick with the BM chassis as designed, possibly with a bit of weight added to the front frames

Jol

Yes Jol it is the Brassmasters kit. The instructions say to fit a substantial 20thou nickel silver piece on top of the bogie/chassis assembly which acts as a rubbing plate on the underside of the running plate. I tested mine with just the running plate (yet to do the rest of the upperworks) and that's how it worked. The rubbing plate, as Will states, presumably provides sufficient friction to control sideways movement. More bodywork should provide more weight.

However, the Experiment and Precursor do not share the same bogie particularly as to how it fits to the frame. Rather than hijack this thread I'd probably better start my own workbench thread as previously threatened.

Peter

[grovenor-2685]

You don't need a workbench thread, that just removes it from the Chassis and suspension section, just start a new topic here.

[Will L]

As you say a key purpose of the bogie is to keep the load on the driving wheels within the limits of the permanent way and structures.

Another key reason to have a bogie is that it limits the potential for yaw, side to side movement of the front and rear caused by the vehicle rotating about its centre...

Which on our models seem to be adequately covered by friction between loco and bogie

...If this is to great then there is the potential for the buffers to move sideways to such an extent that buffer lock occurs or sticky out parts of the loco such as cylinders foul infrastructure, in particular platforms and through girder bridges.,,,

So “Do you need side springs” - Yes - because it avoids buffer lock...

You may have a point. However when I have a loco which may indeed have enough end throw to cause buffer lock if uncontrolled it tends to be 4-4-0s and the like and my solution has always been to have a fixed pivot on the bogies and a bit of side play on the driving axle next to the bogie. A solution not available to the prototype.

Remember we are talking practical solutions to model performance here, not prototype fidelity

[jim s-w]

I use side control as it’s easy enough to add as you go along. The big difference I found is it makes putting the loco on the track easier as the wheels are where you expect them to be. This is very noticeable on peaks with their pony trucks.

[Will L]

There has been quite a bit of traffic on the forum over the Christmas/New Year period suggesting many of us haven’t got anything more pressing to do, so perhaps it’s a good time to indulge in a little theoretical, possibly contentious, discussion.
So, despite knowing there will always be the occasional exception, it is best to save yourself work and potential problems by forgetting side control springs on bogies and pony trucks.

Nah, while it's on the bench, if in doubt - do it!

Saves a lot more work later..

Au contraire Ted. When ever I've tried fitting side control springs to a bogie or pony truck,I seem to spend half my life trying to get it adjusted so it works. Exactly how strong the spring needs to be to achieve the desired effect is very hard to judge. Get it too strong and you have nothing but trouble. Too week and its doing nothing and you might as well leave it off as everything seems to work OK anyway. This is exactly why I was tempted to start this discussion.

As you know I'm a firm believer in the value of springs in loco suspension. However the difficulty lies in knowing what spring rate you need for a given job, knowing what spring rate you have actually delivered and knowing if the two match in any way. The beauty of CSB is that they give an answer to this problem.

[Will L]

... I've always understood the intent of a lead truck to be guiding the loco into the curve, not carrying weight (or, not solely carrying weight).

Exactly what I was questioning. I know people are inclined to think that, but I reckon it is, carry weight first, prevent oscillations (and buffer locking!) next and any help go round corners is just a bonus.

The UK standard work horse, the 0-6-0, seemed to manage pretty well without. In a world where 4-4-0s were king I suppose you may have more of a point but as I've suggested elsewhere, on a model other solutions exist that don't rely on guessing the right spring rate.

[Will L]

I use side control as it’s easy enough to add as you go along. The big difference I found is it makes putting the loco on the track easier as the wheels are where you expect them to be. This is very noticeable on peaks with their pony trucks.

Fair point, but, as the lightest spring capable of keeping them central will do, not side control in the steering round corners sense is it?

[jim s-w]

who said anything about the lightest spring?

[Crepello]

There's also the question of yaw moment applied by buffing or tractive forces,
particularly on curves. While the centring spring force might be too difficult to
judge in controlling these, the springs will help restore the loco's fixed wheelbase
alignment with the track after the yaw moment reduces.

[davebradwell]

....but they can only do restore alignment with track if side control is designed properly with pre-load or notch effect as described earlier. Without it you'll always get a residual amount of yaw as, perhaps, the chassis has slight errors so turns to one side or wheels are just slightly bigger one side....etc

DaveB

[Stephan.wintner]

... I've always understood the intent of a lead truck to be guiding the loco into the curve, not carrying weight (or, not solely carrying weight).

Exactly what I was questioning. I know people are inclined to think that, but I reckon it is, carry weight first, prevent oscillations (and buffer locking!) next and any help go round corners is just a bonus.

The UK standard work horse, the 0-6-0, seemed to manage pretty well without. In a world where 4-4-0s were king I suppose you may have more of a point but as I've suggested elsewhere, on a model other solutions exist that don't rely on guessing the right spring rate.

Well, North American practice used a leading truck under almost every loco, with 0-6-0 and the like being used only as low speed shunters. 4-4-0, 2-8-0 and 4-6-0 were king in the early days, the 2-8-2 was king later. Buffer lock wasn't a concern, because knuckle couplers were standard, yet it was done anyway. Some 2-8-0s were converted to shunting service, and their lead truck was removed. So I respectfully draw a different conclusion.

I definitely agree that a model is different and should not attempt to copy solutions. Square Cube law, giant insects are impossible, likewise locos don't just scale down.

Stephan

[Will L]

There's also the question of yaw moment applied by buffing or tractive forces,
particularly on curves. While the centring spring force might be too difficult to
judge in controlling these, the springs will help restore the loco's fixed wheelbase
alignment with the track after the yaw moment reduces.

The coned wheel effect can and does restore and maintain the fixed wheelbase alignment with the track without the assistance of side control springing. It is true that this isn't perfect and in the wrong circumstances when they hit a resonant frequency it can cause significant hunting from side to side. When I was a kid I used to train-spot on Bletchley station, watching the early electric pre class 86 locos bounce from side to side at 100 mph was actually a bit frightening. But while the full size job may have needed springs to control this tendency, I've never seen any evidence that a 4mm scale model loco behaving like that, probably because they never hit a resonant frequency at the speeds we drive them at. The whole point of this thread being, why fit them if our models don't need them..

[Will L]

who said anything about the lightest spring?

I did. You seem to be installing a spring to help keep the wheels in a strait line when re-railing them. Perfectly reasonable but surely fitting the lightest spring possible is all that is needed for that purpose, particularly as fitting an over strong spring will cause derailments on curves.

[jim s-w]

That’s a complete misinterpretation on your part Will. I said I install side control and a noticeable benefit is is makes rerailing easier. I said nothing about I install it primarily for that reason.

[zebedeesknees]

Looking at the question above - Need? - no.
Sub. Desirability? - it's your train set!

Ted.

[down_under]

There's also the question of yaw moment applied by buffing or tractive forces,
particularly on curves. While the centring spring force might be too difficult to
judge in controlling these, the springs will help restore the loco's fixed wheelbase
alignment with the track after the yaw moment reduces.

The coned wheel effect can and does restore and maintain the fixed wheelbase alignment with the track without the assistance of side control springing. It is true that this isn't perfect and in the wrong circumstances when they hit a resonant frequency it can cause significant hunting from side to side. When I was a kid I used to train-spot on Bletchley station, watching the early electric pre class 86 locos bounce from side to side at 100 mph was actually a bit frightening. But while the full size job may have needed springs to control this tendency, I've never seen any evidence that a 4mm scale model loco behaving like that, probably because they never hit a resonant frequency at the speeds we drive them at. The whole point of this thread being, why fit them if our models don't need them..

I'm a relative newbie to this, but if we look at the real thing, they reasons for fitting side control (in no particular order):

a) control side thrust
b) hunting / oscillations from bad track / poor springing (the famous Cox example from India)
c) carry weight
d) assist steering / guiding the coupled axles around curves, especially big pacific without radial boxes at the front, or tank engines operating at reasonable speed

Did I miss any?

When I think of a model

a) doesnt apply
b) depending on track quality / weight of loco and speed - how many layouts actually get up to a speed where track undulations are likely to cause a problem > yes there are a few large continuous runs
c) yep, they do in a model
d) perhaps related to b) or becomes more of a problem at speed.

So, in summary, maybe not all locos with bogies need to have side control in model form, but in some instances I can see that is could be beneficial

[Julian Roberts]

Will I don't exactly qualify to reply having built only one loco with carrying wheels, a 2-6-0. I've often seen mentioned the derailments of front bogies and pony trucks. Isn't the first issue actually keeping the things on the rails?

I found that it's a question of getting enough weight onto the pony truck, that there isn't room for enough lead for it to run as if independently, and it has to take some of the weight of the loco simply to stay on the track. There was a ratio - the worse the track the more weight needed.

I supported the body through a roller. I think this relatively free movement still imparts some sideways thrust which is probably quite enough. I've always thought a rubbing plate would present far too much friction on the model, and just cause the bogie or truck to derail. I managed the relative weight on the pony by adjusting the front driving wheel springing.

I've made three 4-4-0s or the inverse. All of these locos have a roller on the bogie to support the loco. The bogie is taking (say) half the loco weight - the Sharman tender trick restores some of the traction - and steers the loco through a pivot between it and the driving wheels.