wobbly locos!

[sb67]

Been running in some stock on my new rolling road and I've noticed, mainy the smaller locos wobble like mad! I havent noticed it much on the track, is there anything that can be done to correct it?

Steve.

[Lady_Ava_Hay]

Been running in some stock on my new rolling road and I've noticed, mainy the smaller locos wobble like mad! I havent noticed it much on the track, is there anything that can be done to correct it?

Steve.

The answer is really 'no'. It is a function of the rolling road being less than 100% accurate in quality. I would imagine that there are rolling roads that are better made, i have seen some in Germany ( where else?) but with quality comes cost and we know where that leads.

 

That wobble can, at high speed, actually derail the loco.

[TRAILRAGE]

I have an old H&M rolling road that makes everything wobble. 

Just don't run them to fast!

 

Cheers Trailrage

[The Johnster]

Film of real locos running on the Swindon 'home trainer' and the Rugby Test Plant rolling roads suggests that things were not always completely steady with those either, a function of unbalanced reciprocal masses I assume; I wouldn't worry too much about it, so long as the locos perform ok on the track and you can do whatever function you need to on the rolling road satisfactorily.

 

It might indicate that your wheels are not perpendicular to the axles, which you should be able to detect fairly easily with a visual check , but if this is not the case, and it probably isn't, then there is no real problem. 

[TheSignalEngineer]

I've several locos which will almost throw themselves off the rollers but are perfectly OK on the track. A slight imperfection, mis-alignment or slack in the rollers can magnify minor wobbles in the loco if they happen to co-incide. In real life if you've been on an 08 going down a siding at top speed you soon realise that it can be a bit like a bucking bronco.

[sb67]

No need to worry then :-) Thanks for the replies guys.

Steve.

[Dunsignalling]

If you are using the sort of roller units that clip to a section of track, they work best on set-track rather than flexi unless the latter is firmly fixed down.

 

I use a self contained-rolling road (de Locloods) which is of excellent build quality (with a price to match) on which most locos run very smoothly but I think the waddle exhibited by short wheelbase ones is inevitable, at least in OO where there is a lot of slop built in to the wheel/track interface.

 

John

[34theletterbetweenB&D]

What is seen on the rolling road is a dynamic summing of the errors in both moving systems as they interact, and there may even be excitation of resonances in either or both systems. In comparison, the single moving system on the stable and well damped rails shows very little effect. If the rolling road unit can be run on similar rails, it too would probably run steadily.

 

Film of real locos running on the Swindon 'home trainer' and the Rugby Test Plant rolling roads suggests that things were not always completely steady with those either, a function of unbalanced reciprocal masses I assume...

 Observed on the steam test plants elsewhere in the world also. Too late for the steam age, it required better understanding of dynamics and the availability of fast computing to model what was going on; technique that developed in the aero industry particularly, and is now widespread in engineering. The out of balance forces that direct drive steam locos create are well damped by the track, which has no significant resonance with the peak frequencies the locomotive generates. On a frame with rollers for the loco to run on, it all becomes rather too 'exciting' (pun intended) as resonances are pretty much bound to occur. (The old marching column of soldiers ordered to break step for safety when on a beam bridge: and see also wind excitation of resonance, the famous Tacoma Narrows suspension bridge failure, flexing to destruction once the wind speed was 'just right' to drive the bridge at a resonant frequency.)

[coachmann]

I disagree completely because no one has explained why some locos run like a Rolls Royce on rolling roads! I have a rolling road and my locos run as steady as a rock because the wheels are mounted accurately on axles. In fact, there are classes of loco that I won't buy because I know that mass-production has turned out some really poor chassis over the years. I pay as much attention to chassis as I do track, so when testing a loco I look for a steady gait as well as drive characteristics.

[34theletterbetweenB&D]

I disagree completely because no one has explained why some locos run like a Rolls Royce on rolling roads! ...

 Here's a description of the explanation. If you have a loco mechanism and a rolling road which are both made a decent degree of precision, and such errors as are present in each do not excite resonances in the other system, then the smooth running result will be observed. Put on a different loco which does excite a resonance between rolling road and mechanism and it will show up, as on your 4F.

 

Here's an example which you may have experienced or seen, running a car onto a road surface with a regular pattern, like stone setts, or the infamous Belgo-French Pave. At some speeds it is possible to obtain a resonance between the unsprung mass of wheel assemblies and the road, and the normally reasonably well damped and thus 'smoothish' ride becomes a very bouncy experience, Alter the  speed, and the bouncing goes away; or if a significantly different vehicle design is tried, the bouncing resonance will occur at a different set of speeds.

[The Johnster]

I've several locos which will almost throw themselves off the rollers but are perfectly OK on the track. A slight imperfection, mis-alignment or slack in the rollers can magnify minor wobbles in the loco if they happen to co-incide. In real life if you've been on an 08 going down a siding at top speed you soon realise that it can be a bit like a bucking bronco.

Except that the bronco is a steadier ride.

[Killian keane]

Ahem... Brings to mind a certain gaiety pannier tank...

[Dunsignalling]

 Here's a description of the explanation. If you have a loco mechanism and a rolling road which are both made a decent degree of precision, and such errors as are present in each do not excite resonances in the other system, then the smooth running result will be observed. Put on a different loco which does excite a resonance between rolling road and mechanism and it will show up, as on your 4F.

 

Here's an example which you may have experienced or seen, running a car onto a road surface with a regular pattern, like stone setts, or the infamous Belgo-French Pave. At some speeds it is possible to obtain a resonance between the unsprung mass of wheel assemblies and the road, and the normally reasonably well damped and thus 'smoothish' ride becomes a very bouncy experience, Alter the  speed, and the bouncing goes away; or if a significantly different vehicle design is tried, the bouncing resonance will occur at a different set of speeds.

The crux of the matter is that, by definition, nothing in OO is made to a "decent degree of precision"; our locos can only go round silly train-set curves because they have loads of slack deliberately built in to the mechanisms.

 

If you watch EM, P4 or O Gauge locos running on rolling roads, the action is much better. There is proportionately much less slop involved so the amount of possible deviation from true running is accordingly reduced.

 

Your resonance argument does hold water, however. I do have locos that waddle like mad on the rolling road but are well-behaved on a layout. However, I've never come across the converse, one that is OK on the rolling road and wobbly on a layout. What one does have to watch for is that a rolling road doesn't show up back-to-back errors - they only become apparent through measurement or by encountering real pointwork.

 

John

[bertiedog]

I am afraid it is down to the wheels, the rollers are usually ballraces and completely accurate to within extreme accuracy, a few microns.

I am not getting at the makers, but most wheels on commercial locos are not accurate to within 5 thou, let alone microns.

Evan a few thou of run out, or out of roundness, will set off the wobbles.

 

Add it to side to side play and you have the answer, the wheels themselves are to blame. It will not show so much on the track, it just makes the body move side to side a bit, but on the accurate rollers there is no give and it sets off vibration and wobbles.

The only answer is to change the wheels, but if it runs all right on track, why bother?

Just run in a slower pace, where it should stay firmly on the rollers.

Stephen

[Dunsignalling]

I am afraid it is down to the wheels, the rollers are usually ballraces and completely accurate to within extreme accuracy, a few microns.

I am not getting at the makers, but most wheels on commercial locos are not accurate to within 5 thou, let alone microns.

Evan a few thou of run out, or out of roundness, will set off the wobbles.

 

Add it to side to side play and you have the answer, the wheels themselves are to blame. It will not show so much on the track, it just makes the body move side to side a bit, but on the accurate rollers there is no give and it sets off vibration and wobbles.

The only answer is to change the wheels, but if it runs all right on track, why bother?

Just run in a slower pace, where it should stay firmly on the rollers.

Stephen

It's not just the concentricity and truth of the wheels, either.

 

Many r-t-r loco wheels have surprisingly rough treads when new, which will set off the collywobbles on a rolling road and limits pulling power too. 

 

I have several older (particularly Bachmann) locos that, having polished the roughness off through some serious running, will happily haul almost twice what they could manage when first bought. Collett goods, N mogul, WD and BR5 in particular.  The issue still occurs, but it is much less pronounced with newer examples, so I presume the wheels are made smoother nowadays.

 

John

[DCB]

Most likely problem is the rollers are flat and wheels are coned, unless they are 1950s Triang.   The wheels will wobble and hunt as the wheels try to find equilibrium.   The prototype had similar  issues when heavy flat bottom rail sections on concrete sleepers with much less give were introduced towards the end of steam.

 

EDIT  I believe the Swindon "Home Trainer," had the locos securely attached to the frame using the tender drawbar and the rollers were braked so the power could be measured.  I wonder if its possible to make a set of roller sleeves  to code 100 (or 75, 87 what ever) rail profile.

[SRman]

The effect is exaggerated with short wheelbase locomotives such as 0-4-0 and 0-6-0 types, but also shows up with even large types with lots of slop in the driving wheels/axles and also lots of slop in their bogies and ponies (e.g. in pacifics). On the real thing the bogies of larger locomotives lead the loco into curves and control the side to side movement a bit; in our models, they usually swing or slide freely from side to side and offer no extra stabilisation compared to running the same locomotive minus its bogies. Add to that the factors mentioned by others earlier, regarding the way the rollers are usually flat not coned, and also possibly not being exactly to gauge, and you have many ways of allowing an engine to 'float' a bit.

Watch a video of a small industrial shunting engine (say, an 0-4-0ST) and you'll see them hunting and wobbling all over the place, not helped by the usually poor track in the places they were to be found.

My Hornby Sentinel diesel wobbles a little even on the rails; the effect, when I videoed it head on was actually very reminiscent of some of the videos of real locomotives I referred to before.

[bertiedog]

The wheels will hunt as it is impossible to cone the ballrace, without adding a sleeve. This would reduce the self steering and hunting, but there is so much play in 00, the wheelset will hunt on a flat or cone.

If this is a problem the use a cradle to support the loco dead on the line of the ballraces, to stop side to side wobble. But you must be sure the wheels are basically running true first.

Try to run the Loco at about 100 RPM at most, no higher, it is just asking for trouble to go faster. And always run the whole thing on a bed of firm foam, with side blocks to catch a falling loco. And do nor run it on a table edge!!! or narrow shelf....

 

Stephen

[Il Grifone]

Ahem... Brings to mind a certain gaiety pannier tank...

 

Very variable beasts! Some are so bad you can see the wheels are eccentric without having to run them and others run true. There are also differences in the chassis which can aggravate the problem.

[bertiedog]

The mystery of the Gaiety was the pinion wear from the steel contrate gear and why was such an advanced motor design used? If the armature is removed and fitted with modern magnets the motor is superb. I suspect that the split stack armatures were made by Taycol, who were very experienced in making model motors.

[Il Grifone]

I must try some new magnets. I have a couple that run OK, but the brush gear soon runs hot.

I think they had hopes of sales of the motor as a separate item - all the parts were allegedly available separately, but probably the long thin shape was not the best choice. The fact that the body was available fitted with Tri-ang or Dublo chassis (neither of which fits properly) suggests all was not well.

 

The pinion gear is easy enough to replace (turning it around gives extra life) but sourcing a replacement is difficult today.

 

The wheels look like they were copied from the Dublo N2 tank (same excessive number of spokes for the diameter), but unfortunately they didn't achieve Meccano's accuracy of manufacture and assembly.