Atomic Systems

I have looked through 40 pages of threads and can't find anyone else who has had this problem , and I couldn't work out how to phrase it for a search. The picture shows the problem

Click the image to zoom in


The train follows the track and does not derail, but the wheels don't go where the rails are. Can this be prevented

the train doesnt follow the tracks it follows an invisable data line that can only be seen in Activity editor.

Do you mean the front and rear of the loco follows the track which drags the wheels over the rail. cos I'd like to know if theres a cure.

kevin

You're not the only one with the problem

http://forums.uktrainsim.com/viewtopic. ... highlight=

It looks like a problem with tight radius curves.

MSTS uses the pivot point for the ‘MAIN’ or parent object for object placement and rotation in MSTS.

The model is moved and rotated around this pivot point in the MSTS world.

The track that MSTS uses to guide the rolling stock is not the 2 rails that you see in game but an invisible centre line on each track section. This ‘centre line’ is not curved on curves but a number of straight sections (like a chord (http://en.wikipedia.org/wiki/Circle)). On a sharp curve, if the number of ‘centre lines’ are not close together the centerline cuts across the inside of the curve. As the pivot point for the model follows this line it makes the loco ‘cut’ the corner as shown in the screen shot above.

Being a 0-4-0 my suggestion would be to make sure the pivot point for the MAIN part is centred between the two wheel sets rather than the centre of the model itself. It may not eliminate the problem but may help minimise the amount the loco side slips on the track work..

Another possible option (untested by myself) is to add invisible bogies to the model front and rear. I suspect bogied stock uses the bogie pivots (as opposed to the MAIN pivot) to traverse the invisible centreline. If you created invisible bogies these may follow the track work a little better minimising the side slip.

If I’m not mistaken, if you view the track work in the Route Editor the grey line overhead the track work is and indicator of what MSTS uses to move objects around in the MSTS world. If it is the case then have a look at the short radius curve in RE and you will get a better idea of what your model is trying to traverse.

Hope this helps.

Thank you for your responses. I have learned a few things, and will see if I can apply it all when I get home tonight

And then an uninformed plonker was heard to say.....

Are these invisible straights(chords) tied to the track geometry or to the track section?

If the track section, then making smaller sections to complete the same curve or maybe even 'smoothing' (multi-section in the tsection.dat file), may help on these tight radius curves.

Probably already thought of and dismissed, sorry to butt in.

This is all getting very technical! I wonder if the train follows the path defined in tsection.dat.

Anyway, I observed the behaviour of bogie vehicles over the same piece of track, and whilst the loco cut the corner, the wagon did not. I borrowed the trailing truck from another engine and put copies of it on the centre of the leading and trailing axles, the loco now follows the track, so I will now have to modify all the 0-4-0s I have in progress!

Thanks for your help, I would not have found the solution without it.

As it was explained to me, MSTS uses the bounding box to determine what part of the model actually follows the rail...ie each end of the bounding box. For 0-4-0 and similaqr configurations where the wheelbase is considerably shorter than the bounding box, you get the wheels running off the track.

One solution is to shorten the bounding box down to the length of the wheelbase. This will then force the model to track where the wheels are.

One problem.....with a short bounding box, the first wagon will run into the engine. This can be resolved in one of two ways...either build a short "invisible" wagon to act an interface...you'll need one front & back...or to extend the bounding box of the tender forward to counter for the short bounding box for the loco.

I've not tried this yet....I have a PC problem at the moment..

Doug

MTST seems to put the centre of the wheels midway between the extreme end and the centre.
This works on a short wagon but gets out of step on anything longer. Using bogies solves the problem.
I make the two axleguards as the bogie but you could also use the springs or hide a single poly inside the axle.

Parent parts like axleboxes, springs and brake shoes to the bogie so they move with it.

Actually the 'cure' is simple. With any locomotive without leading or trailing bogies, such as an 0-4-0 or 0-6-0 you need to give the locomotive a 'false bogie'. The driver wheelsets are then linked to the 'false' bogie - which is called bogie1, so it's bogie1, to which wheels1 and wheels2 are linked. The 'false' bogie is linked to the main object. The movement of the bogie around curves is hardly noticible but you will find your locomotive will track correctly.

You should do the same for any wagon that has no bogies, ie, 4, 6 or 8 wheeler, although the latter two would more likely be tenders.

Note that Ian's method of two bogies also works and is useful for longer wheel bases.

Cheers Bazza

I tried that Bazza and my tram caught a bad dose of wheelpopusowtasidus! Particually on 25m radius curves!

Hmmm, that is strange, it's always worked okay for me. Bring the bounding box back to the buffer beams and adjust the couplers with the Size and InertiaTensor settings.

If it jams/binds on tight radius curves, then the problem is likely to be related the BB, and the above should fix that.

I've used the twin false bogie trick for a four wheeler tram, with each wheelset linked respectively to bogies 1 and 2. I also attached the tram's link bar/s to the bogie/s, thus the couplers visual match around tight curves.

Cheers Bazza

I've finally been able to try reducing the bb size down......and it works!!

I suppose I'll now need to create an invisible spacer to allow another tram to couple up...

dkightley wrote: I suppose I'll now need to create an invisible spacer to allow another tram to couple up...

Already done.

35ft Spacer and Coupling Bar. [9470 bytes] - CouplingBar&Spacer.zip File ID: 13935 Date: 27 Jan 2006 - 1003 Downloads