Banks and Curves

[TheTazman]

Hi

Just been trying to do some searches on older posts regarding banks on bends etc. Has there been any mention of this in KRS?

Regards

Tazman

[AndiS]

If you mean the sideways banking of the track to compensate for the radial force moving a fast train off the track in curves, search for superelevation.
Short summary: It is likely to be there, going by the Leipzig demo, but there is no official confirmation yet and I cannot remember a report that someone managed to build a curve that definitively had it.

[keber]

Hi

Just been trying to do some searches on older posts regarding banks on bends etc. Has there been any mention of this in KRS?

Regards

Tazman

Sabrina (or was that Claire?) said somewhere on this forum, that superelevation will be included. Also on screenshots from Leipzig option for superelevation was seen in new route editor, but didn't function.

So probably it will be included (if BVE has it for years ), but it still needs to be seen in action.

[TheTazman]

ahhh searching on banking, and tilt not a good idea then.

I did not realise it had special name for itself.

I saw a good video on youtub.com that had a HST banking on a bend which looked cool.

Cheers

Tazman

If you mean the sideways banking of the track to compensate for the radial force moving a fast train off the track in curves, search for superelevation.
Short summary: It is likely to be there, going by the Leipzig demo, but there is no official confirmation yet and I cannot remember a report that someone managed to build a curve that definitively had it.

[pgmetcalf66]

Banking is the name of a train for helping locos pull loads up steep gradients. Maybe double headed and doubled tailed.

[bigvern]

Superelevation AKA cant.

[jamesinbolton]

Superelevation AKA cant.

Don't talk to me about cant! I'm out surveying tonight all because of defective cant!!

[pgmetcalf66]

Here is worth something talking about though. How will superelevation be modelled in conjunction with catenary gantries and wire?

[keber]

Question is: how will superelevation be modelled at all? As real 3D banking or simulated one as in BVE. I'd prefer real 3D, but I think this will present quite a challenge to Kuju to simulate this correctly.

[pgmetcalf66]

Question is: how will superelevation be modelled at all? As real 3D banking or simulated one as in BVE. I'd prefer real 3D, but I think this will present quite a challenge to Kuju to simulate this correctly.

Maybe standard radii fixed track

[timbooth]

I think superelevation is something they will either do properly, or not at all. I'm not sure how BVE handles it, presumably just banks the camera, but that won't work in KRS for external views - it will have to bank the train as well. Its not really hard to do, just rotation values along the track path. Superelevation transition may not be prototypical though, but thats not really a problem - it just needs to change from level to banked in a smooth step. Hopefully there is a realistic superelevation limit, so banking is not unrealistic.

Not sure KRS will have any fixed track shapes, so it should be a function of the track tool - if you set out a curve it should allow you to set the superelevation for it, and should do this automatically as most curves would have it, even if a small amount.

[keber]

Having superelevation doesn't mean just rotating the track, but it's a bit more complicated. As I can see in screenshots, KRS in contrary to MSTS uses 3D track bed. You just can't rotate track bed, because left and right side of this trapezoid always need to have constant inclination, which is 1:1.5. Upper side then can have arbitrary rotation, but it needs to be connected with other two sides. And then you have lower side, which is also some sort of variable, depending on underlying terrain.

There are two algorithms for modeling this sort of banked trackbed:
- when you define a curve with lenght, radius and superelevation, route editor (or subcomponent of it) computes and defines new object depending on these parameters and terrain and makes custom fixed object.
- you have a dynamic sort of object (defined only with parameters, like dynamic track in MSTS), which is probably more difficult to implement, but much more flexible at creating track bed, especially for transitions and possible use of different trackbeds (like ballast-less track)

I would prefer latter option, because of possible flexibility. But anyting is better as no superelevation, even BVE style (rotating camera).

[AndiS]

From all the talk about procedural track nearly a year ago I take it that superelevation will also be defined in the rule base for each track set. Therefore, even if Kuju don't get one detail right, it can be fixed as it is only a configuration parameter in some file -- provided that the rule language is powerful enough to express everything. E.g., the pivot point of the rotation along the track bed need not be in the middle of the track. For practical reasons it is often the inner rail (at least for steam era).