Superelevation and pointwork

[memphis455]

Hi Folks,

May be pushing the envelope here, but does anyone have experience of laying pointwork on superelevated curves?

I've three locations on my MML northwards extension - Wellingborough, Kettering and Market Harborough - where the fast lines were banked even back in the '60's, and pointwork exists so close to, or on, the curves that I can't run the superelevation out to a suitable stopping point.

What are the limitations of laying pointwork in such places? I've arranged matters so that the transition curves are clear of where the pointwork needs to be, so the curve radius is constant through the points locations.

Is it even possible?

If it isn't, I'll have to dispense with the superelevation, which'd be a shame.

I'd appreciate comments from anyone who has tried this.

TIA,

Memphis

[AndiS]

May be pushing the envelope here, but does anyone have experience of laying pointwork on superelevated curves?

Yes.

Is it even possible?

No.

If it isn't, I'll have to dispense with the superelevation, which'd be a shame.

Yes.

They never invested into the effort to implement this.

[deltic009]

Yes.

They never invested into the effort to implement this.

I think that is a little harsh. Super elevation has to try and be all things to all levels of user, otherwise only a select few would be able to function with complex calculations - they didn't want to give people complete freedom to define how easements are implemented because it would just end up with a lot of trial and error before getting it right. Yes, I agree there are limitations to it and perhaps it could be easier and more flexible to use, but without having taken part in any of the programming for the implementation we have I'm certainly not going to assume the lack of something is down to laziness.

[AndiS]

I meant effort as in financial effort, not as in attempt. So it was not harsh. They made a decision based on that they perceived as addition in game value and as cost to get there, and we have to live with it. We are not required to applaud.

I must say that when there was first talk about superelevation, I was not convinced it is needed. Later I was not convinced you need superelevation at crossovers. But lots of people came up with lots of examples of crossovers in curves.

The bad news is that they chose to hard-link superelevation to track radius. This not only brings along the need for a different track rule for each speed rating (and the official advice is to only have one track rule) much worse it means that after any pointwork you need to have small piece of straight track and start a full size easement from there to get superelevation.

So I cannot see why the cant of track could not be a track property that you freely define, with transitions at arbitrary length. If and only if you go the full way like Zusi does, then it is a hard exercise to get the tiny bits really right. But if you permit arbitrary track ribbons to link up, like RW does, it would not be harder to form a crossover in a superelevated curve that it is to form a connection between straights that are not at exactly the same height. There, too, you would desire assistance in placing the needed gradient in the connecting track but all you got is gradient change points that you drag about until the frog looks good enough. If this is fair enough, then it could be done in superelevated curves, too. But only if the cant is not computed straight from the radius. And of course not, if you switch off superelevation as soon as there is a track node in the track piece.

[memphis455]

Oh Well....no cant it is then!

Andi - thanks for the info, better to know than waste time thrashing about trying to make it work!

Ta,

Memphis