Re signalling...........a real pain!!!

[40058]

Having spent the best part of three days re signalling Pat Dalton's S & C 1930 route using Edmund Kinder's excellent LMS UQ signals, I was looking forward to running services on it until to my horror most of the "distants" were not working. I've read the "A Problem with distant's" file that John Yelland wrote but being an amateur have failed to sort the problem out.

I'm now faced with resignalling again changing like for like on the route after painstakingly noting down the co-ordinates and signal types AND not being able to use "ground signals" because there's no sigcfg or sigscr with

Edmund Kinders LMS UQ, Default S & C 1930 Signals (Herbert Feyerbend ???) and ground signals in them.

and to make things worse no matter what I do if I edit the sigcfg file to put these signals together I get a line error all because of a poxy bracket position.....ahhhhhhhhhhhh!!

Can anyone with expertese in signalling please help me out and do "sigcfg" and "sigscr" files with the three mentioned signals? I would
be really grateful If someone could.

Thanks in advance

SteveS

[isambardkingdombrunel]

You have mail.

regards,
IKB.

[johny]

Steve,

I don't think the causes I gave in " A problem with Distant signals" are necessarily causing you problems as that problem only occurs in about four or five places on the S&C route.

However, one thing I have discovered recently, is that there appears to be a limit on the number of signal shapes you can have in the sigcfg.dat file. I am waiting on the return of Edmund Kinder from his holidays and will discuss this with him.

I am in the process of compiling a signal kit of LSWR LQ semaphores (to scale not 2/3rds as Hubert's and MSTS's) for issue at some time in the near future, and as I see it I will have to confine myself to only issuing 18 or so shapes as Hubert Feyerabend did.

One solution maybe is to do what I understand IKB does and delete all the installed signals before attempting to install such a large kit as Edmund's. If you do this make sure you do delete them all. Another way would be to decide exactly which signals you require and only install those.

One thing to remember is that some distants require junction links just as the stop signals do, especially those on junction bracket arrays. Single distants should work OK as they are just literally installed. Also remember that a distant preceding a branch stop that is cleared will be on as it is an instruction to the driver to slow down for divergence, but if the main stop is cleared then so should the preceding distant be.

John

[40058]

Steve,

I don't think the causes I gave in " A problem with Distant signals" are necessarily causing you problems as that problem only occurs in about four or five places on the S&C route.

John

Sorry John...no I was'nt suggesting that the causes you gave were causing my problems, I think generally (as it's my first attempt) it's my
inexperience and the fact I've put too many signals in. I've removed a few
home/distant and a few distants and more of the remaining distants seem to be working. I will be resignalling again replacing like for like as per the S & C 1930 route especially now as IKB kindly sent me those sigcfg and sigscr files with the ground signals in.

May be worth noting that one or two of the original distant's in the Llangollen-Bala route are not working too.

SteveS

[uknut]

What a coincidence. Today i also started resignalling my route with these signals from the oter ones mentioned. IKB would it be possible to send me those files with the ground signal included?

I have also had problems with the distants not working and i have only so foar included a few signals. Any ideas?

[johny]

If it's any consolation I get problems with the dat files when checking out a signal just created, and the line numbers that the RE throws up have nothing to do with that line in the dat file, they are some sort of code saying there is an error, thanks very much for that one MS and Kuju. These files are a right pain and there is no quarantee that if everything is OK for me it will be OK for someone else.

John

[40058]

If you want my opinion I think Edmund kinders CL & Semaphores plus Goingnorths set too should be made default in MSTS 2 after all they're so authentic and I think all types of UK signal used should be made available in one neat sigcfg and sigscr file set....no guarentee distants will work though.
Still I'll resignal S & C 1930 again using like for like now that I have the sigcfg and sigscr files to do it.....might be an idea to use CL's at Carlisle then semaphores second station onward up the S & C line.

SteveS

[Goingnorth]

Well I'm all for a project group getting together and producing a really authentic set of signals. 2/3/4 aspect with feathers, plus also lot's of the older semaphore signals, including pre-grouping like the GNs somersault signals.

Then produce several signalling packs...and one big one.

Signalling fans apply below...we also need programmers to sus out the scripting!

[johny]

As far as semaphores are concerned, other kits are in the pipeline, I have already mentioned LSWR LQs which were still in use in BR days alongside the much more evident UQs. To go with them would be SR UQs, Edmunds semaphore arms would be alright but the posts would need to altered. The SR used posts made of redundent rail, and lattice posts and also tho older LSWR wooden posts.

Yes, Edmund's UQs are brilliant, he has certainly pointed the way, but to use those and nothing else say on the S&C, a default route in the first instance, would be as incorrect as the original MS/Kuju signalling. In BR days the older MR LQs could still be found on their wooden posts, I think Garsdale was one such location.

It is the colour-light signalling where there are problems with the scripting, and as Rob says we really need a programmer to come up with the goods.

John

[Goingnorth]

The problem with the MSTS scripting is it's a 'C' derivative. After sitting staring at it for a bit, I was able to get my head around it. I'm sure you could do more, but not being an expert in C there are limits.

I could learn C, but even then it doesn't necessarily match up with the MSTS script.

You've got a limit of signal logic I/O functions, which leads me to believe that you can't do approach control:

block_state ()...Gets the state of the Block ahead of the train. Returns true/false. The current block state would obviously be occupied (if you could return such a value), but really what you need is some sort of marker to 'approach release' the signal as per reality.

Various functions allow you to get the state of other signals and whether a certain signal is present. I thought about use a signal as a marker, but it messes up your block logic. You could recalculate that, but I'm not sure how given the variables and the flags from the running database you have to work with.

[micksasse]

Can anyone recommend a good basic tutorial covering the principles of signal placement in MSTS - how to link them with junctions, etc etc. This is one thing which is obviously essential to any route - and, I think, the only thing of any importance which isn't dealt with in Michael Vone's otherwise excellent Route-Building Guide.

Any recommendations?

Thanks a lot

[johny]

There are no tutorials as such on this topic.

Place the placement cross in the track, click and the signal and its red pyramid shape will be in the middle of the track. Move the signal to one side and position it. Right click the signal or select the pyramid and right click that and the signal properties dialog box opens. Click on the tab labelled Signal and if the link button is greyed out that signal does not require a link. If links are required there will be two or more link buttons, click on each of these in turn and a route map opens on which you click which route or siding the signal is set for. Just place the cursor, which is a circle, over the line and it will turn red, then click and it will go green.

That basically is it.

John

[micksasse]

Thanks for this. My problem is in trying to work out linking - that is, using signals to differentiate between access to various diverging tracks. I'm confident that this is a deeply stupid question, but would like to work out how to do this - and also how to allow the signals to link with speed limits, so as to allow e.g. linespeed to be maintained on the straight through line and impose a 20 m.p.h. limit on the diverging track.

A second question: with Xtracks 3.x it's fairly easy to create some quite effective-looking catch points, which I'm wanting to use to protect the entrances to my single-track sections - is it possible to arrange the signals to show a red aspect when set for one exit of the points (i.e. for my purposes when the catch-points are set against the train) and a clear aspect when they're set to enter the single line? (I realise that this is kind of logically putting the cart before the horse but can't imagine it working the other way round so that a red signal automatically sets the catch points against the train).

Apologies if this is all staggeringly obvious!!!

[johny]

The original scripting for the UKLQ and UKUQ semaphores is arranged such that a distant preceding a clear branch stop signal is always set to caution and the driver therefore has to slow down. The distant will show clear only if the main stop and the next three blocks are clear. Edmund Kinder kept this arrangement for his kits.

Not too sure what you mean by catch points. If you are talking about overruns say at a station on a single line where the track is double, a small semaphore (siding signal) would be used for shunting into the overrun and a disc signal for backing out, the main signal controlling access to the single main line.

John

[micksasse]

Cheers John, I think you've answered my first question - in any case hopefully still have some time left to work out signalling as not yet finished track-laying.

What I meant by catch points (a.k.a. trap points) was those used traditionally to protect conflicting traffic paths, such that when the signal to enter restricted section is (was? Railtrack seem less and less into these basic safety features!) on (red), if the train proceeded the points would just derail it, thus having a minor derailment rather than a head-on crash - you can simulate these visually quite nicely with the modular points in Xtracks.