Switch & Smooth curves

Hi all,
I have recently decided to use smooth curves and switches on the Barrow - Carlisle. However, there is a slight problem. The places where the switch meets the straight track just doesn't look right as shown below. Can anybody help?PS Sorry to any copyright holders whose work appears below

Derek kaye

http://bvesimulator.topcities.com
kayedj@hotmail.com

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

RE: Switch & Smooth curves, ferrovipathe, May-31-04, 02:12 PM, (1) RE: Switch & Smooth curves, Uwe Post, Jun-01-04, 05:52 AM, (3) RE: Switch & Smooth curves, cjh401, May-31-04, 02:16 PM, (2) RE: Switch & Smooth curves, kayedj, Jun-01-04, 07:47 AM, (4) RE: Switch & Smooth curves, Stooopid_person, Jun-01-04, 09:25 AM, (5) RE: Switch & Smooth curves, kayedj, Jun-01-04, 10:15 AM, (6) RE: Switch & Smooth curves, Stooopid_person, Jun-01-04, 11:55 AM, (7) RE: Switch & Smooth curves, kayedj, Jun-01-04, 12:59 PM, (8) RE: Switch & Smooth curves, ferrovipathe, Jun-01-04, 02:15 PM, (9) RE: Switch & Smooth curves, kayedj, Jun-01-04, 02:47 PM, (10) RE: Switch & Smooth curves, TomBoyRacer, Jun-01-04, 04:13 PM, (11) RE: Switch & Smooth curves, kayedj, Jun-02-04, 07:49 AM, (12) RE: Switch & Smooth curves, Technical Numpty, Jun-02-04, 08:52 AM, (13) RE: Switch & Smooth curves, ferrovipathe, Jun-02-04, 10:28 AM, (14) RE: Switch & Smooth curves, Technical Numpty, Jun-02-04, 10:56 AM, (15) RE: Switch & Smooth curves, kayedj, Jun-02-04, 02:06 PM, (16) RE: Switch & Smooth curves, TomBoyRacer, Jun-02-04, 03:02 PM, (17) RE: Switch & Smooth curves, Technical Numpty, Jun-02-04, 06:35 PM, (18) RE: Switch & Smooth curves, GyzmaCZ, Jun-03-04, 06:34 AM, (19) RE: Switch & Smooth curves, kayedj, Jun-03-04, 09:09 AM, (20) RE: Switch & Smooth curves, Technical Numpty, Jun-03-04, 09:49 AM, (21) RE: Switch & Smooth curves, kayedj, Jun-03-04, 10:08 AM, (22) RE: Switch & Smooth curves, GyzmaCZ, Jun-04-04, 04:45 PM, (23) RE: Switch & Smooth curves, Technical Numpty, Jun-04-04, 09:36 PM, (24) RE: Switch & Smooth curves, GyzmaCZ, Jun-05-04, 01:36 PM, (25) RE: Switch & Smooth curves, Technical Numpty, Jun-07-04, 06:28 AM, (26) RE: Switch & Smooth curves, GyzmaCZ, Jun-08-04, 02:18 AM, (27)

Conferences | Topics | Previous Topic | Next Topic

>Hi all,
>I have recently decided to use smooth curves and switches on
>the Barrow - Carlisle. However, there is a slight problem.
>The places where the switch meets the straight track just
>doesn't look right as shown below. Can anybody help? That's a hard point of BVE. You must compute everything if you want to create realist switches and curves.Your track (on switch) is going to left with some radius, and follow ( after the turnout) a line with an X angle from track 0.
Your problem, here, come from length (or, it's mean the same, the radius) of your switch.
Compute first your angle of deviation (after the switch).
Then, create a switch with the same angle.
As this one seems small, your turnout must be longer than 25 m, as your angle (on switch) reach the same as the angle of following track.Angle of deviation = 360°* (Length of switch)/ (2* Ð*Radius)
And Radius = (Length of switch)^2 / (2* Deviation ) (in Radian. 360° = 2* Ð radian)If you want some explain, look at:http://ferrovipathe.free.fr/Docs/Computes.doc.It was just a quick writing, I am writing a better "How to...", and put it on web soon.All the best.

http://ferrovipathe.free.fr

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

LAST EDITED ON Jun-01-04 AT 05:53 AM (EDT)

>That's a hard point of BVE. You must compute everything if
>you want to create realist switches and curves. Or use RouteBuilder http://routebuilder.bve-routes.com

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

LAST EDITED ON May-31-04 AT 02:17 PM (EDT)

HiYou need to make a smooth curve on the incoming track/outgoing track.
What i mean is, if teh x offset of teh points is 2m, then the track effectively leave the junction at a radius of 625m. So, if the points leave to the left, then make a right curve of 625m radius to use directly before/after the points. Hope that makes sense. That way the new track will curve gently to point in the correct direction and not sudddenly as in your pics!HTHEDIT: Sombody beat me to it!

Chris H
------------------------
Ramsbury Route for BVE
www.freewebs.com/ramsburyBVE/index.htm

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

LAST EDITED ON Jun-01-04 AT 07:47 AM (EDT)

It doesn't really make sense I think I will just leave it because you can't really see it from the cab Thanks for your help though.
Derek kaye

http://bvesimulator.topcities.com
kayedj@hotmail.com

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

>It doesn't really make sense I think I will just leave it
>because you can't really see it from the cab
>
>Thanks for your help though. He means, make the switch longer, e.g. make the switch length 50m instead of the present 25m.

-----You are reading my signature now, don't waste your time reading it!-----
Or probably, maybe you should waste your time reading it.
Eh, where did my signature go?
Oh, here it is, the one you are reading right now? Can't you see it? I know I know, it's probably too bright to see it.
Hey, look at my avatar, nice right? I drew it with my own hands holding the mouse in Paint, with the pencil tool(and the Filling tool!)!
It's a picture of me in Planet Stooopid, where everything is the same as what you call "Earth" except that we are called "aliens" in your planet.
Anyway, I warned you not waste time reading my signature! You have just wasted probably about 30 seconds reading it, where you could have spent the time doing better things, like maybe reading the next post below if there is one...

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

>He means, make the switch longer, e.g. make the switch
>length 50m instead of the present 25m. ok, i'll try that...as can be seen, it still meets the end rail at the wrong angle. (I have left the old line in to illustrate the original angle )Derek kaye

http://bvesimulator.topcities.com
kayedj@hotmail.com

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

That means, it'll have to be even longer! Maybe 75m!

-----You are reading my signature now, don't waste your time reading it!-----
Or probably, maybe you should waste your time reading it.
Eh, where did my signature go?
Oh, here it is, the one you are reading right now? Can't you see it? I know I know, it's probably too bright to see it.
Hey, look at my avatar, nice right? I drew it with my own hands holding the mouse in Paint, with the pencil tool(and the Filling tool!)!
It's a picture of me in Planet Stooopid, where everything is the same as what you call "Earth" except that we are called "aliens" in your planet.
Anyway, I warned you not waste time reading my signature! You have just wasted probably about 30 seconds reading it, where you could have spent the time doing better things, like maybe reading the next post below if there is one...

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

>That means, it'll have to be even longer! Maybe 75m! It cant go any longer for two reasons:1) there is another switch on rail 3
2) rail 0 starts a curveI'll just leave it as its not really noticable
Derek kaye

http://bvesimulator.topcities.com
kayedj@hotmail.com

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

If you want some more help, we need to know:- deviation between the 2 tracks (on the left)
- and distance between the 2 switches.Exemple: deviation = 4 m, and there are 75 m between the 2 switches.I'll compute tomorrow for you.All the best.

http://ferrovipathe.free.fr

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Is this any better?

Derek kaye

http://bvesimulator.topcities.com
kayedj@hotmail.com

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

you are probably better off using a 25m point with a -3 metres offset, then using 25 metres of normal track and then a right hand point with an offset of 1.5m to lead into the first siding if you want it to be parallel with the straight bit of the main line. using such a long point with a wide offset means you have to have an equally wide offset at the other end.

How can you eat your pudding if you don't eat your meat?

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

>you are probably better off using a 25m point with a -3
>metres offset, then using 25 metres of normal track and then
>a right hand point with an offset of 1.5m to lead into the
>first siding if you want it to be parallel with the straight
>bit of the main line. using such a long point with a wide
>offset means you have to have an equally wide offset at the
>other end. Like in my first post then?
Derek kaye

http://bvesimulator.topcities.com
kayedj@hotmail.com

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

LAST EDITED ON Jun-02-04 AT 09:30 AM (EDT)

Hi Guys,Part of the problem that I found when I first started playing with Switch is that it produced track of a slightly different gauge than the 'plain' rails used elsewhere. I see that this is also happening, but it's not really the problem you're facing.It would appear that part of the problem is that your switches turn out further than you need them to. This means that you have a choice. You can either custom build a switch for every situation or produce two or three "standard" switches (tight, medium and open, for example) and standardise all your junctions into these three 'grades'. You'll lose a small amount of realism, but you won't have to recalculate every single instance and have a dozen or more switch objects. Where a track comes back parallel, you will also need a corresponding curve of the same deviation in the opposite direction.The others have covered pretty much all the other problems associated with these formations, but if I think of anything else I'll add it in.EDIT:Just had another thought.Correct me if I'm wrong, but the siding appears to run straight off the mainline rather than arcing away from it. If this is the case, I would be inclined to start Rail(0)'s turn to the right earlier and have the switch the other way around. It might be a bit of a cheat, but it might help to make the calculations a wee bit easier./ / / //
| / / //
|/ / //
| / //
|/ //
| //
| //
|//
||...as opposed to.../ / / //
| / / //
|/ / //
| / //
|/ //
| //
\ ||
\||
||...if you see what I mean.HTHTechnical Numpty

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

>It would appear that part of the problem is that your
>switches turn out further than you need them to. This means
>that you have a choice. You can either custom build a switch
>for every situation or produce two or three "standard"
>switches (tight, medium and open, for example) and
>standardise all your junctions into these three 'grades'.
>You'll lose a small amount of realism, but you won't have to
>recalculate every single instance and have a dozen or more
>switch objects. Where a track comes back parallel, you will
>also need a corresponding curve of the same deviation in the
>opposite direction. Yes, that's the better way to build complex tracks map.About your problem, we got:
Sinus(Angle) = 3 /25 = 0.12 => angle = 0.12 radian = 6.87°
Radius of turn out = (25^2)/(2*3) = 104.1667 mSo, create a switch of 25 m long, 104.166 m radius, or -3m left offset.
Then, append a straight track between +25m and +50m, going from - 3m to -6m.
Then, you can a second switch (right one) of 104.666m radius, and just after that one, the same switch as the fiste one, but turn by 180°.
You may have your three switches, one on rail(0), second on rail(1) (the first on left), third on rail(1) (second on left).
But, it is more simple to not turn on right on rail(0) until the end of third switch, like Technical Numpty draw it.All the best.

http://ferrovipathe.free.fr

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

>But, it is more simple to not turn on right on rail(0) until
>the end of third switch, like Technical Numpty draw it. Hi There!!Oh. My track plans didn't come out very well. They looked much better when I was typing them.What I was actually suggesting was that Rail(0) starts it's turn earlier rather than later. The first switch that is causing all the problems could then be made to deviate to the right rather than the left so that the path into the sidings is straight and remains so until all the switches are passed. That way the siding is the tangent to the arc described by Rail(0), making the calculations simpler and not changing the scene as it exists too much.Just a thought...TN

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

LAST EDITED ON Jun-02-04 AT 02:08 PM (EDT)

I kind of understand.Technical Numpty:Do you mean have rail 0 turn all the way e.g.
|........//
|.....//
|..//
|//
||or turn then straight e.g.|....//
|...//
|..//
|.//
|//
||(these probably wont turn out well)Ferrovipathe:I'm not sure here either Do you mean:..|..|..//
..|..|.//
..|\.|.||
..|.\|.||
..|..\.|| }Stright
..|...\|| }Switch
..|....|| Again sorry about the diagrams <EDIT> to make the diagrams look neater </EDIT>
Derek kaye

http://bvesimulator.topcities.com
kayedj@hotmail.com

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

so basically you want it to look like this

How can you eat your pudding if you don't eat your meat?

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

>I kind of understand.
>
>Technical Numpty:
>
>Do you mean have rail 0 turn all the way e.g.
>|........//
>|.....//
>|..//
>|//
>||
>
>or turn then straight e.g.
>
>|....//
>|...//
>|..//
>|.//
>|//
>||
>
>(these probably wont turn out well) LOL!! Don't worry Derek. You're nearly there.It's a little difficult to get an accurate idea of exactly what layout you're trying to achieve at the moment. It would appear from your first screenshot that you've got three sidings coming off Rail(0) that run parallel to Rail(0) as it rounds a curve (and then straightens...?). I'll try outlining my suggestion again and hope that it displays better this time. ./././..// }Continue curve or straighten Rail (0)
|././..//
|/./..// }RH Switch (3) for Siding
|./..//
|/..// }RH Switch (2) for Siding
|..//
|.//
|// }Start curve on Rail(0) with RH Switch (1) here (See Below).
|| }StraightKeeping one track straight while the other curves at a constant radius keeps things nice and tidy from the mathematical perspective. It is possible to calculate the distance between two curving tracks, but it's a right faff. For your project, Rail(0) would follow the RH deviation of the switch with the straight on heading into the sidings.This screenie is from a still-born project that may eventually see the light of day at some point (maybe) and sort of illustrates what I mean. Apologies to copyright holders.Rail(0) is straight up until the switch, and carries on curving at the same deviation as the switch for as long as you like, but at least until past RH Switch (3) to ease the calculations. If you can imagine the sidings coming off this straight piece of track and curving parallel to the mainline, you've virtually got what I'm trying to convey.I hope this makes sense. I've read it a few times myself and I'm not entirely sure. Any problems, give me a shout.Technical Numpty

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Hi,
I can make something with your problem with Switch. If you want, send me your route and I repair this problem with angle and many others you have.Bye GyzmaPS: Don't worry, I wil no distribute your route.
PS2: Sorry my bad English------------------------------------------------
http://www.bve.wz.cz

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

LAST EDITED ON Jun-03-04 AT 09:20 AM (EDT)

heres the track plan and scene (sorry about my lack of paint skills )PS. what do you think of the new track textures?
<EDIT> For spelling </EDIT><EDIT2> The above makes sense now but If I want to keep it vaugly prototypical I can't do that. I hope the above track plan / screenie shows this</EDIT2>
Derek kaye

http://bvesimulator.topcities.com
kayedj@hotmail.com

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Hi Derek,Thanks for that. It makes much more sense. It also indicates to me that my suggestion might not work.I always hate trying to replicate junctions and other complex track arrangements, but an investment in time works wonders and gives much better results than guesswork or trial and error. What you're trying to pull off looks a bit of a pig given that there are curves, straights and switches, and that all the rails are moving away or towards each other at various points. I'd try and simplify it a bit if at all possible.The way I go about it is like this.Step 1:
Work out how far I want each rail to be from Rail(0). Don't be afraid of getting down to millimetres if that's going to look best. Try to standardise your distances as far as possible to cut down on the need for lots of extra switch objects.Step 2:
Create your switch objects using Switch remembering that, for a long crossing, you'll need two 25m objects to complete it. The switch that you're currently using that is causing the problems (50m, -6m deviation...?) mated with an opposing switch (if that's the correct phrase) would be fine if to use if you wanted to create a 100m long crossing between two tracks 12m apart.Calculating this for your junction might be a problem as you have a number of complicating factors, chief of which is that Rail(0) does not remain straight. This can be overcome by using an imaginary datum line that goes where Rail(0) would be if it was straight. All you then need to do is to calculate how far from that line Rail(3) is at 25m intervals and add it to the distance that Rail(0) is from the datum line at the same intervals. You will get some error due to the fact that Rail(0) curves, but these can either be corrected by eye using Track Viewer and "fiddling" the numbers, or ignored if they're out of view.Step 3:
DON'T PANIC!! It always looks worse when it's half done.All I can really suggest is that you spend as much time on it as you feel you can afford, but to concentrate on getting the numbers right. BVE can handle track placement down to millimetres, and getting it right on paper will make it look really good on screen. If you'd like some help, I would be happy to have a stab at it for you if you'd like to PM or e-mail me with the distances and dimensions of the layout, together with the seperations you would want to use.HTHTechnical Numpty

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

You have e-mail Derek kaye

http://bvesimulator.topcities.com
kayedj@hotmail.com

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Hi,
I read your mail now. I have this route and I have no problem or error 53 with this route.I make something with Swith. You can see on picture. It is't completly rebuild. Only small part. It is difficult an hard work. I thing-It need much timeGYZMA

Attachments

http://www.train-sim.com/dcforum/User_files/40c0df2e61935213.jpg

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

>I make something with Swith. You can see on picture. It is't
>completly rebuild. Only small part. It is difficult an hard
>work. I thing-It need much time
>
>GYZMA Oh. Righto, then.Do you want me to carry on looking at this as well or shall I stop?Technical Numpty

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Hi Technical Numpty. I don't want you to carry on or shall you stop. It is only one possibility how repair this route with nice curve and switchGYZMA

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

...and you got there first.Thanks, I get the picture.Technical Numpty

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

>...and you got there first.
>
>Thanks, I get the picture.
>
>Technical Numpty I´m not here first time. I was here much time ago I´m interesting about BVE for 3.5 year!(since 2000) But my Englisch is not good so I read this forum only and no writeGYZMA

Alert | IP

[ Home | MSTS Forum | MSTS FAQ | File Library ] Copyright © 2004 by Nels Anderson. All Rights Reserved.