Sub assembly composer does not work well (rail)

[Anonymous]

Rail_cant:example.pkt  is a good example to design a typical section of ferrocaril. However in the aid itself it can be seen that when applying AOR, the length of the benches (L40 and L43) is modified. When it should move at point P40 and P41. As the assembly must be configured for this to happen, especially if the sub-base plane is not horizontal, but has slope.

Thanks!

[wang890]

if it gets too complicated logic it is worth it to hire a professional to do it in .net

our provincial government programmed their own subassembly with custom conditions for subgrade superelevation.

SA composer is nice but has limitations.

I made some subassemblies for rail subgrade grading. but haven't get involved with real railway design much. I am only doing highways.

try start from scratch and define your own logic in SA composer. if you can define those cant with multiple if then if you have to it may work.

[doni49]

You're free to make any changes to the PKT file that you like.  But just as you need this change, I'm guessing that OTHERS do not.

Do you need help in figuring out HOW to change it?  If so, what trouble are you having?

[Anonymous]

Exactly, I have made the necessary changes in this assembly to adapt it to the needs of my project.

But in the curves when the section is tilted, the length of the lines (L4 and L6) is modified and the position of the points (P52 and P53) calculated as the intersection of the subbase line and points (P8 and P 10) with an established slope is fixed (in a relative position when the section is normal).
Points P8 and P10 are labeled AOR apply, while the subbase line is not.

What I need is that the points P8 and P10 in the sections in curve adopt the position that the cant determines, keeping constant the length of lines L4 and L6 and that points P52 and P53 are displaced keeping the design slope that is fixed and independent of the cant, as well as the slope of the subbase plane.

In straight sections the assembly works well

Attached is a file with which I am working for better explanation.

[doni49]

Please give me a screenshot showing a cross section that this produces.  And also a copy of the screenshot that's been marked up to show me what you want it to look like.  I'm looking to see what it DOES and what you want it to do differently.

I'll look at your pkt file in SAC but opening your dwg file won't do any good unless I install the pkt file first which I'd rather not do.

[doni49]

I took a look at your pkt file and I do see something in there and I think that's what your complaint revolves around.  P10 & P8 are both set to go from P9 & P7 respectively a distance specified by the Parameter "Banquina" "Banquita".  But yet they're not the same width (except P8 uses the NEGATIVE version of Banquina Banquita).  Correct?

As best I can tell you've got it coded in a way that SHOULD result in them being the same.  I'm not sure what's causing this.  I'll keep looking but in the meantime, hopefully someone with rail design experience can chime in.

[Anonymous]

Exactly, the "banquina" parameter whose value is 0.45m must always be constant, even in sections with a slope.
However this does not work well on the curves, because the assembly modifies this length instead of modifying the position of the points P52 and P53 of the assembly.

Attached three screenshots, the first is from the normal section (straight) all parameters are correct.
The second is the section that produces the assembly (bad)
And the third is the same section as it should be according to the project parameters (cyan)

[h.malgarezi]

mr1181: I also tried to create this kind of rail subassembly and had exactly the same problem as you did but coudn't figure out how to solve it..

I hope someone can help us with this.

Anyway, the workaround we use here is: create a corridor for the subballast and a second corridor for the ballast that targets the surface of the first one.

But this way won't let you have a fixed 0,30m 'Banquina' between ballast and subballast. It will be necessary to keep a constant width for all the subballast (7m, 8m..) that will fit the ballast at maximum cant (superelevation) for both sides.

[Anonymous]

Thanks for your answer, the possibility of doing two corridors is the solution that I am taking temporarily but I would like to be able to design a section that works correctly in the whole project.
Greetings.

[BrianHailey]

@doni49 asked me to take a look at this so I did a little bit of playing around. It appears that the numbers are being calculated correctly but something is still really weird.

Here's what I did, I created two new points in the subassembly (just to test things) and I set them as a delta x delta y from the origin. I set the first one to be a delta y of 0 and a delta x the distance between the x values of points p7 and p8. I repeated the process for the other point and set it's delta y to 0.05 (just move it up a little bit so they aren't exactly on top of each other) and a delta x the distance between the x values of points p9 and p10 and they line up exactly.

If nothing else, it appears that you are doing things correctly, it's the software that's messing things up. I'll keep looking at it, perhaps something else will become apparent.

Quick thought, is the subassembly behaving the same way in the file? Perhaps it's just a display issue in SAC.

[doni49]

Thanks @BrianHailey for looking at it.

[h.malgarezi]

When I tried to create this kind of subassembly, the behaviour was the same in Civil 3D and SAC.

[Anonymous]

hello, I'm new here so I would really appreciate your patience... 

I'm not sure how to create a corridor for the subballast and a second corridor for the ballast that targets the surface of the first one, thats why I've tried to set offset targets to P8 and P10 using two different alignments for each side, also setting the points properties but it didn't work out 😞

Could you please explain me how to generate a corridor for the subballast and another for the ballast ? do you use the same pkt file or have you modified it?

thanks in advance!

---

@h.malgarezi wrote:

mr1181: I also tried to create this kind of rail subassembly and had exactly the same problem as you did but coudn't figure out how to solve it..

I hope someone can help us with this.

Anyway, the workaround we use here is: create a corridor for the subballast and a second corridor for the ballast that targets the surface of the first one.

But this way won't let you have a fixed 0,30m 'Banquina' between ballast and subballast. It will be necessary to keep a constant width for all the subballast (7m, 8m..) that will fit the ballast at maximum cant (superelevation) for both sides.

---

[Anonymous]

I will try it!

thank you very much, sorry for my late response

[Anonymous]

Hello,

There is any chance for patch which will solve this problem?

Regards