Am I doing track laying on DEM right?

I heard that you should closely follow the terrain. I figured that the ballast should always be showing but never be floating. I got to work using 100M and 10M (a2t has no 20, 30, or 40 mstrs). The result? It looked really good just scrolling along, but I think the terrain is too sloped. Although the track variation looks realistic, I find it hard to believe my route has a %1.5 grade, even for short distances. And having the ballast always touching but not floating requires a lot of small (10M) pieces. Am I doing it right or is the DEM not perfect? I still haven't figured out how to get 10M DEM for Geneva, and it would really help because of the river bed. Track grade charts are expensive, but do I absoulutley need one?Thanks!
METRAUP

Chicago-Geneva Metra Line

Alert | IP

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

RE: Am I doing track laying on DEM right?, amtkfan79, Oct-06-03, 11:16 PM, (1) RE: Am I doing track laying on DEM right?, nwman, Oct-07-03, 08:42 AM, (3) RE: Am I doing track laying on DEM right?, DRGWK36, Oct-07-03, 12:18 PM, (4) RE: Am I doing track laying on DEM right?, metraup, Oct-08-03, 09:43 PM, (7) RE: Am I doing track laying on DEM right?, amtkfan79, Oct-08-03, 10:21 PM, (8) RE: Am I doing track laying on DEM right?, metraup, Oct-09-03, 09:48 PM, (9) RE: Am I doing track laying on DEM right?, amtkfan79, Oct-09-03, 10:16 PM, (10) RE: Am I doing track laying on DEM right?, DRGWK36, Oct-07-03, 07:47 AM, (2) RE: Am I doing track laying on DEM right?, geoffmayo, Oct-07-03, 05:12 PM, (5) RE: Am I doing track laying on DEM right?, metraup, Oct-08-03, 09:41 PM, (6)

Conferences | Topics | Previous Topic | Next Topic

Well you do have grade charts... they show the elevation.
There are no 1.5% slopes, and it will make your route highly unprototypical if they are there. And remember, those charts show gradients in percent, the RE shows it in degrees. Big difference there.
Of course to get the grades, curves, and elevations absolutely perfect takes a builder ten times as long and requires A LOT of calculations. Trust me, been there done that, and I'll probably never release the route I did it on(BNSF Ottumwa Subdivision).

Alert | IP

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

LAST EDITED ON Oct-07-03 AT 08:43 AM (EDT)

How do I convert the percents to degrees?Thanks for any help in advance.

Alert | IP

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

This is it! This is the answer to that question you asked back in geometry - why would I ever use this stuff in the real world? Well, OK, it's not exactly the real world, but in any case:Percent = rise in elevation divided by length of track run (rise/run) x 100%The tangent of the angle between level and the grade is the inverse tangent of the rise/run, or:deg = atan(rise/run)Grade = tan(deg)So a 4% grade (0.04 rise/run) has an angle of 2.29 deg from level - the closest to the MSTS coarse settings is 2.25, which works pretty well.The easy way to do this, however, is to download mstsgrad.zip from the file library, which has done all of this math for you.Andy

Alert | IP

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

LAST EDITED ON Oct-08-03 AT 10:03 PM (EDT)

Thanks! I didn't realize you had thrown them in there!Edit:
I think I am getting the hang of them! Just one thing, what are the 740.0 or so numbers between grade changes? And are the 0.70 and 0.10 etc above the tracks miles? To bad KUJU didn't do imperial.Thanks!
METRAUP

Chicago-Geneva Metra Line

Alert | IP

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

Numbers in the 600-700-800 range that are at grade changes would be elevation. Divide that by 3.28 and you get the Metric value.
The numbers on top of the "grades" are the gradients, expressed in percent.
In the Route Editor you rotate track by degrees tho... Somewhere in the F/L is a table showing a rough conversion of percent vs. degrees. There's an exact formula you can use for it, it involves some big calculations if you're up to that.

Alert | IP

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

Great! Thank you! I am also wondering, is there some scale? Like it it 4 inches to a mile or something like that?Thanks!
METRAUP

Chicago-Geneva Metra Line

Alert | IP

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

No scale when they're scanned and resized...
The mileposts are on there if you want to estimate your own scale, those would be the numbers above the diamonds at the top that are in order across the page.

Alert | IP

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

I was one who recommended following the terrain, but as with most advice, there are limits. I agree that you shouldn't have a 1.5% grade for the track, but it seems likely that the terrain itself would have grades like that, or even higher. When I recommended following the terrain, that had to do with differences between markers and terrain. For track, you want to have smooth transitions between different track grades, and use the "Y" key to cut/fill the terrain to meet the track. I didn't have an alignment diagram to work from, but wanted to keep from changing grade every 50-100m. So I would find a marker 1/2 mile or so down the line, figure out what elevation change I needed to get there from where I was, and calculate the grade from current to target points (elevation change divided by distance between points). I then place a 250m road section at the end of my track, elevate it by the correct amount (changing to degrees from grade, since the rotate tool uses degrees), and then swing the road around until it intersects the terrain - that often happens within 50m (in some cases, it may stay above or below the surface, but usually I can find an intersection). I then place a tree at the intersection point, and rotate the road section again to find the next intersection point. These are all at the same grade along the terrain, which results in minimizing the amount of cut/fill I had to do.Now that's for a mountain route, which will be different than a plains route. You're looking at long straight stretches, probably at a consistent grade. So you probably don't need to use the swinging road section, but figuring out an average grade increase over a 1/2 to 1 mile stretch, then laying track at that grade, and using the "Y" key to cut or fill is probably your best bet. When you get to a point where you have to make some grade changes, then use your 10m sections to change in small increments. If possible, change no more than 0.150 per 10 m (the smallest "coarse" increment). Even better, change about half of that using the "end" key - it really makes the transitions much more smooth than an abrupt change from 0.00 to 0.450.Sorry for the long-windedness. Hope this helps and doesn't confuse.Andy

Alert | IP

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

Bear in mind that DEM data is highly innaccurate when you're down to the sub-metre level of MSTS. Even 10m DEMs basically mean that the height in that cell is the average over a 10m square of earth (~30ft), which on a railroad line can vary by 1-2m just for the roadbed. Then add the fact that MSTS uses 8m squares (ie a different resolution) and that the DEM data has to be averaged and resampled to fit the 8m squares.Basically, you can assume that the DEM data is only right to within a couple of vertical metres. That is not enough to lay track according to the ground. You need to work out the grade over a length of track, say 100m, and try to make most of it sit on the ground. Then use the Y key to fill in the holes.It won't be perfect. There's nothing you can do to make it so.Geoff M.

Alert | IP

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

THank you for clarifing that! I guess I kind of always thought it was almost exact.Thanks!
METRAUP

Chicago-Geneva Metra Line

Alert | IP

[ Home | MSTS Forum | MSTS FAQ | File Library ] Copyright © 2002 by Nels Anderson. All Rights Reserved.