Route Building Ambitions--East Tennessee Railway (ETRY, former ET&WNC "Tweetsie")

[yellowdog]

I would like to learn route modification. I figure the best way to learn how to do that is by building a route. Route building has always been intimidating to me. But Eric and Scott have put so much effort into updating our tools over the last year, that it seems a shame not to use them. So here I am.

The route I'm considering is a short one. I'm looking at the East Tennessee Railway (aka ETRY) as it existed in the early 90s between Johnson City, TN and Elizabethton, TN. 100 years ago the line was a narrow/dual gauge line called the East Tennessee and Western North Carolina Railroad (aka the Tweetsie) running from Johnson City, up and over Roan Mountain, NC, to Boone, NC. (As an aside, that would be an interesting modelling endeavor, but is too much for me to undertake for sure, given all the complexities). Today it is little more than an interchange yard between the CSX (former Clinchfield) and NS (former Southern) yards in Johnson City. But in the 1990s it was an 11 mile long standard gauge short-line, serving small customers in Johnson City and Elizabethton (as well as the larger Rayon plant in Elizabethton) in addition to being an interchange between CSX and NS. That seems like a right-sized project to undertake.

So here I am. I've used TSRE for years as a consist editor, and currently am on the trainsim.com stable fork (8.005) as my daily driver. But I know I'll need some other tools. Which is why I'm posting here--looking for advice on what I'll need.
Here's what I have so far:

TSRE: Eric's 8.006ndds version, Scott's latest GenX version, Goku's vc_0.7.620 version.
Scott's SCO LIDEX tool
Pete Willard's latest TSRE guidebook

This is an abandoned line, but I found this map generator

https://www.frrandp.com/p/the-map.html?m=1

and then found some track charts and other maps (and just general modeling tips, mostly for HO scale) here

http://cfordart.com/20yearsofthestemwinder/info.html

and I also have a print book coming in the mail.
I also have the Clinchfield track charts/profiles, which include the interchange point with the ETRY.

I also have various track and road packages downloaded (XTracks, DBTracks, MTracks, ATracks, ScaleRail, ScaleRoad, NewRoads)

I know I need to download terrain files for the area. I'm vaguely familiar with KML files, and I know I've used a key before in TSRE. But I only have a vague notion now of what that actually entails...

Beyond what I've listed above, anything else I will need to get started? (Not worried about buildings and structures at this point )

Thanks!

[eric]

Awesome... More TN routes...

I'd lean towards Scott's build right now.

If this is going to be a true not-100%-fully-automated railroad, which I suspect it was and is, you'll want switchstands for the visuals. Scalerail will get you working switch stands out of the box, but might have more of a learning curve than ATracks and XTracks. You can also add a switchstand manually as a signal object, which has its own learning curve but opens up using any of the track systems.

[yellowdog]

Good point, Eric! I hadn't considered switch-stands to be separate items, so I'll go look at options. On a related note, I think the only signals would be in and around the CSX and NS yards. I can't remember if there are signals controlling the actual interchange points. But I was thinking it would be nice to build enough of those respective lines to be able to simulate run-through traffic, which would probably mean installing a few signals.

[yellowdog]

Well, I tried to dive it. And smacked my head on the concrete bottom of the shallow end. lol

I went to the Google Earth website, and (as far as I can tell) generated a path along the 11 miles of ETRY ROW (not including sidings, yards, spurs, industries, etc). This goes from the NS (nee SOU) connection in Johnson City to the literal end of the line in Elizabethton. That didn't seem too hard, and having multiple years to choose data from was very handy.

Google Earth Paths.jpg
Then I exported said Map to kml, using the File menu in Google Earth. That downloaded the kml to my local machine. Then I opened up Pete's guide to TSRE route building and tried to follow along. I opened TSRE (Eric's last stable version, 8.005), and clicked New. It took me a minute to figure out how to convert my Google Earth Map settings from DMS to straight decimal, for the TSRE required entry. But I got it (I think....see below). Then closed TSRE and dropped the kml file in the new route folder that TSRE generated. I then re-opened TSRE, opened the Navi Window, and used the drop down to select the kml. It had a couple of options I could select and Jump to.

Navi Window.jpg
Then I booted up Scott's SCO LIDEX tool. It took me a bit to figure out how to work it, but I selected my kml options (I think....see below) then conducted the Scan and Run features. That took a few minutes to complete, but generated no errors (that I can tell).

LIDEX.jpg

Then reopened the route in TSRE, opened the Navi window and jumped to End of the Line, and this is what I get:

end of the line.jpg and this is what I Looks like some terrain, and some path things.

But following the black line, it very quickly ends (exactly where you see it end in the picture above) and then there's just a void:

void.jpg

And jumping to the other end (SOU Connection), I'm completely in the void (but still with some pieces of marker/path):

SOU connection.jpg

Jumping back to the End of the Line and then rotating the camera, it looks like I only have 1 square/tile/whatever of green. Otherwise it's just void. So clearly I did something wrong. Cause the KML does not load up as much data in TSRE as it displays in Google Earth. The "good" news, I guess, is that my 1 square has terrain data in it. lol. So maybe I got the LIDEX to work?? I'm attaching the LIDEX and TSRE logs, in case they help

SCOLIDEX.log
tsre-log-20260818-2047.txt

[eric]

When you encounter those blank areas, press B to create a new tile.

[scottb613]

Hi...

You're running the original 1.000 release of LIDEX.

There was an issue with creating tiles that didn't exist - that was fixed in v1.200 - we're now on v1.300.

Also - I would recommend setting Tile Radius to 2 when creating a new route from KML.

https://github.com/scottb613/SCO_LIDEX

Capture.jpg


Regards,
Scott

[yellowdog]

Hi...

You're running the original 1.000 release of LIDEX.

There was an issue with creating tiles that didn't exist - that was fixed in v1.200 - we're now on v1.300.

Also - I would recommend setting Tile Radius to 2 when creating a new route from KML.

Regards,
Scott

Indeed I was! Thanks for the heads-up--I downloaded the latest version and ran it again, to include Tile Radius of 2. Much better results!
View from End of Line shows a significant increase in tiles:
Lidex 1.3 1.jpg

When I follow the black path, it seems to go all the way to the other end (i.e. SOU Junction), based on the lat/lon location indicator. However.....if I jump to SOU Junction, no black line/path in sight. And there's a huge monolith just off in the distance:

Lidex 1.3 2.jpg
There are a couple other free-standing pillars scattered around, too. I have a vague memory that artefacts can sometimes crop up in the terrain data, but I seem to recall those are spikes rather than what I'm seeing. I'll search for that topic to learn the solution.

[scottb613]

Hi...

Can you post the log from the run - your KML file - and lat lon of your start tile - I've never seen a spike like that, and I'm supposed to be handling it... I've ran it on a number of routes.

Thanks.

Regards,
Scott

[yellowdog]

Hi...

Can you post the log from the run - your KML file - and lat lon of your start tile - I've never seen a spike like that, and I'm supposed to be handling it... I've ran it on a number of routes.

Thanks.

Regards,
Scott

Hi, Scott
The forum won't let me upload a kml file (either in the forum or in the messenger). I can potentially share the Google Earth Web link, but the Messenger is also giving me an error that it only accepts 85 characters, which doesn't leave much room for any text (much less the link). And seems like a very odd restriction.... I can at least attach the LIDEX log til we figure out a work around.

[yellowdog]

Also, just fyi, I created the terrain via the Overwrite function of LIDEX.

[scottb613]

Hi - thanks - if you put everything in a zip file - you should be able to upload it

[yellowdog]

Hi - thanks - if you put everything in a zip file - you should be able to upload it

Whoops, didn't think about that.... Been a long day

[scottb613]

Gracias amigo...

[scottb613]

Hi...

The problem was:

• An overlapping 1 m USGS file contained only NoData in part of the tile.
• LIDEX incorrectly treated that as a download/source failure and stopped instead of continuing to 5 m and 10 m fallback data.
• The Tennessee 5 m OPR elevations were stored in US survey feet, but were previously interpreted as meters—creating enormous terrain walls.

LIDEX now handles those issues - I built your route clean on my machine, and it looks good now - gotta luv those mountains.

I am pushing the fix to Git now - just discard your old pack and download a new zip. Same version - the timestamp should be updated when you look.

Thanks for the bug report.

Regards,
Scott

[yellowdog]

Hi...

I am pushing the fix to Git now - just discard your old pack and download a new zip. Same version - the timestamp should be updated when you look.

Thanks for the bug report.

Regards,
Scott

Hi, Scott
I just downloaded the update and reran the program. Looks like everything is all good now! Thanks!!