Subroadbed, Roadbed and Track

[jbvb]

I'm starting this thread for the whole Forum membership.  Demonstrations, experiments, Q&A, new materials are all appropriate. I'm going to start with my own oddball practices, but if you've figured out how to repair "over the throwbar" point contacts, keep old cork roadbed from getting brittle, repair broken Maerklin track connectors, or effortlessly fasten down Code 55 rail, let us know.

Full size or model, trains need track.  These days the major scales have multiple ready to lay options for most common RR track and turnouts. However, minor gauges, trolley fans and modelers of specific prototypes are often required to hand-lay what they want.  I learned to hand-lay when it was the standard of my college club. I find its durability and flexibility justify the extra effort to build it.

Roadbed ranges from molded-to-the-track (E-Z Track & competitors) through cork, sticky plastic, plain and profiled Homasote to pine lattice stock. I've used all except the sticky plastic; my experience with other people's modules hasn't encouraged me.

Subroadbed is a tradeoff between light and sturdy. Careful planning and construction can produce light but stable support, but the extra effort required usually limits it to modules. Home and club layouts today often use strips of plywood instead of the second-growth lumber most readily available.

The NMRA Achievement Program's Civil Engineer certificate used to require detailed track. Here's what I did with mostly Detail Associates parts on a hand-laid Code 100 HO turnout. Yes, Code 100 is really large rail in HO, but that was what the HUB Module standard required.



Joint bars and a strip applied to model a bolted frog (much less common these days than before 1980).



Post-1900 turnouts have special slide plates to support the points, and other special plates to support and position the closure and stock rails where they're too close together for conventional tie plates.  I didn't try to do physical tie plates, as they're normally only 1/2 inch thick; I painted them on.  Proto-48 and the related projects for other scales offer physical tie plates these days.

[deemery]

Marty McGuirk posted this good video on his replacing a turnout causing derailments.  Some good hints in here, but he didn't say how long, or how much cursing, it took to get the throw rod into the throwbar hole....   

dave

[KentuckySouthern]

I use N gage cork on plywood for the majority of my roadbed. Atlas C-100 track and combine Atlas turnouts and a few Fast Tracks hand built turnouts in my main yard.

I heavily ballast them after rail rust from initially Krylon or Rust Oleum camo. Then I changed to airbrushed brown after bad air concerns.

I don't detail my track like James discussed. Too much of it in 25'x35'.

KS

[jbvb]

Dave, I don't see a link. I watched Marty's "Track & Wiring Tweaks 2" video but he solved that problem by de-gunking his rail and frogs to fix a really messy ballasting job.  My handlaying scheme places and bonds the ballast before I begin laying rail.

Karl, the lines I've walked (and in this case, modeled) haven't been very high tonnage, so my ties, ballast and rail are three different colors.  That one switch is the only place I've detailed, though I am using laser-cut joint bars painted yellow to mark rail gaps for visiting operators. And I like the effect of painted "tie plates" enough that completing a couple of scenes is on the To Do list. It's been on it long enough that I doubt I'll ever do the whole layout.

My current track project is learning to build street track with simulated girder rail. This XTrackCAD screenshot shows the GE River Works in-plant track that remains to be built, highlighted yellow:



Whenever possible I start building new track where it connects to existing track. This avoids many kinds of Revolting Developments possible when "meeting in the middle". I could have used a single slip in place of the leftmost diamond crossing, but that would have put locos running around left of 6th Street, limiting car spots on that track.

[deemery]

Here's the link:  https://www.youtube.com/watch?v=mlo6sJuO75E

dave

[ReadingBob]

Great idea for a thread. I glue down the ballast before adding the rails as well. I stain the ties after they been glued down and sanded level rather than before gluing them down.. 

Got my popcorn ready to follow along on this one!

[jbvb]

At the Tech Model RR Club, I learned to hand-lay on 1/4" x 1 1/4" pine "lattice stock", usually supported on 3/4" plywood.  That layout was built so the track could be walked on when the overhead fluorescent tubes needed replacing. I did my Rowley Modules that way too, but the rest of this layout is mostly 1/2" Homasote on 1/2" plywood. I have maybe 30' of Homabed, but usually cut the corners of plain sheet down as below for a ballast slope:



Next to the styrene bridge piers you can see some of the 1/4" lattice stock roadbed.  I found that very reliable for years, but then bought one length that was much harder than I expected.  Of course, I didn't find out it bent spikes till it was glued down and ballasted, so I got out the pin vise and tiny drills.
 

[jbvb]

Recently, I've been exploring street track with simulated girder rail. General Electric's big River Works in West Lynn had a lot of in-plant track. Prior to WWII, it was operated with 600 VDC locomotives running off overhead trolley wire, e.g. using poles rather than pantographs. But by the time I'm modeling it had been dieselized. GE 44 tonners were in use around 1970, but GE's industrial 45 ton side-rod centercabs were used earlier.  Until the end GE's plant track had tight curves, tracks entering buildings and street running.

I know some trolley modelers and I watched several work at the Tech Model RR club. But I've never built any trolley track before. I didn't need 45 scale foot (6") curve radii or single point switches, so I chose Code 100 running rail with Code 70 guard over the Orr/Customtraxx cast parts. I decided to try building the Code 100 running rail parts first, then soldering the Code 70 guard in place once I'd tested operation.



I started by building the Code 70-Code 100 transition (left center) with some partial ties because the pavement edge will be angled. Then I continued locating the closest track, with its curved spur into the building. This will be crossed by two other tracks before it reaches the building, but crossing geometry is simpler than switch geometry.  I'll build the switch first.

I laid out track center lines. Because there won't be any ties, the straight track has more marks to show where its rails will go.  The center line of the curved spur turns toward the upper right corner. I'm using a plastic slotted gauge I probably got from MicroMark to locate the curved stock rail.  The Vertex Bend is at the line drawn across the straight track where the curve begins.  The Vertex Bend is an important point in a prototype switch: It's where the diverging stock rail is actually bent 3-4 degrees to protect the switchpoint from approaching wheels.

[jbvb]

Here's the link:  https://www.youtube.com/watch?v=mlo6sJuO75E

dave

Watching that, I see:

1. He didn't check the track gauge, or the gauge and wheel profile of the three cars that 'bumped' near the frog.

2. Blunt points, which I'd touch up with a 4" file.

3. The throw wire from the ?Tortoise? is forcing the throw bar to the left, which opens up gaps at the joiners connecting the points. I'd try bending the wire a bit from below the layout to see if it closes up the gaps.

First thing I'd look at is the RH truck of the MEC car. Might be broad gauge, or wheel not perpendicular to the axle, or maybe truck screw too tight.

Then I'd look for ballast in the flangeway approaching the frog, or between the guardrail and the stock rail across from the frog.

I'd only junk the turnout if the gauge was wrong through the frog. I think that's unlikely, because the loco and the first few cars went through without a bump. Only the last three boxcars bumped, and only the leading truck.  If the frog is blunt or the flangeway slightly misshapen, I can often correct that with a 4" Knife File.

[jbvb]

Here's the first street track turnout with the frog and points "sketched in". The Vertex Bend location is marked "VB" in pencil. I won't solder anything together till the adjacent track is ready to push cars back and forth. If something bumps or binds, I can adjust before making it permanent.



One rail of the 2nd track is visible above the turnout. I had started to sketch in the first curved diamond before I realized I needed this photo to show the steps of building a turnout.

[jbvb]

This shows the stock rails of the second turnout. The Code 100 rail is coming from 36" pieces of fiber-tie Atlas flextrack from the early '70s. I've left them full-length because fewer joints usually improves the alignment of handlaid track.



I've started making the closer diamond crossing by cutting the closer stock rail into the curved spur rails.  Next I'll do the farther stock rail. Once both stock rails are lying flat and loosely spiked, I can start on the closure rails and the second switch's frog/points.