• Welcome to AnyRail Model Railroad Forum. Please login or sign up.
November 17, 2019, 11:56:25 am


Due to heavy spamming attempts on this forum, automatic registration has been disabled. We will approve registration requests as quickly as possible (unless you're a spammer of course :) )

Katten Koffietafel Spoorweg

Started by webmartians, December 13, 2018, 07:29:55 pm

Previous topic - Next topic


...finishing what I started 30 years ago ... maybe.

Attached is the .ANY file for my Cats', Coffee Table Railroad.

  • Coffee Table -  Sure, it IS a Z gauge, after all

  • Cats' - More like, somehow, prevent the cats from playing "Godzilla Wrecks Tokyo" (Plexiglas is your friend!)

The layout is inspired by the David Mitchell KidmorZ Mountain Railway (https://kidmorengauge.weebly.com/kydmorz-mountain---z-gauge.html), but with so many modifications that any lineage is nebulous, at best (How did he maintain clearances and 3% slopes?). The name is Nederlands (Dutch) because ... well ... why not?

The work space is 110cm (43") by 60cm (23") by 18cm (7"); inside of a glass-topped, coffee table. I'm going to see what happens with a 3D printing of the base with an AnyRail produced .STL file and a subsequent cap to effect tunnels (stay tuned for tales of glee and sorrow).

There are three sections:

  • Main Run - Three loops a la KidmorZ but with a lot of rearrangements to obtain the 3% maximum slope

  • Reversing Loop - "Hours of mindless fun" with the loops requires some novelty to reach the "fun"

  • Switching Yard - While the main and reversing sections will be automated, hands on folk can use a small switching yard to assemble and disassemble trains

Opinions, even  :P, are solicited.


Tom Springer

Interesting layout.

Do all of those S-curves cause any operational problems?  Not a Z-scaler, so I don't know if the problems with S-curves in other scales applies to Z-scale. And the 4.4% grade at the top of the layout on the 'green' straight track is meant to be there?
Tom Springer

(Unintentional Pyromaniac)


December 14, 2018, 02:07:19 am #2 Last Edit: December 14, 2018, 02:22:38 am by webmartians
THANK YOU! This is exactly what I hoped for (but feared). How I missed that... There's an additional slope disaster in the lower right, maybe connected to the one you spotted.

I'm reminded of the real world debacle  as reported by Der Spiegel (http://www.spiegel.de/international/germany/german-railway-admits-massive-stuttgart-station-was-a-mistake-a-886099.html) for the proposed changes to the Stuttgart Hauptbahnhof.

I've resloped (Is that even a word?) and am attaching the .ANY file.

Regarding the S-curves, if you mean the green or Reverse Loop section, it was a necessary evil to get the tracks to align. That is part of the automated area (red, yellow, green, and blue; not under direct control of a human) and speeds can be force-limited (not yet certain whether by a current mirror and lower voltage or via DC chopping).

The orange ("tan" in AnyRail) section is the Switching Yard. While "curated" by the electronics, it is the closest to fully human controlled. It has a general slope from left down to right so that uncoupled cars will drift into the appropriate spur and (hopefully) come to rest in the middle of the spur, neither at the bumper end nor the switch. Yes, I can just imagine what happens if the layout is not properly aligned. Maybe I should include a little, bullseye, bubble level and adjustable feet.

Again, thank you.

(Arrrgh! ...found another glitch: replaced the attached .ANY file)

Tom Springer

So the switching yard is styled similar to a US "hump yard" with the entering track raised slightly and a small downward slope for cars to roll along. Putting a single-car-length track at the end that is sloped back up a little will help somewhat with the problem of cars running back to the bumper area; that's what I saw on the one hump yard I studied years ago.

Are you using roadbeds underneath your track?  If yes, what type and width? Would you need a sub-roadbed for support?  Would these impact any of the overlaying/underlying track areas?  Or possibly impact with those turnout motors (squares in row 1, columns 4-5)?   Think about how you would place roadbed/sub-roadbed for supporting track (if needed), width of any needed support structures for these, and where they might encroach on any tracks below - if you think of these support structures being very similar in form to wooden tunnel portals and then visualize putting them on your layout to hold up the overcrossing trackage, then you can ask yourself how do thinks look and work.

The closest I've come to anything Z-scale, other than looking at a few layouts, is having a piece of Z-scale track as a "mine" track to give visual perspective in a background setting.  I'm looking forward to the pictures of the layout when finished to get new ideas on someday perhaps making another visual perspective of a track line running on a mountain ridge as a background if I ever rebuild an area. Hoping I can learn new things from your efforts.
Tom Springer

(Unintentional Pyromaniac)


Regarding the roadbed, I have no idea. I was sort of figuring on just painting the base some kind of gray color.

I found a Union Pacific document (https://www.up.com/cs/groups/public/@uprr/@realestate/documents/up_pdf_nativedocs/pdf_up_reus_wireline_rdbed_sec.pdf) that I a-l-m-o-s-t understand that mentions 12" (~30cm) depth. At Z-gauge's 1:220, that translates to less than 1/20" (~1mm). Thus, just painting is probably enough.

The hump yard slopes are just guesstimates. I'd better do some experiments to see what kind of slopes really work. I'd hate to 3D print the layout just to find that cars are stuck in position.

I'm planning an IR trap near the end of each spur: a diagonal beam across the tracks (so that it won't "see" gaps between wagens). If the beam is broken and the track's voltage is such that the train is headed for the bumper, the voltage is cut. The idea is that a train can approach but never reach the bumper. Such sensors are pretty easy with an Arduino controller.

Supplying track power is NOT so easy with Arduinos (or PICs or ESP8266 or...). The opportunities for overcurrent are legion; it's really easy to fry a controller. I'm predicting just a few cubic cm for the controller and a big honkin' brick for isolation circuitry. The variety of voltages (3.3V, 8V...) means I'll have to watch for multiple ground issues and even "ground loops."

BtW- I hope to power everything from a removable, rechargeable battery that will reside somewhere inside of a detachable hillside.

Tom Springer

I asked about the roadbed, because all track needs something to rest on, whether wood, plastic or any other rigid material, so it doesn't sag and has the support to stay in place.  Generally another material is placed on that support surface to dampen sounds - cork roadbed or foam is usually the choice as either help with the noise a running train generates.

In particular, regarding supports, is the area with 3 levels of track in row 1, grid square 4; the yellow upper track will need support, and so will the light turquoise track above it; the 3 levels may make support structures "interesting". Add in that the tracks are above a turnout with a switch machine attached, and complications regarding placement of the support structures may arise. Hence why I asked about roadbed and supports.  The track arrangement there may be very hard to construct.

Questions are easy; answers often aren't.  I tend to put out the questions so folks know what to think about; I don't know what the answers should be, that's up to each builder.  If it works for you, then it works.
Tom Springer

(Unintentional Pyromaniac)


Thanks Tom.
I have this hare brained idea that I can use the AnyRail .STL file export to build a base (probably resin plastic) onto which I can "glom" (technical term :P) the tracks. (Any tunnels would be effected by building some kind of hat that would cover parts of the layout.)

Well ... anyway ... that's my fantasy. It's turning out to be a lot more complicated than that. You can see all sorts of spikes and pinnacles in the 3D image where AnyRail tried to provide a foundation. You call out a few of them. My answer is "Oops!" and I'm playing with the design to separate each level.


Also, while this may be a "Bridge too Far," I'm going to shrink the footprint to 100cm x 58cm. That latter dimension is the available depth of the coffee table.

Tom Springer

Not sure about an .STL export working, so I tried it just to see if you could get just a "roadbed" printed for the tracks.  (I do a lot of 3D designing and printing these days; "a lot" is probably an understatement, as I'm now using 3 different machines... no real time for layout work anymore...)

I exported your layout (track only) as an .STL file, then put it into Cura, the 3D slicer I generally use, and what I got was pretty bad.  The .STL file may have representation issues with dimensions/sizes.  My Cura configuration was for a Monoprice Select Mini that uses a 120mmx120mm bed size; the bed shows as a series of squares in a grid pattern, each square being 10mmx10mm.  When the .STL file was imported into Cura, it occupied less that 5percent of a grid square (as a very small structure), and was rotated so that it stood vertically on one side, not flat as I expected.

Once I rotated it to lay it flat on the bed. the size was so small, Cura would only see it as a single layer, meaning 0.2mm in height - the normal layer height I use for the Mini with a 0.4mm (standard size) nozzle.

So I tried importing the .STL file into other 3D programs, and I got pretty much the same result - something so small it initially couldn't be seen on the screen.

If you want to pursue this effort of exporting as a .STL file, I think David may have to look at the export parameters.  What I got couldn't be done on a 3D printer.

Even if you get an export the right size, 3D printing with the necessary supports is probably only going to get you a mess that isn't usable.

You'd probably be better off adding an AnyRail 'trace' with the appropriate width (not an AnyRail 'roadbed'), switch to just the 'centerline' for the track, no sleepers, and printing that 1:1 on a stiff cardboard sheet (having to segment the printing, probably), and cutting that result so you have the base for the track, then use the proper supports underneath the raised parts.  With the 3 levels of track you have, you probably would have to place each level (color) into it's own AnyRail layer, so you could isolate each and print each independently.

You may want to find other z-scalers and ask them for ideas.  I know ModelTrainForum.com and RailServe.com have z-scale forums.
Tom Springer

(Unintentional Pyromaniac)


Ugh! Oh well... heavy sigh. Looks like this project may have to be put on hold until some "engineering" is done. For archeological purposes, I'll post the latest .ANY file soon.

And THANK YOU Thomas for saving my tush!

Your idea of the big cardboard print out is my best hope right now ... either that or a series of layers. I'll see what those other forums say.


December 31, 2018, 09:08:34 pm #10 Last Edit: December 31, 2018, 09:10:45 pm by webmartians
I held off posting the latest because I expected to have to make modifications ... and, sure enough, there were many. The new design is 1m (39+") by 56cm (22+"), which barely fits into the Koffietafel (coffee table) and leaves a couple of centimeters for acrylic panels to keep the rolling stock safe from kattenpoten (cats' paws).

  • Clearances are at least 3½cm

  • Slopes are no more than 3%

Frankly, those values are a bit tight; I would have preferred 4cm clearances and no more than 2% slopes but the footprint just does not provide enough area for the required approaches and departures.

I have not specified any hidden track in the belief that I'll make some kind of cap to effect tunnels. In case of a mishap, I'll lift the cap and rerail whatever has gone off.

Assuming nobody discovers any difficulties, I am going to explore further the use of .STL files and subtractive 3D printing. I'll post my adventures, here, starting, probably, some time in middle February. Due to site limitations, I could not post my .STL file; I have attached images and the .ANY file.

Thanks to all who have commented. I look forward to your further observations; don't be strangers!

Tom Springer

Other than the S-curves in the blue and pink track sections, there's not much else I would worry about at this point, given your limitations on space.

The cork roadbed for Z-scale that I'm familiar with comes from Itty Bitty Lines and is 18mm wide; even if you don't plan to use that type of roadbed, you might set the trace option to 18mm and see what the tracks look like where they are close together and if there are any clearance issues between them.

If you need to put a .stl file in the forum, just add the .txt filetype id at the end (file.stl.txt) when posting it and that can be removed after downloading to view the file.
Tom Springer

(Unintentional Pyromaniac)


Thanks, Tom
I'll look into Itty Bitty. Maybe their curved bed has some provision for banking?

Also, I have a lunch planned with a friend and his colleague who is a "maker" for his workplace. I intend to pick his brain about printing layouts.


Maybe not do hump yard and keep it flat? Might make for better operation and storage? Just a thought.


OK ... I'm back with a newer layout (KatPuntSpoorWeg) and some adventures in pre-construction.

SheWhoMustBeObeyed laid down some rules:

  • Layout Must Fit in Our Family Room Coffee Table - 140cm x 55cm x 12cm available volume

  • Layout Must NOT Cannibalize the Noch Bärenweiler Suitcase - bought decades ago and have just stored (in a plastic bag) for all of this time

  • Layout Must Use Only Components I Have Already Acquired - purchasing exceptions for "inexpensive" curved and straight track - but no new switches, bumpers, decouplers, and so on (sigh)

Attached, for your amusement, is an image (with perspective rendering and parts list) and the .ANY file. You'll note that while there are bridges, there are no tunnels; some covering caps, not described here, will be used to effect tunnels. Because the caps can be removed, catastrophes can be corrected without major surgery.

There are two ways to delight in a model railroad: passive observation and hands-on operation. This layout delights (I hope) the passive observer by performing a form of shell game: the train moves in a difficult to predict pattern. Additionally, by bridges over tracks and scenery, it delights by highlighting vertical layers. Because of Koffietafel restrictions, I was not able to exceed three layers (two track layers and a base, scenery layer); I would have been thrilled to have one spot with four layers.

Computer (an Arduino) control is a requirement in order to perform for passive observers. An interesting layout is shown in this short video of "Lynn's Railroad:" https://www.youtube.com/watch?v=J04qXfDdkfg. I'm impressed by the controller's motion detector that starts the train when it detects observer(s). The woodworking skills are impressive, too. Like my Bärenweiler, Lynn's was purchased decades ago and has lain moribund for years. Unlike Lynn's, KatPunt does not yet work. :P

To satisfy hands-on users, KatPunt incorporates a minimal switching yard. It is computer "supervised" and performs an automated change-over once a new train of wagons is assembled. During this time, a Märklin, cleaning, rail bus traverses (only occasionally) the track in order to reduce corrosion. You can see in the image the positions of seven occupancy detectors that use infrared light to detect trains (not just locomotives but wagons too). That Z-scale wagons are difficult to emplace, using the switching yard assists in replacement trains (also, the length of the assembly spur ensures that train lengths are manageable).

Now as to the adventures ... advice solicited ... I'll put that into a subsequent posting so as to not run afoul of session timing restrictions.