martink

That size & weight of van seems reasonable. IDL's website is www.teenytrains.com.

With the turntable, that is exactly what I meant by a "cradle" - something that physically grabs and turns the car without doing anything with the track. My fault for not explaining it more clearly. I build points as two pieces of track, one on top of the other, then switching power between them. As always, there are a number of complications. The main one is that the lower piece is further away from the car's magnets so the drive force is roughly halved, a possible showstopper for a heavy HO model, which I solve by doubling the current (which quadruples the generated heat). Other ones are getting the track surface back to the same level with packing pieces or another overlapped pair, ensuring that the coils line up perfectly where they do overlap so that the model doesn't jerk when you switch tracks, keeping the joins at the end of each track piece from getting in the way, and limiting the track heating problem. Unfortunately, I am not able to sell you any of my track, so you would need to adapt IDL's. IDL's track pieces are designed so that each turns on as a vehicle approaches and off after it departs, and are hard-wired for one-way running. You would need to tap into and override the on/off control wires on each piece and drive them from an Arduino or similar. You may as well use the same Arduino to completely replace IDL's controller and drive the track directly. One nice thing about the linear motor approach is that everything runs in lockstep, so the vehicles move a fixed distance for every drive pulse from the controller. This makes automation remarkably easy. The end result would be a program built around a movement script where each step turns on two specific sections (the one the car is in and the one it is going to), then generates a certain number of drive pulses either forwards or backwards until the car is fully in the next section, then optionally delays for a few seconds or immediately continues with the next movement. Solid mechanical, electrical and programming skills are all needed. Bluntly, it is certainly doable, but it would be a big and challenging job, especially for a first-timer. Back at the beginning I did look at adapting IDL's track for my own projects, but the associated difficulties and limitations meant that starting from scratch and completely reinventing the wheel was actually the better option.

I do exactly that with the turntable on Penzance (there is a video in the T Gauge section of the forum), but it has its challenges. The track has overlapping coils on the top and bottom, so joints and the turntable need special arrangements to provide an extra coil to span the gap. There are also more polarity reversal issues than a conventional system. I had to make some custom track pieces to do the job; trying to adapt a standard piece would be tricky.

I was deliberately pushing to find the upper and lower size limits with this round of experiments. My version of the track was optimized for T scale, so N scale plastic car-sized vehicles really are as big as it can go. For HO, IDL's track could probably do it - it is inherently about twice as powerful and you can also use larger magnets which boost it further. However, you would have to rip out some of the electronics and build a new control system from scratch - not a simple task. I know somebody who has done a simple version of that using IDL's track with a Z scale front-end loader, doing a 3-point turn back and forth between a coal stack and an unloading point. The turntable would be straightforward, but would have to use a brute-force approach (drive into a cradle which then forcibly turns the car around without moving the track). Overall, doable but non-trivial.

I find I need to plan one layout ahead, so with Penzance coming along nicely I have been trying out some preliminary ideas for next year's project. This includes some experiments with new prototypes and/or scales. These are all just rough test builds, but hopefully might be of some interest here. https://youtu.be/uBcmRts7tYk

It works! Trains are now running, although only in manual mode or with limited record/playback automation. This is just a rough work-in-progress video, but it should give an idea of what it is all intended to look like when finished.

It is proving difficult to get motivated in this era of few or no exhibitions, but things are slowly progressing with Penzance. All the electronics have now been built and tested, and their installation and layout-side wiring is under way. The road loop is working, and the next task is making and fitting 100+ cables linking the track to the circuit boards for the rail side. While this layout is primarily intended for fully automatic exhibition running, the more powerful computer I am using for it gives me additional possibilities. There will be a simple control panel that will allow full manual operation, including shunting, as a test-bed for using this technology on a future home layout. Also very useful for testing.

The new electronics have been designed, and some basic tests done, but I am now waiting for the blank circuit boards to arrive so I can move further towards getting things running. In the meantime, I have started work on the structures, beginning with the main station building. Some deliberate distortions here, mainly widening the building and overall roof to cope with the over-wide track spacing.

Another spurt of activity has moved things along a few steps... The basic timberwork is complete, with only the backscene still to be added. All the track is done, including the road along Chyandour Cliff behind the station, but final installation won't occur until after the new electronics have been built so that it can have some all-up testing. The first parts of the scenic structures have been printed and assembled: the east end of the sea wall and the beachfront viaduct. The next jobs are the backscene timberwork followed by a complete revamp of the electronics, which will allow for significantly more complex layouts.

The trackwork for the station area is now complete, all soldered together into a single piece. These are the live tracks, and additional dummy tracks for the rest of the loco depot and the headshunt / carriage siding next to the main line will be added as part of the scenery. The next tasks are cutting out the plywood trackbed to support it and attaching the electrical connectors to the underside of the track.

After a break over the new year, I am back at it again. The reversing loop is up and running, and so is the first attempt at a turntable. https://youtu.be/PXay57VdjaI

It may very well just be due to dirt. The gears and wheel tyres are obvious targets to clean, but the most insidious problem tends to be the axle-boxes. The brass bogie sides have small dimples that fit into hollows on the outside face of the wheels, and gunk tends to collect in there, usually as a mat of lint and fibre that acts as a very good insulator. Tease each wheel slightly out of its correct position and probe into the wheel-face hollow with a pin or fine tweezers. If this is the problem, it will be very obvious. Be very careful handling the bogies - DO NOT pull them away from the body, or you can stretch and damage the bogie springs. Keep pressure on the bogie pushing it up into the body by holding the model between thumb and finger on the body top and bogie.

I actually spent a couple of days trying out various ideas for working semaphores, but could not achieve anything repeatable. On the other hand, I've done colour lights successfully in T Gauge, so the next time I do a modern image layout..., or maybe Paddington with pre-war GWR searchlights....

With the layout basically complete, here is the last planned video of it. I let the layout do its own thing, while moving around to different positions with the camera.

John - I have thought of doing a modular system, but even in 1:480, railways are just too big and I still have to compress the track plans. I did some preliminary experiments a while back for a 1:10000 scale model using a completely different approach (and that number isn't a typo), looking at Plymouth-to-Saltash as a baseline, but even there I was running out of space. Penzance has been at the back of my mind as a project even when I was still modelling in N, so it was a prime candidate here. Assuming this one works out, I will probably look further afield for the next one, possibly LMS or MR. Too far in the future...

After considering several possibilities for the next layout, I have settled on pre-1914 Penzance. At this time the station had only 2 platforms, but still had its small loco depot with turntable and the single-track timber viaduct running along the beach. The size will be 6' x 2' and the scale 1:480, the same as Dauntsey Lock. It will be set up for automated exhibition running, both passenger and goods. Chyandour Cliff will have working road traffic, mostly horse-drawn at this period. The track plan shows both the real and dummy tracks, almost complete except for one crossover and two short stubs off the turntable. The main technical challenges with this design are the reversing loop (already up and running, and just as tricky with a linear motor as with normal DC), the turntable, and coupling/spacing issues with reversing trains in the station.

The scenic detailing is mostly done, just waiting on a few extra bits and pieces to arrive (people, better signals, etc.). I'll do a final video or two then.

About 2/3rds of the planned goods wagons have now been built. This is enough to run the layout, but with shorter trains than it is designed to handle. I'll do the rest at the very end of the project. They are built the same way as the buildings and passenger stock: simple 3D printed body shells, roof and underframes hand-painted, with paper sides for the detailing. I tried decals, but this approach proved more effective. One of the big advantages of the linear motor drive is that it can run very long trains. One of the big disadvantages is that I then have to build them! Next step: the software, bringing it all to life.

After a fair bit of work, all the locomotives and passenger stock have been built, including milk tankers, siphons and TPOs. These are not the final train formations, but do represent all the major types the layout will handle. All the models were 3D printed on a common FDM printer, then hand-painted and with paper sides on the coaches. The milk tankers use home-made decals for the lettering and ladders. For the period, the locos should all have variations of the "GREAT WESTERN" legend on their tenders and tanks, but I was unable to produce an acceptable version so had to go with a golden blob to represent the shirt-button logo instead. As usual, the close-up shots in the video are cruel and highlight all the defects.

I have had a couple of people ask me how the canal lock works, so here are a couple of CAD images. Both the scenic bits and the working parts are 3D printed, and the whole thing is driven by 3 servos. The structure and supports were designed for 5 servos to handle the more common double gates, but that was just getting too crowded. The assembled view gives the green elevator piece stretched legs to show how it all fits together. Progress is continuing on the layout itself, with the trains currently under construction. The locos (19 of 10 classes) and passenger stock (63 of 19 types including milk tankers) are nearly finished, and I am plucking up courage to start the goods wagons (180-200 in total). These figures should be about 20% more than I actually need, but allows some choice when setting up trains and sequences. Why oh why didn't I pick a one-train-a-day branch line? With this one nearing completion, I have pretty much settled on the basic concept for layout #4 - Penzance circa 1913 on a 6' x 2' board. This will feature basic terminal operation including the small loco depot and turntable that closed the following year, running to an out-and-back reversing loop. I can even fit in most of the long-vanished Penzance viaduct. And, most importantly, reuse a lot of the new GWR stock!

Seriously, the limitations of the lightweight trains, poor electrical pickup, less-than-satisfactory points, fixed couplings, etc., really do preclude anything other than the back-and-forth and continuous-run options. Even getting reliable running in those cases takes a lot of work, for example using multiple locos per train and permanently wiring them together. Plus lots and lots of maintenance. Some people are now using DCC, which improves things a bit, but does not change the fundamentals. That is why I have chosen to head off on a complete tangent with my own linear motor stuff, because I really DO want to have a fully functional T Gauge model railway. Unfortunately, that isn't a generally available option.

It depends on what you mean by end-to-end. A simple automatic back-and-forth shuttle using multiple unit stock is very easy to do, since the standard controller comes with a pair of reversing sensors. For a terminus that requires shunting, well, that is a completely different story...

The cars are temporary, and completely wrong for country & period. It all looked looked a little empty, so I put them there just for the video.

With the scenery mostly done, it is time to get things running again. First, the canal...

A bit more progress on the scenery - trees, hedges and a bit of a general cleanup. I'll save the fences and the rest until the final detailing phase, so for now it is back to the moving bits...