1:480 scale canal with lock

[martink]

A major scenic feature on my latest and almost-finished T Gauge layout is a working model canal, half a scale mile of the old Wilts and Berks canal in SW England, including a functional lock.  The canal is driven by a linear motor system, and the lock by three small servos.  The video shows the canal boats racing along at 6mph to shorten the video's run time, but they normally obey the 4mph speed limit for Britain's inland waterways.  It takes about 20 minutes for each boat to make a complete circuit.  At the time this video was taken, the trains were posed and the scenery still incomplete.

[DKS]

Most excellent. Upvoted (would that it was possible).

[aikorob]

way cool!!!

I do have a question-----the outbound gate opens into the lock--is that prototypical? It would seem to make more sense if it opened outward

[C855B]

I do have a question-----the outbound gate opens into the lock--is that prototypical? ...

Yes. Lock gates open into the natural flow so the pressure assists in sealing the gates closed.

[wcfn100]

Not only do they open 'in' they don't close flat, keeping a slight 'v'.

Jason

[Maletrain]

Not only do they open 'in' they don't close flat, keeping a slight 'v'.

Jason

This model uses a single gate closure, not a double gate closure.  I would expect that seals against a cutout in the lock side that is across from the swing hinge.   You can see those cutouts in the video.

As long as we are picking nits:
1. I don't see any seam in the water where the water in the lock that changes level meets the static level water on the upstream or downstream side.  That is nice, but I am wondering how it was accomplished.  Is it modeling or video image editing?
2. It looks like the bottom part of the upper lock gate is not modeled - just a dark recess.  I would think that the part of that gate (which is hidden when the lock water is up and exposed when the lock water is down) could be modeled as the downstream face of the upper water level, because we would only see it when the top part of the upper gate is directly above it.  Or, would that somehow interfere with the apparently seamless water in nit #1?

[tehachapifan]

Wow, that is awesome!

[martink]

That was my first attempt at canal modelling and, as usual, there were plenty of mistakes and lessons to be learned for next time.  I only know of one other similar working model with a boat passing through a lock, and that is in OO scale!

Most but not all canal locks use double gates for symmetry and balanced forces on the structure.  This model was originally designed that way, but it was simply getting too crowded underneath to squeeze in 5 servos so I simplified.  The lock chamber is only 36mm long, after all.  It should be a bit longer (ideally 48-60mm) but the elevator portion has to be a complete piece of track and my only available choices were 36mm or 96mm. I have since made some 48mm pieces, but too late to retrofit.

These are visible gaps at the edges of the moving section, but I made them as unobtrusive as possible and the gates almost totally hide them when closed.  The only video editing was at the end of the second lock cycle, shortening an unwanted pause of about 5 seconds between the elevator reaching its lowest position and the lower gate opening.  I should have done that to the first cycle too.  The real fix is to recalibrate the stop positions in the software, which I will do when I also get around to fixing the initial jerk when the water level starts to fall.  The dark recess at the bottom was there to let me maneuver the track pieces into position during assembly, since they have awkward joiner flanges on the sides.  The intent is to cover that hole as well during the final cleanup stage.

[ Guests cannot view attachments ]

[ Guests cannot view attachments ]

[Chris333]

I watched your other videos with the trains running. How is that done? I see the track is just an image, but the trains seen to run fairly well and at realistic speeds.

[martink]

I watched your other videos with the trains running. How is that done? I see the track is just an image, but the trains seen to run fairly well and at realistic speeds.

It is a linear motor system, my own variant of a commercial product from IDL Motors (teenytrains.com).  The track is made up of double sided PCBs with three strings of interleaved coils built into them.  As each string of coils is powered in turn, the magnets under the trains/cars/boats/whatever jump to align themselves over them.  With a bit of ingenuity, I was able to do better than the actual 2mm coil spacing, and get down to an effective 0.25mm step size which allows for reliable low speed running.  The pulse frequency is fully controllable, and everything moves in lockstep, so accurate speeds are simple.  This implementation is set up with 50 speed steps up to a maximum of 67mph in 1:480 (62.5mm/sec), but that is just an arbitrary software decision.  Since the track does not look anything like a real railway, it is covered with a layer of paper (self adhesive label paper) printed with suitable imagery to provide a better appearance, smoother run and long-term protection for the PCB.

[wm3798]

That's really cool.  Having kayaked in the canals in Wales, I'd say you've nailed the overall concept!
Lee

[Dave V]

Just realized we can't upvote in this forum.  Neat-o though!