Shelf Marshes (first attempt at a cameo layout)

[sb67]

I did a small test  sticking some track down using that method. I tried on a bi of c&L finescale track and some peco code 100.

I really liked the results but I think you do need to cut the sleeper webbing away and it was easy on a small bit of track but wonder how it would be trying to lay a length of track on a layout. 

It definitely looks better on thinner sleepered track. 

I'm going to try the method of goods yard rack next. 

 

 

 

 

[47137]

I had a second go with the "Ballast Magic" using less water. More of the ballast has stayed stuck but the result is not at all robust. To my mind it is good way to tack the loose ballast in place, but it needs flooding with dilute PVA the next day to hold it solid.

 

- Richard.

[ISW]

6 hours ago, sb67 said:

I did a small test  sticking some track down using that method. I tried on a bi of c&L finescale track and some peco code 100.

*snip*

 

 

Steve,

 

Thanks for showing the results of that method on Code-100 track. It may not have been a success, but it has convinced me not to use the method on my Code-100.

 

I think I'll be sticking with the good ol' water/pva mix on the ballast.

 

Ian

[sb67]

I tried the Barry Norman method of creating goods yard trackwork. I think it suits the finescale track better but with a bit of work I think it's ok for an ash/grime covered yard, again I wonder how difficult it would be on a length of track. 

I didn't have the same colour paint as was used on the DVD clip.

 

 

 

 

[47137]

I have set my heart on including a length of Magnorail in the layout, this will let cars from the Classic Vehicle Restorers go out on little test drives. I am building a prototype away from the layout. To this end, I have bought myself a palm router - £35 from Aldi. To get the Magnorail roadway down to the height of the railway at one end, where a connecting road goes onto a level crossing.

 

Now - I laid all of the track flat on the baseboard (no underlay) because most of the ground surface will be up to the sleepers, but there is a "main line" or rather a "through route" through the layout and this could be ballasted to a higher standard. So I could have a go with the router to make grooves for the cess along the sides of these tracks.

 

I ought to be doing the ballasting now but the Magnorail has proved an interesting diversion. I ought to get it installed before I do any scenic work.

 

- Richard. 

[ISW]

20 minutes ago, 47137 said:

To this end, I have bought myself a palm router - £35 from Aldi.

Richard,

 

I'll be interested to hear / see how you get on with the router. It's something I've considered for a while but not really found the need for one yet. My only 'excuse' would be to use it to create inspection pits inside my (future) MPD building, but that's only 2 cuts ~100mm long. 

 

Ian

[sb67]

That Magnorail looks interesting Richard, I've seen some good youtube clips of it. Look forward to seeing how you get on.

[47137]

On 26/10/2020 at 15:16, sb67 said:

That Magnorail looks interesting Richard, I've seen some good youtube clips of it. Look forward to seeing how you get on.

 

I have finished my prototype. The car is a plastic one by Brekina (for 1:87 scale) and is very lightweight, but I am sure the system would have enough power to move a heavier one such as the 1:76 models by Oxford Diecast. The maximum number of such cars would depend on the length and curvature of the chain.

 

This is the innards. The return loop at the right (with the pulley) is modified to make for a more compact turn, and the curve on the left is pretty tight - a 38 mm radius, quite a lot less than the minimum of 50 mm suggested in the instructions. The motor assembly simply drives the chain along by gripping the chain between two O-rings on gears:

 

This has run reliably for several hours, so the plan now is to dismantle it, retrieve the paper template underneath it, and glue the template onto the railway layout.

 

- Richard.

[ISW]

2 hours ago, 47137 said:

This has run reliably for several hours, so the plan now is to dismantle it, retrieve the paper template underneath it, and glue the template onto the railway layout.

Richard,

 

Interesting looking system. I'd seen it before on the Everard Junction YouTube channel. Do you need to lubricate the 'track', or does that cause problems with the O-ring drives?

 

From what I understand, you press magnets into the 'track' at your desired intervals to push/pull the cars along. Would it be a good idea to make an 'access panel' in the road surface such that you could access the magnets and shuffle them about occasionally for a bit of variety in the 'traffic'?

 

Ian

[sb67]

Good stuff Richard, will you have to be careful with the thickness of the roadway when you put it on your layout? 

Edited October 30, 2020 by sb67

[47137]

20 hours ago, ISW said:

Interesting looking system. I'd seen it before on the Everard Junction YouTube channel. Do you need to lubricate the 'track', or does that cause problems with the O-ring drives?

 

From what I understand, you press magnets into the 'track' at your desired intervals to push/pull the cars along. Would it be a good idea to make an 'access panel' in the road surface such that you could access the magnets and shuffle them about occasionally for a bit of variety in the 'traffic'?

 

Lubrication is an absolute no no. This would probably stop the chain running - no grip at the O-rings. The chain links are moulded from a material like a solid and shiny nylon, and the channel is another shiny plastic, like a polythene. The mouldings are shaped to minimise friction. There is very little friction in the system, except of course at the contact point with the O-rings.

 

An access point is possible. The magnets are a press-fit into the chain links, so to remove a magnet you would want to remove its link and put in a fresh one. To do this, you need to turn the link to be at 45 degrees to both of its neighbours. The simplest way to do this would be to make a lift-off section above a 180-degree return loop. The pulley would lift out, leaving you enough space to manipulate the chain. The instructions suggest removing the motor assembly to reach the chain - I tried this once, never again! Gravity would be a major problem here on a permanent layout, and the clips holding the motor mount don't like repeated undoing.

 

- Richard.

[47137]

20 hours ago, sb67 said:

Good stuff Richard, will you have to be careful with the thickness of the roadway when you put it on your layout? 

 

The recommendation is a roadway about 0.5 mm thick. I have some oiled stencil board (0.4 mm) and some Faller self-adhesive road (another 0.2 mm or so) and I think these will work fine.

 

I could do some experimentation. I've seen a video with the system driving Lego with normal Lego roadway sections. I suppose, the main thing is to support the roadway well so it cannot warp.

 

- Richard.

[47137]

I have had a lot of advice and help for this project from Clyde Humphries, the Magnorail distributor in Oz. Clyde has taken some video clips and photos from me, and created this amazing video:

 

https://youtu.be/IaN384o5cv0

 

I could offer a prize to anyone who can identify the location in the opening clip, but the lat and long figures in the footage are too much of a giveaway!.

 

- Richard.

Edited October 31, 2020 by 47137
Sorry for the link but I cannot get the video to embed.

[sb67]

Great stuff Richard.  The mini looks really good and drives pretty smooth. Look forward to seeing it built in your layout.

 

[47137]

I have transferred my prototype Magnorail system onto the layout:

 

This has taken me most of a week, due almost entirely to my desire to put the roads at the mini RA level with the tops of the rails and not a scale foot above them. The "woodwork" to do this took me absolutely ages and began with the router I bought in Aldi's going into the dustbin. This was the most awkward and dangerous power tool I have ever owned, this made worse by it having a gaping hole in its guard in exactly the place where I want to rest the fingers of my dominant hand!

 

So, I made my recess in the baseboard using the familiar tools: a hole saw, a jig saw, and finishing off with a chisel and the craft knife. Underneath there is a panel of 4 mm ply, and I built up the new levels with some 3 mm ply. It's not pretty, but the levels are actually pretty good:

 

The upward gradient continues through the length of the motor mechanism, and the Magnorail channels level off onto a panel of 3 mm ply for the last part and the loop into the workshop building:

 

So, I have a gradient with a rise of 6 mm over a distance of 290 mm. This is just about steep enough to be noticeable and add a bit of visual interest to the model, and it shouldn't look out of place in the setting.

 

The system all still seems to work. I am so glad I built a prototype first. It was a lot harder to install the Magnorail at the back of the layout (quite awkward really putting the chain in) and the prototype gave me a lot of confidence.

- Richard.

Edited November 8, 2020 by 47137

[47137]

The Magnorail motor is bulky and noisy, and its mounting lets it swing from side to side when I move the layout and turn it onto its back. The instructions do not mention any kind of additional support for the motor, but this is what I have done.

 

The default support for the motor is the plastic drive mechanism. This is so flimsy, most of the motor noise must be coming directly from the motor, possibly reflected from the underneath of the baseboard. The drive mechanism doesn't have enough mass or rigidity to transmit much vibration into the baseboard structure.

 

So I have wrapped the motor in a soft material (some neoprene foam left over from the servo installations) and clamped the assembly onto the side of a baseboard support with a strip of Meccano:

 

My baseboard frame here is 50 mm deep (two lengths of 25 mm strip wood glued together) and the motor terminals extend a few millimetres below this.

 

So I made a "guard plate" to protect the motor and its wiring from knocks. There is a strip of 6 mm plywood fixed onto the baseboard frame (also in the photo above) and the guard plate is more Meccano:

 

The final installation of the motor looks like this:

 

The rest of the noise from the Magnorail system is coming from the chain running in its channel and this will lessen when I eventually add scenery to fill the gap below the road surface.

 

The Meccano came in a box of mixed parts from the Farleigh Hospice shop in Boreham, near Chelmsford. The nuts and bolts always seem to come undone for me so the three cap head screws and nuts here got a dose of Loctite.

 

- Richard.

Edited November 8, 2020 by 47137

[47137]

Thinking about the last few weeks, Magnorail is really a hobby in its own right. I may well not do any more "railway" modelling this year!

 

I see this Magnorail system as a mini-project in four stages:

1. The prototype - built, proven and dismantled

2. The Magnorail installation on the baseboard - underway

3. The control system

4. The scenics

 

The control system could be a drill speed controller (for the motor) or something much more sophisticated and based around an Arduino. I don't want to promise much because the Arduino would be my first microcontroller project. It might be too difficult. It would be sensible to install some reed switches and their wires beside the Magnorail channel now, and bury them in scenery for future use.

 

I have some Magnorail parts left over and maybe I could have a mini digger moving back and forth near the chemical plant. This would be something else to maintain viewer interest between trains; and of course another bit of modelling :-)

 

- Richard.

[47137]

At the start of last week I realised the servo I had set aside for the Magnorail project was unusable because I had cut off the two mounting lugs for another project. So I ordered up some more servos and diverted myself to put a coat of paint on the backscene:

 

This has rather transformed the look of the layout and it should make photography a lot easier - bare plywood takes the life out of everything.

 

I painted both sides of the backscene to try to avoid warping, only time will tell whether I have succeeded. For the record, the primer is "Leyland Trade", this is a sort of high-build primer for wood. It has the consistency of treacle (maybe I should have thinned it a bit?) and it is the most evil-smelling paint I've ever used, so it must be good! On top of this there is a coat of Dulux Trade undercoat, and the top coat is Rust-Oleum satin white. I chose the Rust-Oleum because it is a long-established brand and maybe it won't yellow too soon. It has a very fine pigment. I used a foam roller for the undercoat and the satin and really I couldn't ask for a better finish. I know I ought to have a scenic background but I really do like the plain white and I'd like to keep this for a while.

 

I painted the framing around the backscene in Rust-Oleum dark grey and this seems to define the edges of the model and make a frame for it.

 

The servos arrived on Friday.

 

- Richard.

[47137]

This probably looks like a staged photograph but it isn't:

 

The Magnorail installation now has three reed switches, some wiring and an Arduino. The switches trip the Arduino to slow the car as it approaches the mini RA, speed up the car as it leaves the RA, and stop the car in its garage. The car starts its journey on a random timer, and also when tripped by a PIR sensor I can bury somewhere in the layout. The software for all of this is working but I seem to have blown up the motor driver hardware attached to the Arduino so at the moment the chain won't run.

 

I have been reminding myself why I never took up programming as a profession. I can do quite a lot of what I want, but from time to time I get really stuck. This is my first go at "C", but fortunately I did some Object Pascal years ago and the two have quite a lot in common.

 

I have a new motor shield on order.

 

- Richard.

[47137]

I have pulled the plug on my Arduino project as I conceived it to be. The trouble was, I found myself making a project where the hardware (the Magnorail) depended on the software to make it run; and the software could not run properly without the hardware. I could of course attach a spare motor to the Arduino and connect some push buttons to emulate the reed switches, but success depended too much on timing as the motor drove the chain around the system, accelerating and decelerating and stopping on the way. The sketch did run for a while but with the fluidity of one of those clockwork programmers they used to put on washing machines.

 

To be honest, I think I made a major error of judgement by dismantling the prototype and rebuilding it on the layout before I built the control system. I could not tip the layout on its back to alter connections to the Arduino, because the chain then dropped too far from the reed switches to let them trip. With the layout horizontal, I had to have the road surface removed (to see where the chain was) but every mistake sent the chain spooling out of the channel and coming apart into its individual links. It was all "too difficult", at least for someone who is new to both  Magnorail and Arduino.

 

To move forward, I have installed a manual controller for the Arduino, this is a PWM drill speed controller. It works fine from a 12V supply. With the Magnorail running I have stripped down my software so the Arduino can sense and respond to the reed switches, but it isn’t controlling the motor any more.

 

Here is a photo of my Arduino installation, this is underneath the site of the mini roundabout on the model. The three pairs of white and grey wires are for the reed switches, only one is in use as I rework the software and decide what it should best do:

 

The motor shield here is now unpowered; I am simply using it as a breakout board to connect wires to the Arduino underneath it.

 

- Richard.

Edited November 30, 2020 by 47137
Added the second paragraph.

[47137]

Yesterday evening I sent some fresh video clips and photos across to Clyde Humphries in Australia and he has produced a marvellous new video:

 

https://youtu.be/R2t9kOa41GY

 

This shows how my mini roundabout works far better than I could describe in words.

 

- Richard.

[ISW]

9 hours ago, 47137 said:

a marvellous new video:

Richard,

 

Now that's what I call innovative! Very clever and interesting.

 

Ian

[47137]

On 01/12/2020 at 17:24, ISW said:

Richard,

 

Now that's what I call innovative! Very clever and interesting.

 

Ian

I think the main thing is to not be too ambitious. I was thinking of a young mother with her pram waiting to cross the road and jumping back (using another servo) to get out of the way of the understeering car. But really, this part of the layout is getting pretty congested underneath the baseboard. The cable trunking is almost full of wires and it is time to stop adding electrical features. I want to refine the software (quite a lot!) but no more animations near here.

 

- Richard.

Edited December 3, 2020 by 47137

[47137]

I have added my tenth servo to the layout. This one is for an animation: a man hidden behind a bulk storage tank and coming into view to take a pee:

 

I have used this report to try to make the timings prototypical:

https://www.livescience.com/46625-most-mammals-take-21-seconds-to-pee.html

 

I'm assuming a half bladder to keep viewer interest. The servo takes about four seconds to make its sweep each way.

 

The other nine servos are eight for the points and one for the Magnorail. I want to add more animations: a mini digger, a barrier, maybe some gates. I want to use these as point repeaters when the layout is at home, and to make an animated display when the layout is at a show. There will also be a semaphore signal, for the exit to the fiddle yard.

 

With growing complexity I have added a second Arduino to the layout, this is a "scenics processor" and it leaves the first Arduino to support the Magnorail. As a newbie to servos, Magnorail, and now Arduino, I am thinking of new ideas faster than I can work out how to do them.

 

I haven't forgotten about "roads and fences". These set up the overall balance and define how the whole layout will look. Somehow, I want to sort out my mini digger and my barrier; and rewire the Magnorail system; before I move on to the physical side of the scenics.

 

I am incredibly pleased with Clyde's second Magnorail video (link above about a week ago). I like his production, and the use of video fills a gap where forums like RMweb can contain masses of narrative and ideas but it is really hard to show things actually working.

 

- Richard.

[47137]

The Arduino has had a pretty profound effect on me. My desktop computer (Windows 10) is in my office, some distance from the layout. This is ok for starting to write sketches and build prototypes, but hopeless for testing and fine-tuning on the layout. The latest updates of Windows 10 have ruined my Panasonic laptop, so I borrowed a newer laptop from my partner. She hardly ever uses this but it was still a pain to me - Windows kept on losing track of the COM port for uploads to the Arduinos. I spent more time in Device Manager than in my code.

 

So, in desperation I installed Ubuntu on my own laptop, alongside Windows. Ubuntu is quite simply *fabulous* and the laptop is restored to something probably better than new. My only regret is doing the installation as a dual boot - I might as well remove Windows altogether.

 

This success has prompted me to put Ubuntu onto my desktop PC as well, and I am trying out "Shotwell" here as a photo manager to replace iPhoto on my Macbook (2008 vintage), this blew its battery pack apart a week ago.

 

All this has taken most of a week but I do now have a working barrier on the layout, and here are the photos to go with it - my usual Fuji camera but processed by a PC running Ubuntu/Shotwell instead of OS X 10.5/iPhoto. This is my second animation after the "man having a pee":

 

I don't know who made this barrier? I bought two at a show, the first one went onto "Fairport" layout.

 

I am using hot glue to hold the servo into place. This seems to be ok for a lightweight load like this:

 

The linkage here is a thin piano wire - 0.4 mm diameter. This has a fair amount of "give" in it and maybe the barrier will survive if the servo ever runs amok. The trick here was to set up the software to hold the barrier at its mid-point, and then adjust the end stops in the software to suit the barrier.

 

The barrier has its own Arduino, so it can run when required and I don't have to learn how to create multi-tasking code:

 

 

 

There are more notes on the Arduino here.

 

My task now is to work my way back through the diversions and distractions over the last few weeks. I think it will be best to try to make some more animations and connect up an LED display to an Arduino; and use these to learn more about coding. Then use the experience gained from these to have another shot at a "self driving car" on my Magnorail.

 

Ubuntu 20.04.1, full install with all the trimmings.

 

- Richard.

Edited December 18, 2020 by 47137