Helston Revisited

[Andy Keane]

On 29/01/2022 at 09:58, Andy Keane said:

I have now found a thread on the jmri users group that tells me there is a known issue with some pins on NCE AIUs such that they have a different sensitivity that means the LED lights up but the circuit does not trigger! A fix is in the offing but is not strictly needed on this board as yet.

It turns out the fix is to bypass the output resistors on the DTC8a and then they work just fine with the less sensitive pins on the NCE AIU - problem solved. I just need to source some more AIUs - these seem as rare as hens teeth.

[Andy Keane]

I am now wondering what colour to paint the carriage shed. GW Way suggests it should be mid stone and dark stone but also notes that timber sheds were often preserved with tar or creosote. The B&W photo in Jenkins suggests a light colour with a darker paint on the boards at the very bottom. So I am currently planning to poaint the majority of it in mid stone (Phoenix stone number 1), the base boarding in dark stone (Phoenix stone number 3), and the glazing and intetior in white. If anyone has any thoughts please let me know.

thanks

Andy

Edited January 30, 2022 by Andy Keane

[Andy Keane]

On 29/01/2022 at 09:56, Andy Keane said:

Phil, which 48xx would you advise? Also loving your coach - where is that from?

Andy

Should also have mentioned that YouChoos squeezed two fat tantalum capacitors into my 14xx which really help it run over points. Heaven only knows how they managed - I have not dared take the smoke box door off in case it all falls out.

[Andy Keane]

I have now completed most of the parts on the carriage shed bur rather than glaze and assemble I am going to take the parts as they are to the Gret Missenden weekend where I am down for the weathering course. Thought they might think weathering the bits before assembly would be wise. Not an option I have on the other three buildings done so far but I thought it would be nice to have the choice.

This shed had a corrugated iron roof and I have gone for the very rusty look as it would have been about sixty years old when I am modelling. Probably needs toning down and some moss growing on it as well!

Andy

 

 

Edited April 6, 2022 by Andy Keane

[Andy Keane]

I have also spent some time pottering with Arduino electronics and coding so I now have an Ardunio Mega plus a Raspberry Pi RS485 board doing the job of anNCE AIU courtesy of Alex Shepherd's coding of the interface ( https://mrrwa.org/nce-cab-bus-interface/ ).
The little RS485 Pi board is from the PiHut: https://thepihut.com/products/rs485-board.
The only subtlety is that as shipped the wiring of the socket on the RS485 board is the wrong way around (RS485 does not specify which way they should go for any given application).

The Mega has more capability than this application strictly needs but it does nicely separate the RS485 from the connections to the programming PC and I had one lying around. Until such time as NCE start shipping AIU boards again this works for me. I may never replace them with AIUs and think of some other stuff they could do on the layout.

Andy

 

 

Edited April 6, 2022 by Andy Keane

[Andy Keane]

Alex has now shipped me a code for the Mega that allows it to behave like a pair of AIU boards with two separate CabBus addresses and 28 sensor inputs, but using just a single RJ11 socket and cable to connect. This is really helpful to me as the far end of the fiddle yard has a very dense set of points, track and sensors. It’s cheaper than two AIUs and neater. So now I can have JMRI draw me a really pretty and active track layout from which I can see occupancy and also change points or set whole routes. I have bent the fiddle yard around to lie above the station in my JMRI plan to better fit a screen than an L shape would, see below. I had planned a board with physical switches and LEDs on to do this for the yard but I think a simple tablet will be much neater and also easire to change. This will contrast with the signal leaver approach I plan to have for the station itself, though the JMRI board will still display all that goes on and provide a backup interface. I can see myself getting sucked deeper into coding up auto running if I am not careful.

Andy

 

Edited April 6, 2022 by Andy Keane

[Andy Keane]

I have now finished the first half of the fiddle yard with all the dropers in place, 12 track occupancy current sensors and four laser /photo diode sensors mounted in 3D printed mounts.

The sensors are all MERG kits feeding my Arduino Uno which feeds on via the CabBus to the NCE kit and ultimately to the live track diagram shown above courtesy of JMRI.

A lot of wiring underneath as you would expect.

Andy

 

Edited April 6, 2022 by Andy Keane

[St Enodoc]

7 hours ago, Andy Keane said:

I have now finished the first half of the fiddle yard with all the dropers in place, 12 track occupancy current sensors and four laser /photo diode sensors mounted in 3D printed mounts.

The sensors are all MERG kits feeding my Arduino Uno which feeds on via the CabBus to the NCE kit and ultimately to the live track diagram shown above courtesy of JMRI.

A lot of wiring underneath as you would expect.

Andy

Just a suggestion before you get much further - place the sensors at an angle across the track so that they don't give a false indication through the gap between two vehicles.

[Andy Keane]

10 hours ago, St Enodoc said:

Just a suggestion before you get much further - place the sensors at an angle across the track so that they don't give a false indication through the gap between two vehicles.

Oh yes - good point - had not thought that through. Hopefully as I am just trying to use them to get an input when shunting stock back towards the points I may get away with it - the idea is the laser warns when I need to stop. These sections of track also have MERG current occupancy detectors to pick up locos, gaurds vans and brake vans so that will help. But if needs be I can move them - currently they are just a push fit. I am pondering the idea of computer controlled run around when a loco and stock enter the fiddle yard but this would also entail accurate decoupling - maybe a bridge too far.

[Andy Keane]

I now have the first half of the fiddle yard all wired up and working - there are 16 track occupancy sensors and four laser sensor plus eight point motors under this.

It allows me to move a few engines around while I press on with the second half. I am still amazed by the sound chip installed by YouChoos - I am fully hooked on sound.

The only problem I have is my hattons 14xx really does not like running over the points backwards - the pair of trailing wheels don't seem to be held firmly enough when going in reverse and they jump about too much - some further investigation required there - does anyone else suffer this on one of these locos?

Andy

 

 

Edited February 23, 2022 by Andy Keane

[Graham T]

Perhaps check the back to back measurement?

[Andy Keane]

2 minutes ago, Graham T said:

Perhaps check the back to back measurement?

I think the issue is the little bogie the trailing wheels run in is just too light and being pivoted towards the front of the loco means the bogie is being pushed from behind when running backwards which is bound to be less stable. I think my first move will be to see if I can epoxy some lead shot onto it.

[5BarVT]

Another suggestion (not to do with the 14xx).  Put a temporary end piece on the fiddle yard just in case something gets a mind of it’s own and heads off the edge.

Paul.

[Andy Keane]

18 minutes ago, 5BarVT said:

Another suggestion (not to do with the 14xx).  Put a temporary end piece on the fiddle yard just in case something gets a mind of it’s own and heads off the edge.

Paul.

Yes, it is a danger. I am going to make some end boards that will do that and also allow me to stack the base boards for storage. 

[Andy Keane]

More wiring - the second fiddle track board is nearing completion. This time a bit neater as I do more.

 

Edited April 6, 2022 by Andy Keane

[Andy Keane]

One gotcha this wiring revealed was rather obscure, at least to me.

As there are no NCE data boards currently for sale to convert track occupancy signals onto their CabBus I have been building my own using Arduinos. To allow daisy chaining the boards I added pairs of six way RJ12 connectors to these (you can see these in the photo). All great when tested baseboard by baseboard. But when I tested the boards coupled together as intended all fell apart with garbage showing on my hand throttles. Much research later revealed my ignorance. It seems that CabBus uses the RS485 protocol that requires terminating resistors at the far ends of signal lines to deal with termination impedance. But such resistors must only sit at the ends. The wave share connectors I am using have them fitted by default. So you have to cut them off on intermediate boards, and of course this does not show when boards are tested in isolation. Resolved in the end but it makes me realise I am straying into strange territory.

Andy

[Andy Keane]

Have been on the Missenden weathering course this weekend. Made good progress on the carriage shed and also some work on a few wagons as well as learning how to drive an airbrush.

This is the shed assembled, glazed and weathered:

This with the roof just resting in place - these are pre-coloured Redutex sheets mounted on card and then toned down with an airbrush:

I think this is better than the out of the packet finish and non-weathered shed sides:

A few Bachmann wagons made a bit grubby:

In all a very productive weekend plus loads of very friendly people.

Andy

Edited April 6, 2022 by Andy Keane

[Andy Keane]

Roof cemented into place. Only a few minor touchups now needed. Next comes the main station building.

 

 

 

 

Edited April 6, 2022 by Andy Keane

[Andy Keane]

I have finally finished wiring the second fiddle yard board and put the two together with the corner board.

All works fine electrically and it also all talks to JMRI - 33 track blocks with MERG current sensors, six MERG laser sensors, two Arduinos and 18 DCC point motors - I now need a significant rest from soldering. Also if you are of a nervous disposition my standards of wiring may upset - it works but it aint pretty.

However mechanically I have a problem - I don't have enough clearance around the MERG laser sensors with my standard track spacing if I set them high enough to read rolling stock body sides.  There is just not enough room for wider tenders and near bends, bogie carriages run into them. If I lower them down all will be fine but I will then be detecting just the wheels - this is a puzzle to be thought about. It may be for my automation purposes detecting wheels will be enough though.

Ho hum.

Andy

(ps - three of the four locos on the boards here have GWR lettering so really need changing over for my 1930s period - something to worry about another day)

(pps - the latest YouChoos sound chip is, I think, significantly better than the standard setup in the Oxford Rail Dean goods)

Andy

 

Edited April 6, 2022 by Andy Keane

[Andy Keane]

I have been looking at the Hornby Collet 57' bow-ended non-corridor sets of coaches and wondered if a rake of four might ever have gotten to the west country on a summer special train. According to Hornby they got as far as Bournemouth but as to Helston who knows. Might that give me a legitimate excuse to buy a set? If anyone knows I would appreciate input.

Andy

[Andy Keane]

I have finished programming JMRI with routes for the fiddle yard so that either from the PC or the hand throttles a full route through the yard is a single operation. Turns out I need 14 such routes for the yard. I will make up a laminated sheet with all this on so I don't forget which is which!

I have also added the sides to the first of the fiddle yard boards. This is 3mm clear acrylic sheet. The two sides will stay on all the time and the end will be removable to allow the legs to be attached and also to aid loading stock into the yard. When I am done I will peel of the paper cover off the sheet so it will then be transparent.

 

Edited April 6, 2022 by Andy Keane

[Andy Keane]

Glued up the track bed for the embankment last night, ready to be mounted on the baseboard. I plan to trim to the final shape after the track is down and the carriage shed offerd up. It's six mm ply with laser cut joints and expoxied, so light but very stiff.

 

Edited April 6, 2022 by Andy Keane

[Andy Keane]

I see RMweb is back online. I was getting withdrawal symptoms. When I get the time I will potter back through this thread an reinstate my photos - luckily it seems the file names survived so I will be able to put the right images in the right posts.

Meantime I have been to Helston and gone all through the photos in the museum and also stood on the remains of the platform. Around fifty feet remain but it is a very forlorn sight.

Hopefully others can rebuild their threads as well, as I really value the photos as a source of ideas and inspiration.

Andy

[Andy Keane]

I have finally been through and put back all the images I can onto this thread. Some early ones I don't anylonger have as I reused the file names - ho hum. If any of those who have contributed their own images to this thread can put them back it would be much appreciated if you can put up with the tedium.

Andy

[Andy Keane]

Some pics from my Helston visit:

and Truthall halt