Along the Chapeltown loop

[HillsideDepot]

I've been trying to use the mail merge function in MS Word to fill out all my wagon cards from an MS Excel spreadsheet - with limited success. So, while the field titles make this look odd, this is my idea for the cards, printed on A5 card. 

 

 

The pocket is formed by folding the bottom of the print up, while folding the rear around adds to the strength of the card. The neat way of fixing the pocket would be to use glue, but I'm thinking Sellotape would be quicker to assemble, and provide reinforcement across the pocket if one piece of tape fixed both sides, although shiny tape probably wouldn't look as nice as internally glued seams.

 

I've made a few prototype cards, but not progressed much further than that, other suggestions are certainly welcome!

[Ray H]

I like the cards.

 

Have you considered what I think is called Invisible tape - a 3M product I think, but I have seen others.

 

Didn't the OP's cards also have maintenance/repair work details on the back? Would it be worth adding some lines to the back to facilitate this?

[Lacathedrale]

Hi all,

 

Firstly, I'd like to say what a revelation this layout has produced in me - all my Christmasses have come at once and it has provided a very thoroughly enjoyable morning's reading. I was redirected to this fantastic top thread from discussions I'm having elsewhere on the board about a very similar question and I'm very happy to see that prototypical wagon-load freight in a British setting is not only possible, but both prototypical and interesting.  My research indicated that this kind of work (a Wagon of X to Company Y) had dried up by the 1960's, but clearly that doesn't appear to be the case. Is that a geographical thing, or simply a reflection on your own desires for era?

 

Can I please ask a few questions about your implemetation of RR&C TrainController?  I appreciate this discussion has mostly focused on the operational aspect of the layout, but that appears to pivot around a robust system for automating train movements. You obviously have a USB-DCC interface, and so I can gather how train and point operation work by addressing individual decoders, but I'm far less certain about other tactile elements. One of the problems I've had with layouts in the past where I've reached too far in terms of size and complexity, was that with only myself operating it was pretty bloody quiet, so some automation would be excellent and most apropos.

 

What kind of block detectors did you use? I can see how Traincontroller or JMRI would set routes and even control trains - but I'm not clear at all about how occupancy detectors (such as http://www.coastaldcc.co.uk/products/teamdigital/dcc-block-detector)would be able to return data back to the computer. Are they constantly being polled for their status or is there some kind of 'return to controller' configuration you need to do on the detector itself?

 

I can't seem to find any information about how the push-button timetable/routing works on the RR&Co website. I gather the momentary contact switch is triggering some kind of pulse or in the DCC bus, and the controller is somehow getting that information back and using it to set a route in the same way one could click a button on the PC itself, but I'm not clear how it actually connects up. 

 

Lastly, do you have automated uncoupling (hence your choice of Kadee's) or Mk. 1 hand? I gather these would be yet more relay-controlled electromagnets under the trackbed controlled by DCC accessory decoders?

 

Many thanks,

[northolland]

I am happy to explain how I made things work. However, the operations on the Chapeltown Loop that I describe below concern a layout that was demolished in 2013. I am in the process of building a new Chapeltown Loop which will make use of new technology for a number of reasons that I will explain below. This post also allows me to provide a bit of an update about where I currently am with the new Chapeltown Loop. I’ll start with the technology behind the old Chapeltown Loop first and then explain where I am with the new Chapeltown Loop.

 

Operating the old Chapeltown Loop in the way that I did required hardware and software. The software was Railroad & Co version 4, running on a Windows XP desktop, all very ancient; giving you the reason why I have to update for the new Chapeltown Loop.

 

For hardware, apart from the desktop computer, I had the original Uhlenbrock Intellibox. The Intellibox has an in-build computer interface that allowed me to connect it with the PC by way of a RS232 cable. In this way, RR&Co was able to communicate with the DCC system.

 

Train detection and Feedback

 

For train detection, I used Littfinski RM-GB-8 occupancy detectors. This unit is a currency detector. If an engine with a DCC decoder is on the track, it draws a current and it will tell RR&Co that the block is occupied. You will have to create a block physically on your layout by electronically isolating it from other blocks and you have to define a block and identify and add the block address of the occupancy decoder in RR&Co to make it work. You connect DCC track current to the RM-GB-8. There are eight outputs that connect with eight separate blocks. The RM-GB-8 therefore can watch 8 blocks. Feedback from the RM-GB-8 about occupancy to the Intellibox goes through the S88 bus.

 

I also used Littfinski RM-88 units. This device provides feedback to the Intellibox via the S88 bus if one of the contacts is connected to earth. It has sixteen contacts and if you connect a push button switch or an on/off switch that makes the earth connection, it provides an input to RR&Co, which can be used to operate something like a point, a signal, or to start a train. But what exactly each switch/contact operates has to be programmed into RR&Co.

 

So feedback for and inputs into RR&Co was provided by connecting Littfinski RM-GB-8 and RM-88 feedback units with the Intellibox S88 bus.

 

Point and signal operation.

 

Points in the visible part of the layout were operated with Tortoises through NCE Switch 8 decoders.  I had one servo-operated signal, as I had not come round to building more signals, which was connected through a ESU SwitchPilot. The points in the fiddle yard were operated with Peco PL-10-W low current point motors with Littfinski S-DEC-4 decoders. This means that everything was connected to the Intellibox and through the Intellibox computer interface with RR&Co.

 

Train operation

 

 

All engines had DCC decoders fitted from various makes. Some were sound decoders others were not. What has to be done to have trains run automatically is that the decoders have to be calibrated for speed, braking distance and acceleration. This requires tweaking the decoders’ CVs to create maximum (scale) speeds, and braking and acceleration characteristics that suit the layout. Once this is done, RR&Co needs to be told about these. There is a function in RR&Co that allows you to do this.

 

The next step is to programme schedules into RR&Co. A schedule – when activated – drives a train over a certain route on the layout. The schedule also operates the points and the signals and can make the sound decoder make noises when and where you want them. RR&Co is smart enough not to make trains run into each other as long as it knows which blocks are occupied. RR&Co knows the blocks and the routes – as these have been programmed in by me - and while RR&Co drives a train, through the feedback from the occupancy decoders, it keeps track of where a train is.

RR&Co also has a timetable facility. I had trains travelling the layout out and back from their spot in the fiddle yard at fixed times, in accordance with the schedules that were added to the timetable.

 

I also had a number of goods trains that I could shunt with. This required more programming as well as designing the layout in a way that facilitates this. With regard to layout design, you have to have designated reception/departure sidings with occupancy detection. I have found that not all sidings need to have occupancy detection but you do need to add at least as many blocks with detection in the sidings as the maximum number of engines that you expect to be present at any moment in time.

 

If you look at Astley Down Sidings in the diagram in an earlier post, the No.1 arrival/departure line was two blocks. One block was from the entry point up to the crossover with No.2 line. The other block was the crossover point plus the bit of track beyond it towards the stop block. Apart from the shunting neck, the other sidings did not have occupancy detection.

 

A schedule would be programmed, which drives a train from the fiddle yard from - for instance - the Elsecar & Hoyland direction into the No.1 line, stopping at the stop block. The engine would be in the block beyond the crossover. RR&Co knows which engine this is, as it keeps track of where an engine is on the basis of the feedback from the occupancy detectors. I had a Kadee magnet there and with the engine stopped with the coupler over the magnet, if I then attach the Astley pilot (waiting in the shunting neck) to the back of the train, the coupler would release and I would be able to pull the wagons away from the train engine, clearing the no.1 line.

 

I had another schedule programmed in RR&Co that would drive the train engine that was left at the end of the no.1 line across to the loco facilities, setting all the points, turning the engine (the turntable was motorised and driven by an engine decoder) and leaving it their in one of the engine facility blocks. This schedule was activated by a push button that was connected to a Littfinski RM-88 feedback unit.

 

I would then do my shunting and when I had finished, I left the wagons to be picked up in the no.1 line at a particular spot (marked by a little Airfix/Dapol men). Pushing another button would activate a schedule that drives the train engine that was waiting in the loco facilities to the front of the train. The schedule was programmed to stop the engine in the no.1 line block exactly at the spot where Airfix/Dapol men was standing and the engine would couple to the waiting wagons. Another schedule was programmed that was activated by a button, which would set the points, drive the train, etc. to its spot in the Fiddle Yard. This is just one example and I had other schedule schemes that did similar things.

 

I am to recreate all of this again with the new Chapeltown Loop. However, I had to ditch the computer with Windows XP. It was too ancient; I could not upgrade it to a later Windows version. As I also switched over to Apple, I tried to find software that would run on a Mac. There are Mac versions of JMRI and Rocrail and I tried both. In JMRI, as far as I understand it, automatic running of trains requires you to start programming in Python. I did not fancy spending the time to learn this. I played around with Rocrail, found it not easy to learn but more importantly could not find a facility to have trains running at scale speed. Eventually I bought RR&Co Silver and run it on an old laptop I had lying about, which just about manages to run Windows 7. Eventually I expect to have to buy a new (but cheap) desktop. I have most of the new fiddle yard up and running and am getting to grips with RR&Co Silver. It differs from RR&Co version 4 in many ways and I have to find new ways to programme and run the layout how I want it, but I am confident that RR&Co Silver will allow me to do this.

 

I also planned to ditch the Intellibox because the RS232 connection has become obsolete. I decided to go Digitrax. I bought a DCS51 Zephyr command station and a PR3xtra computer interface. I have found that the Zephyr lacks the power to run some of my engines at their maximum (scale) speed. I suspect this is because the maximum voltage on the track is lower than I had with the Intellibox. I might need to buy a booster. Does anybody know?

 

For occupancy detection in the fiddle yard, I now use the Digitrax BDL168 units connected through a PM42 power manager, connected through Loconet. I plan to re-use the Littfinski occupancy detectors and feedback units for the visible part of the layout, connected to Loconet with an Uhlenbrock S88 Loconet adaptor.

 

The points in the fiddle yard are operated with digital Cobalts. With the lack of power of the Zephyr, I put the Intellibox back in use to drive the Cobalts. The old laptop has a RS232 connector and RR&Co allows you to connect more than one command station. The Peco point motors to my taste are too crude and too fiddly to mount. I had 99% reliability on the old Chapeltown Loop, with the Peco point motors responsible for the 1%. The Tortoises and NCE Switch8s are for the visible part as well.

 

Finally, as explained above, I did not have any automatic uncoupling apart from a few strategically placed permanent magnets. When engines stopped with the coupler above it, the coupler would be sufficiently released so that I could pull away the train, leaving the train engine standing there. When I build Pennyfields Sidings, I added Kadee under-the-track permanent magnets. I found them a nuisance because – I think this has been documented in other threads – the couplers tend to uncouple also if you don’t want it. I did buy an electromagnet un-coupler to experiment with but never came round to doing this. I did not mind using the little yellow Kadee stick to uncouple wagons while I was shunting.

 

Finally, I have the bench work for the visible part of the layout in place and expect to start laying track soon.

[Lacathedrale]

Thank you sir - I'm starting to grasp this idea that one fundamentally needs either a system with feedback inputs (such as the Lenz LSV101) or a dedicated add-on for such  (such as the Loconet module for your Zephyr set).  The LDT product pages are very opaque, it seems as though they don't want people new to the hobby to have a chance! 

[McRuss]

Hello Northholland,

 

instead of the Zephyr Extra you could use a Intellibox Basic, or Intellibox 2. The first is an  original Intellibox, but with an USB Interface. I've used the Zephyr Extra in the past, but was not entirely convinced, because, I can't run more than 10 trains at the same time. I know Digitrax told something different, but on my it won't work.

 

Markus

[danstercivicman]

All very interesting stuff!!

[Dave Goddard]

This thread is wonderful - a fascinating account of a remarkable railway system (much more than just a model railway) with a quite superb method of operation.  Piecing together all the factors of different industries and their co-operation, integrating transport within them and factoring in what goods was needed for each one and at what times...this could almost be a module in a Business Studies or Logistics degree.

 

The paragraph headed "The layout is no more..." actually quite saddened me, but I am looking forward to following the resurrection of this superb project - best of luck!

[northolland]

I agree that LDT's product info is mainly catering for the initiated. It probably does not help that the English text appears to have been translated from German. The LDT equipment I have has been behaving very well. Digitrax info - and especially the manuals - are not the clearest either. However, Digitrax does have a facility on their website through which you can ask them questions and they provide quick and good responses. I have done that a couple of times.

 

The main reason why I have to do away with my old Intellibox is the RS232 interface. Current Intelliboxes having USB interfaces would negate this problem. I was actually thinking to use my old Intellibox as a booster in conjunction with the Zephyr by linking it up through Loconet rather than to the computer via the RS232 connection. If I could use the Intellibox to feed the DCC signal into the Digitrax PM42 power manager and then into the Digitrax BDL168s occupancy detectors that I have; with the Zephyr driving the digital Cobalts, then I would have enough power to drive the trains. However, I could not make it work, presumably because the Intellibox cannot be switched into booster mode. Another problem is that the BDL168s need to be connected to earth. The Zephyr and other Digitrax boosters/command stations provide an earth connection. The Intellibox does not. One of he questions I asked Digitrax concerned whether I could feed the DCC signal from my Intellbox into their PM42 and BDL168s while connecting the BDL168 earth leads to the Zephyr. They advised me against doing this. I think that prevents me from using newer versions of the Intellibox as well.

Edited December 6, 2017 by northolland