Computer Automated Loft Layout

 

Note that the smaller pictures can be enlarged by clicking on the images.

 

Introduction:

The loft layout has not moved forward for quite a while due to the ongoing locomotive and rolling stock projects listed in the Model Rail index page.  However, this has enabled me to revise some of my original ideas and (more importantly) to clear a lot of junk out of the loft, making it a more inviting part of the house to visit!

The story continues......

 

The current layout plan:

The revised layout has an outer horizontal circuit with up and down lines and a long passing loop, coupled to a large freight yard. Sharing the main station is an inner circuit, also with up & down lines, which incorporates a slope down from the main station, to run below the level of the outer track at the bottom of the layout diagram above. The inner circuit is however an inch or so above the base board at its lowest points, to allow for a river/canal feature adjacent to the tracks. Interconnections are provided in a way that I hope will allow for some interesting automated schedules, with several trains running under computer control, simultaneously.

 

Some Issues with Diamond Crossings:

The new layout brings back the use of diamond crossings to enable trains to move from the up line of one circuit to the up line of the other and similarly from down line to down line. However, the crossings have a few issues which of course, I discovered the hard way when first evaluating code 75 track.

First for simplicity, I tried the insulated frog Peco small crossover. Running some representative locomotives over the crossing revealed a few problems:

1) The prolonged discontinuity in supply as the train passes over the crossing requires multiple pick-ups, to maintain rail contact. This is not too much of a problem as I've added additional contacts on many of my DMUs, whenever supply issues occurred.

2) The killer however is demonstrated by my Hornby 142, which when traversing the frog, consistently shorts out the DCC supply by touching the adjacent track as well as the wanted track as it rolls over the frog. (Perhaps the wheels are wider or have a simpler profile compared to other locos, or maybe the back to back dimensions are out.) Whatever the cause, it illustrates that switched polarity electrofrog diamond crossings should be a better way forward.

The original

 

The electrofrog replacement....some isolating joiners required here!

 

Traincontroller software includes a switched polarity cross-over within its track plan elements, so I can use a points decoder to switch the frog polarity via a two pole changeover relay, but I'll need another relay to provide feedback to the Lenz control system to confirm the operation has happened (in the absence of a three pole relay). The Lenz controller and the PC then know which way the crossover is polarised and this should protect the trains from collision when schedules result in trains approaching the crossing on both tracks. I better confirm this on a temporary test track before committing to the definitive layout.

Probable Crossing frog supply circuit including RS feedback signal.

 

Yup! That completely eliminates the running problems! (Just driving the relay from a bench power supply to verify loco behaviour over the crossing)

 

The LS100 is reprogrammed to set output 1 to continuous...... then:

LS100 output voltage between C and + is measured at 15.5V so I'll add 220 ohm in series with each relay coil to keep them cool. 

Then I have to set up a simple Traincontroller layout and corresponding schedules to see how it handles crossing conflicts.

Then we'll see if we get a train wreck or a controlled stopping reaction by the lower priority train!

I've found a window in Traincontroller that I have not previously filled in, concerning "route" detectors. (I don't find the programme very intuitive) We'll soon see if this enables the programme to detect and deal with conflicts on the crossover. (It may explain some undesirable behaviour of the original test circuit, where I've occasionally had a point switch under a train, which should never happen!)

The Process I used for registering route monitoring contact indicators is as follows:  In edit mode, select the track element concerned (in this case, the crossing). Call up properties. Under the General tab, enter a name for the element. Under the connection tab, enter the address of the controlling auxiliary decoder output. Under the Indicators tab, use the create command to generate a new named  route monitoring detector, which should be entered in the right hand column as the assigned indicator. Highlight this, and press the properties button below. This opens the contact indicator window for the new detector. The General tab includes the new detector name, the connection tab enables the address for the the new detector to be added. Open the operations tab and from the "operations" menu, select "Routes". Add the required routes to the right hand side column and set them to active via the small box below. This seems to work on a simple crossing element. Of course things may get more complicated for routes with more connections!

Excellent!  No train wrecks!  I had to make a minor change to the default schedule rules to enable a train to initially remain stationary in its start block until it had a clear path to move into. Once this was done, I ran both trains in shuttle mode and started the the schedules close together. The trains ran over the crossover alternately and faultlessly, with polarity and protection all working without a hitch! 

I was impressed enough to make a short video:      Please click on the picture below to start the Youtube video.

 

Wiring up the associated points on the real layout:

Showing how I plan to hook up the circuit transfer tracks to the occupancy detector outputs.

 

 

 

 

Points switching arrangements: (Proven earlier)

 

Revised Track Plan:.................

 

Phase 1 Build:

The outer up & down line circuits with the passing loop at the bottom of the diagram.

2nd Feb 2016:

The entire outer circuit with passing loop, is now in place with all track connection wires in place at the track end, but so far no external connections!

Next:

The common track is the outer rail. All outer rail drop-down wires need connecting to the common bus bar.

The inner rails are separated to form the blocks. The inner rail drop-down wires need connecting to each block bus bar.

The common bus bar & the block bus bars need connecting to the occupation monitoring PCBs on the appropriate interface boards.

Then points P1 to P4 have to be hooked up. (Remaining points and the cross-overs will be permanently switched & hard wired to the through routes. These will form part of the adjacent blocks for initial testing.)

10th Feb 2016:

All now connected up and working:  Two points switched the wrong way round and were corrected via Traincontroller adjustment. One point switched correctly but indicated that it was switched in the opposite position to reality. The feed back wires were swopped at the points decoder to fix this problem.

All now working correctly when operating manually via the computer.

Interface board controlling the 4 points and monitoring track occupancy for blocks 1 to 4 & the points.

 

Interface board monitoring track occupancy for blocks 5 to 12.

Next:

Write a schedule to automate the operation of a DMU including station stops and appropriate sound sequences.

11th Feb 2016:

First schedule on the layout:  Bachmann Northern Rail Class 150 mark 2 pair successfully orbits the inner track, making station stops including staggered hazard light operation and all the appropriate sounds. Engine power sounds are also altered around the circuit, by programmed speed changes! 

Next:

Tidy up the wiring and then try to tidy up as much of the accumulated Bits and pieces on the track base-boards as possible.

Decide upon a better lighting regime. Its probably too dark up there for good video results!

13th Feb 2016:

Decided to try a triple GU10 spot fitting as the base can be fixed straight on to the sloping ceiling and the individual spots can be pointed at the layout as required. (A LED tube light would face the wrong way if mounted directly on to the sloping ceiling, so a support bracket arrangement would need to be fabricated, to point the on-board LED strip at the layout.)

Experimental lighting unit.

 

I've also ordered up some 5W cool white GU10 LED bulbs.      If this works OK I'll need a few more fittings to cover the entire layout.

 

A quick tour around phase one of the loft layout:

 

 

 

 

Northern Class 150 about to depart for Leeds.

27th Feb 2016:

Small Ikea unit now in place and the DCC control kit is now off the floor and installed in/on the unit (together with a short length of programming track). More LED spotlights are now fitted to the roof around the layout, with wiring up to junction boxes in the ridge and mains connection via a 3A fused mains plug. (Unfortunately Mr Portillo wasn't available for the big switch-on, but fortunately it all worked first time!) Wife has kindly helped to remove remaining non-required storage items, leaving a largely unobstructed central floor space  up there!

 

Now operating some longer complete trains around the tracks: Gresley P2 with 6 Mk1 coaches, Cross Country HST with 6 mk3s & MOD nuclear flask train.

Good to finally see the complete trains in operation.... all very smooth and behaving well. Just one derailment, due to the DAB antenna of the stereo coming unstuck from the wall and falling on to the track just as the flask train was passing :-)

DRS Royal Navy flask train

The computer is getting confused due to long train lengths and flashing rear lights, so I need to apply conductive ink to enough rolling stock wheels for the occupancy detectors to give a clearer picture on the location of the entire train. Just done this to the DRS flask train & we'll see if I've done enough wheel sets to restore clarity to the PC software!  (I've applied between 5k and 8k of resistance to each of the wheel sets in around 50% of the train's unpowered 4 wheel bogies.)

28th Feb2016...... The PC is now happy with the DRS flask train and I'm no longer getting error messages..... so I'll give the P2 mark one coaches & the HST Mk3s, the same treatment!

P2 Prince of Wales

 

Cross Country HST

 

Ongoing.........

 

Updated layout plan, including simplified Phase 2 inner loop:

 

The revised layout has a modified inner circuit with no turn-outs apart from those in the vicinity of the main station. A canal is now planned below the outer track embankment, requiring the space originally occupied by the country station passing loop. A two way working regime will be applied to the inner circuit tracks instead of the original up-down line interface points.

 

Some scenery enhancements:

A grassy embankment and canal are now in place.

 

The track bed for the inner loops and modern country station are taking shape. (Junk below deliberately blurred.)

 

Dent Station platform & main buildings are now located on the outer loop. (The overflow pipes may end up as mill chimneys?)

 

The inner loop bridge deck is trimmed ready for side walls and arch support..... to be followed by the valley walls & backscene.

 

Swimming not recommended!

 

 

 

Just one block temporarily wired to my PSU, to run the D16 load test on DC.

 

New board 5 with Cross-over control relays on-board.

 

Phase 2 Part 1:

I've now connected up the inner loops...... I had to re-lay the tightest curve to increase it's radius, to permit longer wheelbase steam locos to negotiate the curve safely. Initially I've only activated the two bay platform points, and I've hard wired the frogs on the remaining points and crossovers in the through direction. (I'll gradually work my way through the remaining point and crossover wiring over the next few weeks, adding them to the PC control system.

Running test trains, its interesting to see how the BEMF drive system copes with the gradients on the new circuits. A major worry on the inner circuit is its proximity to the layout edge. This looks like a fairly disastrous accident that's just waiting to happen........  So the next priority is to create an extended inner rim that would "catch" a derailed loco & train before it plummeted to the floor.

The story continues......

The saga continues......................

 

Supplier website links (work in progress) :

 

Railroad & Co Traincontroller software is used to control the layout
DCC Supplies Helpful & well informed supplier of all things DCC. The Lenz 100 starter set came from here
Digitrains A very helpful DCC specialist company. The test track LDT and other Lenz parts came from here.
Lenz German DCC system supplier
LDT Specialist DCC module supplier

The photos of real track were taken on the Settle-Carlisle line at Ribblehead during 2009.

 

Click to move to Model Rail Index Page