That Effing(ham) Railway

[Trofimow]

A bit more progress....

 

The basic structure of the second third of the bottom level storage loops begins to take shape..

 

 

 

Edited September 9, 2022 by Trofimow
Restoring lost pictures

[Trofimow]

Time for a bit of an update on the progress on Effingham Mk2....

 

Around two thirds of the lower storage deck has been built so far and track laid.

 

This is the track plan.

 

 

What is not apparent from the plan is that there is a difference in heights in the top right hand corner, shown more clearly in this 3D view.

 

 

This is the lowest of what are planned to be three identical storage levels, with a return loop on yet another lower level. The whole multi level storage system is on a ruling gradient of 1 in 100 throughout.

 

The principles mentioned in previous posts have been incorporated into the design, the simplest possible pointwork, resulting in standard length storage loops in a parallelogram pattern. A standard length train is a pacific and 6 BR Mk1's. A standard length storage loop holds one such train, and double length loops will hold 2 of these or one 13 coach train.

 

DCC concepts power base principles are being used throughout, even on falling gradients, though a home brewed alternative to the proprietary system is used on non visible straight track for reasons of economy.

 

Initial testing of track laid so far indicates that a magnet fitted Hornby Bulleid pacific can comfortably handle the maximum length 13 coach train, including restarting on the 1 in 100 gradients.

 

 

 

[Trofimow]

The correct and reliable functioning of these 3 storage levels is essential to the success of the finished layout. The design is a little unusual in that the storage is in effect a one way system in that it is entered and left at the same points regardless of the route or direction that a train used while "on stage".  Entrance and exit routes are grade separated.

 

All this has been designed in SCARM, which has been a great help in ensuring that necessary gradients can be achieved. Ensuring that the gradients as built are actually as designed is vital. to help in this, I am using these aids;

 

 

The first item is a spirit level with a digital readout of slope as a percentage.

 

The Piko Messwagen is most useful in discovering unexpected humps, dips or twists in my tracklaying. It gives a real time readout to a smartphone as it runs around the layout.

 

 

 

[Trofimow]

It's a couple of months since my last update. There has been steady progress in the meantime, but not a lot to show you.

 

A great effort has gone into wiring and basic systems infrastructure for the whole layout, not just the small portion built so far.

 

One of the principles instituted as a result of experience with Effingham Mk 1 is to absolutely minimise under baseboard wiring, which had proved to be a PITA for all sorts of reasons. It is, in fact, all the more necessary to do this on the rebuild as the clearance under the baseboard at the lowest point of the layout is barely 7 inches.

 

All the main wiring, including ring mains and busbars for the main electrical services is now carried along the front of the baseboards in cable ducts. Circuit boards for occupancy detection, the Megapoints system, etc. are mounted above the baseboard in 3D printed holders and identified with printed labels.

The DCC control is split through a total of 10 PSX circuit breakers for the whole layout, and for the storage levels currently under construction, each PSX output is further split down through a control panel so that each RS-8 occupancy detector board and each point ladder can be individually switched out. These panels serve no operational purpose, but are purely for maintenance and to aid fault finding, which was previously a nightmare.

 

All the main systems and infrastructure services are Alexa enabled and can be switched on and off by voice control. Why on Earth would I want to do this?

Well apart from the obvious answer – because I can – there are practical advantages.

 

Effingham Mk 2 is going to be a large and complex layout and already has multiple systems which have to be started up and closed down in the correct sequence. I could build a control panel with lots of switches that need to be operated in the right order. Instead, one voice command will start up or close down everything correctly. Additionally, whilst working on wiring a particular system, such as soldering feeds, that system can be switched off by voice, and the switched on again for testing without having to extract myself from a tight space and walk across the room to a panel.

 

Provision has also been made to give some basic voice commands to RR&co Traincontroller, such as to pause or resume operations, once things progress that far.

Edited September 9, 2022 by Trofimow
Restoring lost pictures

[Trofimow]

Next in this occasional series of progress updates...

 

The final set of storage loops on the lowest deck have now been installed, which completes phase one of the build.

Next will be some intensive test running to find the bugs before work starts on the next deck.

A temporary return loop has been installed to represent the rest of the layout.

 

 

Edited September 9, 2022 by Trofimow
Restoring lost picture

[Trofimow]

Another occasional update on progress with construction of the layout....

 

Testing on the level 3 storage is now complete. Trains have been runnng under computer control from their assigned storage loop down to the return loop below level 3, back up to level three and then up to the temporary return loop at level 2 height, a vertical range of 12 inches.

 

The rails are graphited and with the magnetic adhesion enhancement, full length trains of 13 bogies or 39 slu goods stock have presented no problem, and a loco can restart a train at any point on the layout.

 

The intensive testing has certainly shown up those places where my tracklaying was not as good as I thought it was, all such instances have now been investigated and fettled accordingly. A few wiring errors have also come to light and been corrected.

 

The temporary return loop has now been removed and work can start on installing the baseboard structure for level 2 storage.

Progress beyond that is going to be dependent on the future availability of points and track in these uncertain times.

[Trofimow]

The proposed Alexa voice control additions have now been implemented and tested. These are useful in that it is not necessary to get to a control panel in order to action them, although they can also be operated by a hardware panel control.

 

"Alexa Emergency Stop" puts Traincontroller into freeze mode, cutting the DCC power and making a hard stop of all trains and layout functions. This can only be reset at the computer.

 

"Alexa Pause Operations" instigates a soft stop of the layout. It stops the timetable clock and locks the exit of every block, having a similar effect to putting all signals to danger.

 

"Alexa Resume Operations", as you might expect, restarts the clock and unlocks the blocks, enabling trains to continue.

 

Each of the above will  also trigger a voice announcement by the computer and put a message on the track display.

 

The pause function will also be triggered automatically by the PSX circuit breakers if a short circuit occurs, such as a derailment, with a voice announcement of which breaker,  indicating the general location of the problem. When the short is cleared, the breaker will reset and trigger the resume function.

[Trofimow]

Once again,considerable time has slipped by since my last update on the layout...

 

Very little happened for a while, because life got in the way, as it tends to, and because essential supplies were difficult to obtain thanks to the pandemic, but recently things have been moving on quite quickly.

 

All of the level 2 storage loops have been completed and thoroughly tested.

 

The baseboard for the last of the 3 levels of storage loops was completed at the back of the room, for level 1 B loops.

The track and wiring for those loops has also been finished and tested.

 

Baseboard is in place for the top level C storage loops on the centre peninsula.

 

[Trofimow]

Track laying is done for the first of two sets of storage loops on the centre peninsula and wiring is in progress.

 

 

Access to the peninsula has the stair well in the way, which needs to be crossed twice on each of the three levels, once on the way in and again on the way out. Two bridges are therefore needed on each level, passing between the bannister spindles.

 

Here are the bridges put in some time ago when the first level was constructed.

 

 

 

Now there are bridges for all three levels. It's quite a drop to the ground floor below.

 

[Trofimow]

Although Effingham has survived thus far using a single Lenz Set 100 for the DCC supply, it was always expected that in due course one or more boosters would be needed to support the number of trains that the layout will be able to accommodate, and the wiring was planned accordingly.

 

It seems that time has arrived, as it was noticed during the wiring of the most recently laid set of storage loops that the layout now draws over 4 amps of DCC power just in standby and with only a few trains parked on the layout, but not moving.

 

A second handset would also be useful to reduce the amount of unplugging and moving around of the solitary existing one, but a cost effective way of achieving all this needed to be found.

 

I obtained a reasonably priced second hand Lenz Set 90 on ebay, which gave me a LZV 100 command station and a handset, and from a dealer, a Lenz transformer to go with it in as new condition, but offered cheaply as it has a continental mains plug.

 

When the set 90 arrived I was not greatly surprised to find it reported Error 17, internal battery fault – perhaps why it was sold on. No matter, since I am only going to use it as a booster, in which mode the battery is not required. A simple internal mod is all that is necessary, and now the power available to the layout has been doubled.

[Trofimow]

With all the train storage on Effingham being on the hidden lower levels, a cctv system is necessary to see what's going on.

 

Each set of storage loops has a camera on the entrance and another on the exit, panable to view the approach and departure routes.

 

Here are the cameras at the end of the centre peninsula storage, one for each of the three levels.

 

 

The cameras are mounted on servos using 3D printed fittings. With three sets of storage per level, there's a total of 18 of them.

 

Pan control was done using arduinos (in the green boxes) and pots mounted to the control panel.

 

This pan control worked, but has never been entirely satisfactory. There are of necessity long wiring runs from the panel to the cameras, and control has always been twitchy, and subject to noise. Attempts to damp out the instability merely introduced latency into the control response, which merely replaced one problem with another.

 

With the penultimate set of storage loops now nearing completion, a better system was needed.

 

Effingham uses considerable quantities of servos for various purposes, and I have one of these little testers to use when installing them.

 

 

It was temporarily tried in place of the arduino control on the most remote of the camera servos and worked perfectly, so now these little units have replaced all of the previous arduino controls.

 

[Trofimow]

All three levels of storage are now completed at the rear of the room

 

 

and also on the centre peninsula

 

 

All the tracklaying and wiring is done so things are nearly ready for test running this phase of the build, however another system constraint might delay things a little.

 

I mentioned in a previous post that I had reached the limits of available DCC power, and how a booster was needed to grow things any further.

 

This time I've hit the limits of the point control system. Servos are used, controlled by a Sprog over a Megapoints network, which has now reached the maximum capacity of 192 turnouts. A second network will need to be installed before any more points can be linked to the computer.

 

 

 

[5BarVT]

4 hours ago, Trofimow said:

This time I've hit the limits of the point control system. Servos are used, controlled by a Sprog over a Megapoints network, which has now reached the maximum capacity of 192 turnouts.

That’s a lot of points!

Paul.

[Trofimow]

16 hours ago, 5BarVT said:

That’s a lot of points!

Paul.

Yes, it's scary how the numbers mount up!

 

Still the end is now in sight, only the uppermost storage level at the front of the room remains to be done.

 

One through road and seven storage loops in each direction, plus an entry road and an exit road  for the storage system, but that will eat another 31 points.

[Trofimow]

Intensive testing has been in progress of that part of the layout thus far completed.

It's been instructive to say the least, but the issues have been identified, and faults with the layout itself rectified so that construction can begin on the next phase.

 

We have four levels so far, from the bottom up, a return loop, level 3 storage and level 2 storage all completed, plus level 1 storage two thirds finished. The speed limit throughout the storage system is a scale 45mph. To circumnavigate at this speed takes around 14 minutes.

 

I have set up a representative selection of 20 or so trains to test with, and the problems so far identified fall into 2 groups, issues with the layout and issues with the trains.

 

Layout issues have been relatively few, starting with the least common, some programming errors where a section has been assigned the wrong DCC address for the occupancy sensor, or a turnout operation has been programmed in the reverse sense.

 

Next, there have been a few wiring errors, reverse polarity, poor connections or a feed not being put in at all. All easily rectified once identified.

 

Finally, there have been a handful of places in which my track laying has fallen short. These were soon found by the half dozen or so locos in my fleet that are most intolerant of poor trackwork. The necessary improvements have been made so that all these locos will now circulate round the entire layout reliably.

 

You might think that faultless running should now follow. Unfortunately not, which brings us on to the second group of issues, those with the trains.

 

Train lengths are long, up to 13 bogies or 39 SLU wagons, the layout is mostly on a gradient of between 1 in 80 and 1 in 100. There are curves of up to 180 degrees at 3rd and 4th radius in the storage areas. This results in a lot of weight and drag in the train and contributes to the issues to be dealt with.

 

The first and quite common issue has been wheels and bogies. Back to backs of course, but there has been a major issue with trains, particularly goods trains, derailing for no apparent reason. This was traced to vehicles from a particular manufacturer, made a few years ago. The wheels have a particular profile which is visibly different to the norm and have caused no end of derailments. Once re-wheeled with suitable replacements, the derailments cease.

 

Bogie stock has sometimes given problems with a given bogie always derailing at a particular place were nothing else has a problem. This has usually been down to too much or too little freedom of movement of the bogie in question, rectified by adjustment to the pivot.

 

Which brings me to the greatest cause of trouble, tension lock couplings. They tangle, they part spontaneously and they override one another in a heavy train descending a gradient and then derail it on the next curve. Sometimes they just fall off, particularly those with dovetail fixings. Where possible they are being replaced with alternatives in the longest trains.

 

All the test trains now perform satisfactorily, but it has been quite a game to get there.

 

That's it for now, who knows what more fun awaits with the next phase...

[Trofimow]

So the test trains that were giving trouble have now had their tension locks replaced with alternative couplings.

Most trains will run in fixed rakes so standardisation on one type is not necessary.

Reliability of running has been transformed and the longest trains can now be left to circulate continuously round the layout without problems.

 

This means that the layout itself as so far constructed has passed its testing and construction can continue with the next phase, which is the final set of storage loops on the top level, a vast expanse of plywood awaits...

 

[Trofimow]

Well, it didn't take long to hide all that bare plywood under some track.

 

Wiring it up might take a little longer though.

 

[Trofimow]

Wiring on these loops is now complete and tested.

 

There is rather a lot of it - certainly not the mythical DCC two wires only.

 

It seems that depending on your point of view, I over complicate the wiring, over engineer it, or arrange it for ease of operational fault finding.

 

[Trofimow]

Well, that's the three levels of storage fully completed, now the various ramps up to the scenic level need to be installed.

 

To start with, there will be two single track branch lines running into the main station at Effingham, these will share an access ramp into the centre of the top level storage loops.

 

This ramp is now in place and has been tested. Trains of up to nine coaches have been successfully started on the gradient using a pacific loco, which is far more than will be required in practice.

 

[Trofimow]

Some more progress has been made, the entrance and exit routes from the storage area are the current tasks.

 

The storage tracks across 3 levels are basically a one way system, with one way in and one way out, regardless of the routing or direction of a train coming or going from the “on stage” scenic level.

 

The scenic level as planned will consist of a BR Southern Region main line with 4 tracks at the London end diverging via a flying junction into 2 double track main lines at the country end.

 

The main station also has 2 single track branch lines, on on each side of the 4 track main line formation.

 

There are thus 10 tracks going on or off stage in 5 different places in the room which all need to arrive at the same storage entry/exit points and these need to be grade separated on the way. This was “fun” to achieve at the design stage and is proving challenging to actually build.

 

The access to storage for the 2 branches was shown in the previous post.

 

Here are the arrival/departure tracks for the main lines, the four to the right are storage entry, the four on the left are storage exit.

Ignoring the stray curved track on the left-

 

From left to right, they are:

 

Up central/eastern section

Up western section

Down fast

Down slow

 

Down central/eastern section

Up fast

Up slow

Down western section

 

 

 

And the view from the opposite end, showing how the central section tracks drop down before they turn to pass beneath the bannister rail to cross the stairwell.

 

 

 

[Trofimow]

There are now 7 bridges crossing the stairwell to and from the centre peninsula, two for each of the three storage levels, plus the L shaped access to the ramps from the rest of the layout.

 

As seen looking down the stairwell

 

 

And the view from below

 

 

The tracks are very well boxed in as it's a very long drop in the event of a derailment!

 

Once out of the stairwell, the climb up to the scenic level of the peninsula has now been installed, in the form of two turns of an oval ramp above the storage levels.

 

 

[Trofimow]

It's now just over 3 years since the great rebuild started, and at last some baseboard is beginning to appear on the scenic level, on the centre peninsula at least. There is still a fair amount of work to be done before the same stage is reached with the main baseboards around the perimeter of the room.

 

This will be the site of Forest Road station, a through station on the central / eastern section route.

 

[Trofimow]

Here is the current view of the centre peninsula now that things have moved on a little.

 

The 4 platform tracks through Forest Road station heve been laid and wired with their occupancy detectors for Traincontroller,

and the points have been fitted with their operating servos, connected to a new fifth zone of the Megapoints system and the whole lot programmed into the computer.  It's not much to look at since the last picture, but a lot of work has been done.

 

 

On the other side of the peninsula, track has been laid for a cement terminal. It's still at the trying out stage at the moment, the scenic items are just being tried for position to see what works and what doesn't.

 

The track has been wired so that trial operations can be carried out before anything is set in stone.

 

[Trofimow]

Gosh, 6 months since the last update on Effingham, time flies when you are having fun.

No new pictures this time, as there has been no new construction to show. That does not mean though, that nothing has been happening.

 

Intensive testing of what has been built so far is under way before the scenic level can be put on top of the now complete 3 levels of storage.

 

To do this, each train set has to be commissioned, have its storage location assigned and be thoroughly tested over every route that it is going to be required to take operationally.

 

Because of the nature of the layout, with multiple levels and automated running, each train set must be 100% reliable, or as close to that as I can get it. Trains are expected to be able to run up and down gradients, round curves and across complex pointwork without issues, sometimes even more than one of those things at the same time.

 

Achieving this with RTR models is proving to be hard work. Some of the things that have been found to lead to derailments and have to be dealt with, in order of frequency, are couplings that are unreliable or incompatible, close coupling mechanisms that stick or jam, bogies with insufficient or excessive articulation, wheels that are incorrect b2b or profile, and vehicles that are too light to stay on the track in a train of any length.

 

All these things are fixable, but take up time. So far 65 of the 109 trains in the roster have been dealt with and passed fit. In the main, tension locks have been replaced on bogie stock, with either rigid bars, roco type or magnetic couplings, and this has made the biggest improvement. Long freight trains are more of an issue and retain tension locks, but these have to be matched and tested wagon to wagon.

 

Progress is still being made, but not as quickly as I would like.