Automated signal schematics

After my initial success with the basic signal control circuit, I needed a second to work the two signals at the other end of the station. As with most of the signals on the layout at the moment (apart from the two automatic ones on the shuttle) their aspect control is governed by the routing of the points, ie when the points are changed, the corresponding signal also changes and I want this feature to remain.

The design of the new circuit means I can wire it into the existing wiring and simply divert the current signal feed wires into the PCB. I've also designed a configuration for working a pair of signals from one control PCB.

After some head-scratching around how to configure the circuit, I acquired a circuit design tool for my Ipad, which also allows me to run animated simulations of how the circuit will perform - changing the values of the capacitor and resistor shows pretty accurately how long the timing will last. I might splash out on some variable resistors so that I'll have the ability to adjust the timings. It still works out cheaper than buying them pre-made!

Using the design tool, I've mapped out the schematic diagram for a two aspect signal as per below...

For the circuit to work, I need a trigger signal to start the timing (charging / discharging) process. For nearly all DC applications, either reed switches or infra-red detectors are used. Neither of these options appeal and for three rail applications there is a much easier way, which uses the dead track sections in standard HD points, which is why there's a second relay in the circuit.

The left-hand relay is momentarily energised by the passing train, this closes the relay contact and allows voltage through the diode into the capacitor to charge it. Once the train has passed the section, the coil is de-energised, the switch contact opens and the discharging process begins. Current now flows from the capacitor, through the resistor to the base of the transistor fitted in the 0v return of the main control relay, which is powered with a permanent 12v supply. The transistor is now energised via the capacitor charge voltage in its base so the relay coil is now powered. The contacts change and power is sent to the red signal (colours not shown here) once the capacitor has fully discharged, the transistor loses its base signal and prevents current flow, de-energising the relay and returning the contact to the green (normally closed contact) position.

The switch shown on the right of the diagram is the red or green power feed that is received from the existing wiring configuration. When the signal is set to red, power is sent via the other diode straight to the red signal, bypassing the relay control altogether.

After even more head-scratching, I worked out the circuit for two signals, controlled by the same signal feeds (red and green aspects swap according to the points position)

I'll be relying on the track voltage from the running train to provide the power to the coil of the first relay, I know from previous experience that this will work. I don't need to add any isolated track sections, I can use the dead sections of the points (circled)

The two short dead sections circled have wires soldered underneath which will feed into the first relay coil. Two additional diodes will prevent track shorts from occurring. Thankfully when we built the layout, I soldered extra wires on all the points as a precaution!

Here's a quick mock-up using a three rail pick-up from a three rail loco to demonstrate how this works...

As the locomotive approaches this section, the third rail pick-up is momentarily conducting power from the centre rail and makes contact with the isolated section.

As the train continues to move the process is repeated again as the front of the pick-up makes contact again with the main rail, as the back of the pick-up connects to the isolated section. This effectively creates two very short pulses of power into the relay coil, which starts the process.

 

As the pulse is so brief, it's hard to capture in the simulation software but here the capacitor has been charged and is now powering the main relay, closing the contact and turning the signal red (first lamp)

Now the capacitor has fully discharged and the relay is disconnected, power returns to the green signal (second lamp)

If the points are selected in the other direction, power is shunted straight through the diode to the red signal, bypassing the relay control altogether, keeping the signal red at all times.

 All I need to do now is build the circuits...

Automated signals (again)

A while ago, with the old layout, I experimented with automating the signals using a couple of latching relays, which in fairness worked quite well.

The theory was to use the isolated pieces of rail that existed at each end of the station on the old layout - the signal changed to green on entry and then back to red as the train passed over the points at the other end. With the new layout I decided against overly complicating things and now that the points on the station approach on the outer loop have been removed, I need a better solution.

Of course you can buy signal aspect controllers but I'm tight careful with my money these days, I'm also developing a bit of a hobby for electronics and after building the pulse control for the shuttle points out of a transistor, relay and resistor, coupled with making lots of points control PCBs and a new control panel, I found myself with quite a few spare components left in the spares box. After a bit of internet research my idea of using the discharge of a capacitor to create a basic time delay circuit seemed like a reasonable approach, so I set to work...

The relays mounted on the PCB in this picture were originally controlling the track isolators in the station, now that this has been simplified this PCB is redundant so I carefully removed one of the relays.

The diodes were bought as a job-lot for various other signalling control ideas and the resistors were supplied with the LEDs I bought to make the control panel. I also have a few capacitors left over from the points control PCBs and several transistors...

And this is the first attempt at making a time-delay signal aspect control.

There are two 1K resistors here, due to the size of the capacitor and even with the briefest of supply pulses, the circuit took nearly a minute to discharge the capacitor fully and release the relay so the second resistor is bypassed, which gives a time of around 20-25 seconds. That should allow even the goods train to pass through the signal and clear completely before the signal reverts to green.

An extremely cheap and seemingly quite effective method of basic signal control, which I can hopefully integrate into the layout without too much grief!

Video of the circuit being tested:

https://www.youtube.com/watch?v=Dho5HM39KAk&feature=youtu.be

Fiddle Yard Modification

I wasn't planning to start work on the fiddle yard until February but as there are no more scheduled outings for a couple of months I decided to begin lifting the track from the rear fiddle yard sections from one of the three boards.

The current arrangement has proven to be too awkward, particularly when trying to access the outer loop from the fiddle yard. This involves stopping whatever is running on the inside loop while the changeover occurs, I didn't install any safety devices to halt the trains and the Deltics, with their rubber tyres and long rakes of goods wagons, don't appreciate being stopped suddenly.

Even the inner loop has been troublesome, the twists & curves associated with accessing the fiddle yard, and the additional curves due to the way the points are fitted means that changing locos is normally done by manual intervention, ie lifting the loco off the track by hand. 

So it's time for this configuration to go - no more twisty bits and two crossovers removed from the layout...

 Barely a few minutes later and the bulk of the track is stripped. I'll be leaving the straight tracks on the edge of the board in-situ, the amended fiddle yard will connect with the original tracks here for two reasons. Firstly because it will reduce the amount of modification work needed to the centre rear board (as the tracks will still line up) and secondly because of the location of the run-around loop, I need to ensure that the track layout remains in the same overall location so that the connecting track still lines up properly.

Beginning to loose-lay the modified points, the overhanging track will be cut to fit...

With the tracks secured into position and the new PCBs and wiring in place, time to start wiring the points and track sections...

Once all that was done and all the track sections tested, time to add in the manual points for the new loco sidings at the edges, these will be used to accommodate locos only. The plan is that the new, long, straight sidings will accommodate whole trains - three on each side, six complete rakes stored ready to run. The outer sidings will be used for prepping trains, swapping locos and making any changes to the train rakes. I'll add some uncouplers to aid with this.

Although it's no longer possible to access the outer loop from the rear of the layout, I can use the run-round at the front to change locos from one side of the yard to the other. I'll probably keep the outer loop for EMU / Diesel locos and keep the steam locos on the inner side due to the access to the turntable at the front being much easier from this side.

One down, two to go...

E3002

I have a wish list of Hornby Dublo locos that I still want to collect, sadly after over eight years of collecting (or is it nine?) the locos I've yet to find are on the more expensive side.

I've bid on, and lost at least a dozen of these electric locos, so when I spotted this boxed, two rail loco (as they never actually produced the three rail version) during Saturday's exhibition, I decided that I had to have it.

So I'm now the proud owner of E3002, a little playworn, the numbers are faded on one side but I think it adds to its charm so I have no immediate plans to resolve this. Three railing one of these should be easy, in fact it turned out to be easier than expected, in fact there are some clues that suggest that this two rail loco isn't all that it seems...

First of all there's the box - it's mint, in fact it's too mint. I suspect it's a reproduction box, while I'm happy to be corrected I'm pretty certain that the original box would have had in internal flap and some packing rings.

Next there's the wiring, again it's very good and the many tales I've read of wires breaking and wires being too tight to work on the loco properly, don't seem to be issues here. The wires look very clean and the solder work very shiny for a 50 year old loco.

Then there's the wheels - pretty certain it wouldn't have left Binns Road with mis-matched wheels, the only logical reason that I can come up with is that at some point in its life it's been three-railed and then put back to two rail again.

So I've stripped the loco by this point and removed two of the isolated wheels, replacing them with some spare black wheels I had in the spares box...

Needless to say, regardless of whether it's spent time previously as a three rail model, it'll spend the rest of it's days in that state! For now I've fitted a Märklin skate as these models, like the EMU & Co-Bo, use a shorter pickup than the version found under most steam locos. Eventually I'll replace it for the right version.

Up & running in under an hour - result!

Beaminster show - January 17th

Our first outing of 2015.

This is the last outing for the layout with the current fiddle yard configuration, work to modify this will be completed in February. The points are modified ready and the electronics built, the control panel work is completed and I finished the modifications to the front board wiring the day before the exhibition, this unfortunately meant that there was no opportunity to test the layout first but thankfully there were no issues!

After a relatively painless setup the night before, we arrived on a rather icy Saturday morning ready to run. I decided to experiment with the 0-6-0 diesel shunter on the shuttle track hauling the small goods train, we'd previously identified that a small rake of goods wagons looked rather good running up and down the front of the layout and the shunter is ideal for this, not only because the loco is so small but also as it isn't happy hauling long rakes of wagons or coaches, so this little rake is ideal...

I kept a 0-6-2 tank on the shorter platform with one passenger coach, ok so in real life you probably wouldn't see trains propelled like this, but who cares?

The new feedback controllers also allow a much more sensible running speed and collisions with the buffers no longer occur, we even moved the isolators nearer to the buffers!

The red 2-6-4 LMS was put in charge of the mail train and Deltics Crepello & Royal Scots Grey hauled another ridiculously long goods train.

Dorchester hauled a five coach rake of Pulman coaches, unfortunately some warpage has occurred to the rear sections of the upper loop, so we avoided running the usual brick wagons as the risk of uncoupling was too great.

The back of the goods train, which just about fits into the long straight section on the front of the layout.

The outer loop now runs much smoother and straighter with the two redundant pairs of points removed, now that the section has been re-ballasted there is no evidence that they were ever there.

I managed to complete the work to modify the points on the upper loop to electric operation, this with the addition of the changes to the switch panel means no more walking to the other end of the layout to change trains - just one throw of a switch and the points at both end throw. These are plugged into the fiddle yard boards via six-pin plugs.

Here the warpage on the rear section is quite visible, even after attempts to level things up, however all of the coaches managed to stay together.

I also managed to pick up some more Southern Suburban coaches at the exhibition so quickly added these to the back of the rake being hauled by the two trusty 8F locos. 

Another view of the shuttle goods train.

I finally remembered to un-box the breakdown crane, and this was left parked in the shuttle siding with the Southern 2-6-4 attached to it.

A final shot of the goods train (you may be thinking that I love my little goods train!) in transit.

We were very well received, attracting lots of attention, some nice feedback and at least two more bookings. Sadly I forgot to bring the video camera that I got for Christmas so these are the only pictures!

All locos ran well, the only minor issue is with the turntable, which I'll look at while the other work is being done in February. The new switch panel stand also made life much easier, a much better operating height, and no back ache!