Interested in a signal building group?

[Eddyg]

Hi Alan,

If you can't find them, I will be happy to make some replacements, but Dave will have to send me some stiles if he has any, since I don't have any left here.

Eddie

[Ben Lyons]

Hi Alan

I have I have one bit Not stepped in but I guess you can do, it 6 rungs long, May well have off cut as I trim the long lengths for my upper posts.

Ben

[Ben Lyons]

Hi there was talk of wether the spectical operation lever should allow the cable to run in front or behind the arm, having spent some time on the net there seem to be as many in front as behind, however looking at brackets it looks as if in front is a good option as the cable needs to pass the platform beam and get to the transfer arm without rubbing on the steelwork.

So looks like l need to change mine around more 14 BA bolts oh well.

Ben

[Richard Phillips]

Graham did ask me to write up something about the way the base wiring works - I'll try and write something up after Easter is over. I thought I'd write a bit about what happened at Shildon commissioning some of the signals that were complete (or nearing completion).

I think most know there are three yards, called Shildon, Middleton and Noble. Shildon being the simplest with 2 working arms. I arrived on mid-Saturday and the bells and block tokens were up and running at that point. I had a look round the museum and exhibits and didn't interfere with what looked like a fairly harmonious system!

On Sunday I decided to install two of the signals that were complete or nearly complete in Shildon yard after talking to Jonny about where to take power - I made my first mistake trying to shove lots of wires into the same terminal block I allowed the positive and negative to touch and blew the inline car fuse Jonny had made wise provision for - so before I'd done anything I'd broken a working system and went off to get some more fuses from the local garage - and I didn't have any in my car so was probably a useful thing to do anyway...

By the time I'd got back Keith had bridged out the fuse but I wired in a take-off to another terminal block to drive the electronics board that takes the inputs from the switches/signal levers. As I'd finished the bases the Sunday before and I'm sure Jonny had been burning the midnight oil, we'd never tried it all together before and I soldered on some terminations to fit the logic board.

Much to my surprise S1 did seem to work as advertised! However at this point I/we discovered every time the crossing buzzers sounded there must have been a voltage drop somewhere and the logic board reset. I think we'll need a separate regulator and reserve capacitor as the battery was putting out 13/14 volts which is a bit on the high side.

I modified one of Eddie's signals with a longer horn and set the appropriate base address but S3 didn't seem to work consistently. Jonny explained this was to do with the issuing of a token and you couldn't issue more than one, hence repeating wagging of the switch didn't do what I expected... I don't understand it yet but I think it was working as designed! These were just set up on the table as we hadn't yet made up appropriate interconnects.

I moved onto Middleton and started to add signals there, and these didn't have the degree of movement required without longer horns so I'll need to modify the running program and adjust this for the length of servo horn we're all using. We tried firing up Pauls' laptop but ran out of battery! Since the point detection wasn't in place some inputs to the control weren't working and Jonny and myself will need to sit down at some point and simulate all the inputs and get check functionality.

I started to make up temporary interconnects so we might put out signals, but this took a very long time. I was melting the 5 pin DIN plugs, although plugging them into a socket acted a bit like a heat soak and kept the pins in place. I eventually abandoned this as it was taking a long time and I made the criminal error of wiring the pos/neg line of one of the leads the wrong way around and blew a voltage regulator in Ben's base which I had to swap out. I think a session of wiring these at our leisure and testing them is probably in order.

I suspect we'll probably become the owners of one cable per signal, and that will be the length of the longest interconnect so we can bring signals and leads to Shildon and just connect them up no matter what the position, and coil any excess where necessary. I'm making up a little board that you can plug in the end of the lead with a female socket that has switches to control the arms manually or you can send either PWM for the Megapoints system or modbus/manual lines for the Arduino bases, and we'll need at least two of these for the Signalboxes anyway.

[Ben Lyons]

Hi Rochard

See you are still progressing with signal electrics once you feel happy and Jonny is on board is it worth as many as we can getting together at wotton and having a big cable up, no one watching. As to cables l have palugs so if we get cable we could wire and try.

Ben

[Keith1500]

Ben,

Sounds like a good idea. Cable wise; I can bring along the drum of CAT5 cable.

One observation from this years Shildon is -
we need to think about how we get a cable to cross the tracks bearing in mind the thickness of a sleeper is not very much.

Look forward to doing this big cable up sometime soon.

Keith

[Dave Noble]

Keith,

For Noble Yard, I just ran a router across the underside of three boards (near one end) to give a slot big enough for the cable, in my case about 9mm dia. Seemed to work OK. The turnout also acted as a crossing for the cable by putting a socket in each side.

[Ben Lyons]

Hi Dave,

I it worth doing as you say a few boards with built in cable ie socket each side maybe 4 would do, I guess it does not matter where we cross.

Ben

[Dave Noble]

Ben,

Putting double sockets in is a real pain, they need to be in a plate, and the back of the socket recessed into the board so they don't stick out, for handling and storage. Then the cable needs to be in a groove, and either fixed in , or ideally, covered over, a lot of work. Then there is the problem of the plugs being kicked and damaged in use, it did happen. I'd strongly advise just going for a simple groove.

I think with either system it is important where the cables cross the track. With Megapoints, we're limited to 5 moving parts (arms or points) per cable, and with the Arduino system, if it needs to be linear as Richard suggested (which rather defeats the object of putting 3 sockets in each signal base), placing will still need to be planned. Would the cable run up one side of the track from Middleton towards Shildon, cross and return down the other side, or zig-zag it's way up, having taken a yet to be planned, route around the junction itself?

Richard, Keith,

What were the results of your experiments to determine the maximum length of cable (presumably CAT5) which would work?

[Ben Lyons]

Hi Dave
how about doing some 1 or 2 foot bases with your slot would make it obvious where to look for crossing point.
Whilst on should we consider some sort of easy out base so we can split track at some mid points.
Ben

[Dave Noble]

Ben,

On the bases which I slotted, I just drew a line across (felt tipped pen) where the slot was to identify it. No-one seemed to notice, so it's not too obtrusive. I think they need placing in specific places, rather than placing randomly and then trying to fit the wiring to them.

I did design a splittable base and have the bits, just haven't got round to welding and assembling, but I certainly think we should make some. Time, time, where does it go?

[Richard Phillips]

I'd strongly advise just going for a simple groove.

To throw my two'penneth in, making a groove available in a standard place in a board would offer the prospect of not putting specific boards in specific locations, so would seem to be a good solution, and every connector is a possible failure point.

if it needs to be linear as Richard suggested (which rather defeats the object of putting 3 sockets in each signal base)

Power and manual lines are extended to the third socket, so permitted the teeing off of power/manual control.

Would the cable run up one side of the track from Middleton towards Shildon, cross and return down the other side, or zig-zag it's way up, having taken a yet to be planned, route around the junction itself?

At this stage I imagine a zig-zag if permitted by regular slots in the boards, uses the least cable. Either would work.

Richard, Keith,

What were the results of your experiments to determine the maximum length of cable (presumably CAT5) which would work?

[/quote]

I've only tried the long cable which Keith made up at Ben's which worked, the standard says 4000 feet, depending on cable impedance and speed - and we're not running anything like the high-end speeds. So we're probably okay up to half a mile depending on the quality of the joints. If we could put a connector on the end of a reel of that CAT5 cable Keith we can test a drum length. I don't have anything like 4000 feet of cable to test with.

[Ben Lyons]

Little up date. ladders added to 2 signals along with handrails on all 3. lamps still to fit.
So getting there!

s3.JPG (56 KiB) Viewed 38193 times

s2.JPG (39.18 KiB) Viewed 38193 times

s1.JPG (32.16 KiB) Viewed 38193 times

[Richard Phillips]

servoarms.jpg (20.06 KiB) Viewed 37938 times

Just a note on servo arms when fitting. Attached is a photo of the various servo arms that have come with the servos. From bottom to top we have:
1) Long arm (not used)
2) Arm with large holes (will probably create slack and lost motion depending on diameter pull rod used)
3) Thicker arm with small holes
4) Thin arm (7mm first hole from centre)
5) Cross arm (7mm first hole from centre)
6) Trimmed cross arm
7) Trimmed 7.5mm arm

My second cheap vernier has given up the ghost so I can't measure accurately, but the thinner arm/horns are 7mm from center and the thicker arms are 7.5mm. My understanding is the intention to fit the thicker of the horns with 7.5mm from center. On this basis I think the calculated range movement between lid and top of the base is 110 degrees. I think you'd be okay with the 7mm horn, but the longer horn does need to the shortened and converted to single arm. I've adjusted the program now to have a larger arc of movement. So at some point these boards will need to be quickly re-flashed (takes about 30 seconds when set up).

I think this is the plan anyway if I'm wrong about the servo arms, someone please mention.

[Dave Noble]

Richard, yes, the plan was to use the arm with the hole at 7.5mm. On my drawing I assumed a useable angle of 102 degrees which gives 11.4mm of movement which seems to work fine without catching anything. The two pull wires can then be connected straight through, so that there is plenty of scope for altering the throw at the arm if for some reason that should be necessary.

[Dave Noble]

Oops, can't completely delete an accidentally submitted post apparently.

[Dave Noble]

It's that time of year again, mad panic before Shildon.
Ben suggested I post something on my progress with the latest signal. To make things quicker, I decided to have a go at a colour light signal, after all there's nothing to them, mistake.
My earliest drawings are dated 1974, by which time health and safety had reared it's ugly head, and therefore the signal has a platform and handrails. The latter look very simple, but are quite fiddly to make, I used 3 x 1mm brass strip for most of them. This came from Metalsmith model supplies who do a really good range of such things. 12BA bolts were just a bit big, so I used M1.2 which have a disproportionally small head. The base is the same size as the other signals, but the electrics are far simpler as you can see, though there is inevitably still a lot of wire in there.

The attachment P3110169.JPG is no longer available

P3110169.JPG (65.87 KiB) Viewed 35201 times

The post base is separate from the wiring, this is to try and keep water off the electrics to avoid the problems we had last year at Lincoln. If it is reliable, I intend to spray the inside and the sockets with wax.
I'm still growing the signals themselves, I was going to use two 2 aspect heads and one set of feathers, but our tame signal engineer told me that they wouldn't do it like that on the real railway so I have to fit a position light signal as well. A PL signal is essentially a ground signal, and as it happens I've already made one, but as it is fixed to a base, I'll have to make another, but this time with only two lights in it. The ground signal was made partly for the hell of it, but also partly because it will fit below footboards so can be used between our standard portable tracks without getting in the way.

[Ben Lyons]

Sorry for the late reply as normal really nice work, can we make this into a joint project, will catch up on this at cinderbarrow