More Chudleigh Road

For those of you in need of more rambling about my layout, there is a thread on the RMWeb forums about Chudleigh Road.  I'm trying not to exactly duplicate the content between blog and thread, but it turns out that I'm not terribly good at that.

Guess what? Soldering!

I want to have the control panel finished before I start any scenic work on the layout.  I'd rather not have to keep turning it upside-down with too much built on top of it.  This means that I have to crack on with the control panel.

The next stage is to solder up all the switches and LEDs.  The LEDs I bought came in mounting assemblies but without ballast resistors.  Each LED had a resistor soldered on, followed by wires long enough to reach to the chocolate blocks from the 25-pin connectors.  The switches just had wires soldered on.  By the time I'd got through this my trusty wire strippers had collapsed and I had to nip round the corner to Maplins to buy a more robust pair.

All in all, this took several weekends.  I hope it works.

The control panel takes shape

After taking a lot of care while building the baseboard, I got a little overconfident.  The control panel is a story of measure once, cut several times, then squeeze everything in as best as possible.  I planned the mimic diagram to be A4 in size, which I had imagined would give me a lot of room to fit things in.  After I'd bought and cut a sheet of perspex it became apparent that this was a little optimistic.

In the end, I managed to squeeze things in.  The main constraint was that the switches and LED mounts I had bought needed 6mm holes in the panel; this meant that I couldn't shrink things too much without them getting crowded together.

The idea was that the top of the panel would be made from a sandwich of 3mm perspex, a paper printout of the mimic diagram, and 3mm plywood.  Looking the LED and switch mounts, I discovered that they would fit through something so thick and still have enough thread visible to attach the mounting nut.  Replacing the plywood with 1mm plasticard solved this problem.  The next step was to ensure that I could actually make 6mm diameter holes through the sandwich.  Some internet research suggested that a good way to drill through perspex without cracking or melting it is to use tile and glass drill bits.  Off I went to  Amazon to buy a set of bits .  I used some off-cuts to test the process.

Success! Now to drill 54 holes in perspex without cracking it.

Remarkably this worked as well.  By the end the bit seemed to be getting much less effective.  Either these bits are of poor quality or perspex takes its toll on them quickly.

I made up a frame from some spare 2"x0.5"ish timber that I had knocking around from an old bookcase and used this to mount the diagram sandwich.  My hope was that this would give enough space underneath to accommodate all the wiring.

More soldering

The control panel with mimic diagram and switches for controlling the point motors and track feeds will have to be separate from the layout, otherwise it won't fit under my son's bed.  Here the large number of points makes for doubly hard work - the panel is going to have connections going to operate the point motors as well as connections coming back from them to feed LEDs.  As a result, I need three 25 pin connectors on the panel and baseboard to connect the two.  This doesn't include the connections from the Gaugemaster twin track controller or the power feeding the track, for which I am going to use some 6-pin DIN connectors.

All this amounts to yet more soldering, which has kept me busy over several weekends.

The 25-pin D-Sub connectors will be fitted into the side of the control panel, and the wires from them will go into chocolate block connectors.  The hope is that if something goes wrong this will make it easier to diagnose and fix the problem.

Why do I have so many points?

So I now have 17 SEEP point motors with dropper wires soldered onto them.  The next task is to fix them to the baseboard.  This turns out to be harder than I imagined.

The problem with using SEEP motors with N-gauge points is that the points have a very small throw as they switch.  Not a problem if you're just using them as a motor but if you're using the inbuilt switch to change track feed polarity or operate indicator LEDs you can be in for a rough time.  The switching operates by a sprung metal disc moving along the surface of the PCB making and breaking contact between different tracks on the circuit board.  Of necessity there is a small dead area in the centre.  For the switch to operate reliably the motor has to be positioned with pinpoint accuracy otherwise the disc will be in the dead area at one end of the throw or another.

One often suggested method to ensure accuracy is to make a small jig out of cardboard or a carefully cut and filed clothes peg to ensure that the motor is at the centre of its throw.  Hold the point in the centre of its throw with a little Blu-Tack, fit the point motor pin through the point and screw the motor into place.

This didn't work too well for me.

The method I came up with instead involved creating a portable switch/LED circuit. I attached this to the point motor which I then held in place under the baseboard as I carefully adjusted its position to ensure that the point changed properly and the switch operated reliably.  I marked this position on the baseboard and screwed the motor in place, checking again to ensure that the switch still operated.  For some motors, the act of screwing it down changed the operation of the switch.  This turned out to be a very fiddly process.  Nonetheless all point motors apart from those in the as yet unlaid siding area are now fitted to the baseboard.

I'm hoping the switches will be working by the time I've constructed the control panel.

Soldering Sunday

Having Seep point motors on all 17 points (barring the catch point leaving the sidings) seemed like such a good idea.  I've just spent a large part of Sunday afternoon trimming 5cm lengths of wire, stripping the ends and tinning them with solder.  For 16 point motors that's 96 pieces of wire.

A packet of glass & ceramic tile drill bits arrived during the week.  I hoped to use these to drill holes in the acrylic that will form the surface of the control panel.  Having just tested them on an off-cut of 3mm perspex sheet I can confirm that they work well, certainly much better than the normal HSS bits that cracked the acrylic without fail.

Of course, now that I know that the control panel construction to go ahead, I will need to solder dropper wires to the 17 point switches and solder resistors and dropper wires to the 34 indicator LEDs.

When is a semicircle not a semicircle?

If it all works as intended, four long curved pieces of Peco code 80 Setrack should form a semicircle.  The pieces that I have don't.

Tightly assembled, the ends of the curves don't line up.  The curve isn't quite of the right radius and when you do pull it to radius, the ends aren't quite parallel to the diameter of the circle.  This has made track laying more of a trial than it should be with Setrack.