Return to Model Railways

Yet more wiring has been done, and it’s nearing completion now. I’ve added in the point control units I got from Heathcote Electronics. I fitted them to the base board using some computer motherboard standoffs I had spare. I did have to very carefully enlarge the holes on the PCBs to fit the screws. I did the same with the CDU. It keeps them off the board, and it makes them a little easier to get on and off of the boards. The standoffs are just screwed into pre-drilled holes on the base board.

I’ve also added in the power droppers for the electrofrogs (the red wires on the left), which meant some VERY careful soldering top-side. ]

Some additional power droppers were added and joined up too, as well as the interconnecting cabling between the boards. Finally, all the cabling for the point motors themselves was added in, and labeled up accordingly.

I knew I needed to get connectivity to the control box I had designed. Originally I’d picked up a little pre-built unit from Maplin, but it just wasn’t big enough. I needed to fit in point control and indication, as well as eventual signal control, so there just wasn’t space on its A5-sized top. I fabricated a new control panel from some 6mm Ply, and some offcuts of the battens I had. Simply screwed together, it does the job. The top of the box was drawn up in Xara, then printed out and stuck on. I drilled out the holes for the buttons and LEDs, but the edges were a bit rough, so I purchased some small LED holders. The switches are fairly well presented anyway so they’re fine (the nut covers the rough edge). Only half of the top edge of the panel is used, the other half will be used for signal control. I probably should have fit the lot into one section, but I’m happy splitting them out.

Connectivity back to the main boards is via two female 25-pin D-sub connectors, and two straight-wired cables. The indicator and point control wires from the point control PCBs end on female 25-pin connectors, so the cables between are just standard male-male.

All told I ended up soldering 49 connections across the lot, plus all the connections within the control panel. Inside is, quite frankly, a mess. The layout didn’t lend itself to any sort of tidy wiring. I did try, but despite my best efforts, it just wasn’t going to happen.

The wiring used within the control box, and from the control PCBs is actually from a leftover section of 25-pair telephone wire. It’s single core copper, so ample to work the LEDs, and I think OK to work the point motors for short bursts.

It has taken me a couple of days working on it to finish all the wiring. First tests, it worked – ish. A few motors were firing opposite to what I wanted, a few LEDs likewise, and some were totally wrong. Fixing them was simple – I just swapped the wires on the PCB! I also found one motor seems to have either burned out, or just doesn’t work. I have spares though, so it’ll be a doddle to swap out.

There’s a few little wiring jobs left to do. The frog polarity feeds need connecting up, and checking they’re the right way round (don’t want any shorts!). I also have a dodgy joint which needs replacing. Other jobs to do include replacing the dud motor, and one seemed to be sticking, but may have freed itself. I could also do with making some sort of base for the control panel, otherwise I’ll only end up knocking out a wire. The LEDs need gluing in properly, as they tend to pop out now and then. Finally, I’ll finish the main track power feed and I might actually get some trains running round the thing!

I’ve been wiring things up today. Most of the power droppers were already in place, so it was a case of making sure they connected to something. I mulled over a few ideas of how to connect them up. I was wary of running loads of cable around as it would be horribly wasteful, and a pain to try and diagnose faults with. I wanted to use some sort of bus, a common method I’ve seen being to use a copper strip – one for positive and negative (or A and B for DCC, as it’s AC), then soldering all the droppers to it. I couldn’t find any such strip, so I went with another plan. I’ve basically joined droppers in twos and threes into a single wire. The joint is simply wound together, then soldered. Finally a heat-shrink is put on top to keep it protected.

These bigger main wires (there’s a few of them) were then soldered up into pairs, which went into screw terminal blocks, which all led up to a main point where the controller will connect. I need to add two other main power feeds which will go out to the addon boards, which will have a similar setup. It’s still slightly more cable than I’d like, but it’s fairly tidy, and it makes sense.

A few cable ties and screws held everything in place, and it’s done. I plugged up the DC controller and ran a loco up and down all the bits I expected to have power and they all work fine, even at lower speeds, so there seems to be good conductivity all through. I might check a multi-meter on it and see if there’s much voltage drop.

I’ve got one more point motor to fit, and then I’ll start running the power leads for the point motors. I’ll run them all to a central location. I have a feeling the labeller will be given plenty of use with all those wires!

I’ve been busy faffing about with a few things lately.

Firstly, I’ve gotten the bulk of the track glued down now. All the code 55 trackwork is down. The Code 80 track round the back is still to be done, but that also requires the copper joiner strip fitting, so I’m in no rush to do that yet!

The bulk of the power feeds are also in. Some are a bit messy, but it’s such a pain soldering to the track that they’re staying as is, and will just be hidden somehow. In most cases they’ll be buried in ballast, so should disappear. As much as possible I’ve but them on the underside of the rails.

I’ve also converted one of my Class 168 DMUs to DCC using a guide from RMWeb. I also added some feeder wires directly from the bogies to the DCC unit, bypassing the body block entirely. I don’t think it’s made any difference to the pickup as it stands, but long-term it might help. I’ve still got a few to convert, including a Class 60 and Class 66 both from Bachman which could be fun!

I might next embark on trying to fit the point motors, and get them wired up, as well as building the control panel. The box I have is a bit too small I think, so I might need to find another one, or make one up out of wood or something.

I’ve run a Class 67 up and down all the tracks, and there’s a few iffy bits, but I think that’s just dirty track as I have managed to get dabs of PVA on the rail tops occasionally, as well as the odd bit of Sharpie ink!

My biggest issue right now is storing stuff, so I could do with picking up some sort of storage pots, rather than the random assortment of boxes I have.

All good progress though, so I’m pleased so far!

I decided against going to the Warley show at the NEC yesterday. I considered paying £8 to park, and then a further £10 to get in a bit steep. Plus the fuel to drive there (I’d guess about £10) then it’s the best part of £30 just to get in the door. The RMWeb forums suggested it was also very busy at the event, which whilst not neccesarily a bad thing, was something I didn’t really feel like doing, so I vetoed the plan and drove over to A&H Models in Brackley.

Fate had conspired against me though. I’d phoned ahead to make sure they were open (having wasted a trip over before to find they were shut). I drove over, and found they didn’t hactually have everything I needed – they’d taken a load of stock to Warley!

DAMN!!!

I managed to buy two points, and some set track, and slinked home with my purchases. I took a scenic route, which added about a million miles to my journey, so probably spent almost as much on fuel. But never mind! I later ordered the other bits I needed online, so they should arrive in the week.

Today I got into some more track laying, and cabling up. The extension board had no power, so I soldered wires onto the copper strips that form the join. I then joined them up to some Molex connectors, to allow for easy removal. I did a similar thing at the back in the fiddle-yard area. Unfortunately, I wasn’t thinking, and wired these back to front. I’d been using Black wire for the furthest rails, and Red for the closest. I did the same in the fiddle yard forgetting that it flips seeing as the track forms a circle. Rather than re-soldering, I just swapped them round in the terminal blocks, and all was well.

I finished off with laying out the other sections of track as much as I could, and pinning things into place.

I ran a few tests with my two class 67 locos. I ran the full W&S train around the end curve, with the EWS 67 going the other way with a couple of wagons. Thankfully, neither struck eachother. One issue cropped up with the join between Code 80 and Code 55 track causing the W&S train to occasionally de-rail. I beleive that can be filed down to make up the difference. Also, the fiddle-yard track seems to be dirty as it has gained a few dead spots.

One thing I have decided on, is that the rear fiddle yards will be partially obscured by a low-profile multi-storey car-park, modelled on the one in town. Along with the depot building, that’s two buildings I now need to scratch build. I just need to learn how to scratch build now!

I’ve been having an issue with the double slip on my layout. I’d wired up the Seep switching motors, and followed the leaflet that came with the point, but I had this strange issue with a short. Every time the loco I had got over the points, it’d short out. I’d hear it spark, and the controller would show an overload protection kick in.

I re-wired bits countless times, and followed along with my multimeter but couldn’t fathom it. Then I suddenly noticed what was up:

Can’t see it? No, neither could I. Took me a whole week to suss it. Basically, the curved point at the bottom is not set to turn out; i.e. the top blade is touching the top rail, making both the rails leading into it the same polarity! Because there’s insulating joiners between it and the double slip it’s all fine until the loco runs over onto the curved point. As soon as two pickup wheels are on it, there’s a short.

So once I spotted this, I flicked the point over and ‘lo and behold, it magically works! Well, the loco will run until the wheels are all on the curved point, at which point it stops dead, as there’s no power that end yet. I’ve been slowly working along, ‘debugging’ and fixing before moving onto the next bit.

I just need to add more power feeds, then wire up the point motors. Then wire in the CDU. And make the point control panel. But then it’ll be operational, and it’s on to gumming up things with glue and ballast.

I’ve been doing a little bit of wiring today. Nothing amazing, just getting some power into the rails. I’ve been around with insulating joiners, and I think I’ve got them all into the right places now. I’ve also been following up with my continuity tester, basically checking places where I think there should or shouldn’t be continuity. I found a couple of places where the opposite rail was carrying the same charge due to the electrofrog on the previous point being switched. I think I’ve got them all now though.

After working it out, I realised I needed 20 power connections! I’ve found the easiest way to accomplish this is to bare the end of a wire, and bend it to 90 degrees. Once tinned, it can be attached directly to the underside of the end of a rail, OR a metal joiner. A carefully drilled hole below it takes the wire away. The theory being that once ballasted, it’ll be unnoticeable. I did attempt to fix one to the side of the rail (a technique I know is perfectly valid) but in N gauge it’s just not do-able. In OO perhaps, but N is far too fiddly.

So far, I’ve got power into a siding. Woo! My carefully drilled hole wasn’t so careful though, and as a result the track doesn’t fit properly. I’m not worried though, as I’ll just re-drill the holes.

I’ve managed to knock quite a few random holes into the board, and I’ve decided I’ll get around those just using some sort of filler. There’s a couple of bigger holes which a piece of card would be able to bridge. I’m intending on adding a little bit of undulation to the whole base, so I’ll probably do that with a mix of filler and other fun things.

But there’s still another 18 power connections to do yet…

So, the weather has changed, and naturally, so have my plans.

I got rather fed up with the scissor crossings. The wiring was ‘fun’, but no impossible, but that wasn’t the deal breaker. The real issue was that I couldn’t get the point motors to align in a sane way. They either clashed with eachother, or with the board framework. So, I decided enough was enough, and they have now been replaced with ordinary points. The motors for those fit in fine, and the wiring issue is gone; now I just need insulating joiners to work with the electrofrogs.

The joiners ended up costing me a lot more than I planned. It seems Dapol have had a class 67 model coming for a while, and whilst looking around on the web for the joiners, I found the pre-orders for these fine new models. So I’ve got a 67 in EWS livery, and, the one I really wanted, a 67 with MK3 coaches and DVT in Wrexham and Shropshire livery! So that’s pleased me no end. Hopefully it’ll fit in the platform.

Progress overall has slowed a little, but now my issue with the scissors is gone, I can plod on far more easily. I need to solder power connectors onto some more point motors, and re-lay some of the track, but I’m not a million miles from wiring up the power. Next it’ll be onto ballast, and we’ll actually be getting somewhere.

Now there’s a slightly more definite track plan, I’ve started work on point motors. I read up on some stuff, including this nice article on the N Gauge Society website. I pondered things a bit, and here’s the decision I came to.

Point motors

Going with Seep, as they can mostly hide under the baseboard, they’re very low profile, and they’re simple. Peco’s motors require hacking great big holes in the board, whilst the Seep motors need a tiny hole for the armature.

Wiring

Colour coded, naturally. I found some stuff in Maplin which looks like it’ll do the job.

Power

I’ve got an old controller in the loft (I think it’s a Gaugemaster, or something) which has an accessory supply, so I’ll bring that down and use that for now. Or at least, until I look into DCC. I’ve also got a Gaugemaster Capacitor Discharge Unit (CDU) which will help fire the motors. For the scissor crossings I’m planning on making up a couple of push-button diode matrices, with relays switching the electrofrogs, but I’ll do that after I’ve got the basic points working.

Connections

I pondered this for a while. The above article mentioned using terminal blocks, which is a nice idea, but I’m not a great fan of them. I pondered some more, then realised I already knew of a perfectly great connector. The Molex Hard Disk connector! They can take the amperage, and the wattage I need, so they’re quite suitable. Plus, pretty easy to fit (crimp on pins), and it all just pushes together. If I want to remove something, I just pull the connector out. For bigger connections I can use the motherboard connectors, or multiples of these.

I’ve made up eight point motors so far with the female Molex connector on (technically, it’s back to front to how it goes in a PC, but I’m not fussy), and they look pretty good.

The screws I bought to fix them on are just the right length – they’re a bit long actually, but with the thickness of the point motor’s PCB, it should almost fit without sticking out the stop. So what if it does though, it’ll all be hidden under scenery!

I might have a bash at screwing one of them in and testing it out. Once I’m happy with all the point motors I’ll move onto the other electrics.