Treasure Hunt at Amberley

Last weekend I was at Amberley for their annual railway gala, helping out with the 009 Society stand in the model railway exhibition. Making up 009 wagon kits and chatting to punters wasn't very demanding so it left time to have a good look round. The format was much the same as previous years, some good NG layouts (though little I hadn't seen before) but the real action is outside with lots of trains and locos running. Visiting loco this year was Jack, an Andrew Barclay originally built for the Edinburgh Gas Works, and looking superb.

However this year I was on a mission - I knew that three of the Thakeham Tiles locos ended up there, and it would be great to see if I could find them, so camera in hand I went for a wander. First up is No 4, which was built by the staff at Thakeham, clearly no expense spared!

Nearby was No 5, also built at Thakeham. This time a little more thought had gone into the design and it had a (removable) cab, it almost looks like a loco! Almost, like No 4 it has more in common with a lawnmower than a railway loco.

The final loco was more elusive, but a quiet word with someone in the know pointed me in the right direction (probably a direction I wasn't supposed to be in), to where I found Hudson Hunslet 3653 of 1946. Now this loco was significantly modified to fit under the Thakeham hoppers, so first here's a picture of a similar Hudson Hunslet looking as the factory intended:

So finally here is the Thakeham Hudson Hunslet. The chassis and radiator are recognisable but the driving position has been dropped behind the chassis, with the driving controls modified and extended for sit-down operation. Note that like No 5 there is no coupling on the "rear" of the loco - the cab would not be strong enough - but the operation at Thakeham only required coupling at one end.

To complete my model of Thakeham Tiles a model of at least one of these locos would be great, though none would be easy. The enclosed No 5 would probably be the easiest to motorise, and although there is a good kit of a (standard) Hudson Hunslet from Nonneminstre Models that could be modified, the Tenshodo motor bogie which it is designed for is unsuitable for 14mm gauge, and the open sided prototype makes motorising it a real headache. In any case I will be struggling to get the layout finished for EXPO as it is so building another loco is out of the question!

Still an enjoyable and productive day at Amberley and perhaps some projects for the future!

Track and Wires

It has taken over three weeks, but finally I have finished the trackwork. Well, almost - I should add the nice plastic fishplates KB Scale supply with the sleepers! Also the section over the bridge is not yet fixed as I need to complete the bridge and walling first.

I'm pleased with how it looks, although at about a week per foot of track it has been pretty tedious! At the ends the track is hidden inside structures, a few copperclad sleepers add strength and I have ensured there is no chance of a train leaving the board! Track feed wires are visible here too, but every rail section in the scenic area has a feed wire soldered to the bottom of the rail.

I have also fitted a control panel using the method I've applied to a few layouts now. The panel is an offcut of 2mm Aluminium from the guillotine waste bin at work, onto this is stuck a print-out of the schematic (Word is OK for drawing this), then a sheet of clear plastic (Wills kits bubble packaging in this case!). The toggle switches bolt through and hold the layers together, then the whole lot is screwed in place behind the facia.

The position of the panel allowed me to incorporate the manual point actuators - these are the two dowels visible in the panel. Behind the scenes these slide through a block of wood, and are screwed to the plastic lever of slide switches. The point operating wire is also attached to the slide switch, then goes to the point via a tube as seen in the last post. The slide switch is then used for feeding the vee of the point and provides the latching mechanism. I'm rather chuffed with this idea - operating the wooden dowels in the panel is nicer than operating the slide switches directly, and they have a nice "clunk" action.

Also visible in the photo above is the main terminal block for the wiring. I used to use tag strips, and would prefer to but these terminal blocks are much cheaper! As well as joining wires together they act as a point to route and test wires.

A wider view of the under-board wiring. The power comes in from the controller via a socket bottom right, the panel is centre right. Wiring is colour coded for ease of understanding, and despite being a simple layout the need to feed each rail length and switch point frogs makes it surprisingly complex:

- Red is main track power (controller fed direct)

- Blue is the common return to the controller

- Black is for switched point-frogs and any rails joining to them

- Yellow is for the three isolating sections. 

And yes, it did all work first time! Sketching it out in advance, and drawing the layout and labelling feed wires on the underside of the layout helped. Incidentally the yellow wires are the only ones a DCC user could skip, along with the three switches. But then you'd have more wiring in each loco. Who says DCC is two wires?

A pair of points

I'm finally nearing completion of the track-laying, so here is the story of the most tricky part - the points of course! I'm very glad I only have two to build for this layout, I'm sure I'd get better with practice but point building is very tedious to me, especially using the spiked construction of the KB Scale track representing the sectional industrial track.

 

The rails were cut and shaped onto plans of the points. Without soldered construction and with so few sleepers the crossing vee (frog) area of the point would be very weak. I decided to assemble this onto a small, thin sheet of brass, justified as these type of industrial points often had steel plate under the crossing. Track spikes help with the assembly here.

 

The point blades have wire droppers soldered to both the hinge end and the tie-bar end, these will form the pivot and the attachment to the tie bar. The "proper" construction with tiny brass tabs under the blades riveted to tie bars is well beyond me, so I have a cheat method. Once ready, and not forgetting to add feed wire droppers, the rails were then transferred to the plastic sleepers on the layout.

 

The rails were carefully spiked in place checking the gauge. Slots had been cut under the tie bar for the wire droppers, and holes through the sleeper and foam-board at the hinge point, so the blades could be dropped in place. The tie-bars are made from 40" plastic (they will be cut down) through which the droppers pass, although these are largely cosmetic they do help keep the blade spacing. At present I have yet to fit the check rails.

 

The real action is underneath. The real tie bar is a strip of PCB (sliding on a piece of black plastic), through which the droppers pass, then small pieces of PCB are soldered to each dropper to act as "nuts". The droppers are therefore free to rotate relative to the tie bar as it moves sideways. The same method of PCB soldered to the wire is used to hold the pivot wire in place too. Note that the blades are electrically isolated from each other and a wire feed will be attached to the bottom of the hinge wire.

 

The points are operated by wire-in-tube, a steel wire through the tie-bar passes through a plastic (PTFE?) tube which is held in place by lots of hot-glue and pieces of foam board so it cannot move sideways. A Z-bend in the wire by the tie bar acts as a spring for any slack. The other end is attached to a slide-switch, which not only will switch the electrics for the point vee, but also acts as the latch to hold the point over. Actuation is by dowels to the side of the layout, attached to the slide switches with a screw, more on this when I get to the wiring.

This isn't precision engineering but so far the points seem to work, which is the point ... if you excuse the pun. That said, they've not been tested by a loco yet! Time to finish the last bits of track and connect up the wires.

Integrated Lighting? Not Quite...

While I'm working on the points I thought I'd update you on how the layout will be lit. Rather than go for a full lighting rig for such a small layout I figured I'd use the same angle-poise type work lamp I used for Landswood Park. This relies on a G-clamp fixing, so rather than rely on it fitting to the table (table designs vary!) and to ensure it is high enough, while building the baseboard I incorporated a lighting attachment fixture.

 

By which I mean a block of wood! It fits into the recess behind the curved back-scene at one end, and as you can see this allows the lamp to sit at a good height over the layout providing an even light.

 

Sadly IKEA don't seem to sell this design of lamp any more, although this model appears quite similar, relying on an LED bulb instead of a fluorescent tube.

Spikes

First here's an explanation for the limited progress for the last couple of weeks! The view from our Hotel room in Crete. It's rather bright because there was a funny yellow ball in the sky...

 

Since I've been back I've been concentrating on building the track for the layout - after all it won't be much of a model railway without rails! The track at Thakeham was pre-fabricated industrial track with steel sleepers, essentially 2' gauge train set track. The KB Scale Hudson track sleepers are ideal for this, albeit with a different profile but good enough for me. However looking at the photos it seems the track was laid on top of wooden sleepers, so I am using wooden coffee stirrers underneath the plastic "steel" sleepers.

 

The approach is then to spike the rails (Peco IL-115 code 82 flat-bottom) to the sleepers using KB scale spikes. Although the sleepers have the holes pre-drilled, the coffee stirrers don't, so a drill in a pin vice was used, tedious but easy enough so far. Then I discovered the spikes didn't fit! The head   is too deep, meaning it didn't clear the head of the rail, and if it did it pushes into the web moving the rail out of gauge, and making the second spike impossible to fit. They fitted if turned about 80 degrees, but then they did't grip the rail well either... After some experimentation the answer was the big pair of side cutters seen above!

 

This close-up shows an unmodified track spike (left) with one that has had the head cut down (right), only about 1mm needs removing but it has made fitting them much easier, and the track stays in gauge. Also visible are the pencil marks used to locate the plastic sleepers as the wooden sleepers obscure the printed template. With plain track now sorted, and I have to say looking quite good from this close, it is time to move on to the points!

Scary Chassis Stuff

The scary part of the Hudson Hunslet kit for me is building the chassis. For the most part I've used commercial chassis under my locos in 009, and the two previous 014 locos I have built have had pre-assembled chassis blocks requiring very little work. The chassis in this kit though comes... well, flat packed!

The etched Nickel Silver parts assembled reasonably easily, and soldered together as square as I could (using the inside of a square). The sand-pipes were somewhat fiddly, and I think they are a bit short, but eventually I got them all soldered in place. So far so good!

 

At this stage I masked the bearings and applied a little primer to any parts that might be visible later.  The instructions suggest the brake-gear could be made removable, which I thought was a good idea as it makes painting easier, and avoids soldering to a completed chassis. Soldering the nice brake-shoe castings to pieces of wire was fun.

 

The next stage was assembling the wheels and gears. The idler gears are held in place by small brass collars which need super-gluing to the shafts, as the instructions say "do not allow the glue to wick along to the gear"! After a nerve-racking few moments, somehow they all still rotate. The axles need pushing through the wheels and drive gear while being passed through the chassis frames - yet another tricky job! I used small pieces of lead as packing while tapping the axle through. I have to admit the wheels are not entirely perpendicular to the axles though.

Having achieved a rolling chassis it was time to attach the pick-ups. As with the O&K I decided to substitute the wire provided for phospher bronze-strip, this meant attaching the small pieces of PCB  to the chassis side rather than on top of the sand pipe bracket as instructed, but that was easy enough. Finally the motor is attached and wired to the pick-ups.

 

To my amazement the chassis not only works, but runs smoothly and sweetly! The gears will get a drop of oil in due course, and of course the body weight will help, but even without that performance is good. The photo below shows the brake gear loosely fitted, after painting a drop of epoxy inside the frames will secure them.

 

So, really scary bit completed, and I think I can call it a success!

Starting to set the scene

Alongside finishing the baseboard and building the bridge I have been experimenting further with the block walling. The Scalescenes printed paper sheets looked good, and will save a lot of time in painting. I printed them out on the laser printer at work - this gave a slightly darker and more defined finish, and is resistant to water drops (bubblejet ink smudged and ran when wet). I also understand laser prints are more resistant to fading.

Once the paper is applied to the card or foam-core model I have added a little relief by scribing with a blunt metal object - actually one point of my tweezers. It is easy to follow the printed courses and although the relief is barely visible, it does make a difference, especially when viewed at an angle. The depth of mortar courses is very small in reality so I think this is enough.

 

These photos show the embossed finish, that has also had some colour variation added using brown and grey weathering powders. Again this helps bring the prints to life, and allows dirt or rust staining. I don't have a green powder but I need to find a way to represent the moss on the real walls under the bridge. Finally a coat of Testor's Dullcote matt varnish should protect the paper and weathering powder, and ensures a nice matt finish. The piece of stone wall is polystyrene pizza base embossed with a pencil. So far I am not happy with the colour on this, or the render on the sloping buttresses.

 

You will also see the bridge has been painted. It was given a coat of red primer, then painted with brown rust colours. Dabs and smears of PVA glue where applied, and once dry, it was painted green. The glue patches were scraped off, and soaked off (leaving the bridge in water) to reveal lots of rust, final touching up and some weathering powders finish it. The result is not as rusty as in the photos of the real bridge, but the model is set 20+ years ago when the bridge was much newer!

 

Loosely put in place the bridge scene is coming together. I have bought a Gaugemaster sky backscene, but I'm not sure how best to fix it - any ideas?

Finally I've been looking for a fork-lift truck or other suitable vehicle to add to the scene. I haven't had much success so far, but I did find this in my son's toybox*. It is missing the seat and (obviously) the forks, but it might have potential with a bit of work. However it is a little over-scale, and I think much too modern for the late 70's/early 80's era of the model.

 

*He has given me permission to use it!