Scratch Building a Modern Bracket Signal - by Tony Sissons

Reproduced with Permission - Thanks to MRJ and Wild Swan

After building the gantry for Charlotte Road (See MRJ 161 - page 231).  I was asked if I would build a full complement of signals for the Widnes Vine Yard layout.  This consisted of seven single post signals of which four of them were fitted with functional subs and stencil indicators as well as multi aspect colour light heads.  The bracket with its two x 3 aspect colour light heads also carried two subs and best of all, two route indicators each mounted to one of the colour light aspect heads.  Looking at the requirements and specifications of the assignment I accepted, I learned that the bracket signal project was based on designs currently found at Chester Station which were installed during the 1980's with the majority of them commissioned during 1984. The signals I built had some changes and additions to suit the operation and track layout of Widnes Vine Yard, for example, the feather route indicators, which do not exist on the equipment at Chester station.  That fact alone immediately led me to accept the assignment, route indicators, what a challenge.  I felt I probably could do a better job from what I learned making the single posts and gantry mounted signals for Charlotte Road.  I knew I’d be able to get a lot more detail photos from a variety of different angles as well, even measure some dimensions of the actual equipment at Chester.  What did come as a surprise was what I thought would be a reasonably easy bracket signal to make, comparing it with the model of Faversham gantry which I made for Charlotte Road, turned out to be a far more difficult project.

 I have focused this article on the bracket but with reference from time to time to the single post signals where I used similar or identical construction techniques for common assemblies, such as the aspect heads and the functional subs being common features fitted to single post and bracket signals. 

 Looking at the basic design, a simple right angle, two vertical legs and an enclosed walkway seemed so easy to me.  But scratchbuilding is a little more complicated than that, you have to do the design work yourself and make each part, only then does it become similar to a purchased etched brass kit.  The more I looked the more I saw my problem, how do I get all these wires from the signal lenses down through to the underside of the baseboard so that they can’t be seen.  Unlike many r-t-r models every single one of these lights had its own LED, no single LED shining up through a piece of clear plastic here, that’s a cheat.  But I did cheat a small amount, well, not so much cheat but I had help, from a member of WFRM, Mike Turner, who showed his skill as he cut out the main feather shapes, drilled the holes for the LED’s, mounted them, soldered a bunch of wires to the back of them and mailed them to me in the US.  It’s the first time I have ever shared a scratchbuild and it worked a charm.  More later on this but to start.

Legs & Main Structure:

I started the bracket making the horizontal frame first.  I cut two lengths of 3/16 x 1/8 channel section and filed the required taper along their length.  Next I made the large angle that connects all the main load bearing components together.  It is represented as the bolted plate at the ladder end of the bracket and this was fashioned from 0.010" thick brass.  Then soldered these two beams to an end piece and an angle to make a stable rectangle from which I could work with.  Fairly straight forward and not much to tax the brain but now I had to work out a method that would permit the passage of twenty two wires (11 per signal) through the main structure.  It meant that the vertical support legs had to be hollow for a start.  Reviewing my ACAD drawing of the prototype and the dimensions scaled to 4mm to the foot, meant that I needed a rectangular hollow tube to the dimension of 0.131" x 0.157" which I knew immediately I would have to make from scratch.  Nothing especially difficult or technical here really but I was forced to use a length of brass channel .157" X .125" and fill the open sides in by soldering a strip of brass sheet 0.010" thick x 0.157" wide making a complete tubular box to a dimension of 0.135" x 0.157".  I then removed 0.004" off the 0.135" dimension so that my final rectangular dimension matched the prototype at scale.  Once these were cut and cleaned up to the final scale length of 2.558" they didn’t look half bad.  I was then able to solder these legs into position on the inside of the end angle that was previously soldered to the horizontal beams ensuring that each leg was at the correct angle and position.  The final piece to this part of the construction and to ensure strength in the model I soldered a chequered plate rectangular section to the bottom of the legs which made the legs assembly very strong. 

The next step was to cut and fit the walkway platform which I fabricated from a piece of chequered plate.  It does not run the full length of the horizontal signal beam but stops short close to where the ladder is situated where it mates up against the end angle plate.  Two rectangles are cut out in the walkway plate to permit the sub bases of the signal heads to be fitted before final positioning onto the beams.  Then either side of each aspect head a kicker plate is fitted made from a small length of brass angle, four in all.    Similarly, two bases must be made and located to support the position light subsidiary signals and these are fitted above the level of the chequered plate walkway.  At a later stage when all wires have been passed along their respective paths, to complete the assembly, the underside of the two horizontal beams are plated over and here I use 0.010" styrene sheet.  My reason is that if ever I need to access the wires that run underneath the beams, which become captured inside once plated over, it is not a big deal to remove and replace styrene than four plates made from brass and soldered to the underside of the beams.

3 Aspect Signal Heads: 

The unit is made up of 6 components, all of which have to be made before soldering the parts in a specific sequence.  These are the main light enclosure, target board, shrouds, base, top cover, rear inspection door.  Plus fitting and wiring the 3 LED’s in place before the unit can be fitted to either a bracket, gantry structure or to a single post. 

I always start by cutting the main light enclosure to length from 3/16 x 1/8 brass channel followed by cutting out the target board from 0.010" brass sheet.  I then mark out three holes in the channel, drill through with a # 76 drill and then solder the target board too the channel.  This permits me to continue the hole drilling from the rear of the channel through the target board preventing any hole misalignment between these two matching components.  Then I solder the three shrouds in place onto the face of the target board.  Next comes the base with the bolts previously affixed in their respective locations.  After all of this soldering I do a big excess solder clean up, filing and fiber brushing until I am satisfied that all is clean.  At this stage I paint the inside of the shrouds black and when the paint is dry I superglue into place the red, yellow and green LED’s.  Wiring is straight forward, an earth/common wire is soldered to all three LED anodes leaving the cathode on each for the supply power.  A photo showing the LED wires soldered is on page 235 of MRJ 161.  However that lamp enclosure is of a different design than those for the bracket signal described here.  The top cover and rear door I make from 0.015" thick styrene sheet to eliminate any potential short circuit problems and at this stage only glue into place the top cover into the channel.  The rear door is left off until very much later so that access can be maintained for any LED problems later when all the wires have been passed through the complete assembly and the units are tested for short circuits and correct wire orientation.  Thereafter the rear door is glued into place with the merest amount of adhesive so that it can be prized off later if service is needed to replace a dud LED later in the models life. 

Subs: Position Light Subsidiary Signal

These small heads I constructed completely from styrene by first making the front and rear of the unit from a 0.015" thick sheet.  It is an odd shaped piece and I made a brass template to use to cut the shape with repeated accuracy.  The shape of the front and rear plates are identical the one difference being the front plate has the holes for the lights and the rear a cut out for the LED.  I start with the front plate and drill the two 0.046" dia holes for the lenses.  For the lens aperture that is usually covered by a blank on the prototype I make a similar circular blank at this stage and glue it on.  The back plate is the same shape but in this I cut out a rectangular aperture for the LED.   The Main housing is a whole different exercise.  I start by cutting several styrene blocks from 0.080" square styrene stock.  I glue small pieces around the perimeter of the plate outline and make a cavity for the LED and leave it all to dry for 24 hours.  The images show this progression.  The next stage is to file the overlapping styrene down to the profile of the rear plate which effectively forms the enclosure casing wall.  I then drill #80 holes through the body of the enclosure in each side and one through the top on centre.  Into these holes I superglue a short length of 0.012" dia wire, snip off and file the end square to represent the conduit fittings. The final work on the build of this rear plate is to cut out a rectangle so that the LED will fit snugly into it. 

The lens shrouds are1/16 dia styrene tube.  The length of it as well as the angle of the sloped underside cut away is dependent on the location for the post.  Consideration must be taken into account of the drivers angle of view on approach and the backlighting according to the time of day and season.  These factors will then determine the height of each piece of equipment fitted to the post and the angle each piece is orientated.  The prototype permits a plus or minus lamp angle of view, or tilt, of 5  degrees in the ‘Z’ plane.  In simple terms its all about adjustment.  How much a lens enclosure can be tilted forwards or backwards.  How many degrees a shroud can be rotated about their axis and also the length of the shroud, all of which to achieve  maximum brightness intensity from the lens within its enclosure directed toward the driver of an oncoming train.  Think of the shroud as a baseball cap with its peak preventing the sun shining into the lens therefore blocking the sun to affect the signal lens brightness.  All of these considerations contribute to the correct design of your signal so as to present to a driver in an oncoming train the maximum field of view, longest sight line and brightest light intensity possible.  This applies to all lights fitted to the signal.

It is not paramount that the shrouds have to be of a certain length, they could easily be longer by twice the length, the important thing is that you make them dimensionally the same.  I felt that longer shrouds made my model look unbalanced so opted for the shorter design.  Once made I glue these onto the front face using a small length of brass of the same inside dimension, a shroud on a stick idea, so as to align and rotate them concentric about their axis to the parent holes in the front plate. 

Before moving on to make lenses I paint the inside of the shrouds black.  The inside of the assembly needs to be white to help reflect as much white light as possible through the lenses when fitted.

At the time I made my own lenses and for this I used as many single 0.015" dia fiberoptic strands as I could stuff into two 4 inch lengths of 1/16th I.D. brass tube.  Once accomplished I pull the tubes apart approximately 1.1/2"  which leaves a clump of fiberoptic strands ready to be bathed in superglue.  Dosed with a liberal quantity and ensuring that none of the adhesive contaminates the brass tubes I leave it to harden completely over a period of 24 hours prior to touching it again.    Thereafter I pull both the brass tubes away and I’m left with a bunch of fiberoptic strands nicely glued together with each end all frayed.    I prefer this home made option so that I have control over the O.D. of my lenses.  I cut the lenses like slicing bread with a scalpel and cut off the loose ends until I reach the glued solid section.  I now carve off a length of this glued stick of fiber optics for my lens and glue it into the lens hole in the front face.  These lenses must be long enough to place in the hole but not too long that they make contact with the LED that will be fitted later, nor that they poke out of the shroud.  I try to cut the lenses as short as I possibly can, say no longer than 0.040".

Then I trial assemble the front and rear plate using finger and thumb to keep the two pieces together checking that I can insert the LED into the rear enclosure and there exists no interference anywhere.  Once confirmed that I can, I glue the front and rear assemblies together to make one component.  From this point I solder the two wires to the LED and affix it into the rear aperture of the enclosure that I had previously filed.

Route Indicators: Position Light Junction Indicator

The route indicators or PLJI per their technical nomenclature, were not all my own work.  Mike Turner, another member of the Wirral Finescale Railway Modellers,  made the target board shapes for me from a slice of double sided copper clad PC board, drilled the holes for the individual LED’s, fitted them and then soldered the required wires to each.  He then mailed these naked pieces over to me in the US.  I took them from there.  However, to make this text complete I have included three paragraphs which I asked Mike to type up for this article, they follow: 

First job was to make a CAD drawing of the PLJI’s based on information kindly supplied by the prototype manufacturer. This gave the basic outline of the backboard plus centres for the lamp units which could then be overlaid onto the AutoCAD drawing produced by Tony and adjusted as necessary to take account of the materials we were using etc. The fact that Tony and I live 4000 miles apart is not a problem with modern technology!  Once satisfied, the drawing was printed out on photo quality inkjet paper using the ‘superfine’ setting on my printer. This allows the rendition of nice crisp fine lines with the absolute minimum amount of bleed.

My idea was to use 0.3mm thick double-sided PCB for the backboards so that I could solder the LED’s on one face and Tony could solder the shrouds on the other.  We soon realised that the heat transfer was such that this wasn’t possible and the hoods ended up being superglued. However, getting back to how they were made…. The printouts were stuck onto the PCB using Pritt adhesive and the light unit centres marked through taking great care to keep them all in line. This was aided by the old draughtsman’s trick of moving a straight edge up against the scriber whilst still in the first mark and using this as a guide for the rest.  Once this was complete the outline was marked through using a scalpel, the paper removed and the glue cleaned off with lighter fuel. After checking all was well under an eyeglass the holes for the LED’s were progressively drilled out to size and the backboards cut and filed to shape.

Thoughts now turned to how I was going to get them to work. Each arm of a PLJI shares a common pivot LED so this had to have it’s own supply. The remaining 4 LED’s on each arm were wired in parallel so another supply was required for these.  Lastly a common return was required for the whole shebang. The copper on the rear face of the backboard was masked out using an etch-resist pen so that the LED’s would be interconnected as required and the surplus etched away using ferric chloride. After this the LED’s were pushed through the holes from the rear and soldered in place with the aid of a magnifier – well they are only 0.8x1.6mm.  Five enameled wires were required for each PLJI and after fitting, the units were tested and cleaned up for shipping across the pond to Tony.

Sub denotes Position Light Subsidiary Signal

PLJI denotes Position Light Junction Indicator  

Then it was back to me.  My task was to complete and fit all the details to the boards to make them look something like the prototype.  The shrouds I made from 1/8 O.D. x 1/16 I.D. brass tube, hand filing the distinctive cut away shape.  They were all painted matte black on the inside prior to gluing them to the face of the target board using CCA and keeping them in alignment using the edge of a steel rule.

The covers at the rear of these took some time and they were made 100% from styrene scrap pieces.  I started with the hexagon shaped box at the centre of the rear enclosure assembly and made two of these.  Six sides and a flat top.  I used CCA to locate this in position and then made matching rectangular boxes with a short taper at their ends and glued these into their positions.  After this I detailed the back of these enclosures with the tapered strengthening fillets as per my prototype photo.  In the field the enclosures are aluminium castings with the tapered fillets providing additional strength. 

Handrailing:

Current safety handrail design is definitely not made for the modeller to replicate easily.  It is fully welded and no longer do they use round bar for the handrail anymore, it’s all flat bar on the prototype these days.   The use of all flat bar doesn’t make it especially difficult to model, but what it does is reduce the rigidity of this assembly and make this area of the model very fragile.  Soldering the very thin edge of the vertical upright to the large flat face of the horizontal bar, I knew immediately that as soon as the solder was cleaned up, there would be no significant solder to maintain a rigid joint.  A road to disaster if ever there was one.  So it was compromise time, but I do have a modellers license, and decided to make only the vertical uprights from flat brass section and the horizontal bars from 0.012" diameter brass wire.  Being able to drill two holes in the face of each vertical flat bar and thread both upper and lower horizontal handrail wire through and thereafter soldering it to each post, made for very strong handrail sections.  I added some extra detail to the upright handrail supports.  The prototype vertical handrail supports are fitted to the main horizontal beam of the structure with a pair of small angle brackets located at the bottom of the upright on each side.  Some signals have them either welded to the bottom of the vertical post and bolted to the main horizontal beams or all bolted.  To make these I superglued small rectangular pieces of styrene each side of the post to replicate these brackets.

Ladder: