Pragmatic approach to throw bars

[highpeak]

This is not a description of how to make a throw bar/stretcher bad/tie rod that is the ultimate in prototype fidelity, it is a pragmatic approach to perhaps moving beyond the "moving tie" method. It is not my idea, I came across it in issue 181 of Scalefour News. That issue included two articles describing ways of modeling stretcher bars, a high-fidelity approach described by Howard Bolton, and a more pragmatic approach that sacrificed fidelity in favor of ease of fabrication and robustness used by Mike Norris on his P4 Preston layout. I have used the latter approach and thought that it might be worth giving it a mention to an audience that might not be members of the Scalefour Society. It might be of use to people who are looking to improve or repair proprietary turnouts as well as people who are building their own.

 

The moving tie method is long-established at least in the world of those who dabble in copper-clad/printed circuit board technology. It is very easy to make since all you are doing is soldering rail to the copper clad tie. Iain Rice describes it in his "Pragmatic Guide" to PCB track and it's been advocated by many, and criticized probably by about as many. There are those who claim the solder joint to the rail is too highly stressed to be reliable, and this may well be true, especially with the one piece switch/closure rail approach. The other camp will downplay that and point out that even if the joint fails, it's a relatively easy field repair, at least if the turnout is accessible. I think both camps are probably right.

 

I don't think anybody could argue though that the moving tie, even when it's not in motion, is pretty conspicuous. Mike Norris neatly solved both disadvantages of the moving tie by flipping it to a vertical position so it presents a thin edge to the eye, and connecting it to the switch rails by brass wire.

 

This is my attempt at the Norris method. I use thin double-sided copper clad from C&L sold for turnout timbering. They make two thicknesses, the one I use is 40 thou. I originally used a Fast Tracks tie, but they are about 62 thou thick and so a bit more obtrusive. I cut the C&L tie along its length as the original dimensions intended for a 4mm scale turnout would require a bit more vertical clearance. The wire I used is from Detail Associates and is a bit thicker than Mike Norris's choice, he used .45mm, this is a bit heavier at .55 (22 thou). It doesn't seem to make much difference but I wouldn't go any thicker as it might end up too stiff. Photo one shows the materials with holes drilled 14mm apart in the copper clad. I've gapped the copper clad on both sides, the additional gap is to make sure that the end where I attach a loop of wire for the drive will also be isolated. The method allows from adjustment in the brass wire, so the distance between the holes is not super critical as long as its close to the desired distance between the switch rails. The wire has a 90 degree bend for soldering to the copper clad. It will be trimmed to length later.

In photo two, I've soldered the wire into the copper clad using a helping hand to hold everything. 

With both pieces of wire soldered in place, the first bend is a 90 degree bend so the wire points straight up in the air. I've also filed down the wire a bit to make sure there are no clearance issues between the throw bar and the adjacent tie.

The next bend is another 90 degree bend so the the wire is now horizontal and parallel to the switch rail. I use a 6 inch steel rule as a spacer when making this final bend, and I've reinforced things with another dab of solder. It's worth noting that you need to work out which side the drive is going to be attached as this arrangement is handed. Mike Norris following British practice made his stretcher bar so it would just span the stock rails, I've made mine longer, the drive from the switch machine will be underneath the cosmetic switch stand.

I file a small slot in the base of the switch rail for the wire to sit in when it is time to solder it to the web of the switch rail.

This turnout is nearly ready for installation on the layout. I used a Fast Tracks tie here which is a bit fatter and more noticeable than the C&L one. I disguised it a bit with a tie rod raided from the Central Valley details kit, the other cosmetic tie rod is a home made job made out of 20 thou plastic rod, super glued to a noggin of styrene that is in turn super glued to the underside of the foot of the rail. Nothing special but better than nothing at all. I have no idea how well it will stand up to operation.

 

As I said, this is not a route to a high fidelity throw bar, there are other techniques and products available if that is your goal. I liked Mike Norris' approach because it represented a method that somebody of moderate ability could use, but so could a ham-fisted guy like me. To me the trade off between appearance and potential reliability was acceptable. It does take a bit longer to do than the moving tie, but not excessively so.

Once again, this is not my method, it was described by Mike Norris in Scalefour News, I've adapted it a bit but take no credit for it.

[298]

Correct me if I'm wrong, but isn't there still the issue of using a solid joint to solder the rails to the tie bar...? The more recent fast tracks turnouts I've used have brass pins soldered to the switch blades and he's in the tie bars, so a small amount of rotation is catered for.

 

If only there was a reliable way of keeping everything well engineered but still electrically isolated, one solution might be to model a trolley line and not gave to worry about track polarity, but then someone will still tell you your earth bonding wires across fishplate joints uses the wrong gauge wire...

[highpeak]

Quoting Mike Norris in Scalefour News: "there is some flexibility at the tip of the rail, using the torsion in the wire, this helps reduce stresses, but crucially a good joint can be created by soldering a decent length of wire in the web of the switch rail."

 

You are of course quite correct in noting that both the moving tie method and this method have a soldered joint between the switch rail and the component that is transmitting the force to move the switch rail. There is a degree of flexibility in the wire (which is why I think I would not want anything thicker than what I used, which was what was to hand, and a little bit thicker than Mike's wire) and the joint is better than with the moving tie.

 

I think the reputation for the weakness of the moving tie joint might be a bit overstated and has more to do with the type of drive. If you were using a solenoid drive, which was pretty typical some years ago, then the repeated hammer blow probably would over-stress the soldered connection. More modern drives like the Tortoise probably place less stress on the soldered connection and either the moving tie or Mike Norris' method would be reliable enough.

 

Using overhead wire as one side of the circuit eliminates a lot of issues that two-rail systems have to deal with. Those bonds though are pretty tiny and fragile enough in 1:1 scale.! I recall the issues we had with insulated joints and the signal circuits. One of those joints on our line was not quite up to it, and you could sit in the yard on a summer evening after operations had ended as the temperature slowly dropped and the rail contracted and watch the signal  for the section from the yard to the end of track go from green to red and back again, knowing that all the trolleys were safe and sound in the car barns.

[highpeak]

Ah the joys of spill chucker! I think "rude to the pub" started life as "rod to the PCB". If my picturing of it is correct, Iain Rice has an arrangement like that in one of his books on track construction.

That would appear to improve matters in the stress area. One function of MIke Norris' arrangement (like the moving tie approach) is that it prevents any tendency for the switch rail to move in the vertical plane. With the rude in pub arrangement you could do the same if the rod was bent 90 degrees at the bottom of the tube.

As I recall, Iain's solution involved the PCB sliding in some form of carrier which was attached to the roadbed. Nothing wrong with the idea, but it was a bit more complex than what Mike Norris did.

[dave1905]

I solder a brass tab with a 00-90 clearance hole drilled in it under the tip of the point.  I tap 00-90 threads into a piece of PC board, then put a 00-90 screw through the tab into the PC board  the tab goes under the stock rail.  As the switch throws that allows for rotational movement at the point tips, the tab keeps the points from rising up under any torque or thrust from any switch mechanism.  The tab vaguely resembles the metal clip to which the rods connect.

 

A little more work but doable.  I guess one could modify that by standing the PC tie on edge and soldering a couple pieces of brass tube to the now vertical face, then tapping them 00-90 and sinking the screws into them.  That would minimize the "moving tie" effect.