Garden Railway Forum

This is hopefully the first step on the path to a track I can run some trains on........

Commercial 45mm gauge turnouts available in Oz cost a fortune and only look like VR turnouts if you view them in the dark. I figured if I couldn't build my own turnouts I might as well not bother, so this one was started a while ago as an experiment.

Note, this turnout is based on VR NG practice, so the appearance is very different to UK practice and you certainly won't find anything like it in a Peco catalogue.

The VR NG lines were laid with used 60lb rail, recycled from late 19th century programs to upgrade main line rails. I assume the turnouts were assembled from recycled BG components as well, as the NG ones resemble those on BG branch lines which were also laid with 60 lb rail.

This turnout is what the VR Way & Works Dept. called a 'symmetrical contraflexure junction'. I kid you not, that's what is printed in the title block of the drawing. The model railway hobby would call it a Y point.

The gauge of this one is 45mm and the v-crossing is a #4 (~14 deg.) with an equivalent radius of 7'6", or about 2.3 chains in full size. Considerably sharper than anything the VR would have used, but I don't have the space they did.......

The sleepers were cut from Australian Cypress, which may yet be changed, although it is rated for exterior use and can be buried in the ground, e.g. fence posts. Importantly for any wood used outside around here, it is resistant to termites. The rail is Code 250 brass from the US, which scales out near enough for the VR 60 lb rail profiles. The web is a bit thick, but I can live with that, as I needed to drill and tap holes in the web anyway, to bolt the buttress chairs in place.

The first pic shows the rail parts formed, machined and laid out on temporary sleepers to make sure everything fitted. Behind the turnout in the photo, are the plain gauges and roller gauges I machined up to hold everything in place during testing and final assembly. The rails forming the V of the V crossing were silver soldered together for strength. This also worked out conveniently when I started assembly, as the spacer blocks in the v-crossing could be soft soldered in position without everything falling apart. The taper on the point blades and crossing rails was milled, which is much easier than hand filing.


The next photo shows the v-crossing end of the turnout after everything was assembled. The wing rails are bolted to the V with hex head 10BA screws to allow disassembly for future maintenance if necessary. The same with the check rails and stock rails. The number of bolts is much less than you'd see on a dead scale turnout, but this is a working model. The sleepers are noticeably thicker than the first photo. In the time between taking the two photos I decided to use Accucraft flex track for all the plain track and thicker turnout sleepers were used to match the flex track height. I won't live long enough to hand lay that much plain track.


The next photo shows the switch part of the turnout. Again the parts in the switch assembly are made to resemble the real VR parts, without getting carried away. The buttress chairs were milled out of brass bar, drilled and then cut to length to form individual chairs. Not a perfect replica in close up, but easy and quick to make and they give a fair representation of all the lumps and bumps on that area of a VR 60lb rail turnout.

The spreaders are fabricated out of three little turnings and a formed piece of 1/16" rod, all assembled with silver solder, with the spreader lugs on the points riveted for strength. I have used soft solder in some areas to fix parts in place, but I don't really trust it out in the weather, so all the soft soldered joints are reinforced with a mechanical fastener.


The last photo shows the point lever and operating mechanism. Basically similar to the Tenmille one, but scratchbuilt. The VR NG lines seem to have mostly used what they called a spur lever in yards. This was weighted off centre so the points would return to a default setting if the handle wasn't held over and they could be trailed if set the wrong way. On mainline turnouts something called a WSA type was used which held the points in the position set by the lever. This model lever also fixes the points position when thrown and bears a very vague resemblance to a much older VR design called a Ford lever, which was in use before the WS/WSA levers were developed.


Building this was slow, but not too traumatic once I worked out what I was doing. All I have to do now is build another ten or so ordinary #6 turnouts and this backyard track might actually happen.

Regards,
Graeme

Beautifully executed as ever Graeme.

It would be nice if Peco did model the extended sleepers like you have done for mounting a point lever.

That does look the business, it must have been a satisfying build it's come out so well.

That's impressive. I have considered hand building my own turnouts, but not sure I have the skill or tools.

Ian

First rate.

Impressive work, Graeme. They do look like the real thing to my eyes.

Rik

Hi Graeme,

That is quite simply fantastic! I am very impressed. The VR descrition is still how many parts of the modern pway industry would refer to a turnout of this type.

Some on here will already know that I am a Pway engineer by profession and so have a critical eye when it comes down to this aspect of our hobby but I can only praise every aspect of your work. It's a functional and well engineered piece of trackwork, I've no doubt it will perform it's job without a hitch. I very much look forward to seeing more as you progress!

Andrew

Well, it's about 4 months further on and some more progress towards completion of a running track has taken place, as another four turnouts have been turned out by the TVT Way & Works workshop.

For a fleeting second I contemplated handlaying the plain track as well, but at the rate it takes me to spike rails to turnout timbers, it could take a decade to do enough plain track for this job. So there are now two boxes of AMS 45mm narrow gauge track sitting under the workbench.

If my maths is holding up, there is now enough track and turnouts available to lay stage 1, which is the mainline oval, around the back of the block. AMS track currently comes in DIY form locally and I spent today making a simple wooden jig to make sliding sleeper units onto the rail an easier task when track laying eventually commences.

The first photo shows the result after a couple of weeks making turnout components. All the parts for two r/h and two l/h #6 turnouts of more or less VR style are laid out on the template, with the small parts contained in the row of cups at the back and timbers in a range of lengths are in the trays at the far end. The rails for the first of the new turnouts are assembled and can be seen awaiting spiking to the timbers.


The next photo shows the first r/h turnout completed, along with the original 'Y' turnout behind it. The nominal radius of all these turnouts is 7'6", which will also be the minimum radius of the track. Construction method is the same for all the turnouts and the details were covered in the first post.


The last photo shows all four of the new turnouts stacked up on the bench ready to go. Reminds me of what our local station yard looked like when the VR relaid the line back in the mid '50s.


The timbers and point levers still need staining grey and painting white respectively, but that can wait until closer to track laying time.

All up, it took about a month to make the four turnouts, so about a week each. Spring has sprung hereabouts, so the next job is organising materials for the support structure........

Regards,
Graeme

Very impressive Graeme, and beautiful too.

A week for each turnout! Wow, no wonder they look so good. Each one is a labour of love. I do like seeing the construction process in context.

Rik

ge_rik wrote: ↑Wed Oct 03, 2018 7:43 am A week for each turnout! Wow, no wonder they look so good. Each one is a labour of love.

Sort of.......... I regard track as a model too, so worth a reasonable amount of time and effort.

Plus, being retired like me, you would understand the meaning of the saying 'time rich and cash poor'. I could buy a Roundhouse steamer for what five commercial #6 turnouts would cost locally.

Regards,
Graeme

Graeme they are very impressive and I agree a labour of love. They look far far better then anything you could buy. Have you costed them Graeme.

One thing though, I have counted 5. Did you already have one made before mate.

FWLR wrote: ↑Thu Oct 04, 2018 8:42 am They look far far better then anything you could buy. Have you costed them Graeme.

Materials cost around A$40 per turnout, if you have the workshop to make the parts. About 1/6th of an equivalent commercial product. However, I don't have to pay myself an hourly rate........

FWLR wrote: ↑Thu Oct 04, 2018 8:42 am One thing though, I have counted 5. Did you already have one made before mate.

If you read the first post in the thread, the Y turnout was built first as a test, to determine if the methods I used in HO could be scaled up to Fn3 scale. It was successful, so the others were built as a batch some months later.

Graeme

Nice work Graeme, that bench needs some dirt on it though, and some junk. It's just not right.
See you Saturday.
Grant.
PS cleaned some of my track today after another bit of tree fell down, and actually ran the mechanical loco.

Well hello Grant

I have been watching this thread with great interest but the engineering skill level is way beyond my abilities. Well done Graeme and welcome back Grant.

Hi Graeme,

This is proper permanent way engineering, scale is irrelevant with work of this calibre. The same techniques would work just as well in 12" to the foot scale. I have no doubt that all this careful work will repay itself in spades with reliable running. Well done and thanks for the inspiration!

Hi Grant,

Nice to see you posting again.

Andrew

I wish Peco would do what Graeme has done and model the extended sleepers for point levers.

GTB wrote: ↑Thu Oct 04, 2018 1:53 pm

FWLR wrote: ↑Thu Oct 04, 2018 8:42 am One thing though, I have counted 5. Did you already have one made before mate.

If you read the first post in the thread, the Y turnout was built first as a test, to determine if the methods I used in HO could be scaled up to Fn3 scale. It was successful, so the others were built as a batch some months later.

Graeme

I did read the post Graeme…I realised after I posted that you did say you built a test first…..