7mm Switch Blades - forming your own

Dog Star

Western Thunderer
Some months back, maybe even last year, I promised to write about making switch rails for 7mm track using bullhead rail. My interests are pre-group... when most railway companies were using straight cut switch blades although the GWR had started to install "curved" straight cut blades (not a confused ideal... the rail was clamped to the bed of a planer and the area to be planed was then curved to the appropriate radius for the switch. The switch blade would then be machined as a "straight cut" switch whereupon when the clamps were removed the rail would return to the original shape and the tapered portion of the blade would be curved).

The switch blades for a turnout are handed in that there is a LH blade and a RH blade with the taper being, apparently, on one side of the blade. So, given that we need to produce a pair of handed blades, we need to be aware of which side the taper is formed. Here is a photo-sequence of the work in forming the tapers and shape of a RH straight-cut switch rail for a LH turnout. In the photos / captions note is made as to which way the filing is done in relation to the head of the rail.


This piece of rail is going to form a RH blade and the "first" taper is put onto the rail face which is against the RH stock rail... which means that the head and foot of the rail are filed away so that the taper is across the full depth of the rail. This blade is for a LNWR 18'0" switch which is (roughly) equivalent to a REA "D" switch. For this blade the taper is planed over a scale 11'0" so the rail is covered with black permanent marker and the required length marked in the web. Here the face of the blade marked up prior to metal removal - in this photo the foot of the rail is against the ruler.
switch-rail-1.jpg

This photo shows a blade which has had the taper formed on the face which is against the stock rail, the head and foot of the rail are filed away until the taper has just started to cut into the web (where "just" is about 0.5 - 1.0 mm). switch-rail-2.jpg

At this time the toe of the switch rail is greater than the full thickness of the rail web and so metal has to be removed from the running face of the blade. Here is the running face of the blade marked up prior to metal removal - in this photo the head of the rail is against the ruler. switch-rail-3.jpg

Whereas the head and foot of the rail are removed for the rail face against the stock rail only the head of the rail is removed from the running face of the rail. This photo shows the switch blade after the head and web have been filed until the toe thickness is circa 15 thou. There is no removal of metal from the foot of the rail.
switch-rail-4.jpg

In this photo the tip of the running face is filed to provide a lead-in so that a wheel does not "jolt" when taking the diverging route. The permanent marker was applied before this step so as to show the extent of the lead-in.
switch-rail-5.jpg

So now we have a rail with a taper filed on both faces... we shall to change that situation so as to leave the switch blade with a taper only on the face which abuts the stock rail. Place the rail in a vice with the mark on the web against the jaw (this is the mark from photo 1 and indicates the planing length for the switch blade). Gently bend the rail so as to remove the taper on the running face and leave all of the taper on the non-running face - a straight edge against the side of the rail head can be used to check that there is no remaining taper on the running face..

regards, Graham
 

Dog Star

Western Thunderer
After writing the previous part of this topic I thought that maybe the matter of setting the taper "all on one side" might not be too clear so here is the final part of this trilogy - how to set the switch blade with photos to show the way in which the taper is formed, modified and then set. In this sequence I am creating a left hand switch blade and captions are written with the rail head as the visible part of the rail, this puts the running face of the rail to the left of the image.

In the first photo the taper has been filed on the non-running face of the rail, this is the side of the switch blade which abuts the stock rail. The black spot on the rail head marks the end of the planing.
switch-rail-6.jpg

In this photo the rail head has been removed from the running face of the switch blade. The foot of the rail on the running face of the rail is still in line with the rail below the end of the planing.
switch-rail-7.jpg

In this photo the switch blade has been bent, at the end of the planing, such that the head of the rail on the running face of the switch blade is now in line with the rail beyond the planing - to achieve this the rail is set towards the LHS of the image.
switch-rail-8.jpg

regards, Graham
 

Dikitriki

Flying Squad
That's very fine filing for the taper:thumbs: . Have you stabbed your finger with the sharp end while filing yet? I did:rolleyes:

Richard
 

S-Club-7

Western Thunderer
I use a similar technique which works for me in both S7 and EM. All dimensions are full size, reduce as appropriate for your particular scale. The red lines show the original shape and alignment of the rail, the yellow highlight shows the result of the action performed at each stage.
switch-1.gif
1. Plane the back of the rail until the web is just broken into. The prototype only reduces the end of the web by a sixteenth of an inch so don't go too mad! However, the web of most model rail is overscale so you may wish to reduce it's thickness to a scale eleven-sixteenths of an inch to avoid the switch looking too "chunky". It is essential to be as accurate as possible with the planing length as a few thou can make a lot of difference to the switch angle.
switch-2.gif
2. Put a set (that's a bend to you and me) in the rail at the planing length so that the end of the rail is aligned where the running face was (the picture should make this clearer). This location of this set is important since this affects the switch angle.
switch-3.gif
3. Plane the front of the rails at angle to suit the wheel-flanges. This planing should be to the full flange-depth and rail-thickness at the end, tapering to nothing at the planing length. It is this bevelled edge which guides the wheels away from the stock rail. Again, the picture paints better than my words.
switch-4.gif
4. Top planing comprises two parts. Firstly three-eigths of an inch at the end tapering to nothing at 82% of the planing length.
Secondly another three-eigths of an inch at the end tapering to nothing over a distance of about seven inches.
switch-5.gif
5. Radius the sharp edges along the running edge of the rail. Drill holes for fishplates and stretcher bars as required.
switch-6.gif
6. Remove all sharp edges and polish out any surface scratches.

Based on NER and LNER practice, other railways are also available.
 
Last edited:

Dog Star

Western Thunderer
Knowing what to do is a fair way to getting the job done...

Now if you can offer some CAD stuff for GWR switch blades and show how those differed that shall be worthwhile.

regards, Graham
 
Last edited:

S-Club-7

Western Thunderer
Now if you can offer some CAD stuff for GWR switch blades and show how those differed that shall be worthwhile.
For GWR straight switches modify the second sentence of step 4 above to "Secondly put a radius on the end of of the top-planing". Unfortuately I'm stuck at work so cannot access my library but I think it's a one or two inch radius. It's shown in Smith's 4mm GWR track book.

The top planing really does make a significant difference to the running. Even if it doesn't go all the way to 82% it will still make an improvement and is worth the few seconds it takes to do. The bevel planing removes less metal than if the toe is "knife-edged" so should also be quicker to produce.

Go on, give it a go. It'll only cost you a couple of inches of rail and a few minutes of time...
 

Eastsidepilot

Western Thunderer
Can't add any more to what Graham and Dave have said, only to say that sometimes having filed up the switch blades they can take on a twist so sometimes they need a tweek before final fitting.
If your contemplating building your own P&C work in S7 then it really is worth spending the time refining the blades, you'll see the benefit in the running of your stock through the pointwork:thumbs:.
It's worth investing in the filing jig from S7 stores for knocking out a 'rough' blade, and then refining as the Lad's have done above.

Col.
 

ZiderHead

Western Thunderer
Really enjoying these topics on trackwork :thumbs: , even if its a bit beyond my metalworking skills (and patience!) Does anyone know if C&L's switch and V sets are prepared in a similar way? I'm hoping I can get away with decent running in 31.5 using them with the timber/chair set, some rail and a copy of Templot for under £50 a turnout ...
 

ZiderHead

Western Thunderer
Good lord thats an ambitious first attempt! And a great looking result :thumbs:

I can't wait to build some track but until I find a house to buy I'm a bit constrained by having to be able to pack everything away in drawers or display on bookshelves, both of which are ~29x11".

I did sketch out a test/display plank with an A4 turnout but … err …

Bookshelf.png

:oops:
 

Dog Star

Western Thunderer
Its miles easier to mill them if you access to a milling machine...
If I had access to a planing machine then I could replicate the way in which Swindon machined a switch blade held against a curved former. Whilst those of us who rely on hand tools produce straight cut blades Swindon was able to produce switch blades with the planing in a curve to match the needs of a specific turnout (detailed in one of the proceedings of the GWR Swindon Engineering Society).
 

Susie

Western Thunderer
Hi Graham,

Small planers are very rare: I've only seen two up for sale in the last 15 to 20 years, both dating from the turn of the 19th/20th century and about 12" stroke. Large ones are available: there is one for sale at the moment which will machine an area of 20.2 metres by 5.4 metres! Shapers are far more accessible, in fact I know of a workshop not far from my computer where there are two: one hand powered and one motorised!

Do you know what sort of radius the switch rails were bent to prior to machining? You've got my brain working on this now!

Best wishes, Susie
 

Dog Star

Western Thunderer
Do you know what sort of radius the switch rails were bent to prior to machining?
Susie,

I shall try to find the relevant engineering society pamphlet to see what the words and music say about the operation. From what I recall, the bullhead switch was bolted to the planing table in such a way that the switch was curved before the planing commenced... then after planing the switch regained original form, less that material removed in planing, when the clamps were removed.

Trust Swindon to be different.
 

Susie

Western Thunderer
Sorry Graham,

Once again I have asked a question and then found the information on my book shelf. In this case British Railway Track published in 1950.

A B type switch had a radius of 730', a C type: 1130', and a D type: 2000'.

Susie
 

simond

Western Thunderer
Thanks Jon,

It wasn't exactly a first attempt - I did build a copperclad tandem about 10 years back (you can see it in photos in the "current model" thread in my RMWeb signature) and previously, following some good advice in NGIRMR, I built some 16mm scale points from brass Bonds rail, oak timbers and copper wire, which survived a couple of winters in my ownership, and hopefully are still giving their current owner good service - if so they are at least 18 years old, as I built them over a year before my son was born.

You do need good track gauges, a vice, and a couple of decent files, and you can certainly learn to file to a line, so gauges for crossings and blades are unnecessary - though they may be helpful.

Templot & associated forum are a mine of good info & advice. It takes a bit of practice but the outputs are outstanding.

Give it a go!

Referring to the discussion about planing curved switches - is there a piccy anywhere showing the blades clamped in their curve on the planing m/c? I ask because it seems to me that the planing tool must sooner or later remove the clamp, which is unlikely to be helpful!

Simon
 

Dog Star

Western Thunderer
Referring to the discussion about planing curved switches - is there a piccy anywhere showing the blades clamped in their curve on the planing m/c? I ask because it seems to me that the planing tool must sooner or later remove the clamp, which is unlikely to be helpful!
I do not know of any such picture... and until I read about the process in the engineering proceedings I was not aware of what Swindon did when planing BH switches - I need to find the text and try to post some more information.
 

farnetti

Western Thunderer
Graham

Its miles easier to mill them if you acces to a milling machine and you won't stab yourself with one either.

Richard
Hi Richard,

I can't get my head around how you would do this? If the rail is clamped to the table how does the milling head cut the correct plane?

Good to see you and David again at Bristol after a long time.

Ken
 
Top