Cwmhir signaling - Attempting a somersault and rotating discs

[Darwinian]

27 minutes ago, Stephen Freeman said:

It is much easier to hide wires etc in a hollow brass post

True, is Masokits the only source of suitable tapered posts? I can’t find any online suppliers.

[bécasse]

It isn't a hugely difficult task to solder two L-shaped lengths of brass together to form a tapered post, it helps to make up a couple of suitably sized "matchsticks" in wood to go in each hollow end while you make it up. Tack solder the two pieces together first so that it is easy to adjust the joints until you have the taper right, then run solder up each of the seams in turn. Subsequent soldering of detailed pieces on to the post may temporarily melt the soldered joints in their immediate vicinity but this won't affect the integrity of the tapered post.

[Stephen Freeman]

11 hours ago, Darwinian said:

True, is Masokits the only source of suitable tapered posts? I can’t find any online suppliers.

I think if you Google "etched brass signal posts" all will be revealed.

[Stephen Freeman]

11 hours ago, bécasse said:

It isn't a hugely difficult task to solder two L-shaped lengths of brass together to form a tapered post, it helps to make up a couple of suitably sized "matchsticks" in wood to go in each hollow end while you make it up. Tack solder the two pieces together first so that it is easy to adjust the joints until you have the taper right, then run solder up each of the seams in turn. Subsequent soldering of detailed pieces on to the post may temporarily melt the soldered joints in their immediate vicinity but this won't affect the integrity of the tapered post.

Soldering two parts of etched signal posts together is not too difficult as long as you don not completely remove the tags The tags will help to hold the two parts in alignment until you have soldered them together, after which they will need cleaning up of course.

[Darwinian]

Agreed. I am going to try to follow Steve Hewitt’s approach.

Once I have some suitable concentric tunes and fibre optic I will experiment with making the rotating discs work.

[Darwinian]

On 24/03/2021 at 07:39, Stephen Freeman said:

I think if you Google "etched brass signal posts" all will be revealed.

Thanks for that. Posts duly ordered. Now I have all the bits I need to start.

[Darwinian]

On 28/02/2021 at 22:15, Stephen Freeman said:

Making them work is not a problem, adding illumination though would be challenging.

This turned up on a search of the site today:

I'm planning to try something similar but using a fibre optic into the lamp instead of wires and LED.

[Stephen Freeman]

Didn't say it couldn't be done, just not easy. This one is a M&H one, servo mount is by  MERG and it just features one smd led attached to the spindle (hollow tube) the other polarity is fed by a fine wire within the tube. When I get time I might do a video but am otherwise occupied at the moment and no I can't see a way of making the weight bar work, so far....

Edited February 17, 2022 by Stephen Freeman

[Darwinian]

First issue:

 

How do Rhymney Somersault signals work? 

I thought that like the GNR ones on the North Norfolk Railway the board pivot was nearer the top with the operating rod connecting to the spectacle plate below the pivot. That makes it lie along in front of the support arm, making it almost invisible. I assume it's attached lower down on the board (The MSE etched signals have two holes in the board)

However on P 88 of The Rhymney Railway Vol.1 By John Hutton there are pictures clearly showing the operating link rod above the signal arm pivot point. 

How does that link connect behind the signal board?

The pivot point appears to be central on one signal and above the center line on another.

 

[Darwinian]

Well I'm not sure this is absolutely true to prototype but it does work. Components are the MSE etches and bits if micro tube to form spacers and bearings. Posts are etched brass.

 

 

And here they are assembled on the post fittings.

 

[Darwinian]

The lighting is LED to fibre optic fed through the post. Rather than twisting the fibre optic into the back of the lamp I decided to try drilling the lamp out with a smaller tell-tale hole at the back by drilling right through the casting, the front lens hole was then opened out to half way through the lamp . A 0.7 mm hole for the 0.5mm fibre optic was then drilled up from the base. 

 

The fibre optic then forms the diagonal support on the lamp bracket. The end of the fibre has been sanded with fine wet and dry into a dome so emits light in all directions.

 

Here's a test with 9v battery power. The lamp is just placed over the fibre optic. I have damaged the sides of the fibre getting it in position so there is some light leakage in the diagonal bit. I will try again although it will be painted over anyway (White then black).

 

 

There is a little ambient light here, hence the post showing up. The lamp light is not bright enough to show in daylight but then neither were the prototype ones.

[Stephen Freeman]

Painting fibre optic is likely, no certain to, damage it, resulting in more light loss.

I remember, and I think others will too an article in the Railway Modeller from the early 80s, possibly end of the 70's (not sure) by the late Bill Baker, which details the best method of installing a fibre optic cable without it showing.

The M&H lamps are large enough to easily take an SMD LED and you will get more light as well as not being visible.

[Darwinian]

18 minutes ago, Stephen Freeman said:

Painting fibre optic is likely, no certain to, damage it, resulting in more light loss.

I remember, and I think others will too an article in the Railway Modeller from the early 80s, possibly end of the 70's (not sure) by the late Bill Baker, which details the best method of installing a fibre optic cable without it showing.

The M&H lamps are large enough to easily take an SMD LED and you will get more light as well as not being visible.

 

I did wonder about the effect of painting. In theory the fibre optic works by total internal reflection of the light at the internal surface of the plastic fibre strand. BUT it requires the medium outside the strand to be less dense than the strand material (plastic to air) so any painting that doesn't actually affect the surface of the plastic might affect the transmission of light. 

Any damage to the reflectivity of the surface will allow light leakage (as seen where I have damaged it here). However the length being painted is very short so hopefully the losses will be small.

I use acrylic paints which should not damage the surface but I'll do a test to see. White first as it is more reflective but that may not matter.

 

I came up with this approach to lighting the lamps when [pondering how to light the rotating disc signals that are up signal posts.

 

I will report back when I've done the experiment.

[Darwinian]

Right, this didn't quite stand up to my idea of scientific study but for what it's worth.

 

Here are two lengths of fibre connected to the same LED using the socket to make them the same distance from the LED and in the same alignment.

 

On the left is an unpainted length, on the right one with a length of about 5mm painted White with Vallejo acrylic.

 

Not the best comparison as can be seen there is more light leakage from the sides of the fibres where they are taped down, especially on the unpainted one.  However it does show that light is leaking through the painted area to some extent, especially at the end of it causing a distinct bright spot.

 

Then I over painted the white with black. You can see the black painted section on the right fibre as it isn't leaking any light (it's being absorbed instead).

 

 

To be honest I cannot see any real difference between the two and the camera isn't showing much difference either. The bright light at the top is the back of the LED in it's mounting tube.

 

Looks like I'll get away with it, although obviously it won't be as bright as an LED in the lamp or the Steve Hewitt method of bringing the fibre optic in fro the back of the lamp.

[Stephen Freeman]

Hi,

The effect of painting the fibres is not necessarily instantaneous and no doubt depends on the type of paint.

 

The best way of introducing fibres into a lamp is from the side. You can carefully bend the fibre by placing a hot soldering iron close to (but not touching) where the bend is needed. You should ensure that there is sufficient fibre past the bend so that gravity does the work. The lamp will need larger holes to allow entry of the fibres. The size of hole should be large enough to take the insualtion sleeve (black preferred) from single core wire. The sleeve should fitted over the fibre once in the lamp so that it is a snug fit. The excess is removed with a new scalpel blade.

 

It's all in the original article "Pipe your light" (I think), so you should be able to find it in the Modellers online backnumbers, I think it was probably 1979 or thereabouts.

 

Using this method it is possible to have 2 fibres per lamp, one front and one back but these days it is just easier to have an SMD LED instead, which weren't available at that time.

[Stephen Freeman]

Had a few minutes, the article was by Keith Baker and Bill Baker "Light your Lamp" in Railway Modeller May 1983.

[Darwinian]

After a very enjoyable family holiday in the Peak district I am back to signal building. The ladder and safety cage have been added. The ladder is a Wizard models one although as a first attempt it's a bit wobbly, doesn't look so bad from normal viewing distances.  I also moved the spectacle pivot point down 1mm to make the drive rod to the signal board more horizontal. This of course meant taking the mounting arm/bracket thing off the post and re-attaching it to line up with the holes in the post.  This is very much a learning curve for me.

 

 

 

Here is the whole signal as it stands. 

 

[Darwinian]

A question for the signalling experts.

 

I seem to recall reading that the crescent shaped back blinders were fitted so that the signalman could tell at night if the signal had cleared (the lamp light would be revealed). They would not be fitted on signals facing the signal box.

 

Is that correct?

Would the back lens on the lamps still be showing on signals facing the box?

[Michael Hodgson]

The back light blinder was usually fitted regardless of whether the signalman saw the face or the back of the signal, but some companies only fitted them where necessary.  It is on the other side of the post from the arm, so on some designs of signal fitting it may be necessary to prevent the spindle from working loose over time and the whole arm falling off. 

 

It obscured the light when the signal was off and the lamp was visible when the signal was on.  You can thus confirm that the signal was at danger and alight when it was supposed to be, which for a safety point of view is rather more important than whether it is showing proceed when it should.

 

PS,  In the case of somersault signals, the arm isn't attached to the spectacle plate, so the blinder (where fitted) isn't on the same pivot as the arm.  Blinders were more likely with these to be fitted only where the signalman needed them.

Edited September 27, 2022 by Michael Hodgson
add postscript - somersaults

[Stephen Freeman]

Here is one I made earlier

 

 

[Darwinian]

That's very nice Stephen. You are getting more movement of the spectacle plate than I am on mine. I think trying to get the arm operating rod at the top, as per my prototype photo's, is making the spectacle move less on mine. Good job Rhymney signals seem to have gone almost vertical when cleared (maybe that's why too).

[Stephen Freeman]

1 hour ago, Darwinian said:

That's very nice Stephen. You are getting more movement of the spectacle plate than I am on mine. I think trying to get the arm operating rod at the top, as per my prototype photo's, is making the spectacle move less on mine. Good job Rhymney signals seem to have gone almost vertical when cleared (maybe that's why too).

There is (at least in 4mm scale) more than one etch you can use, the newer one (ex D&S) is more prototypical but a little more fiddly as the drive bearing for the assembly is a separate etching which has to be joined to the spectacle, gives a better action though. Wizard have it on their site as S003/03, there is also S003/M for miniature arms

.

Edited September 28, 2022 by Stephen Freeman

[Darwinian]

Well the MERG servo controls have been successfully completed so suitably encouraged I’ve moved on to try a raised rotating disc signal.

 

Here are the main components. Fibre optic is sleeved in micro bore brass tube so that it doesn’t get worn by the rotating lamp support that fits over it. This does mean the support for the lamp

is rather over sturdy but it will have to do.

Just to make it interesting this signal is on the incline so setting up the servo mount should be “fun”.

Too cold in the ages to so much at the moment and I want to paint the sun assemblies of both signals before I get much further.

[Darwinian]

Something I’m not totally clear of is how/if the clear indication was illuminated. The disc shows red at on and moves to side on for off but was there a green (blue) lens on the lamp too. Presumably a clear lens on the back of the lamp acts as the tell tale.

[The Stationmaster]

15 hours ago, Darwinian said:

Something I’m not totally clear of is how/if the clear indication was illuminated. The disc shows red at on and moves to side on for off but was there a green (blue) lens on the lamp too. Presumably a clear lens on the back of the lamp acts as the tell tale.

The colour shown by a running signal at night was usually determined by the spectacle plate which moved in front of the lamp case. In the early days there was only a need fora red spectacle as the 'all right' light was white so generally this was done only using the light from the lamp case with no need for a spectacle plate with two colours.   When green replaced white as the 'all right' colour for running signals the spectacle plate needed two apertures to in order to show either red or green..

 

The only rotating lamp cases I'm aware of were on ground signals but no doubt someone will know if there were ever used for running signals.