Jump to content
 

Semaphore Signals - 4mm Scale (Mainly)


Recommended Posts

Thank you, Steve, that's very clear. Great photos of some fantastic work, especially the detail of the forked connection. Parallel pliers duly ordered, sure can think of other usesin the fullness of time. Hopefully can use reasonably neatly. Varying the offset of the holes in the balance arm seems a good trick to have up sleeve,

 

Once again thanks

 

Jon

Link to post
Share on other sites

  • RMweb Gold

John,

 

Glad the info was of use.

I'm sure you will find the pliers really useful

The parallel action ensures that thin sheet material is held "on the flat" rather than just at the edges.

They therefore are very useful for bending etched components etc.

 

Steve.

Link to post
Share on other sites

  • 3 weeks later...
  • RMweb Gold

Hi Ian,

 

I suppose the easy answer is "No", because I've never built any Slotted Post signals, in 4 or 7mm scale.

 

However, when I'm faced with a model of a prototype I've not tackled before, I try to gather as much prototype information as I can..

Gather as many photos as possible - they are invaluable.

Try to understand how the real ones were constructed, and why.

With that knowledge, you can attempt to make your model by following those prototype principles.

 

I always prefer to madel a real prototype, ideally from photos.

One of the most difficult things I'm asked to do is to create a "typical" or "standard" signal for one Company or another.

 

I know it can be very tempting, but try not to "model a model".

 

Sorry I can't offer any specific help on Slotted Posts, but others on this forum are probably better placed to do so...........?

 

Steve.

Link to post
Share on other sites

  • 2 months later...
  • RMweb Gold

Seems ages since..........

 

Its a long time since I posted any new signal pics , probably because I've done very little modelling in the recent months!

 

My latest pair of signals have been a bit of a challenge.

Not bacause of the signals themselves, which are fairly standard Upper Quadrant brackets, but because they had to be manually operated.

 

Here are the two I've just built:

post-3984-0-51540200-1413107689_thumb.jpg

 

In this shot you can just see the "shoes" which Les Green cast for me.

post-3984-0-55017400-1413107496_thumb.jpg

They are at the base of each doll, and in the prototype secure the doll to the bracket.

 

Unfortunately, the 3D Printed lamp/bearings which Les also had made are a bit lost in these pics.

I'll try to get better ones on my next "work in  progress" shots.

 

I've only ever built signals for Servo operation, and I rely on the servo to ensure all the moving parts keep "within limits".

In this case I had to modify the Weight Bars to allow for "finger poking".

My solution involved several compromises:

  • I used 7mm scale weight bars, double etch thickness, for strength.
  • I mounted each on a separate bearing, to prevent interference.
  • I staggered them vertically to allow finger access.

post-3984-0-40124400-1413107499_thumb.jpg

I hope this shows what I mean.

 

As an engineer I was reluctant to have stops on both the signal arms and weight bars, (one would always be redundant.)

I decided to limit the weight bars, as they were much more robust than the scale signal arms, and will stand up to some finger pressure.

 

The next problem was to introduce some friction to ensure the arms stayed where they were supposed to.

I built this into the pivot for each weight bar.

The pivot is a 14BA steel screw, which screws through a plate on the edge of the "H" section main post, and the web of the post.

The weight bar is a clearance fit on the screw and is sandwiched between two 14BA washers and the plate.

By adjusting the screw, the "stiffness" of the weight bar movement is controlled.

A 14BA lock nut on the opposite side of the main post secures the position, and for good measure is re-inforced with a tiny drop of Locktight.

 

post-3984-0-50710500-1413107500_thumb.jpg

Sorry its not too clear in this shot.

 

Although there won't be any servos beneath these signals, they are still intended to be "planted" through the baseboards.

Here is the signal's foundations:

post-3984-0-24706600-1413107502_thumb.jpg

 

Finally they will have to survive a trip in a courier's van, so this is the packaging:

post-3984-0-60370400-1413107492_thumb.jpg

 

In place, rady for sealing up:

post-3984-0-63707100-1413107482_thumb.jpg

 

This box will then be packed in a larger box with plenting of "shock proofing" around it before it goes on its way.

 

Steve.

  • Like 11
Link to post
Share on other sites

  • RMweb Gold

As always Steve, some fantastic work there. Even with all that packaging I'm surprised you trust it with some of our delivery services these days.:-)

Aren't they supposed to get a signature for something like that? What if it was broken when you opened the package; who'd have to pay for a replacement?

Link to post
Share on other sites

  • RMweb Gold

Lack of action on these signals, but I've not been idle.......

 

You may have spotted elsewhere on RMweb that I've been busying myself with Ground Signals for Liverpool Lime Street.

See here if you've missed it.

 

As you will now be aware, I've been spending time working with Les Green to develop some 3D Printed items.

In addition to the Ground Signal, we have had some sample components made for me to use on Upper Quadrant semaphores:

attachicon.gifRIMG1036.jpg

The "sprue" comprises several little sub-assemblies of Arm Bearing, Lamp Bracket and Lamp on a carrier plate, together with a range of "Shoes" which are used to mount round dolls on brackets, and brackets onto round main posts.

attachicon.gifRIMG1039.jpg

The lamps are a mixture of square and round backed types.

The carrier plate is designed with a recessed rear surface, so it can be attached to a round doll or a wooden or lattice square doll.

attachicon.gifRIMG1071.jpg

 

attachicon.gifRIMG1070.jpg

 

attachicon.gifRIMG1069.jpg

 

These are the first test pieces.

The design has been modified slightly and the "production" batch of the Lamp/Bearing assemblies is on order.

You will notice that the bearings in the tests seen here are actually my usual Brass 1/32in tube.

Although very effective, I'm concerned that the brass will detach from the resin when I'm soldering the back blinders in place, which is usually my final assembly operation.

The printed versons with resin "tubes" were too fragile to use, so the bearing is being strengthened for the next batch.

So far they look very promising, and will guarantee a good alignment of the Lamp Lens with the Spectacles on assembly.

When the Arm Bearing, Lamp Bracket and Lamps are separate components, its can be quite tricky to get the alingment accurate, and a very small mis-alignment shows up when the lamps are lit.

 

Once I get the "production" batch, I'll be using them on these Woodford Halse signals.

 

The shoes are designed to fit on three sizes of Brass Tube which I use for the signals I build, 2mm, 1/8in and 4mm dias.

attachicon.gifRIMG1076.jpg

attachicon.gifRIMG1075.jpg

 

Les is going to try using 3D prints as masters for some cast copies.

This could be a very good way of making masters for castings?

 

More soon, I hope.

 

Steve.

 

 

Seems ages since..........

 

Its a long time since I posted any new signal pics , probably because I've done very little modelling in the recent months!

 

Snipped...........

 

Unfortunately, the 3D Printed lamp/bearings which Les also had made are a bit lost in these pics.

I'll try to get better ones on my next "work in  progress" shots.

 

..................Snipped

 

As mentioned previosly..................

 

Here at last is a picture of the "Bearing 'n' Lamp" item which Les Green has drawn up and had 3D printed for me.

post-3984-0-29220000-1413837057_thumb.jpg

 

Those with a long memory will recognise this signal is one of the pair for Woodford Halse which I've had on the go for several months.

It might be truer to say "on the back burner, but the gas had run out".

 

More when I get the moving parts added........

Steve.

  • Like 7
Link to post
Share on other sites

 

 

Stopping wires from fouling each other can be difficult, but thigs to try:

  • With two balance arms, keep the ends of the operating wires "pointing outwards", so the tight bends of the wires are adjacent.
  • Vary the distance of the wire connection from the balance arm pivot on each arm so they don't quite line up together.
  • Sometimes I use a loop on the end of the wire and pinch it up close to the balance arm to minimise the space taken up.

 

 

Thanks, again, Steve. My solution here http://www.rmweb.co.uk/community/index.php?/topic/26456-marlingford-begbrooke-the-difficult-signal/?p=1639623, using the first two of the above. Not pretty but seems to work alright on the workbench at least. To keep the arms apart I added three washers between them and made my own mount, the MSE ones not being wide enough or robust enough for my lack of finesse. The rods are cranked as the balance arms are too far outward otherwise. A bit of a bodge but I did notice this being done, though less extreme, in a couple of photos in Vaughan's "Pictorial Record of Great Western Signalling". 

 

I ended up retaining the ends of the wire by a dab of superglue, couldn't manage to squash the wire I was using (probably as it was piano wire!). The pliers have proved useful for lots of other things, though, so a good purchase nonetheless.

 

Thanks

 

Jon

Link to post
Share on other sites

  • RMweb Gold

Thanks, again, Steve. My solution here http://www.rmweb.co.uk/community/index.php?/topic/26456-marlingford-begbrooke-the-difficult-signal/?p=1639623, using the first two of the above. Not pretty but seems to work alright on the workbench at least. To keep the arms apart I added three washers between them and made my own mount, the MSE ones not being wide enough or robust enough for my lack of finesse. The rods are cranked as the balance arms are too far outward otherwise. A bit of a bodge but I did notice this being done, though less extreme, in a couple of photos in Vaughan's "Pictorial Record of Great Western Signalling". 

 

I ended up retaining the ends of the wire by a dab of superglue, couldn't manage to squash the wire I was using (probably as it was piano wire!). The pliers have proved useful for lots of other things, though, so a good purchase nonetheless.

 

Thanks

 

Jon

Hi Jon,

 

Well done, I'm glad my advice was helpful.

Rather than just using washers to space adjacent weight bars etc., try soldering the washers to the adjacent part.

This makes assembly much easier (fewer fiddly bits), and the extra thickness of material will reduce the "wobble".

 

Steve.

Link to post
Share on other sites

Those are superb Steve! Have you used those transfers you were showing us on the blades? The weathered stop arm on the r/h bracket looks spot on. If I ever revert to etched arms for my LMS upper quads I think I may try them.

Wonderful stuff.

JF

Link to post
Share on other sites

  • RMweb Gold

Hi Jon,

 

Yes, the transfers have been used.

I agree, they save a bit of time, particularly on Distants!

You still have to paint the edges of the arm as the transfers only cover the flat front and rear surfaces.

 

I'm told the GWR versions will be "ready shortly".

I'll need those for my next signal which is to be a three doll "Splitting Distant".

 

Steve.

 

p.s. Here's a video of those two being tested on my bench:

Woolford-Halse-brackets

 

Steve.

Link to post
Share on other sites

  • RMweb Gold

How the servos are interlocked.........

 

Each servo is controlled by a GF Controls board designed to give semaphore signals a realistic movement.

The instruction to move a signal is given to the GF Controller by an On-Off switch.

When the switch is "open", the Controller moves the signal to the "Clear" position.

When the switch is "closed", the Controller moves the signal to the "Danger" position.

 

Using a simple On-Off switch for each Signal is sufficient to make everything work, and it is up to the operator to do this correctly.

Such a system is minimum cost and complexity, but allows the operator to set the Signal Arm in incorrect or illegal ways.

e.g. The Distant arm could be "Clear" whilst the "Home" arm above is still at "Danger", or both Home arms could be at "Clear" at the same time.

 

By using multi-pole multi-throw switches, and applying a little logic to the way they are connected, it is possible to ensure that the signals are not able to display illegal combinations.

 

In this case, this has been achieved for each of the bracket signals by using Double-Pole Double-Throw switches for each signal arm.

 

This little board has the switches installed, with a connector block adjacent from which the wiring to the GF Controllers is installed:

post-3984-0-36111300-1414611337_thumb.jpg

 

The three switches to the left control one bracket and the other three on the right control the second.

These are installed in an identical manner, with the Left switch for the "Branch" arm, the Middle switch for the "Main" Home arm and the Right switch for the "Main" Distant arm.

 

The GF Controllers use a "common return" principle for the wiring, so the 12v DC Feed and Return are looped through each Controller (black/white pair of wires) at the Connector Block.

Because of the Common Return, the "-ve" power connection is also the "Return" for each switch.

This is made common within the GF Controllers.

 

Looking underneath the switches:

post-3984-0-24422100-1414611339_thumb.jpg

 

Here you can see I have used three colours for the "switch" connection for each servo.

Brown/Yellow for the Branch arms; Red for the Main Home arms and Yellow for the Main Distant arms.

In this shot you can just see a Green cable which is connecting the "-ve" power to the centre contact on each switch.

 

If you ignore all the White cables, this would be the simple Switching arrangement I described above.

The White wires are how the interlocking is achieved.

The switches I have used are "changeover switches".

This means that the centre conect on each switch is connected to either one or the other of the two outer contacts, depending on the position of the switch lever.

 

In the first photo, all the switches are in the "Danger" position.

In the second photo this means the top row of connectors are all connected to the "Common Return" causing all the servos to be in the "Danger" position.

 

If the leftmost switch is changed, this will open the connection between the Brown/Yellow cable and "Return", allowing the Branch arm to move to clear.

However, when this switch is moved, the white wires on the lower contacts of the switch are now connected to "Return", and at their other ends, these wires are soldered in turn to the Red and Yellow wires.

This means that whatever the position of the second or third switches, both the "Main" arms will be at "Danger".

 

The remaining white wires repeat this logic for the second and third swithes, so ensuring that if say the Main Home switch (middle switch) is moved to clear, it will connect the Brown/Yellow wire of the Branch signal to the Common Return, so forcing that signal to be at Danger.

The wiring of the "Distant is slightly different, in that it is done to ensure that the Main Home must be Clear for the Distant to be Clear.

 

When all this is connected to the three GF Controllers, we get:

post-3984-0-90946100-1414611340_thumb.jpg

 

All that remains is to connect a 12vDC supply to the small connector on the left, and plug the servo cables into the appropriate GF Controllers.

 

This was the arrangement I used when making the little video above.

 

I hope this hasn't been too boring........

 

Steve.

 

  • Like 1
Link to post
Share on other sites

Not boring Steve! It's something I may need to get my head round soon. It looks a little more complicated than the MERG boards as they are wired the "other way round"(make a contact to operate signal to off) . In the case of the MERG boards I have achieved simple slotting of stop and distants by using the extra contacts on the DPST on off switches I use.I believe the larger GFcontrols boards are selectable to operate either way but shamefully, I haven't properly investigated mine yet!

Cheers

JF

Link to post
Share on other sites

  • RMweb Gold

Hi Jon,

 

Yes, the GF Controls 8-way servo controller is switchable between Open for Clear and Closed for Clear.

 

Thinking about the method of implementing "Interlocking" as I've described it:

For the "Closed Switch" is Danger method used in my example, it is quite easy to add any condition that should hold a signal at danger.

There is essentially one wire going to the controller, and if anything connects that to the "Common Return", then the signal will be at Danger.

These are electrically "in parallel".

It is therefore a simple matter af adding extra switches which represent the "Danger" condition.

These could be such things as Point Motors, Track Occupancy etc.

 

With the "Open Switch" is Danger method, I think all the "Danger" conditions need to be in a chain (or in series electrically).

Then if any one of the conditions occurs, the chain will be broken and the signal locked at Danger.

With such a system, I think it is probably more difficult to add extra "Danger Conditions" over time, as they all have to be linked together.

Of course without trying it out its difficult to be sure.

I'm obviously influenced by my experience which has been exclusively with the method I described.

 

Steve.

Link to post
Share on other sites

Hi Steve, Ah, I can see how your method works with the GF boards now I've thought it through! Probably a better option for model railway operation.

With the "closed for clear" method that MERG uses, I'm used to looking at the situation where if a connection fails the signals return to danger. 

Cheers

JF

Link to post
Share on other sites

There is no reason why you couldn't use the MERG Servo4 either way round.

You simply set the signal positions to match your preferred switch positions, so 'closed' can mean Danger or Clear depending how you set the servo endstops.

Of course!...doh! I suppose it's just like choosing the setting for upper or lower quadrant.

JF

Link to post
Share on other sites

  • 3 weeks later...
  • RMweb Gold

I'm asking for help again............

 

Has anyone got a good drawing/photo of the Stevens Finial as used on the LSWR.

I require one or two key dimensions as well.

 

I've tried the MSE and Alan Gibson castings, but both suffer from distortions.

 

Hopefully,

Steve.

Link to post
Share on other sites

I

 

I'm asking for help again............

 

Has anyone got a good drawing/photo of the Stevens Finial as used on the LSWR.

I require one or two key dimensions as well.

 

I've tried the MSE and Alan Gibson castings, but both suffer from distortions.

 

Hopefully,

Steve.

Is there not anything useful in the Pryer book, Southern Signals?

JF

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share

×
×
  • Create New...