Jon Fitness' average 7mm signals workbench.

[Guest jonte]

Hello Jon,

 

Just a line to congratulate and thank you for providing this thoroughly detailed and informative account of your building methods.

 

Having recently had a bash at fettling up a couple of signals in oo gauge (they function but are not a patch on yours), I have nothing but the highest regard for your work - which, incidentally, you appear to make achievable for us lesser mortals. At least you've inspired me to have a go when the time arrives.

 

As far as I'm concerned, there's mothing average about your work: it's incredible. Look forward to further instalments.

 

Jonte

[Jon Fitness]

Hello Jon,

 

Just a line to congratulate and thank you for providing this thoroughly detailed and informative account of your building methods.

 

Having recently had a bash at fettling up a couple of signals in oo gauge (they function but are not a patch on yours), I have nothing but the highest regard for your work - which, incidentally, you appear to make achievable for us lesser mortals. At least you've inspired me to have a go when the time arrives.

 

As far as I'm concerned, there's mothing average about your work: it's incredible. Look forward to further instalments.

 

Jonte

 

Cheers Jonte, and thanks for the compliments . Nice to know it's worthwhile.

Most of what I do in 7mm is the same as 4mm as the etched kits are scaled up or down from the same artwork!

For some seriously spectacular signal building check ou Mikemegs "Hessle Haven" signals thread (I'd put a link in but I'm not sure how to!)

Have a go at some signals and post up the results, it might encourage others to have a go1

Jon F.

[Guest jonte]

Thanks for the pointer, Jon and encouragement - stuck for time at present but in the next few weeks I'll give it a go and post up the results as you suggest. Incidentally, it's Midland round-topped I require so perhaps you can point me in the right direction for scale drawings?

 

With regard to those link thingies: that makes two of us !!! Comforting to know I'm not alone on that score

 

Best wishes,

 

Jonte

[Jon Fitness]

Thanks for the pointer, Jon and encouragement - stuck for time at present but in the next few weeks I'll give it a go and post up the results as you suggest. Incidentally, it's Midland round-topped I require so perhaps you can point me in the right direction for scale drawings?

 

With regard to those link thingies: that makes two of us !!! Comforting to know I'm not alone on that score

 

Best wishes,

 

Jonte

 

 

Hmm, I can provide you with drawings for most LMS U/Q stuff but you may have to ask around for Midland .

 

Meanwhile have a look in any books you have covering the MR area. Once you start building signals, pictures you previously saw rolling stock on, mysteriously start sprouting signals you hadn't seen before.

Jon F.

[Jon Fitness]

You’ll have realised by now that the photos in this signal building ramble are of 2 different signals. They are probably slightly out of sequence as well. I remembered things on one that should have been photographed and took later pictures that illustrated a particular bit a little better!. You know what I mean, I'm sure!



We have a post, an arm and a lamp, so it's time for a linkage or drive rod to work the thing.

Drill a hole about 0.8mm in the base-plate, roughly in line with the little operating arm on the signal.



I use some steel wire about 24swg (it's actually guide wire from a Faller Roadway set) but 0.7mm N/S wire is fine for the job. Cut a 2-3mm length of tube that will fit over the wire (you could some of the pivot tube from earlier) and put to one side while you shape the wire as shown.




Couple the wire to the signal arm, thread the piece of tube onto the wire and feed the wire down through the base-plate.





Position the signal arm in its pivot and set to the "up" position so that the LED is in line with what will be the green aspect.

Allow the little bit of tube to rest on the baseplate and carefully solder the tube to the wire (not to the base-plate, like I do sometimes!)



You now have a limiter that will stop the signal being damaged by rising too high. At a later stage a similar one will be soldered under the base-plate so that the drive rod won't bow out if pushed upwards too hard by the mechanism you choose to operate it with.

Generally, a real signal will need a weight bar or balance weight to help it return to danger. In model form, unless your engineering is of a very high quality, a weight bar is of little use other than to simply be there for correct appearance.

Being a lazy sort I tend to make them non working as they just add another place where mechanical "play" can occur. On these signals I have made them moveable but will bypass them for the sake of mechanical simplicity (cue sharp intake of breath from purists!!)

In etched brass kits they usually have to be soldered in layers to build sufficient thickness to look right.

Tin both sides of the weight bar with a thin layer of solder and clamp the layers together or hold them down on a bit of scrap wood.

I have used some mini wooden pegs of the sort used to hang up Christmas cards on string. I got these from a Well Known Pound Shop. Very useful little things for this sort of work.




The weightbar is positioned a scale 4" above rail level and you will need to drill a 1mm hole (at 90° to the signal pivot) right through the post taking care not to damage the wire inside the post!



If you are making a signal from a kit you'll probably have a pivot bracket for the weight bar but if not, make up a small angle bracket from a short length of 2-3mm wide brass as shown in the pictures and drill 1mm for a pivot pin. Assemble the weight bar and bracket with a bolt or pin through the post and solder the bracket to the post.




If you can, solder a nut to the post so you can pass a bolt through everything. It makes maintenance or repairs easier.

Check that the weightbar doesn't foul the operating rod when all re-assembled.

Ladders Next!

More soon

Jon F.






 

Edited June 23, 2016 by Jon Fitness

[Guest jonte]

Jon

 

Thank you for the kind offer of signal drawings - nearer the time I should like to email you for details if it's not too much trouble of course - incidentally, the signals I have in mind are, I thik, classed as LMS.

 

Glad to see further posted developments in signal construction; excellent idea about soldering pieces of tubuing to the operating wire to limit arm movement. Hints on how to retain small bits and piecs whilst soldering are especially appreciated. Feel like getting stuck in right now.

 

Looking forward to more.

 

Jonte

[Jon Fitness]

Ok, ladders.

I am using etched nickel-silver ladders from Scale Signal supply on these signals which are of a fine section but only equivalent to about 20' in height. Normally for anything taller I would use the brass ladders from D&S but as the person I am building for didn't mind a join I lengthened an SSS one (only a little bit!!) with a bit left over from a previous job.

Trim the ladder to be joined and the chosen extension so the pitch is constant between rungs when placed in line. Cut another small piece that can overlap the join and carefully align all 3 pieces. Clamp them together with one of the previously mentioned mini pegs (told you they were useful!)

Apply liquid flux and solder them with a little 188° solder and re-check for straightness.

The ladder stands off from the post and is attached about 3mm down from the top using a U shaped stay. This can be bent up in pliers so that it matches the width of the ladder. The stand off must be sufficient to allow the backblinder to swing behind the lamp casing without fouling lamp or ladder.

Carefully solder the ladder stay to the post with 145° solder but don't linger with the iron or you may disturb the solder on the pivot tube.

Set the prepared ladder against the base and the ladder stay at the top. The actual inclination of the ladder is 1 in 12 but bear in mind that you need space for the weight bar to swing at the bottom. Once you are happy with the alignment and checking the ladder is not bowed, solder it to the stays at the top and base-plate using 145° solder and plenty of flux.


The ladder will need extra supports halfway down so use a short piece of 0.7mm wire or those offcuts from the LEDs mentioned earlier and provide 1 each side of the post.

You can align them against the post and to the inside edge of the ladder for ease of location whilst soldering. I solder using 188° at the ladder end and 145° against the post. Trim any excess off.


At the same position I have added the "Diamond" sign to the post. This sign on a signal indicates to the traincrew that the presence of their train is indicated electrically in the signalbox and is fitted where appropriate to the location.

At the top of the ladder is a loop. This for the safety of the maintenance staff when attending to the lamp or other fittings and is position in line with the back of the lamp.

It is about a scale 20" diameter or about 11mm.

With the SSS ladders is enough etched strip to make about 6 of these so snip a 30mm length of and bend it round a pen or tube of about 10mm diameter. Pinch the ends together as shown in the picture then open the loop out slightly so that it will clip into the sides of the ladder.



Solder the loop to the ladder in line with the lamp. (it should be horizontal) and trim off any excess from the back of the ladder.



You can now have a little "Topping out" ceremony by way of a post cap.

I use either the ones from MSE or SSS depending which ones I have in stock.

This week I will be mostly using SSS post caps! (With apologies to fans of the Fast Show on TV)





Almost there now. ( Did I hear a collective sigh of relief?)

Painting and final assembly next .

More soon

Jon F.








 

Edited June 23, 2016 by Jon Fitness

[Guest jonte]

Thanks for yet another masterclass Jon - Now it's almost complete I can safely say that that's the type I need for what I have in mind.

 

In fact, if there are actually any more hours left in your busy day, I wonder would you mind checking the website for 'Disused stations - Subterannea Britannica' under 'Liscard & Poulton'. There's a photo of the station which shows a round topped signal post just like yours - I'm certain it's one and the same but it would be reassuring to have an 'expert' opinion.

 

Look forward to the next instalment.

 

Jonte

[Jon Fitness]

Hi Jonte,

Seen the picture you have mentioned and I think it may be an older wooden post with an upper quadrant signal fitted to it.

Cheers

Jon F

[Guest jonte]

Yup, that's the one, Jon.

 

Please forgive my ignorance, but could this be represented using brass tube and the same parts as per your build sequence?

 

Sorry to keep distracting you and hope I'm not stalling your progress!!

 

Most grateful for spareing time to assist.

 

Best wishes,

 

Jonte

[Jon Fitness]

Yup, that's the one, Jon.

 

Please forgive my ignorance, but could this be represented using brass tube and the same parts as per your build sequence?

 

 

Jonte

 

 

Jonte,

If it is a wooden post, no, not accurately as the post will be square in section and taper gently towards the top. Scale Signal Supply and MSE both do kits for this type of signal so no problem with parts. And no problem with asking questions and sharing info either as that's (as I see it) what we're all on RMWeb for .

Cheers

Jon F.

[Guest jonte]

Thanks for clearing that up Jon.

 

I had no idea it was a tapered post - this site can really be an education Perhaps I could use modellers license and replace it with a round post ?!!! We'll see.

 

Kind regards and apologies for the late response.

 

Jonte

[Jon Fitness]

Where were we? Ah yes, time to splash a bit of paint on this signal.

Mask the LED off in the lamp casing with either a little square of tape or a blob of Maskol.

The signal arms can be set into a stick using the axle shafts or bolts so they can be painted without having to handle them.

Give the arms and the post several good coats of white primer. I use Halfords white spray primer as this tends to bond well to both brass and whitemetal. It then becomes the topcoat for the white bits and a good undercoat for any other colour you need (especially red!)



Set these aside to dry for a good 24 hours as you will be applying

masking tape next and you need the paint to be hard.



Paint the black areas front and back on the spectacle plate. (The little black oval is to cover the red spec on the calling on arm. Forgot to fit it before painting!)

The white stripes on LMS upper quadrants are 5mm wide and set in 7mm from the end of the arm. Mask that area off with a good quality tape (I use Tamiya) and paint the blade or arm red. Taking off the tape before the red paint is fully dry should leave a nice clean white stripe.



When the paint has set you can fit the coloured lenses. If using LEDs to illuminate things, true red and green coloured lenses should be used. If grain of wheat bulbs are used then a bluey-green lens should be used for green.

These can be cut easily from a sheet with a hole punch (just right for full size U/Q arms) or a leather punch as size dictates.



These can then be fastened behind the appropriate openings on the signal with a small amount of superglue.



The black stripes can be painted on by masking either side of the area (same size and position as the front) and the tape removed as the paint starts to dry.



Carefully mask the post as shown, including a little bit behind the lamp. This will make it easier to paint the lamp without getting black where you don't need it



Back blinders are fitted on the other end of the signal shaft. Their function (on the real thing) is to cover the little light at the back of the lamp when the signal is up or even slightly up. The signalman observing the signal at night can then see if the signal has gone back to danger when he replaces the lever in the frame. If the light shows, all is ok, if no light is seen he must presume the signal has not gone back to danger (or the lamp's gone out!).



It should now be time for the final re-assembly of the now painted parts. Thread the drive rod through the signal arm then down through the baseplate while pushing the spindle through the bearing on the arm. With the arm at danger postion the back blinder as shown and secure with a little EvoStick so that, if necessary, it can be dismantled for repairs or maintenance at a later date.

Should be just about finished now!



All that remains to be done is to solder a couple of short lengths of tube to the drive rods underneath the base-plate to limit upward movement.



Methods of operation are many, ranging from manual to DCC controlled slow action point motors and I’ll leave you to choose your own but...

I have been tinkering with R/C servos and a control board from MERG which I will pick up on next.



Go build..




 

Edited June 23, 2016 by Jon Fitness

[Pete Harvey]

Hi Jon

 

I know that you have been building signal for a very long time now so please do not take what I'm going to say the wrong way.

 

I understood that all semaphore signals has one glass in red and one in blue due that being illuminated buy a oil or electric light being yellow in colour did not effect the red but when behind blue glass was seen as green buy an approaching train?

 

I know that you use white led to illuminate the your signals but would it not be more correct to use a yellow Led and clear red & Blue?

 

I'm not trying to nit pick as I think your work is great.

 

Pete

[Jon Fitness]

Hi Jon

 

 

 

I know that you use white led to illuminate the your signals but would it not be more correct to use a yellow Led and clear red & Blue?

 

 

 

Pete

 

 

Pete

Fair point, but yellow LEDs tend to be too yellow, affecting the red and turning it orange.

I should try the "warm" white LEDs but they are not as easy to come by or as cheap(most important) as the 3mm white ones. (Unless someone knows differently)

Cheers

JF.

PS. whats involved in getting a 7mm scale Pratt Truss LMS gantry etched?

 

 

 

[beast66606]

As Pete said, just my view and nice signals but I think you'd be better using lamps (rather than LEDs) for illumination.

[Jon Fitness]

Finally got back to the servo project after a few distractions so here we go.
The signal itself is a bit of a lash up. It used to be a bracket on the old Talacre railway but is now a gantry, supported on one side by a retaining wall. The main post is tubular instead of "H" section girder but we'll gloss over that. I just hate wasting previous work I guess and it fitted the bill in it's new position.
There are 4 arms to operate and my old method of operation would have been solenoids. I liked them as I could leave the signals "on" or "off" without them requiring any constant power but the downsides were:-
1. Too violent for delicate mechanisms.
2. Too fast a movement, very jerky.
3. Quite bulky under the baseboard.

Once I started learning about using R/C type servos and the progress made in miniaturisation I decided to try these out. I purchased the bits (see previous posts) and made a start.
First job is to make a rig to stand the signal on, emulating the position it's going to be installed in. I've made sure the baseboard is the same thickness as the layout.


I then held the servos roughly in position to establish the best way to set them. I took off the servo arms and moved them round on the splined shafts to a more suitable position



All seems to be in order and I'm using these

to attach the drive rods to the servo arms.

More soon

Edited June 23, 2016 by Jon Fitness

[Jon Fitness]

As Pete said, just my view and nice signals but I think you'd be better using lamps (rather than LEDs) for illumination.

Hmm. Thanks but no thanks.

Personal preference for LEDs.

Bulbs; too big, too hot, too likely to go pop in my experience. (Yes I've tried them before)

Still waiting for my first LED to go.

JF

[Guest Digital]

Cracking tutorial Jon very interesting and very clearly explained.

John

[Jon Fitness]

Cracking tutorial Jon very interesting and very clearly explained.

John

 

Thanks John

 

On the layout where the signal is going to be installed there is insufficient room to take a chunk of the board out and make a complete "drop in" unit.

The servos will therefore be mounted on a bracket, the signal drive rods fed through a small hole below the main post and the servo bracket with all the gubbins on "offered up".

Obviously mucho bench testing will take place first with the signal on its rig but as long as the baseboard and the rig board are the same thickness, there shouldn't be too many mechanical problems .

 

Half an hour in the garage with a few bits of 9mm ply and a drill&bandsaw produced this.

The servos on the right had to be set back a bit to line up with the drive rods (one slightly less than the other) but I still have a range of adjustment forwards/backwards of about 6-7mm. Hopefully that will be enough for it all to line up.

The servos are quite a tight fit in the slots and don't seem to move when the arms are moved manually. I'll have to dig up some tiny screws if they do work loose!

 

 

More soon

Jon F.

[Guest jonte]

Jon

 

Excellent tutorial and inspiring stuff. Thank you.

 

It's clear from your jottings that it was intended, primarily, to encourage those of us wishing to make the leap into scratchbuilding but lacking the confidence and wherewihall to do so. By using easy to follow methods, non-specialised tools and readily available materials, in my upinion, you've succeeded and I for one will certainly be giving your methods a go in the not too distant future - not sure I'll reach your lofty standards but I'll try.

 

Whist no doubt well intended (I respectfully refer to those recent 'luke warm' submissions by more experienced modellers than me) that the authors should bear this in mind before commenting. I think as Jon has made clear, they're not intended as faithful reproductions but to give a flavour of the real thing whilst creating a robust working model - how many layouts are there with 'non-operating signals'?. Numerous articles have appeared in the modelling press over the years but none has succeeded in persuading me to have a bash - the tools alone required to create these miniature masterpieces has brought me out in a rash !! When it gets to the stage of semantics, I think that Jon's target audience will swiftly move onto the next topic which regrettably defeats the object. No wish to cause offence here, but I feel compelled to make the point in Jon's defence.

 

Intrigued by the servo stuff, Jon.

 

Best wishes,

 

Jonte

[Jon Fitness]

Jonte

Thanks for all that but I welcome all comments of any type as long as they are "on subject" of course and I construed the comments as constructive and valid. (Wouldn't be on here otherwise)

That rather long winded tutorial was intended for anyone to extract anything from as they need it.

It's not a definitive lesson, just shows how I go about things.

Carry on ....

 

Jon F.

[3 link]

Jon,

 

Just a quick line to say thank you for a most informative and enjoyable topic, which I am following with much interest.

 

All the best, Martyn.

[Jon Fitness]

Cheers Martyn!



Once I'd mounted the servos on the bracket I offered it up to the signal rig.

First mistake; right hand servos too far over. I kinked the drive rods over by about ¼inch and had another go and it all just about lined up. This had the lucky benefit of making more space around the servo arms and making it easier to couple it all up.

I have to bear in mind that when the signal is installed and I have to install the servo assembly, I'll be grubbing about under a baseboard, not sat at a workbench!



The bracket was clamped in place under the rig, and all the drive rods fed through the connections on the servo arms. I tentatively connected a servo to the driver board and powered it up. Second mistake:-

I had presumed I had correctly found the central position of the servos when I remounted the arms, but the servo board disagreed and tried it's best to bend things! When you first power up the driver board it sets all connected servos to the centre position, so.. off with the arms again, power them all up and re-set the arms where the servo board wants me to set them.

Once all that was done and driver board screwed to the bracket, all 4 servos were re-connected electrically to the board and the board programmed up with the setting box. This device tells each servo connection:-

1.Where the ON position is

2. Where the OFF position is

3. How fast to rotate from ON to OFF

4. How fast to rotate from OFF back to ON

Sadly there is no bounce option with the setting box but I believe there is the provision to program it in if you use a computer with the appropriate software.

Once it was all set up, a test was carried out on each signal and there seemed to be quite a bit of buzz and chatter in the ranks and this was found to be due to odd movements in the mechanics as things found their positions. This caused the servos in one or two cases to be pushing the mechanisms against the mechanical stops. The buzzing was the servos straining to reach their programmed positions but being mechanically prevented

A bit of fine tuning was required to sort this out and all 4 settled down to perform well over about 10 minutes of random testing, all using nothing more than a 9v PP3 battery for now. The MERG instructions recommend using 10-15vdc so further testing will take place at a voltage available at the layout.



More soon

JF

Edited June 23, 2016 by Jon Fitness

[Jon Fitness]

I've installed the gantry on the layout now and left it with the resident "tester" who'll give it a good playing with over the next week or so. Luckily I didn't need to re-set any of the servos with the box but the signal itself may need a few tweaks as the structure has "adjusted" itself as I've tightened the fixing screws. (You can see the small shunt arms are standing slightly off)

I have about 14volts going into it at the moment from a controller but I may reduce this later I have also connected the feed to the signals LEDs to the same feed (with a 1k resistor on it) and there is no trace of flicker while the servos move.

Each activation of the signals brings on a very slight buzzing and twitching in the ranks but this seems to settle to silence after about 15 secs.

All in all, very satisfying project after a shaky start (all down to my tripping over the learning curve a bit)

Ready to do a few more now.

Jon F.