Scalefour Forum

Pannier Tank wrote:Some very nice examples of the Society's Lever Frame in this thread.

I would appreciate some advice on the best technique in laminting the 3 piece Lever Handles and advice on solders / fluxes to use etc.

Hi David,

I built the lever frame and wrote the article that was carried in MRJ 225. This is what I said in the article when I reached the lamination stage:

"Next comes assembling the levers, laminated from three layers of nickel silver. Not only are there two different components, but they are handed as well, so care must be taken to fit them together the correct way round. The tip in the instructions of using scraps of wire to align them works very well, and meant that I spotted the one layer that I had placed back-to-front on the fifth and final lever just before I applied the solder!

My initial thought on reading the instructions and realising that I would be soldering three quite thick layers of nickel silver together was that I would need to be using a Resistance Soldering Unit, cranked up to a fairly heavy duty setting. However I chose to follow the guidance given, soldering around the edges with a conventional iron, With the aid of plenty of flux, capillary action did indeed draw solder between the layers and make up a very strong lever. I should mention that the entire assembly was completed using only 179° solder and a 50W iron set to 375° temperature."

The whole article was written from the perspective of identifying the little tricks and tips that weren't obvious from the (already comprehensive) instructions. You may find it useful if you can track down a copy. You'll find pieces in there from a number of Society members, including both our past and present New Member Support Officers

HTH
Flymo

If it helps, I screwed mine tight through the bottom hole and a really strong clips at the top. The screw should be long enough to go into a block of wood (I had some oak).
Then lots of liquid solder, supplemented at the edges by standard solder.
Plus a bucket of cold water with a bit of ice in it! Gold star for guessing the reason for this.
Once these are done, removing the screw again needs a load of heat, but the rest of the assembly will not come undone without a blowtorch! Then touch up where needed.

Of course, it does help if the layers are the right way around!

I did find that even this meant that I still had to do quite a bit if trimming on the flat to allow the moveable bits - to move.

John

I have built two of these for use on the Society's stand. (Why two? One got pinched so we needed another. )

My technique was to tin (apply flux then apply a thin layer of solder) the surfaces of the surfaces that join, assemble as John says with a bolt through the pivot hole and a clamp at one end. I do not have a RSU but used a standard 25 watt iron and heater the middle section first to melt the solder. I then worked towards t\he ends - the clamped end first, then I could remove the bolt and work towards the pivot end. The final steam was to run solder around the edges. Using small off cuts of wood to hold the parts avoids the need for a bucket of iced water.

Terry Bendall

Many thanks for all the advice, now I have no excuse!

I am just finishing a lever frame bought at Scalefour North. It does take time, though, most of it filing off the etching cusp but the results are well worth that time.

I made the lever by lining up and holding the 3 layers in a vice, before I removed the etch cusp, put on some Carrs yellow flux and ran the iron along the edge with a small amount of 145o solder. I turned it over, did the same to the other side and then got going with the file. I can see no reason for anything more complicated or higher melting point solder.

The instructions take a bit of deciphering. Incidentally, the significance of 1.4mm (mentioned in the instructions) as the diameter of the top of the handle when you have filed it, is that this is the diameter required to make a 10BA thread. I also drilled out the handles (which have a 1mm hole) with a 1.4mm bit and tapped them.

I have a suggestion for the mark 1 leaver Frame. I found the first 10 leavers a real pain to file down to leave an adequate spigot for the turned handle.

As I am now extending this to to 20 or 25 leavers. I have done the following:-
1 using a 1.4mm Drill bit opened out the hole in the Turned handle. I do this by clamping the handle in a vice and using a dremel Be careful they get rather warm!( i measured the neck of the handle and it is about 1.8mm at the thinest point so only went as far as 1.4mm as any larger may weaken the handle too much)

2 when putting the Levers together I cut the shoulder off the outer 2 layers as the spigot is 1.4mm and the 3 layers together are 1.6mm... let alone the standard problem of square peg to round hole.

This allows the handle to be a tight fit on the spigot and as I solder the levers on it allows the solder to flow around the base and into the hole.
I have also noticed that the L sections at the bottom can be a little over length so I have been soldering the quadrant plate to the side plates first then checking everything is square and then adjusting the L's length to fit.

The lock boxes I still find take a fair bit of fettling to work. However I had more work this time as I deepened the fold lines with a file (also did this to the side plates and these formed better 90 degree bends). So after about 4hours yesterday, all 10 levers are made and cleaned up, the quadrant plates installed to the existing frame of 10. Along with the 10 lock boxes fabricated and soldered.

I’m just about to go to a pcb for a microswitch solution for the Mk2 lever frame , this fits under the frame and also provides a base plate for the frame

I’ll post soon with details

Personally with electric switching , I see no point in mechanically locking the frame, electrically interlocking is easier more flexible and easily updated

junctionmad wrote:
Personally with electric switching , I see no point in mechanically locking the frame, electrically interlocking is easier more flexible and easily updated

Hi Dave,

Not sure if you have actually tried this approach yet - or are you just assuming it to be easier?

Have you decided on a way of physically locking the levers yet or are you planning to just have an indicator light?

These are the locks I developed for the MkII frame:-



They point alternately "up and down" so that they can cope with the 10mm pitch of the levers even though the locks themselves are 20mm wide. I confess they were a bit of a design challenge. These locks give both a front lock and a choice of backlocks - "none", "full back" or "'B' position". As you imply, they cannot be fitted to a frame with a mechanical take-off from the lock tails or the top clevis, but you can see that they allow space for the fitting of the microswitches.

And here is the first version of the "release panel" - the signal man pushes the button for a release, and when the lock "picks-up" the green LED lights up. If it fails to light then the signalman has forgotten something...



On this frame there are a LOT of electric locks (32 on 66 working levers) because we are replicating the full block-control functionality of Midland Rotary Block and also full locking via the track circuits of running signals and facing points, and if I were doing the thing again I would do things like the sequential locking electrically as that is certainly easier than Lifting Tappets, and - because it might be easier to maintain - I might yet make the switch.

But how are you intending to provide the electrical logic? I agree that if you can do it in software and have the programming skills, then it is much easier and I did originally look at using a Raspberry Pi or and Arduino to run the logic. But I rejected the idea in the end because I like bright shiny metal! (and the interfacing has its challenges)

But I would not consider the use of relays to be any easier - in the contrary, the number of relays goes up with the square of the number of levers And lever driven contacts are non-starter for obvious reasons.. I know of one 50-odd lever frame, the relay locking for which takes up three drawers - and it is not "complete" locking by any means ...

That particular frame is not physically locked, rather it just has indicator lights and a warning buzzer and one problem is that from time to time, the buzzer goes off and cannot be stopped - it happens because the (inexperienced) signalman moves a "locked" lever, but before he has time to react to the buzzer he moves a second lever - now there are two mistakes and that causes a lot of confusion! Especially when all the other operators are by now taking the mickey!
(how do I know this...)

Equally, I know of another layout with over 400 levers in 5 signal boxes - it is locked in software and that is certainly the only practical way in that case! But the chap who did that lot has a lot of expertise in how to do such things!

Good luck and looking forward to hear how it goes.

Best Wishes,

My intention is to implement simple “mechanical “ interlocking , electronically. The software is custom and runs on a processor directly interfaced to the switches on the frame. All I will use is an independent light ( bi color ) led , at the foot of each lever , moving a lever incorrectly , will cause the led to light red , green meaning the lock is not in operation

Given my points are controlled electronically , using CBUS, hence it’s easy to prevent points or signals actually moving , I see no real point in physically locking the levers

Ps I agree one of the issues is how to communicate multiple mistakes in locking to the operator , but clearly resetting all levers with a red led , will restore the state of the system

I’m an embedded systems engineer , so the technology is fully accessible to me

Dave

junctionmad wrote:
I’m an embedded systems engineer , so the technology is fully accessible to me

Hi Dave,

Ah - that makes a huge difference. Even though that might mean that such a solution will not be available to every one, I hope you will still post details of how it is done as I think plenty of people would like to go the electrical/electronic route, and so far I have not seen a complete write-up of such an approach.

Best Wishes,

JFS wrote:

junctionmad wrote:
I’m an embedded systems engineer , so the technology is fully accessible to me

Hi Dave,

Ah - that makes a huge difference. Even though that might mean that such a solution will not be available to every one, I hope you will still post details of how it is done as I think plenty of people would like to go the electrical/electronic route, and so far I have not seen a complete write-up of such an approach.

Best Wishes,

Sure , Ill do that

Right now the first step is a new PCB to hold microswitches for the Mk2 lever frame , for the 50 lever setup I'm building

Then I have to integrate that into a fully illuminated mimic diagram , that already exists , then add 20 working ground signals ( 3D printed ) and 8 running signals

Then I do the interlocking ........


dave

Well as it has been a challange of mine to get something done on a regular basis 2 months worth of "doing things" in June I finally finished the 10 Lever frame I started years ago. Then looking at my P4 layout which I have been planning and doing things on and off for years I realized this needed to be at least double and may be another 5... so I thought I will buy another 3 kits and build the next 2 and think about the last 5.

Well last month I finished the next 10 and took the build to the local group meeting where every one was impressed but thought it will need to be longer! So the last 5 levers took me about 4 hours to complete compared to the first 5. Any how here is how the frame turned out:-



I have to say that once you get the hang of the old mark 1 frames and make your life easier as I stated above (drill out the turned handles to 1.4mm, and cut the 1 outer laminates tops off) you can get very fast at putting them together!

Would I do more? Yes however there is a practical length with the mark 1 which you really can't exceed as getting the pivots all the way through becomes progressively more difficult along with unwieldy to build. There is also a tendency for the unit to banana up, I think i may have 0.5mm upward camber in mine. The other thing is this beast above starts to weight a fair bit!

Merry Christmas to all.

then add 20 working ground signals ( 3D printed )

Doug,
I've just come across this thread whilst looking for anyone's shared experience on making working ground signals. You mention 3D printed. Did they work out for you? I'm a real novice and have no experience with such things but, having now got my mechanically locked frame up and running I need to get on with the ground signals.
I'm using servos powered by Megapoints Controllers Servo controllers so as to get the bounce and for ease of use. (See 'novice')
The single ground signal that I have working I made from a kit some time ago and can't remember the make. It has a white metal body and etched disc.
Now that I am starting anew I need a further 11 signals and I am experimenting with incorporating LEDs.
I've tried one Ratio kit but removing light leakage with a plastic kit is challenging as is making it robust enough to stand the movement.
The other one I've made from scratch, albeit with the Ratio lamp. I've managed to make quite a decent disc, (surprisingly) but I'm sure an etched one would be better.
Is there any knowledge that you would be willing to share?
Regards,
Tony

Tonycardall wrote:
The single ground signal that I have working I made from a kit some time ago and can't remember the make. It has a white metal body and etched disc.

Sounds like a MSE one (model signal engineering)

jim s-w wrote:

Tonycardall wrote:
The single ground signal that I have working I made from a kit some time ago and can't remember the make. It has a white metal body and etched disc.

Sounds like a MSE one (model signal engineering)

That makes sense. Thanks Jim. I’ll place an order right now.
Tony

Worth looking at Steve Hewitt's topics on RMweb. Excellent signal building and includes Ground Signals
eg. https://www.rmweb.co.uk/community/index.php?/topic/34064-working-ground-signals/&do=findComment&comment=361235

I've built a couple of these recently, starting with the MSE kits. Ended up ditching the whitemetal middle bit, and scratchbuilding that from some U-section brass (cut down some rectangular tube to get the section needed). Much easier than drilling holes in whitemetal !

- Nigel

And there is always this item from the Scalefour Stores
'PM1941LMS'

This is one of my LMS ground signals which Keith is referring to. Steve Hewitt and Steven Freeman swear by these and you can get them via the stores of from the Palatine Models site.

Ralph