New signals for my layout

[G.M.R.]

I showed my previous attempts at constructing semaphores within Steve's topic, but now I'm starting some new ones, with the intention of following some of his methods, especially in regard to using servos. I've purchased a dozen:-

 

 

And a controller:-

 

 

I have yet to try them out, but will put a simple test rig together for that purpose.

 

In the meantime I have gathered some more MSE etches and started on the "ironwork" for the first signal I need. The signal will control a single line branch as it enters my central junction station. Trains on the branch have a choice of four routes:- The Main down platform, the down loop platform, the up loop platform where trains from the branch can reverse and follow the up main or return to the branch, and finally the goods marshalling yard.

 

So the signal will be bracketed with three main dolls, and a small goods doll. This is the main superstructure:-

 

 

Not sure if there is a prototype for this, but I've seen similar pictures. The three main dolls will be on the main deck, with the goods on the small left hand bracket.

 

I'm now the proud owner of a thousand of the pins which Steve uses in many places, and you can see that I have used them for the decking stanchions. I will trim them down to size after I have fitted the handrail, but first I need to build and fit the operating cranks to the deck.

 

The dolls will drop into bushes already in place under the deck. I will start work on them next.

[G.M.R.]

A little progress. I have started to prepare the dolls, and just temporarily assembled them on to the bracket to see how it looks. The main one has arm pivots and lamp brackets in place, and a bracket and pivot for the operating cranks. I have then made up the connection between the lower arm and post pivot to check it will work ok. It will be removed for painting.

 

 

 

[G.M.R.]

I have been trying out the servo controller in its semaphore bounce mode. Works very well when the servo movement is set to max, but not sure about it when set to a smaller movement of the horn. These signals are not going to need much movement from the servo, so wonder if the bounce will be more of a complete up and down of the arm instead of just a bottom bounce. Will have to see when I am able to connect the signals to the servos. Lots to do yet to get to that stage.

[Steve Hewitt]

I have been trying out the servo controller in its semaphore bounce mode. Works very well when the servo movement is set to max, but not sure about it when set to a smaller movement of the horn. These signals are not going to need much movement from the servo, so wonder if the bounce will be more of a complete up and down of the arm instead of just a bottom bounce. Will have to see when I am able to connect the signals to the servos. Lots to do yet to get to that stage.

Hi Geoff,

 

For the bounce effect to be seen you must ensure a minimum "play" in all the linkages between the servo and signal arm.

Also, to get maximum Horn movement, ensure your connection to the horn is at the shortest radius.

 

Good luck,

Steve.

[G.M.R.]

I decided to have a go at preparing the base of the signal for the servo connections following Steve's methods as much as possible. As I do not have access to modelling power tools other than a Dremel, I have assembled things just a little differently.

 

Firstly I attached the MSE supplied base plate to the lattice post:-

 

 

Actually, I lied. The first thing I did was to solder a 12BA bolt through a hole in the base plate, and then attached it to the post. The plate is cut to take the tabs from the post, but it is fairly small, and I wanted something longer to help with assembly to the servos. So rather than trying to cut slots for the tabs in another plate, I simply prepared the plate you see in the photo ready to fix to the smaller base using the bolt.

 

I then made an identical sized plate and used it to provide an anchor for the linkage guidance tubes as described by Steve:-

 

 

The guidance tubes are 1/16", and I will be using 0.8mm tube on the control wires as I have found they slide smoothly through these guidance tubes. I have added a 3mm thin wall tube which allows the nut from the previously mentioned 12BA bolt to exactly locate inside.

 

 

Although these components assemble snugly, I have utilised the longer base plates by adding two more 12BA bolts to hold it all together. I have done something similar for the other signals already on the layout, and simply need to drill a couple of clearance holes in the baseboard for the protruding threads.

 

And finally here is what it will look like with the half inch tube in place. This tube ready as Steve shows, for lining up the servo assembly accurately with the control wires:-

 

 

Now I need to get back to the dolls and do a trial assembly of cranks and balance weights.

[Steve Hewitt]

Hi Geoff,

 

That's coming along very nicely.

I hope it all works as well for you as it does for me.

 

"Imitation is the sincerest form of flattery"

 

Thanks,

Steve.

[G.M.R.]

Made a start with the balance levers and their pivot. In the past I have just used a 16BA bolt soldered to the post and then assembled the levers onto the thread of the bolt. However, I like the look of the finer components used by Steve.

 

So I tried making a bracket for the levers out of square tube. That worked fine, but no matter what I did, I could not find a way of assembling three balance levers inside the bracket. Therefore, I have compromised:-

 

 

 

I have used a 0.45mm shaft but have soldered this permanently into the post. I can then assemble the levers and spacers onto the end of the shaft and finish by soldering the now dummy support bracket on to the end of the shaft.

 

I think it will look okay.

 

Also in the past I have soldered two balance levers from the etch back to back for strength. I haven't done that here, but they are very flimsy so I may have to go back to doubling up.

[G.M.R.]

Reconstructed these balance levers with back to back etches, and fabricated a different end support:-

 

 

I think that looks much better. I now have a slowly growing number of small parts:-

 

 

I hope it didn't take this long to construct the real thing!

[G.M.R.]

Finally plucked up courage:-

 

 

Despite being quite happy at diy plumbing, including using a blow torch for solder joints, I had never contemplated the idea of using a torch for model soldering until I read Steve's posts. I purchased a small torch a couple of weeks ago and have been looking at it each day on my modelling bench, unsure if I could use it without destroying all my previous work.

 

So, as I said, I finally plucked up courage to give it a go. I tried a test piece first, and then the real thing. Simples! I wonder why I was so worried. I used a metal block under the work to help conduct away unwanted heat just in case. The joint is perfect, even if I say so myself, and the other joints are completely undisturbed.

 

I have been a member here since the forum started way back, and I am so grateful to the sharing of methods and ideas that have continued to help me use my average modelling skills to achieve satisfying results. On this occasion, thank you Steve.

[G.M.R.]

Phew! Sometimes I could do with three hands and tiny fingers. The linkage for the left hand goods line signal is extremely limited:-

 

 

 

At least it works. The arm in the photo is just temporary to get the linkage correct. The actual arm, being a small one, is very fragile around the spectacles so I won't put it in place until the last minute.

 

I was going to remove the doll for painting, but it was so difficult getting the linkage in place, it is going to stay put now and will just have to be painted in situ.

 

I'm hoping the linkages for the main dolls will be easier. Using thread over pulleys on my previous earlier gantry was a whole lot simpler than assembling these cranks and links!

[uax6]

Looks very impressive work there.

 

One thing, wouldn't that bracket on the goods line doll, at the bottom, be higher up the post so that the pull between it and the crank for the down pull would be square, not at a really tight angle?

 

Andy G

[G.M.R.]

I'm sure you are correct Andy. Lack of forethought on my behalf and maybe I should have made a sketch first.

[G.M.R.]

A few more hours spent reworking this doll and linkage and perhaps a better look:-

 

 

[G.M.R.]

A quick question if anyone can help. Colour of ladders on top of signal brackets against round dolls? I've seen photos of both black and white. Both being in the Eastern region, which is where my model lies. I wonder if it is a matter of history? Did they change colour at some time? My model period is late fifties, early sixties.

[Jon Fitness]

I generally stick to painting them black.

Jon F.

[Steve Hewitt]

I also paint them black, except the lower four feet or so near the ground where I often paint them white.

Steve.

[G.M.R.]

Whilst there is still a long way to go with this bracket signal, as I have to wait for supply of some of the white metal items, I have been doing some research for another bracket which will control the other line entering my junction station.

 

This line approaches from a tunnel so sighting of the signal is very limited. I know that my layout is a long way from anything prototypical, but something like this:-

 

Attribution: Ben Brooksbank

 

would look half realistic, and would be very interesting to construct.

 

Question - how would the lamp man get to the outer arm? Also, I wonder if a distant arm could be added to that outer doll?

 

This signal, also at Witham, also provides some food for thought:-

 

http://www.davidheyscollection.com/userimages/gec-alstom-d200-witham.jpg

 

Grateful for any thoughts.

[Steve Hewitt]

Whilst there is still a long way to go with this bracket signal, as I have to wait for supply of some of the white metal items, I have been doing some research for another bracket which will control the other line entering my junction station.

 

This line approaches from a tunnel so sighting of the signal is very limited. I know that my layout is a long way from anything prototypical, but something like this:-

 

image.jpeg

Attribution: Ben Brooksbank

 

would look half realistic, and would be very interesting to construct.

 

Question - how would the lamp man get to the outer arm? Also, I wonder if a distant arm could be added to that outer doll?

 

This signal, also at Witham, also provides some food for thought:-

 

http://www.davidheyscollection.com/userimages/gec-alstom-d200-witham.jpg

 

Grateful for any thoughts.

Geoff,

It looks like there is a suspended ladder down behind the doll to a sort of platform?

Steve.

[G.M.R.]

Most of the bits are now coming together:-

 

 

Most of the time when on the layout, only the rear of the bracket will be clearly visible, so I'm not worrying about working lamps this time. My previous ones have used LED lamps from Express Models intended for locos, but look fine on a signal:-

 

[G.M.R.]

Moving to the underside:-

 

 

Hopefully still correctly following my tutor's guidance.

[G.M.R.]

Just realised while looking at the last photo that with the servo board this way round, I will have to work almost blind when fitting it under the baseboard. It needs to face the other way relative to the signal. Maybe not too difficult to change at this stage. Just need to move one servo from right to left and spin the signal through 180deg. I think?

[G.M.R.]

Nope! Doesn't work:-

 

 

The control tubes are off centre front to back. So it's back to the workbench to cut another top panel for this servo assembly. Not tonight though. Too much RW consumed with my evening meal!

[G.M.R.]

After a great deal of blood sweat and tears, I'm getting close:-

 

 

My previous memory wire driven signal gantries had the balance levers mounted on the dolls with thread representing control wires connecting the levers to the memory wire motors under the baseboard. So this is the first time I have tried to mount a number of balance levers at the base of the bracket support post. The issue has been the linkages between the levers and the signal arms.

 

With two cranks required for all but one arm, and not much space, it has taken me several attempts to get everything "tuned" correctly and running smoothly. I originally used 0.3mm wire rod for the down linkages and they work fine for the goods arm and the left loop arm (which does not need any cranks), but for the three arms to the right, the 0.3mm wire was too bendy and so the arms did not return to the off state reliably.

 

That meant a fairly major refit to use 0.4mm wire instead. It's amazing what difference 0.1mm makes!

 

I have a few ladders to make up, and am still in need of a final signal lamp, but apart from that it is just a matter of connecting the operating tubes to the servos.

 

They are all in place ready:-

 

 

I have completed the programming of the Arduino.

 

 

The Arduino board is to the left, and the servo controller is on the right. The board in between contains a chip which will monitor 16 inputs. These are from the 8 turnouts, 7 occupancy detectors and distant signal which form my central station South Junction. There will be 12 semaphores (5 on the bracket I'm building now), two colour light junction signals, and a pair of colour light split distant signals. The semaphores will all be servo driven from the servo controller, and the colour lights will be driven directly from the Arduino as they are all LEDs. The patch board seen in the photo on top of the Arduino contains three LEDs as a test of the software. The two banks of blue switches are also temporary, and simulate the 16 inputs, again to test the software which appears to be working as it should.

 

May be a while yet before it is all in place on the layout!

[Jon Fitness]

Lovely looking job Geoff! A nice characterful signal and well done with all the linking up. It's a big enough PITA in 7mm scale and even then (being lazy) I usually cheat and bypass the weight bars to simplify the job. 7mm also has the size advantage in that I can mount the servos under the base plate and install the signal as a complete tested and working assembly.

Having said that though there's some lovely little servos out there which are less than half the size of the SG90's and are just as effective. They'd probably fit under a 4mm signal (Dave Fenton from Megapoints has one under his Dapol Ngauge demo signal!)

Check this link  and you'll see what I mean.

Have a good xmas and keep up the excellent work

Jon F.

[G.M.R.]

Thanks for the kind remarks and the link Jon. The servos I'm using are from Hobbyking so I might try those micro ones on a single or two arm post. I need to make one for the yard exit.