Semaphore Signal Control - The Bouncer

hughesp87
Posts: 137
Joined: Tue Aug 05, 2008 8:25 pm

Semaphore Signal Control - The Bouncer

Postby hughesp87 » Mon Nov 03, 2008 6:11 pm

The July 2007 issue of BRM contained an article on the use of servos as control mechanisms for semaphore signals, using a product marketed as 'The Bouncer'. I seem to remember a layout at Scaleforum using this product, although little was written about it at the time.

I'm interested in giving it a try. However the website http://www.copnor.enta.net/FS/TheBouncer.htm#The%20Products doesn't seem to have been updated for over 12 months, and attempts to e-mail the contact given have thus far failed to secure a response.

Does anyone know if they are still trading? Has anyone tried the product and are they willing to offer any views on ease of fitting and relaibility for an exhibition layout?

Thanks,

Geraint Hughes
Geraint Hughes
Cromford & High Peak in P4
Danish Railways in P87

DavidM
Posts: 74
Joined: Sun Jul 20, 2008 7:32 pm

Re: Semaphore Signal Control - The Bouncer

Postby DavidM » Tue Nov 04, 2008 7:11 pm

Geraint,

Keep trying - he sometimes takes a while to respond! I took delivery of one of his units about two months ago and it seems to work, but haven't connected it to a signal yet. I have been keen on trying servos for some time, especially for semaphore control and this seem to be a good way to do it. As soon as I have something worthwhile conclusions I'll post them.

David Murrell

hughesp87
Posts: 137
Joined: Tue Aug 05, 2008 8:25 pm

Re: Semaphore Signal Control - The Bouncer

Postby hughesp87 » Wed Nov 05, 2008 12:36 pm

Thanks David - that's reassuring!

Regards,

Geraint
Geraint Hughes
Cromford & High Peak in P4
Danish Railways in P87

Alan Turner
Posts: 643
Joined: Sun Jul 20, 2008 4:24 pm

Re: Semaphore Signal Control - The Bouncer

Postby Alan Turner » Fri Dec 05, 2008 5:27 pm

This looks very much like the MERG servo kits. It must have a modification to the software to make the bounce.

Alan

User avatar
grovenor-2685
Forum Team
Posts: 3918
Joined: Sun Jun 29, 2008 8:02 pm

Re: Semaphore Signal Control - The Bouncer

Postby grovenor-2685 » Fri Dec 05, 2008 10:13 pm

The supplier is a MERG member but a number of these units have sprung up in a similar timeframe.
The one referenced here, a kit sold within MERG by Bryan Knight which includes bounce features, and the MERG Servo4 kit which is aimed primarily at turnout drives but also has alternative software available for signal bouncing. I think at least one other similar unit is or has been on offer, ECM perhaps.
Regards
Keith
Regards
Keith
Grovenor Sidings

Alan Turner
Posts: 643
Joined: Sun Jul 20, 2008 4:24 pm

Re: Semaphore Signal Control - The Bouncer

Postby Alan Turner » Sat Dec 06, 2008 12:36 pm

Keith

where do you get the alternative software? I have been through the MERG site and I can't see it. Certainly not referenced in the TBs or the kit instructions.

Alan

User avatar
grovenor-2685
Forum Team
Posts: 3918
Joined: Sun Jun 29, 2008 8:02 pm

Re: Semaphore Signal Control - The Bouncer

Postby grovenor-2685 » Sat Dec 06, 2008 12:50 pm

Alan,
Refer to http://www.merg.org.uk/forum/viewtopic.php?f=15&t=95

Not really appropriate to continue this on here where we have only a handful of MERG members.
Regards
Regards
Keith
Grovenor Sidings

User avatar
grovenor-2685
Forum Team
Posts: 3918
Joined: Sun Jun 29, 2008 8:02 pm

Re: Semaphore Signal Control - The Bouncer

Postby grovenor-2685 » Tue Dec 29, 2009 10:17 am

I'm interested in giving it a try. However the website http://www.copnor.enta.net/FS/TheBounce ... 20Products doesn't seem to have been updated for over 12 months, and attempts to e-mail the contact given have thus far failed to secure a response.

Tony Witts has apparently moved to a new address and he has now updated his site
http://www.copnor.enta.net/FS/TheBouncer.htm
Regards
Keith
Regards
Keith
Grovenor Sidings

beast66606

Re: Semaphore Signal Control - The Bouncer

Postby beast66606 » Sun Jan 03, 2010 12:05 pm

> you are correct, elasticity in the wire is only relevant when it
>is under tension and hence when the signal is pulled off.
>when the signal is put back, then the weight pulls the wire back
>leaving the arm to drop onto its stop under gravity, how hard it drops
>and hence the amount of bounce depends on the weight being able to pull
>the wire back quickly, so signals near the box will usually bounce
>more. Note that LQ signals bounce differently as the arm itself is
>almost balanced and as the arm is rod connected to the weight bar the
>bouncing of the weight bar controls the bounce of the arm and the
>ensemble has several times the inertia of an UQ arm.



Before getting hung up on bouncing signals it might be worth me repeating a posting I made to the E4um list ... (answering the above comments) (please take note of the "generally" word)

*Generally* lower quadrants do NOT bounce, the arm is connected to the weight by a solid rod, so little or no bounce is possible, the post may rattle a bit though.
Again *generally* upper quadrants will bounce if the wire goes slack quickly enough for the arm to drop under gravity onto the stop, and bounce back and drop again etc but upper quadrants on wooden posts tend not to bounce as the post absorbs most of the "slack" - typically when such a signal is pulled the post will actually lean slightly so it's very difficult to get the lever back quickly enough to get as bounce on the arm.

The distance from the box only matters because the wire takes longer to go slack, but I have managed to bounce an isolated distant in my time, witnessed by the repeater going from off to on to wrong to on to wrong to on.

The best bounce I ever got was an ex LMS bracket, Hooton South, Down Ellesmere Port, Home 2 - which bounced, visibly, 18 times, and I could never get Port Sunlights up starter to bounce, ever.

***Of course if the signalman returned the lever slowly then the arm wouldn't bounce anyway ...***

From someone who has pulled many many levers and returned the same to danger, at speed, and attempted to get nice bounces out of upper and lower quadrant signals, brackets, straight post, steel and wooden.... (mostly LNWR / LMS / LMR but some of those funny GWR things)



Not knocking the product just mentioning that it's not every signal that bounces ...

User avatar
barhamd
Posts: 289
Joined: Sun Jul 20, 2008 2:45 pm

Servo drive unit.

Postby barhamd » Thu Jan 21, 2010 11:19 pm

I've just started playing with a Pololu Maestro USB servo controller. This has the ability to be programmed with a script from my PC and I've now got with an input from a switch which makes a servo give a suitable movement for a signal (with optional bounce). You can then set the unit to operate independently of the PC.

I reckon I should be able to make the unit detect 5 input switches and operate 5 servos.

The input side relies on me being able to remember ohm's law and calculate a chain of resistors in parallel so that 5 switches generate a voltage which gives an analogue representation of 5 bits. The input can handle the resolution well enough as it has 1024 input values, I would expect to need only 64 values.

<Techie Warning>
As the programming is in something which apparently looks a bit like FORTH, has a stack as the only data structure with 1K of storage for the code/stack keeping the code down to size will be fun. Mind you after the bloatware I work with most days it will make a nice change
</Techie Warning>

David Barham

martin goodall
Posts: 1425
Joined: Fri Oct 02, 2009 6:20 pm

Re: Semaphore Signal Control - The Bouncer

Postby martin goodall » Thu Jan 28, 2010 3:15 pm

"Generally, lower quadrants do NOT bounce, the arm is connected to the weight by a solid rod, so little or no bounce is possible, the post may rattle a bit though."

I have observed GWR Lower Quadrants, both as shown in the GWR Centenary Film and also on preserved lines. What happens is that the arm swings down well past the angle at which it will come to rest, then swings back (I am not sure you could really call it a 'bounce') before settling in the usual 'Off' position.

When the signal arm goes back to danger, it bounces very slightly against the stop (which I suggest would be almost imperceptible in 4mm scale).

I have no suggestion as to how this movement may best be replicated in model form. All I can say is that I have been unimpressed by mechanisms depending on a servo or cam. The two bouncing mechnisms that seem to work best are the horizontal pendulum device developed by Dr Mike Walshaw (seen on his Westport layout), which was written up in Model Railway Constructor many years ago, and the Vincent de Bode 'sweeping pendulum' mechanism (as modfied by Ray Hammond) which was decribed in MRJ some years ago. Vertically suspended pendulums don't seem to work so well.

User avatar
barhamd
Posts: 289
Joined: Sun Jul 20, 2008 2:45 pm

Re: Semaphore Signal Control - The Bouncer

Postby barhamd » Tue Feb 23, 2010 8:26 pm

I spent a fun weekend with brass tube and Model Signal Engineering etches.

The results of the servo bounce mechanism can been seen on youtube
sorry for the image quality, it is difficult to hold camera and operate switches at the same time.

2010-02-22 007.jpg


2010-02-22 008.jpg
You do not have the required permissions to view the files attached to this post.

User avatar
Will L
Posts: 2516
Joined: Sun Jul 20, 2008 3:54 pm

Re: Semaphore Signal Control - The Bouncer

Postby Will L » Wed Feb 24, 2010 6:36 pm

Very cleaver, but I'm sorry I'm not convinced. Could be a lot better at a significant multiple of times slower, but even then I'm not sure the bounce would look right.

Do remember you are under no obligation to take any notice of my opinions!

On Knutsford East our signal supremos have been through a variety of motions. This included one of those cleaver horizontal pendulum devices, which worked in prototype but obviously wasn't going to be replicated six times under a relatively small portable layout and which looked far to delicate to servive transport to and from one show let alone a layouts life time worth.

We too ended up using model aircraft servos and an electronic control package which can be programmed for bounce. The honest truth, however, is that because they are demountable and get uprooted each time we take the layout down, we have to reprogram them all each time we re-erect the layout and replace the signals. Any thoughts of sorting out bounce as well have receded rather.

Will

GMaslin

Re: Semaphore Signal Control - The Bouncer

Postby GMaslin » Fri Jun 05, 2015 5:09 pm

This thread seems to have been dead for more than five years but I touught I'd add my twopenn'th anyway.

I have used an Arduino microcontroller and a stepper motor to generate signal bounce. I've written an eight page description in MERG Journal vol 48 issue No 4. I've made the source code for programming the Arduino public domain, and it's available to MERG members via the MERG website. I'll happily add it to the Scalefour website if anybody's interested, or I'll email it to you. I'm g.maslin231 at btinternet.com.

I think the Arduino and stepper approach is better than using servos. Once you've played a bit with programming the Arduino you'll find it's very easy to learn and you can take my code, which provides one small bounce as the upper quadrant goes to clear, and two larger bounces as it returns to danger. You can easily add more or less bounces, introduce overshoot, add a mid way pause if the signalman's a bit weary, or make the arm move faster or slower as required.

Being digital, the Arduino/stepper approach is very precise so you can easily set the movement to within a fraction of a degree.

Arduino clones are ridiculously cheap. You can now get one for less than £3 post free from Hong Kong via eBay. Stepper motors are also available on eBay for about £2 each and come in various sizes. The smallest is only about 4mm in diameter so can easily be built into the base of a signal. Soldering to it is another matter as it's so tiny.

I'm also going to replace my point motors with Arduino/steppers as they are so precise, cheap and almost silent. Next on my list will be powering a turntable, as the digital precision means that it will consistently stop in exactly in the right place without the need for any position sensor. They're also good for level crossings.


Return to “Semaphore Signals”

Who is online

Users browsing this forum: No registered users and 0 guests