Angst Lesspork

[Hugh Norwood]

A Mechanical Semaphore system

Those of you who have seen Angst-Lesspork will be familiar with the optical semaphore tower on the layout. This was a version of the Murray system used by the (UK) Admiralty.

I had really wanted to make one with waving arms - if you search for the Popham semaphore you'll see what I mean. I thought it would probably be beyond my skills to make, but I have recently changed my mind...

Here is a sketch of a Popham tower near Portsmouth

taken from www.portsdown-tunnels.org.uk

And I found the codes that represent letters:


(from an old Encyclopaedia Britannica) Unfortunately, whoever created the lookup table did not differentiate between the upper and lower arm, so I have arbitrarily named the upper arms positions A..F rather than 1..6

arms	char	arms	char	arms	char	arms	char
1 2 3 4 5 6	1 2 3 4 5 6	A B C D E F	A B C D E F	A1 A2 A3 A4 A5 A6	7 8 9 0 G H	B1 B2 B3 B4 B5 B6	I K L M N O
C1 C2 C3 C4 C5 C6	P Q R S T U	D1 D2 D3 D4 D5 D6	V W X Y Z J	E1 E2 E3 E4 E5 E6		F1 F2 F3 F4 F5 F6

I also found that there was no "J" in the table - so I put it at the end.

[Hugh Norwood]

The semaphore station at Portsmouth had a mast between 30 and 50 feet high
The arms were 8 feet long by 1 foot 4 inches wide (32mm by 5mm in 4mm scale) and the arms fold inside mast.
The photos of the Chately Heath tower, however, give the impression that the arms are attached to one side of the post

Now, I recently got involved with the picAXE range of microcontrollers when developing a digital-only PWM train controller. When I bought the development breadboard, it came with a 18M2+ version of the controller.
I realised this had enough interface pins to let me connect up a PC keyboard to provide the "input" and two stepper motors to drive the two semaphore arms.

[Hugh Norwood]

OK, so what's a stepper motor? You ask.

A stepper motor has a shaft sticking out one end - like a "normal" motor, and it does rotate. BUT instead of going round smoothly, it moves in a series of steps, determined by how the current is switched through the coils inside the motor. The direction of stepping is determined by the sequence in which you drive the coils.

A full treatise on these devices isn't needed here; I chose a pair of "unipolar" motors (stocked by the firm that supplies the picAXE range of microcontrollers) These particular ones have 200 steps per revolution.

The advantage to me was that 25 steps would bring the semaphore arm to the lower quadrant position (1|6|A|F), 50 to the horizontal positions (2|5|B|E) and 75 steps to the upper quadrant positions (3|4|C|D).

It's time to take a few photos...

[MickT]

Hugh,

You're obviously well read up on these things, so you may be able to answer my query?
Where I live in Blandford, Dorset, there was a relay station on the site of the current Defence School of CIS on the Admiralty Shutter Telegraph system on the route from London to Plymouth. It is recorded that they were able to get a message from the Admiralty to Plymouth 230 miles away and receive a reply in approximately 4 minutes, which is not a bad feat for a totally optical manual communication system 200 years ago in my book!

Question: was the Popham Semaphore that you described here using the same shutter telegraph system on the Admiralty to Portsmouth route?

[tedp]

There is some interesting information about optical telegraphs and martello towers on Wikipedia:

optical telegraphs

martello towers

As you may know, the two were closely linked in the defence systems of Napoleonic times.

[Hugh Norwood]

Hi Mick,
As far as I can see the term "shutter telegraph" refers to the Murray system of two rows of three (or three rows of two) shutters that are either vertical (showing their face) or horizontal (showing an edge). It appears that the Popham system replaced the Murray, but I think I might have to do some more reading to give a definitive answer.

[Hugh Norwood]

Here is a photo of the two stepper motors fixed together for testing:

[Hugh Norwood]

And here is a short video of the arms in action.
See if you can decode the message. NOTE you are looking from the rear so the arms are back to front...


http://youtu.be/RZwXq13EMoA

:)

[DeadHorse]

Excellent stuff.
Is there a way to speed things up to make them appear more 'manic'?
In my imagination the ones in the book are thrashing about like demented tic tac men at a race course. :)

[Hugh Norwood]

Yes - I purposely put a delay in during development, and have reduced it quite a bit. I think I'll have to reduce it even further.
I have received the brass strip to make the actual semaphore post and have ordered pulleys and drive belts. So we're cooking on charcoal! Dragging the layout into the century of the Fruitbat (or whatever it is at present)

[garrywhiting]

Is there a way to speed things up to make them appear more 'manic'?
In my imagination the ones in the book are thrashing about like demented tic tac men at a race course. :)

I don't think that the arms "thrashed about" at that sort of speed. The arms would have been quite heavy and it must have been a physically demanding job operating the pulleys that moved the arms. I am sure that I read many years ago that there were teams of operators who alternated every 10 minutes because they got tired out so quickly.
Although fictional the second story in Keith Roberts 1968 alternate history fix-up novel "Pavane" called "The Signaller" includes details of the physical effort required to opreate the semaphore arms in this type of signalling station.

Garry

[Hugh Norwood]

Quite so. Also, in the Discworld books, the clacks rapidly evolves from a (supposed) two-arm system to the shutter system with an unknown number of boxes. I thought the 4x4 matrix used in the TV adaptation of "Going Postal" was a bit over the top - they have to be read by human operators, after all. The Murray system of 6 shutters (3 rows of two) probably represents a workable compromise between readability and "density" of code. After all, to transmit 26 characters and 10 numbers, you don't need that many individual codes. It's only when you start to do strange tricks that you need more.

(Monstrous Regiment indicates coding grey-scale images for example)

And remember the original ASCII code was 7-bit and only coded A-Z, a-z, 0-9, a few punctuation marks and some non-printing codes.

[Graeme Dunbar]

The ASCII code is still 7 bits (128 characters) though an 8th bit can be used for an error (parity) check bit in serial transmission or various non-standard things (graphics characters in LCD panels etc).

Predating ASCII is the Baudot code (as in the same person recognised in "Baud rate") which is a 5 bit code:
http://en.wikipedia.org/wiki/Baudot_code

If you get really adventurous in your programming then there are all sorts of possibilities such as signal arm bouncing.

Many years ago when I used to teach HND students including a number of mature students (some really quite mature) from the oil industry we had a multiple choice section in the introductory computer engineering module. As one of the "distractors" for a question along the lines of:

From the following statements select the one which best describes the ASCII code.

I included the (obviously wrong) choice:

The ASCII code was invented by Arthur Ascii.

It was amusing to see one of the older part-timers laugh out loud much to the puzzlement of the younger full-time students in the exam.

[Hugh Norwood]

I used to be able to read punched tape (by eye, I mean!) Sad innit?

[tedp]

I used to be able to read punched tape (by eye, I mean!) Sad innit?

Don't tell me you go back that long, Hugh...    ;D