London Banter - View Single Post - Train-numbers on the LU

Richard Griffin · February 25th 04, 12:37 PM posted to uk.transport.london

(Thomas Crame) writes:

Richard Griffin wrote:
I have the answer, Roger: it is Binary-Coded Decimal.

So how come I've got an LUL engineering standard, last revised in 2000,
saying the codes are in octal numbering?

Because it's true for the Train Numbers themselves. Train Numbers ARE
Octal, being in base 8: as you know, they run from 0 (theoretically) up
to 7, and then jump straight to 10, and so on. I presume octal notation
was chosen to save having a nearly-redundant fourth bit (for the 8s
column) in two extra tracks in the programme machine.

However, at programme-machine level they are encoded in what I suppose
must be called Binary-Coded Octal. Binary, because a "hole" or a
"no-hole" can only store and provide information in binary form; and
"-Coded Octal" because each digit of the octal number is encoded
separately rather than treating the whole three-digit number as one unit.

Sorry for calling it Binary-Coded Decimal, where the Decimal bit was
misleading.

I'm speculating that using binary-coded digits (rather than pure binary)
may also have been chosen as it would interface better with Positive
Train Identification and its forerunners. Using primitive crew-operated
electromechanical devices to transmit a variable train number,
binary-coded digits would be much easier than pure binary (the latter
requiring either the operator, e.g. motorman, to enter his train number
in Binary (!) or for a computer to be present). The Westinghouse PTI
transmitting-equipment as fitted to 1959/72 stock was very simple.

Rewind time (or end of traffic day) on all LT centralised control
systems is 0300,

Thanks for that.

The programme machine is not fail-safe, and will not cause routes to
clear. What it will do is call routes, the decision as to if they will
clear or not is up to the safety signalling.

Sure. I should re-word the article to make it apparent that a hole for
"plat 3 to SB" doesn't actually clear the signal directly, merely
attempts to pull the relevant lever in the IMR when the electrical
interlocking allows. Is that correct? (Though AIUI, only some of the
Northern Line's ElectroPneumatic IMRs are still in use, others being
solid state replacements -- is this true?)

The other thing I was wondering is how does a programme machine at
Finchley Central call routes for trains departing from Edgware?

I didn't actually say it did, though I can see how you were mis-led. BTW
it was East Finchley (NP) IMR we visited. Roger has the answer I was
going to give.

Cheers,
Richard.
--
http://www.squarewheels.org.uk/