"Peter Corser" wrote in message
...
....
As an aside you may be aware that the train running numbers (set numbers)
shown in the time table were the direct digital equivalent of a hex
number - 477 digital was stored as 477 hex and 477 was the highest number
normally used (there was also nothing between x78 and x99) which also
saved bits (ot sequence machine fingers).

You may mean octal, not hex?
--
David Biddulph

"David Biddulph" groups [at] biddulph.org.uk wrote in message
...
"Peter Corser" wrote in message
...
...
As an aside you may be aware that the train running numbers (set numbers)
shown in the time table were the direct digital equivalent of a hex
number - 477 digital was stored as 477 hex and 477 was the highest number
normally used (there was also nothing between x78 and x99) which also
saved bits (ot sequence machine fingers).

You may mean octal, not hex?
--
David Biddulph

David

Oh Dear - brain still not in gear! Yes.

Peter
--
Peter & Elizabeth Corser
Leighton Buzzard, UK



----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==----
http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups
----= East and West-Coast Server Farms - Total Privacy via Encryption =----

"Peter Corser" wrote in message
...

The half minute timing accuracy was down to the programme (sequence)
machines stepping every half minute. The timetables on the machines ran
in half-minute time (from 0300 to 0300 the next day) with 0 at midday -
half minute time was the most you could do within the limits of a computer
integer (8 bit - +/-32767 IIRC).

The Central Line computer control ran internally to quarter minute
timings, but the Timetable software used in developing and printing of the
published timetables (and also the computer control timetables) could only
cope with half minute resolution.

Peter
--
Peter & Elizabeth Corser
Leighton Buzzard, UK

Think of the problems 'extended hours' is causing on programe machine sites.
(ok it not huge but running a train pass 3am is not that simple)
Andy

Joyce

Good point, but the time tables were still only half minute resolution even
though printed paper!.

The working timetables are prepared partly to attempt to run trains to a
specific timing (which allows crews to be rostered and train maintenance to
be planned), but also to inform the operator which trains had specific
priority over a junction. Timings were usually only given for sites with
signal cabins (later Interlocking Machine Rooms) where there was a specific
choice of route. This was usually at a station, but some "country"
junctions (e.g. Woodford Junction on the Central) also had timings. The
rest of the line would be worked by, in LUL terms, automatic signals
actuated purely by the passage of trains (some signal cabins were left with
their frames almost permanently in King Lever and unmanned). LUL practice
was to signal a station platform (and have a cabin if there was a junction)
using basically 2 aspect signals with repeaters as required. The layout was
somewhat different to main line practice with every signal having a
calculated (if nominal in real life) sighting point and overlap. At the
time of the 32 train service it is likely that the sighting points and
overlaps were calculated to the same "average performance" (nominal rush
hour traction voltage - it varies depending on the number of trains in
section and sub station settings and full load) curve which was a "perfect
driver" curve - full acceleration to max speed, maintain max speed and a
full, single brake application to rest in the platform.

The nominal signalled headway of the Central Line was always stated as 90
seconds which in normal LUL experience should allow running of a 2 minute
service taking into account service variations - normal driving rather than
perfect, traction voltage variations, station stops longer than the defined
value - usually 30 seconds allowed, but Bank & Liverpool Street "may" have
been 40 & 45 seconds). As long as the timetable stated that a train was due
it was telling the signalman to clear the signals - simply reduce the
interval between trains in the timetable to 90 seconds several times in an
hour at each point (not necessarily the same trains at every point) and you
have solved the problem. I suspect (without doing the sums) that 6 off 90
second intervals per hour would suffice. 32 tph was probably pushing it a
bit, in reality, if time was to be kept.

I was heavily involved (from the LUL end) in the resignalling of the Central
Line in the 1990s and, as part of deciding the strategy for the interim
signalling I did a survey of the headways on the whole of the Central Line
and found that a nominal 90 second did apply although most of the inner
areas were closer to 80 second and St Pauls and Liverpool Street were more
like 70 seconds with 30 second station stops. There were, however, several
section without station stops, which calculated to well in excess of 100
seconds (the max was 114 OTOMH) whic was probably due to permanent speed
restriction being imposed subsequent to th signalling and the latter not
being modified. We did the interim signalling with a max headway of 100
secs (although we usually achieved much less) through the "Central London"
area. The result amazed the Operating Department - running a full service
before we did the interim would usually result in late running of 30 to 45
mins at the end of the peak (they relied on cancellations to be anywehere
near time!), but our revisions allowed a full service to run pretty much to
time (and had most trains stopped in platforms if there was a problem), We
decided that this was probably due to our design holding trains in the
platform longer and then getting a clean, unchecked run to the next station
which prevented or reduced the cascading delays which could easily occur
with the earlier layout. I have since simulated it (on a non LUL simulator)
which tended to prove this.

The resignalled Central Line computers were set to receive timetable data in
15 second resolution (although the computer, themselves, ran on a much
faster set of clock cycles), but the timetables were still printed in half
minute resolution! The system worked in roughly the same way as the old
system with trains only being held to time at specific positions (e.g.
junctions) with the "automatic" signalling working on track occupancy in
between.

Peter
PS Clive Feather has a pretty good description of the CL resignalling on his
site.
--
Peter & Elizabeth Corser
Leighton Buzzard, UK
"Joyce Whitchurch" wrote in message
...
Peter Corser wrote:

The half minute timing accuracy was down to the programme (sequence)
machines stepping every half minute. The timetables on the machines ran
in half-minute time (from 0300 to 0300 the next day) with 0 at midday -
half minute time was the most you could do within the limits of a
computer integer (8 bit - +/-32767 IIRC).

A good job they didn't have computers back in the 1920s then - the Central
London Railway used to run its off-peak trains at intervals of one and
seven-eighths of a minute. (Or 32 trains an hour.)
--
Joyce Whitchurch, Stalybridge, UK
=================================



----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==----
http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups
----= East and West-Coast Server Farms - Total Privacy via Encryption =----

On Wed, 14 Mar 2007 11:11:01 -0000, Tim Roll-Pickering
wrote:

Mizter T wrote:

In the past I've picked up leaflets containing timetables for the
Metropolitan line, though I can't remember what stretches of the Met
they have covered - I suspect it was just the extremities, though they
also helpfully contained a line diagram that showed the pattern of
stopping, semi-fast and fast trains - something I can't find on the
TfL website.

I'm not sure about south of Baker Street, but I recall in the mid to late
1990s the Met used timetables on the boards not the "time between
stations"
posters used for other lines. I think they also incorporated Chiltern.


I concur. I've seen leaflets specific to Met stations West of Rayner's
Lane, but only ever seen them available at the stations concerned. They
are quite detailed about train times and destinations served. I also
think, though, that the Met is probably an anomaly, combining the sparsest
and most varied service. I think.

--
Fig

On Wed, 14 Mar 2007 11:11:01 -0000, Tim Roll-Pickering
wrote:

Mizter T wrote:

In the past I've picked up leaflets containing timetables for the
Metropolitan line, though I can't remember what stretches of the Met
they have covered - I suspect it was just the extremities, though they
also helpfully contained a line diagram that showed the pattern of
stopping, semi-fast and fast trains - something I can't find on the
TfL website.

I'm not sure about south of Baker Street, but I recall in the mid to late
1990s the Met used timetables on the boards not the "time between
stations"
posters used for other lines. I think they also incorporated Chiltern.

I would also like to confirm that I have seen "proper" Met timetables at
Wembley Park post 2004/5 modernisation

--
Fig

"Clive D. W. Feather" wrote in message
...
In article , Peter Corser
writes
As an aside you may be aware that the train running numbers (set numbers)
shown in the time table were the direct digital equivalent of a hex
number -
477 digital was stored as 477 hex and 477 was the highest number normally
used (there was also nothing between x78 and x99) which also saved bits
(ot
sequence machine fingers).

Actually, I think you'll find that numbers on programme machine controlled
lines only went up to 377 - there were 8 bits for train number.

The PM data I've seen wasn't organised by 8-bit byte, but simply had a
number of bits for each field. I thought that there were 6 minute bits, 5
hour bits, and a half-minute bit, but I could be misremembering here.

--
Clive D.W. Feather | Home:
Tel: +44 20 8495 6138 (work) | Web: http://www.davros.org
Fax: +44 870 051 9937 | Work:
Please reply to the Reply-To address, which is:

Clive

You are right on the 4xx series numbers - some lines did have them for empty
trains, etc. but not on PM controlled lines.

The train numbers were octal (what's the correct term fo 2 bits giving
0-3?), but displayed as decimal using the same numerals, so 377 would be 8
bits overall.

It was 2 second time I was thinking about with the byte format. Your bit
numbering for the half minute time held on the machine seems familiar. I
think my brain is definitely getting addled - it's a long time since I
actuall did PM design (original Heathrow Extension), computer control system
with PM monitoring (Cobourg Street), LUL style computer control (Met,
Jubilee & Bakerloo) and Central Line. I think things have got a little
confused and blended with time!

Even when LUL had moved to computer control there was a tendency to to
define data structures using bit masks on the data - the GEC (later GPT)
4xxx series central control computers had their own data laid out like this.
I remember that the date information internally within the machine was based
on 0 in 1972 (I think Jan 1st) and included a year field which counted up -
this was a 6 bit field. We all thought that 5 bits would probably have been
enough!

Peter
--
Peter & Elizabeth Corser
Leighton Buzzard, UK




----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==----
http://www.newsfeeds.com The #1 Newsgroup Service in the World! 100,000 Newsgroups
---= East/West-Coast Server Farms - Total Privacy via Encryption =---

In uk.transport.london message , Wed,
14 Mar 2007 23:12:08, Andy posted:

Think of the problems 'extended hours' is causing on programe machine sites.
(ok it not huge but running a train pass 3am is not that simple)

Have you a reference for the use of days ending at 03:00 ?

--
(c) John Stockton, Surrey, UK. Turnpike v6.05 IE 6.
Web URL:http://www.merlyn.demon.co.uk/ - w. FAQish topics, links, acronyms
PAS EXE etc : URL:http://www.merlyn.demon.co.uk/programs/ - see 00index.htm
Dates - miscdate.htm moredate.htm js-dates.htm pas-time.htm critdate.htm etc.

"Dr J R Stockton" wrote in message
. invalid...
In uk.transport.london message , Wed,
14 Mar 2007 23:12:08, Andy posted:

Think of the problems 'extended hours' is causing on programe machine
sites.
(ok it not huge but running a train pass 3am is not that simple)

Have you a reference for the use of days ending at 03:00 ?

--
(c) John Stockton, Surrey, UK. Turnpike v6.05 IE
6.
Web URL:http://www.merlyn.demon.co.uk/ - w. FAQish topics, links,
acronyms
PAS EXE etc : URL:http://www.merlyn.demon.co.uk/programs/ - see
00index.htm
Dates - miscdate.htm moredate.htm js-dates.htm pas-time.htm critdate.htm
etc.

John

No - It was like that whilst I was working in LUL (actually did programme
machine design and circuits, had a go at converting programme machines to
site computers, but couldn't get my head round it, did centralised computer
control for Northern, Vic, Met Main, Jubilee and Bakerloo plus major
involvement in Central Line resignalling). The logic was that all scheduled
services should be in depot by then and none of the next days scheduled
services should have started.

I left LUL signalling design in 1995 (after 25 years) and have no real
knowledge how things have changed since, but suspect that major changes
would be unlikely until lines have been resignalled by current contractors
with their own systems.

Working days have to end at some point in time (think also about ticket
availability not ending at midnight), but where the sweet point is is a
matter for conjecture - some lines did have trains starting from depot not
much after 4 am. It was possible to run extra trains with programme
machines using the normal controls, but I doubt if many programme machine
sites are still in use - spares were becoming extremely difficult to source
(most of the electronic cards used in the circuitry were originall hand
wired using discrete components of an earlier era and the machines
themselves were a throwback to an even earlier period) and they were being
replaced by simple electronic interlockings when I left.

Peter
--
Peter & Elizabeth Corser
Leighton Buzzard, UK




----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==----
http://www.newsfeeds.com The #1 Newsgroup Service in the World! 100,000 Newsgroups
---= East/West-Coast Server Farms - Total Privacy via Encryption =---

On Thu, 15 Mar 2007 11:21:57 -0000, Peter Corser wrote:

The train numbers were octal (what's the correct term fo 2 bits giving
0-3?),

Quaternary.