Signals

THE SIGNAL BOX


Railway signalling discussion

Interesting computer game

For railway modellers to discuss and share ideas on realistic signalling arrangements.

Re: Interesting computer game

Unread postby John Hinson » Wed Aug 24, 2016 12:53 pm

kbarber wrote:(Just to confuse the issue, some non-peggers had a drop-handle intended to be used to flick the needles to give routing information, disconnected by the 1930s I believe but not always removed and some remained in situ until the end.)

Yes, these were used in certain areas where special routing bell codes were later used, i.e. areas with many junctions. I would be tempted to say all remained in situ but perhaps I should cover myself with the word nearly.

But in terms of the computer model, I doubt they would need to be provided so a plain front to the (missing) lower part is probably relevant for that.

John
Image
‹(•¿•)›
User avatar
John Hinson
Site Admin
Site Admin
 
Posts: 6840
Joined: Thu Nov 8, 2007 1:13 pm
Location: at my computer

Re: Interesting computer game

Unread postby John Hinson » Fri Sep 2, 2016 5:20 pm

Elstree box, showing telephone, train register desk, clock and Belling cooker complete with kettle - plus a woolly hat. Don't ask . . .
Image
Photo: John Hinson, December 1975

John
Image
‹(•¿•)›
User avatar
John Hinson
Site Admin
Site Admin
 
Posts: 6840
Joined: Thu Nov 8, 2007 1:13 pm
Location: at my computer

Re: Interesting computer game

Unread postby Mike Hodgson » Fri Sep 2, 2016 5:45 pm

Good to see a photo of the back wall of a box also showing a lot of other little details which would be useful to modellers, or to anybody building a replica, as they show it as a working environment, such as the keys hanging on nails, coat hook, a clipboard of papers and some sort of pen rack above the desk.
The boarding runs vertically on the back wall, but horizontally above the end window.
User avatar
Mike Hodgson
Main line box
Main line box
 
Posts: 2478
Joined: Fri Nov 9, 2007 5:30 pm
Location: N Herts

Re: Interesting computer game

Unread postby JRB » Fri Sep 2, 2016 9:00 pm

You have till October to visit Banbury North and photograph everything. It's all being kept looking authentic until October. Book your visit NOW! (Advt.)
JRB
Double-manned box
Double-manned box
 
Posts: 3208
Joined: Sun Dec 9, 2007 10:12 pm
Location: GWR

Re: Interesting computer game

Unread postby John Hinson » Sat Sep 3, 2016 3:38 am

Banbury North is not a Midland Railway box though. I have been carefully selecting pictures relevant to the original computer model. Each railway company had its own ways of doing things.

John
Image
‹(•¿•)›
User avatar
John Hinson
Site Admin
Site Admin
 
Posts: 6840
Joined: Thu Nov 8, 2007 1:13 pm
Location: at my computer

Re: Interesting computer game

Unread postby JRB » Sat Sep 3, 2016 7:57 am

True, but it is accessible and many features are typical. St. Albans South would, of course, be even better.
JRB
Double-manned box
Double-manned box
 
Posts: 3208
Joined: Sun Dec 9, 2007 10:12 pm
Location: GWR

Re: Interesting computer game

Unread postby John Webb » Sat Sep 3, 2016 8:40 am

JRB wrote:True, but it is accessible and many features are typical. St. Albans South would, of course, be even better.

Extended opening (10am to 5pm) at St Albans South this coming weekend, 10th/11th September, for the national "Heritage Open Days" event.
John Webb
(Member, St Albans Signalbox Trust)
User avatar
John Webb
Main line box
Main line box
 
Posts: 1825
Joined: Tue Jul 8, 2008 1:22 pm
Location: St Albans, Herts

Re: Interesting computer game

Unread postby Paul_G » Sat Sep 3, 2016 10:26 am

Hi Guys

Many thanks for your input, most valuable.

Seeking some clarification on a particular bit of the bellcode sequence.

Train entering section is 2 beats that is acknowleged with 2 beats ?

If this is correct it is causing me a problem with the logic ATM in the particular circumstance of.........

Having accepted trains on both the up and down lines more or less at the same time,
I have belled ahead that train is entering section and he has acknowleged... up line
He then sends me 2 bells that his train is entering my section ... down line
To which I need to reply 2 bells.

The problem that I have is that my logic on the unmanned AI boxes works by auto responce
Whether a responce is req is done by checking the whether the last code transmitted equals the last code received..
If it does then do not respond ( because the incoming message is itself a responce to my last outgoing message).

It all works fine except for the above case.

So my questions are is 2 bells and 2 bell responce the normal practice?
Is there an acceptable alternative eg call attention then send the 2 bells ?

To me it seems strange that the system should have this ambiguity in it..
If the second box gives me 2 bells is he acknowleging my train TES or telling me his train is TES?

Many thanks again.. things are getting closer

Paul G
Paul_G
Crossing box
Crossing box
 
Posts: 21
Joined: Tue Aug 23, 2016 12:57 am

Re: Interesting computer game

Unread postby Chris Osment » Sat Sep 3, 2016 4:14 pm

Paul_G wrote:Whether a responce is req is done by checking the whether the last code transmitted equals the last code received..
If it does then do not respond ( because the incoming message is itself a responce to my last outgoing message).


But what do you do if Box A sends (say) TES, but then B replies with 6 bells ?

If the second box gives me 2 bells is he acknowleging my train TES or telling me his train is TES?

One would expect him to acknowledge a signal before then sending one of his own. Also, if he is acknowledging your TES, then he will turn his block to TOL, whereas if he were not then he wouldn't :-)
Chris Osment
West Country Railway Archives
http://www.railwest.org.uk
User avatar
Chris Osment
Main line box
Main line box
 
Posts: 2251
Joined: Sun Nov 11, 2007 9:32 am
Location: Somewhere in the West Country

Re: Interesting computer game

Unread postby John Hinson » Sat Sep 3, 2016 4:16 pm

2 bells and 2 bell response are indeed the normal practice.

In the real world you would know whether the adjacent box was acknowledging or sending TES by whether he turned the needle to Train on Line or not. But whether that helps with the computer logic I'm not sure, I am no expert in that sort of thing. If you can detect that through needle position you may need to allow a few seconds for the operator to turn the needle before assuming it is not an acknowledgement.

Does that make sense? It is easier to think it in my mind than to write it down!

John
Image
‹(•¿•)›
User avatar
John Hinson
Site Admin
Site Admin
 
Posts: 6840
Joined: Thu Nov 8, 2007 1:13 pm
Location: at my computer

Re: Interesting computer game

Unread postby John Webb » Sat Sep 3, 2016 6:38 pm

Our simulator at St Albans South can cope with TES for both lines in succession - but of course in one case we are responding to the computer's output by both bell code and switching to ToL, in the second case we are giving the computer just the tapper input and it has to respond back.
In visiting Banbury North the other week, they showed a video taken of the box operating before closure, it seemed it was very common for bell codes to be exchanged in both directions in close succession.
John Webb
(Member, St Albans Signalbox Trust)
User avatar
John Webb
Main line box
Main line box
 
Posts: 1825
Joined: Tue Jul 8, 2008 1:22 pm
Location: St Albans, Herts

Re: Interesting computer game

Unread postby Paul_G » Sat Sep 3, 2016 7:20 pm

John Hinson wrote:2 bells and 2 bell response are indeed the normal practice.

In the real world you would know whether the adjacent box was acknowledging or sending TES by whether he turned the needle to Train on Line or not. But whether that helps with the computer logic I'm not sure, I am no expert in that sort of thing. If you can detect that through needle position you may need to allow a few seconds for the operator to turn the needle before assuming it is not an acknowledgement.

Does that make sense? It is easier to think it in my mind than to write it down!

John


Hi John

Yes it does make sense to me, altho most will think we are talkin in riddles :-)
I can indeed take the Train Online output , although I avoided using it because I perhaps wrongly assumed that the 2 bells should be acknowledged before moving the Block Instrument to TOL.
If it is permissable to move to TOL before / at same time as sending 2 bells then this would remove the ambiguity I mentioned.

Maybe just to point out that the logic for a single box is relatively straight forward,
I am attempting to make a modular system that will auto connect boxes as players join and will switch to auto operation if they become unmanned, hence the auto reply logic.

Thanks again

Paul G
Paul_G
Crossing box
Crossing box
 
Posts: 21
Joined: Tue Aug 23, 2016 12:57 am

Re: Interesting computer game

Unread postby John Hinson » Sat Sep 3, 2016 8:11 pm

Paul_G wrote:I can indeed take the Train Online output , although I avoided using it because I perhaps wrongly assumed that the 2 bells should be acknowledged before moving the Block Instrument to TOL.
If it is permissable to move to TOL before / at same time as sending 2 bells then this would remove the ambiguity I mentioned.

I don't think there is a hard and fast rule on the order you do it, some people do it one way and some the other so I don't see any issue with your making it work that way.

John
Image
‹(•¿•)›
User avatar
John Hinson
Site Admin
Site Admin
 
Posts: 6840
Joined: Thu Nov 8, 2007 1:13 pm
Location: at my computer

Re: Interesting computer game

Unread postby Mike Hodgson » Sat Sep 3, 2016 9:30 pm

The logic required for sending and receiving bell codes is quite complex and involve timings of various gaps (between the "dings", between groups of dings, allowing for the other end to respond etc)

In order to recognise any bell code you receive, you have to allow a brief pause to decide that what you have received is 2 bells (rather than say the first part of 2-2, engine assisting in rear). The same goes for distinguishing 3-1 from 3-1-1. So you have to allow a pause anyway to identify completion of the code. After that pause, If your pegging block is showing Line Clear and you receive 2 bells (without prior Call Attention and without your having sent a code in the other direction) you can assume he is sending the code to you (even if your non-pegger is also at Line Clear).

If your non-pegger is at Line Clear and you want to send 2 bells to notify him that you have a train entering the section, you should wait until you have completed any sequence of codes initiated at the other end (slightly longer gap than the interval between the groups of 'dings' which make up a code). In other words, if he has sent Call Attention, you should acknowledge that if you have not already done so, and if he has not yet offered a train, you should expect that (or certain other codes such as 2-1) to be offered within a short period of your acknowledgment of his Call Attention before you send him 2 bells for your train to him. Even after accepting a train, you should not treat his sequence as necessarily complete - he could for example send 2-2, if applicable.

The logic gets even more complex when you add the emergency codes, permissive working/warning arrangement and dealing with situations such as train failures.

A problem software has to be able to handle is an incorrect/incomplete sequence. If he sends Call Attention but fails to offer a train within a reasonable period of your having acknowledged that, you may have to treat that as a mistake, and expect him to call attention again when he does want to offer a train. There is also the question of how you deal with errors such as bell codes incorrectly acknowledged, failure by the other end to move the block indicator as required etc.
User avatar
Mike Hodgson
Main line box
Main line box
 
Posts: 2478
Joined: Fri Nov 9, 2007 5:30 pm
Location: N Herts

Re: Interesting computer game

Unread postby Pete2320 » Sun Sep 4, 2016 11:16 pm

Mike Hodgson wrote:Even after accepting a train, you should not treat his sequence as necessarily complete - he could for example send 2-2, if applicable.

But that would be sent after TES has been sent and acknowledged.

Pete
Pete
Pete2320
Main line box
Main line box
 
Posts: 2114
Joined: Wed Feb 20, 2008 10:50 pm
Location: Barton on Humber

PreviousNext

Return to Signalling - model railways/simulators

Who is online

Users browsing this forum: No registered users and 1 guest