|
|
Post by angelislington on Apr 12, 2009 23:04:04 GMT
Hiya gang, MRFS is exercising his plentiful brain on the subject of overlaps just lately. He tells me that there are no level sections on the system anywhere, other than some platforms. Is this really true? Is the system really totally uppydowny?  xxAI |
|
Tom
Administrator
Signalfel?
Posts: 4,198 
|
Post by Tom on Apr 13, 2009 15:15:01 GMT
I don't think so - I'm sure I calculated some overlaps in the Wood Lane area that included some level bits.
|
|
mrfs42
71E25683904T 172E6538094T
Big Hair Day
Posts: 5,922
|
Post by mrfs42 on Apr 13, 2009 17:04:37 GMT
<ahem> Looks at AI over the table...... I actually meant 'I don't think there are any signal overlaps on the system that are completely level - which has a slightly different import to what was typed!  |
|
|
|
Post by angelislington on Apr 13, 2009 17:05:15 GMT
Um well it's come to my attention that what MRFS meant was that there are no completely level *overlaps* on the system. Ahem. But I guess the question still stands! There have to be some completely flat sections, I can't believe there aren't (other than platforms). (Oh for goodness' sake! Now I'm being told off for having corrected myself at *exactly* the same time that MRFS corrected me! ) Hey ho. Shouldn't try talking about signalling at gone midnight, I'm just not cleva enough  |
|
Tom
Administrator
Signalfel?
Posts: 4,198
|
Post by Tom on Apr 13, 2009 17:47:51 GMT
I've never seen an overlap that was completely level.
|
|
Colin
Advisor
My preserved fire engine!
Posts: 11,348
|
Post by Colin on Apr 13, 2009 20:25:14 GMT
So what constitutes a level overlap then?
I thought I had overlaps nailed, but apparently there's more... ;D ;D....
|
|
Tom
Administrator
Signalfel?
Posts: 4,198
|
Post by Tom on Apr 13, 2009 21:23:38 GMT
Quite simply, the 1982 signal overlap formula (2006 Revision) looks a bit like this:
D = CA + 1.05 [0.278Vtc + 0.077V2/2[fe±(0.1 x %G)] + 7.820x10-7V4]
The only bit we're interested in at this point is %G. This is the percentage gradient, basically an allowance that extends the overlap when going downhill or shortens it if going uphill.
In this context, a 'level' overlap is one where %G = 0.
If you're that curious about the other bits, CA is Cab Allowance (i.e. distance from the front of the train to the first axle), fe is the brake rate (tube or surface), 7.820x10-7V4 relates to brake fade and I think 0.278Vtc relates to brake build-up at tripping. I would have to check my notes though!
|
|
mrfs42
71E25683904T 172E6538094T
Big Hair Day
Posts: 5,922
|
Post by mrfs42 on Apr 13, 2009 22:07:06 GMT
0.278Vtc does indeed relate to brake build up *and* thanks to AI's expert Excel tutelage I've now got a spreadsheet that stabilises!
;D ;D
|
|
Colin
Advisor
My preserved fire engine!
Posts: 11,348
|
Post by Colin on Apr 13, 2009 22:07:16 GMT
Do you ever have a moment when you think to yourself "perhaps I shouldn't asked?" ;D ;D ;D
Naa seriously, cheers for the explanation - not quite what I was expecting but interesting all the same.
I certainly understand now where the level part fits in with overlaps and why you wouldn't have come across a level overlap.
|
|
Tom
Administrator
Signalfel?
Posts: 4,198
|
Post by Tom on Apr 13, 2009 22:18:56 GMT
0.278Vt c does indeed relate to brake build up *and* thanks to AI's expert Excel tutelage I've now got a spreadsheet that stabilises! I could have saved her the effort - I've got one that does that for you! (Although it only does the original 1982 formula - you have to modify the CA bit yourself!) |
|
mrfs42
71E25683904T 172E6538094T
Big Hair Day
Posts: 5,922
|
Post by mrfs42 on Apr 13, 2009 22:23:02 GMT
Well, I've wanted to learn Excel for a while, and AI used to train people professionally in it - so now I've finally found something that interests me, she has been able to teach me....
EDIT: I've just been told to say that it has fancy conditional formatting, coloured cells where stuff has to be inputted *and* it is all in Johnston.
|
|
|
|
Post by citysig on Apr 14, 2009 9:40:24 GMT
Quite simply, the 1982 signal overlap formula (2006 Revision) looks a bit like this: D = CA + 1.05 [0.278Vt c + 0.077V 2/2[fe±(0.1 x %G)] + 7.820x10 -7V 4] Yes. Very simple. The next time the wife reckons I haven't had a hard day at work, I'll show her this, and tell her that all the trains I'm responsible for, all encounter overlaps thousands of times a day. That'll show her ;D |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Apr 14, 2009 10:35:05 GMT
Well, I've wanted to learn Excel for a while, and AI used to train people professionally in it - so now I've finally found something that interests me, she has been able to teach me.... EDIT: I've just been told to say that it has fancy conditional formatting, coloured cells where stuff has to be inputted *and* it is all in Johnston. Is this spreadsheet being used for official Underground business? If so, it is really worrying that calculations with safety implications are being done in an Excel spreadsheet on a PC, especially as the spreadsheet may not have proper documentation or have been formally tested. Coloured cells and Johnston type do not guarantee that it is operating properly, or that someone else could take over if MRFS is hit by a train that over-runs the overlap. Of course, if MRFS is doing the calculations just to annoy Baldrick, my comments do not apply. |
|
mrfs42
71E25683904T 172E6538094T
Big Hair Day
Posts: 5,922
|
Post by mrfs42 on Apr 14, 2009 11:36:29 GMT
Is this spreadsheet being used for official Underground business? Of course, if MRFS is doing the calculations just to annoy Baldrick, my comments do not apply. It isn't being used for UndergrounD business - I'm not a member of staff - I am doing them to annoy Baldrick!  ;D |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Apr 14, 2009 13:39:45 GMT
Is this spreadsheet being used for official Underground business? If so, it is really worrying that calculations with safety implications are being done in an Excel spreadsheet on a PC ... Off topic aside: Calculations with safety implications are often done on a PC with Excel in all manner of industries - whats the problem with that? Providing as you say the calculations have been tested and are proved accurate enough then theres no issue. Using a PC or Excel in itself is not an issue, any inaccuary in Excel will most likely affect any computer based calculation, they all use the same IEEE format for internal number storage. Unless the display settings in Excel are incorrect, but thats a usage issue not accuracy. |
|
Tom
Administrator
Signalfel?
Posts: 4,198
|
Post by Tom on Apr 14, 2009 23:20:16 GMT
Is this spreadsheet being used for official Underground business? If so, it is really worrying that calculations with safety implications are being done in an Excel spreadsheet on a PC ... Off topic aside: Calculations with safety implications are often done on a PC with Excel in all manner of industries - whats the problem with that? Providing as you say the calculations have been tested and are proved accurate enough then theres no issue. Using a PC or Excel in itself is not an issue, any inaccuary in Excel will most likely affect any computer based calculation, they all use the same IEEE format for internal number storage. Quite. While the spreadsheet used by MRFS isn't used by LU, there is a similar version that is used. Any overlap calculation is prepared as part of a three step process of prepare, check and approve. The only person to use excel is the person who prepares the calculation - the others check it manually. I personally have used both methods and tend to prefer the manual calculation - but that's just me. |
|
;D
