Signals for Glenmutchkin

[micknich2003]

Looking at the picture of Kyle signal box the frame appears to be of LMS pattern, note the white plastic lever handles, or are they the "Pull Plates"? Either way, it will be on record somewhere. Whatever, the frame is at the back of the box, that alone suggest an LMS frame. Mick.

[flyingsignalman]

An interior view of Kyle of Lochalsh box taken 15/6/1982.

 

 

(Edit, date amended)

[micknich2003]

 Thanks for photo, and it indeed proves me wrong, the "White Handles" are the Pull Plates. The frame is a M&H No11 or 14 Apparatus, I now wonder why the Highland put it at the back of the box. Mick.

[flyingsignalman]

 Thanks for photo, and it indeed proves me wrong, the "White Handles" are the Pull Plates. The frame is a M&H No11 or 14 Apparatus, I now wonder why the Highland put it at the back of the box. Mick.

 

John Hinson has a diagram dated 1949 on his Signal Box website which he says, in an answer about the box on his forum, dates from the installation of the "new" frame.

[micknich2003]

 The new frame is pressumably made up from "Recovered" material, and will no doubt have been modified to have "Direct Tappet" locking. Strange the LMS didn't fit a frame of their own design. Mick.

[Portchullin Tatty]

With a lot of help from Keith Norgrove (thanks Keith) the locking design for the lever frame should now be concluded.  This is where we are at:

 

The signalling layout:

The frame list:

 

And the locking chart (you can see that Keith has taught me some more accurate descriptions too!):

The lever frame is also ready to receive the beginnings of the locking frame; so guess what i am doing this weekend...........

 

 

 

[beast66606]

Hi Mark,

 

I'm not convinced that 4 should lock 12 and the converse 12 locks 4.

 

Other than that it looks good.

[The Stationmaster]

Hi Mark,

 

I'm not convinced that 4 should lock 12 and the converse 12 locks 4.

 

Other than that it looks good.

You can't have FPL 4 locking FPL 12 (when did that one creep in?) as that would make it impossible to cross trains.

[LNERGE]

You can't have FPL 4 locking FPL 12 (when did that one creep in?) as that would make it impossible to cross trains.

What is the reasoning behind this?

 

Edit to add...

 

IRSE green book No2 is worth a read. Principles of Interlocking (British Practise) by W.H Such.

 

Edit again to add..

 

The green books are available here apparently for free..

 

http://bookos.org/g/Institution%20of%20Railway%20Signal%20Engineers%20(IRSE)

[The Stationmaster]

What is the reasoning behind this?

 

Edit to add...

 

IRSE green book No2 is worth a read. Principles of Interlocking (British Practise) by W.H Such.

 

Edit again to add..

 

The green books are available here apparently for free..

 

http://bookos.org/g/Institution%20of%20Railway%20Signal%20Engineers%20(IRSE)

They're also on my bookshelf as it happens .  

 

Not that the Locking Principles have much to do with 12 locking 4 etc; indeed if you look at Fig.10 in the volume you mentioned you will see that FPL 7 does not lock FPL 10 - different numbers but the exact equivalent of levers 4 &12.

 

Getting back to Glenmutchkin if 12 locks 4  (and vice versa) that means the either points 5 or points 11 cannot be bolted when standing normal, which in turn means a train cannot be accepted from the 'box in rear - so it becomes impossible to accept trains from both directions at the same time, thus it becomes equally impossible to cross trains at this loop unless one of them sits there waiting for a train to pass through an entire block section (which in the wilds of Scotland could take rather a long time).

 

The proper conflicting lock to comply with the signalling Regulations is 2 Locks 14/14 Locks 2 however the older editions of Interlocking Principles doesn't include this lock although technically - in the terms of the Regulations - they are conflicting signals.  It is debatable if 14 should also lock 9 but technically as the two movements would not be trains crossing each other it can be argued that such a lock is not needed - the decision to add it would be dependent, in my view on local circumstance and practice.

 

I suspect that on the vast majority of 'frames 2 and 14 would not lock each other as the Locking Engineer is not familiar, officially, with the Block Regulations and the operating dept on some railways were not necessarily in the habit of checking the locking on schemes (however the Railway would have its own locking principles).  Never having had occasion to work outside the Regulations at a crossing station on a token etc worked line I've never had the opportunity to find out if the locking actually reflected the Regulations or if that was left to the Signalman - even on the Western

[RailWest]

"Never having had occasion to work outside the Regulations at a crossing station on a token etc worked line I've never had the opportunity to find out if the locking actually reflected the Regulations or if that was left to the Signalman - even on the Western"

 

It was certainly the case in all the ex-L&SWR locations for which I have seen the Locking Table, the two opposing Home signals did lock each other (either directly, or by some other means eg running shunts etc). OK, so I never had the 'hands on' chance to see if the actual locking matched the table, but one has to take some things on trust :-)
 

[micknich2003]

IRSE book No3 is also worth a look.

[Grovenor]

You can't have FPL 4 locking FPL 12 (when did that one creep in?) as that would make it impossible to cross trains.

I see your point, far to long since I was involved with the block regulations, and I don't recall seeing  any check on the fpl position in token circuits, but I agree the example from the IRSE book that I showed Mark had each home locking the fpl at the far end to achieve the locking between homes, you need to lock the fpl rather than the opposing signal to provide the route holding. Locking the fpl with fpl simplified the locking a bit and has the same effect so far as the signals are concerned, but as you say, not if the block regs want the overlap bolted up. I'll check with Mark if he wants to operate at that level.

Regards

Keith

[Portchullin Tatty]

Thanks for the various emails that greet me after an evening of "client entertaining" so my mind is definitely not up to the intricacies of interlocking.

 

One thing that I will observe though is that my father always drummed into me that it is illegal (certainly on the highland) for a second train to enter a passing loop after the first was given right of way into the other side.   It is quite legal for trains to leave a loop at the same time but not enter.  This is why I at least “thought” 4 and 12 had a place to be locked and this is what Keith suggested to me.  We do occasionally play the game of trains not entering at the same time but leaving simultaneously at exhibitions on Portchullin, but I doubt people actually notice!

 

That reminds me, when Portchullin was at the Nottingham show in 2012, the organisers ran a competition for the most authentically operated layout.  I did think about entering and when the judges came round to form their view on the authenticity of the operation of the layout, I would run nothing.  You can't knock me - there were three trains a day max so for the vast amount of the day nothing happened……………  I decided that perhaps that was not the spirit of the competition!

[Portchullin Tatty]

I see your point, far to long since I was involved with the block regulations, and I don't recall seeing  any check on the fpl position in token circuits, but I agree the example from the IRSE book that I showed Mark had each home locking the fpl at the far end to achieve the locking between homes, you need to lock the fpl rather than the opposing signal to provide the route holding. Locking the fpl with fpl simplified the locking a bit and has trhe same effect so far as the signals are concerned, but as you say, not if the block regs want the overlap bolted up. I'll check with Mark if he wants to operate at that level.

Regards

Keith

 

In practise, the whole of this locking frame is an experiment for me to try and learn a bit about the principals of interlocking and also about how to make an interlocking frame.  The idea would be then to apply them to the real scheme.  It is fair to say that whilst my signalling knowledge was at least close-ish, my locking knowledge was/is not.  Thus I can certainly say i have "learnt" and I am thankful for the advice that the several of you have offered.

 

Unless there are real howls of derision that this is badly wrong, I am inclined to go with Keith's scheme (as shown above) because I can see why the link is there and why it therefore works - my point about opposing trains entering the loop.   I am not proposing to include a token link (and if I did of course for the Highland I would need two cabins and a separate token machine in the station building!), so I can't quite get this feature in anyway.

[LNERGE]

The main issue is to stop two trains being signalled into the loops at the same time. The 4-12 lock achieves this but at the expense of not being legally able to accept a second train because the route is not set. (i challenge anyone to work my box to those rules, the job will stop). I've no doubt the layout in the IRSE book is correct. I can post a few locking charts for passing loops if the are of interest. I should have Derby Road, Acle and North Walsham at least.

[uax6]

If the signals were more than 440yds from the points the clearing point would be on the approach side to the points, so therefore they would not constitute part of the route, which would allow you to accept to the home, whatever the state of the route in advance of the clearing point.

 

Mark do you know the distances involved on the Highland of the signals from the loop points? This might give you an answer to the issues of crossing trains.

 

Andy G

[beast66606]

Interlocking does not enforce block regulations it (hopefully) prevents mistakes leading to collisions.

 

It's your layout and there is no harm in the FPLs locking each other but typically locking hangs off the points and signals, not the FPLs so 2 and 14 should be exclusive, not 4 and 12

[beast66606]

If the signals were more than 440yds from the points the clearing point would be on the approach side to the points, so therefore they would not constitute part of the route, which would allow you to accept to the home, whatever the state of the route in advance of the clearing point.

 

Mark do you know the distances involved on the Highland of the signals from the loop points? This might give you an answer to the issues of crossing trains.

 

Andy G

 

Andy

 

The signals I've seen were not 440yds from the points so I think the problem existed on the big railway.

[uax6]

Ah, just an idea at a get out of jail free card. Must re-read the discription on how the trains were handled on the HR to see if that sheds some light. (In HR Misc by Mr Tatlow Snr)

 

Andy

[The Stationmaster]

We're getting there, possibly.  A few points (sorry in summary)

 

1. If the two FPLs lock each other you cannot realistically work the crossing loop as a crossing loop because you would only be able to accept a train from one direction at a time towards the loop.  If you're happy with that and don't want to cross trains then fair enough.

 

2. I'm not sure why mention of a 440yard Clearing Point has crept in - there is not and never has been such a Clearing Point at crossing loops on single lines, most of them would be impossible to work if there was.

 

3. Using FPLs to enforce locking which doesn't essentially involve them or to save other locks is a practice which died out in locking design a long tome ago (the GWR did a bit of it in the 1920s, and possibly earlier - and later got rid of it) because in the event of disconnections it means that essential locks can be lost because an FPL lever is out of use.  Hence Chris's point about frames at crossing places where the Home Signals lock each other.

 

4.  Let's get the working of a crossing loop right.  The Signalman can simultaneously accept trains from each direction (unless the FPLs lock each other!) and both trains can simultaneously approach the Home Signals but having arrived at the Home Signals (which are both at danger) one train must be admitted to its loop and come to a stand before the other can be admitted to its loop - hence the Home Signals locking each other to enforce the Regulations and minimise the risk of a train entering a loop at speed and over-running the Starting Signal at the far end of the loop; it really is belt & braces.

 

And it really all is as simple as that.

[The Stationmaster]

I had a surprise at Crowcombe & Heathfield on the West Somerset a few years back.  I was standing at the bottom of the down platform ramp, by the foot board crossing waiting for the King to pull in with an up train.  As 'she' passed, I turned around to walk back up the ramp and was surprised to see a down train rolling into the loop as the up was still in motion.  I had failed to notice the traps at the end of each loop line, which I guess permitted this move.

Yes, it's an interesting way round the problem of delaying one train while the other enters its side, in effect - using modern terminology in a way it was never meant to be used - it can be compared to a swinging overlap.  And odd thou' it may sound the same principle was used at Reading to give acceptance flexibility (as if there wasn't enough anyway  ) on the Up Main Platform Line although I suspect the motive behind that was as much SPAD mitigation as anything else following the sidelong collision at Main Line East.

[RailWest]

In the 1930s the GWR speeded-up the traffic on the Minehead and Barnstaple branches by (a) installing automatic token exchange apparatus and (b) providing facing trap points at the ends of most (but not all) passing loops, the latter enabling trains to entering the loops from both direction simultaneously.

 

When the WSR re-built Crowcombe (Heathfield) they used trap-points as well in order to facilitate concurrent access for passing-movements. As the station is on a summit, and it would be undesirable to try to re-start heavy trains from the Homes on the gradients, the local instructions permit the signalman to clear the Homes once the 'train approaching' buzzer sounds so that they have a 'clear run' into the loops. It helps the time-keeping.....

[Grovenor]

3. Using FPLs to enforce locking which doesn't essentially involve them or to save other locks is a practice which died out in locking design a long tome ago (the GWR did a bit of it in the 1920s, and possibly earlier - and later got rid of it) because in the event of disconnections it means that essential locks can be lost because an FPL lever is out of use.  Hence Chris's point about frames at crossing places where the Home Signals lock each other.

This I don't follow, the IRSE green book explains why you lock the home with the fpl at the other end rather than locking home to home and this was what was taught in the 50s and 60s. If the GWR did things differently that was often the case! And surely if an fpl lever is out of use it would not be acceptable to clear the signal leading over it anyway, you would have to clip and scotch and flag. 

Regards

Keith

[The Stationmaster]

This I don't follow, the IRSE green book explains why you lock the home with the fpl at the other end rather than locking home to home and this was what was taught in the 50s and 60s. If the GWR did things differently that was often the case! And surely if an fpl lever is out of use it would not be acceptable to clear the signal leading over it anyway, you would have to clip and scotch and flag. 

Regards

Keith

If the locking is arranged through the FPL you will lose the security of that locking if the FPL itself is out of use for some reason. Thus if - stupidly - an FPL at one end locks the opposing direction Home Signal at the other that signal will not be locked if the FPL is disconnected or, for some reason the lever which works it cannot be reversed, which was not exactly an unheard of fault/maintenance situation.  Whereas if the opposing Home Signal levers lock each other you have at least removed one totally irrelevant feature from the locking (although obviously if one of the Home Signal levers cannot be worked then the other would not be locked by it).

 

But how on earth does the Signalman accept and work trains into the crossing loops if both his Home Signals are locked by the FPL at the other end of the layout?  I can see some point in doing it in this way if it can be guaranteed that both trains are going to arrive at roughly the same time because then once both have come to a stand at their respective Home Signals the FPL lever for one end or the other would be replaced to normal (assuming they stand bolted with the lever in reverse) in order to release the Home Signal at the opposite end.  Once the first train has come to a stand in the loop apart from the signal being returned to danger behind it the FPL lever is again pulled in order to release the adjacent Home Signal - which no doubt the Signalman will absolutely love when the marks are being taken.  (And which 'when no one is about' is just as likely to encourage sloppy working in order to avoid those selfsame lever movements.)

 

But what happens if the trains aren't likely to arrive at the same time - say one of them is running late or the block sections on either side have unequal running times?  Simples the second train cannot be accepted until the first has come to a stand in its respective loop - so both trains suffer undue, and totally unnecessary delay.  To be honest it sounds to me rather like a case of whichever Signal Engineer came up with that locking didn't understand Rules & Regulations - which is hardly surprising as it wouldn't be his job to understand them, but the Drawing Office Instructions for such a situation would have been agreed with someone on the operating/traffic side who did understand the Regulations and who would, one hopes, have pointed out the major pitfalls in this way of arranging the locking.

 

Incidentally what was shown in the IRSE green books was not necessarily the way in which thing were done in drawing offices - they were intended to illustrate how interlocking worked and to explain the basic principles, not to explain how each Company or drawing office actually did things or indeed to necessarily relate the intricacies of interlocking to the requirements of the Rules & Regulations.  And the use of FPL led locking was, I suspect, drawn from a far earlier publication dating from the 1920s.