WCML Resignalling film (1974)

Does anyone know why BR went with more of a telecom model for its signaling in the 1960's than with a lighter weight CTC model or smaller scale panel boxes?

It seems awfully ambitious for the time with large panels such as these quickly becoming obsolete.

Does anyone know why BR went with more of a telecom model for its signaling in the 1960's than with a lighter weight CTC model or smaller scale panel boxes? It seems awfully ambitious for the time with large panels such as these quickly becoming obsolete.

If I understand the CTC Mike is talking about correctly, it requires each function to be operated individually. Some sort of multiplex system is then used to call the function at the remote location; detection is returned via the same multiplexer and has to be received before the next function in the sequence can be called. Not only is routesetting a lot quicker for the signalman, the time for commands to be sent and detection to be returned can be signficant. Remember we tend to run quite a lot more trains in the UK than is typical in the US, and often with less running lines to put them through. CTC-style operation simply wouldn't cope.

As Geoff has commented, many of our panels lasted 40+ years; I understand 40yr is in fact the design life of a relay signalling system so there won't be many that are going early. Those that are replaced early may well have wire degradation issues that encourage (if not force) early replacement in any case. That is not uncommon in some of the 1980s installations.

It's also worth noting that the large panels allowed much better regulation, with a box supervisor overseeing train running over considerable distances and individual signalmen able to watch situations develop and respond accordingly; all being co-located also made things easier, as instructions and information could easily be communicated in real time without having to pause and phone each other. As a result, West Hampstead was able to cover work previously requiring the best part of 30 signalmen with just four men (numbers are per shift), while improving regulation.

They have certainly completed full service lives, but they were nevertheless functionally obsolete by the 1980's. I respect the ambition regarding the functionality of the panels, but I am surprised that nobody thought that it might be trying to do too much too soon. I know we have the benefit of hindsight, but hundreds of mini-CRT's embedded in the panel? That should have been a red flag. Decoupling the interface from the signaling system (a la CTC) would have future proofed the scheme and allowed for VDU conversion in the 1980's.

The specific communications technology actually isn't important (the 1930's vintage pulse code systems could get backed up, but today it uses fiber optics, data radios or whatever). What I noticed was that CTC style panels can get rolled into an office or tower and be good to go, while PSB's were built on top of what looks like a large telephone exchange. In fact the BR film reminded me of several 1970's vintage Bell System films about new Number 5 crossbar installations.

They have certainly completed full service lives, but they were nevertheless functionally obsolete by the 1980's. I respect the ambition regarding the functionality of the panels, but I am surprised that nobody thought that it might be trying to do too much too soon. I know we have the benefit of hindsight, but hundreds of mini-CRT's embedded in the panel? That should have been a red flag. Decoupling the interface from the signaling system (a la CTC) would have future proofed the scheme and allowed for VDU conversion in the 1980's.

How can they be functionally obsolete when they were still (still are) controlling trains perfectly efficiently?!

Decoupling the interface from the signalling system? Yes. It's called a panel/VDU and an interlocking respectively, and have been functionally separate since probably before the 1960s. It would be nonsense to combine them anyway because an interlocking is safety critical whereas a panel is only safety related. This is one difference from the US: there are three effective safety levels with non-safety being the one I have yet to mention, and safety-related doesn't have a direct parallel in the US (either vital or non-vital).

RE my earlier question about light weight CTC which you haven't answered. Given your comments above I assume it's because you don't know. UK control systems require a direct physical connection to the interlocking which has its benefits in speed. US does not (indeed, would be expensive to do so) require a direct connection.

Talking to a dispatcher controlling the Hi-Line in Montana from the BNSF NOC in Texas, he was demonstrating how slow and unresponsive their "light weight" (!) system was - he could issue a request and up to half a minute later he might get confirmation that something happened. Or not, by the time messages had bounced off a couple of satellites, gone through public telephone exchanges (I kid you not - Pomona near me is controlled that way), off to a local tower, broadcast over the air... - rinse and repeat the request/confirmation. Now, that's not so bad if you have a headway measured in tens of minutes, but horribly inefficient if you're trying to signal trains on 150 second headways in central London (or Yorkshire). Same for NS in Springfield IN.

" said:

How can they be functionally obsolete when they were still (still are) controlling trains perfectly efficiently?!

I've heard elsewhere that the 1960's and 70's era panels were getting unreliable due to a lack of spare parts and wiring issues. They almost certainly have higher maintenance (and alteration) costs than something on a computer screen.

There appears to be some degree of coupling since the panels were co-located with large amounts of interlocking logic and are currently being taken out of service in conjunction with re-signaling projects, instead of re-control projects. Is there some level of integration between the panel and signaling that makes divorcing the two systems impractical? The film, while clearly meant for a mass audience, implies that everything in the PSB was knitted into a single whole.

" said:

RE my earlier question about light weight CTC which you haven't answered. Given your comments above I assume it's because you don't know. UK control systems require a direct physical connection to the interlocking which has its benefits in speed. US does not (indeed, would be expensive to do so) require a direct connection.

What is known as "direct wire control" is a very specific thing that may or may not be employed in the WCML situation presented in the film. As it's been explained to me, direct wire is basically a signaler pressing a button on a remote panel and it acting like he was operating a button on the local control panel. It is generally less popular than "code" systems because you have to physically run a bunch of wires from the remote panel to the remote interlocking. See the neck-thick bundle of cables in this photo? That's direct wire.

Actually Wimbledon ASC has a relay room alongside rather than below, and IIRC is connected by a TEML41 comms link, and even has a slave panel not 30ft away from the normal panel

" said:

How can they be functionally obsolete when they were still (still are) controlling trains perfectly efficiently?!

I've heard elsewhere that the 1960's and 70's era panels were getting unreliable due to a lack of spare parts and wiring issues. They almost certainly have higher maintenance (and alteration) costs than something on a computer screen.

" said:

Decoupling the interface from the signalling system? Yes. It's called a panel/VDU and an interlocking respectively, and have been functionally separate since probably before the 1960s. It would be nonsense to combine them anyway because an interlocking is safety critical whereas a panel is only safety related. This is one difference from the US: there are three effective safety levels with non-safety being the one I have yet to mention, and safety-related doesn't have a direct parallel in the US (either vital or non-vital).

There appears to be some degree of coupling since the panels were co-located with large amounts of interlocking logic and are currently being taken out of service in conjunction with re-signaling projects, instead of re-control projects. Is there some level of integration between the panel and signaling that makes divorcing the two systems impractical? The film, while clearly meant for a mass audience, implies that everything in the PSB was knitted into a single whole.


" said:

RE my earlier question about light weight CTC which you haven't answered. Given your comments above I assume it's because you don't know. UK control systems require a direct physical connection to the interlocking which has its benefits in speed. US does not (indeed, would be expensive to do so) require a direct connection.

What is known as "direct wire control" is a very specific thing that may or may not be employed in the WCML situation presented in the film. As it's been explained to me, direct wire is basically a signaler pressing a button on a remote panel and it acting like he was operating a button on the local control panel. It is generally less popular than "code" systems because you have to physically run a bunch of wires from the remote panel to the remote interlocking. See the neck-thick bundle of cables in this photo? That's direct wire.

" said:

Talking to a dispatcher controlling the Hi-Line in Montana from the BNSF NOC in Texas, he was demonstrating how slow and unresponsive their "light weight" (!) system was - he could issue a request and up to half a minute later he might get confirmation that something happened. Or not, by the time messages had bounced off a couple of satellites, gone through public telephone exchanges (I kid you not - Pomona near me is controlled that way), off to a local tower, broadcast over the air... - rinse and repeat the request/confirmation. Now, that's not so bad if you have a headway measured in tens of minutes, but horribly inefficient if you're trying to signal trains on 150 second headways in central London (or Yorkshire). Same for NS in Springfield IN.

Code systems have vastly different performance depending on the technology. A 504X pulse code system running over a 50 year old pole line will have far more trouble than a fiber optic link. Your story reminds me of the Port Jervis Line running from Suffern, NY to Port Jervis, NY. It was run by New Jersey Transit for Metro North over tracks owned and maintained by Conrail/Norfolk Southern. The telecom grade code line was in such bad shape it would sometimes take 20-30 minutes to get a route lined. That was over a decade ago and the line is only now seeing a new signal system installed.

And it works the other way round as well: A number of relay interlockings have been replaced by computer-based ones, while still being controlled from the original signalling panel.

There have been wiring degradation issues; it seems to me they've affected 1980s installations far more than the earlier ones (Cambridge relay room is full of notices on the racks warning against disturbing the wiring, while the 1960s panels on the WCML continued without significant problems until replaced under the recent resignalling). That would have happened equally in a locally-controlled small installation, or indeed in the electrical portion of a mechanical installation.

(It might be worth underlining a considerable difference between UK and US practice; we indicate all running lines and signal sections, whereas I understand that in the US once a train leaves an interlocking area it effectively becomes invisible until it reaches the approach tracks for the next interlocking - hence the need for 'stop & proceed' working in the US, in contrast to the UK requirement to be talked by failed signals in all cases.)

We do have some direct wire control (a wire pair per function) of remote interlockings. It tends to be used for nearby interlockings and is more common in older installations where a multiplexing system would have been more expensive. But even where we use multiplexing systems the signals are run through railway-owned cables laid at the lineside rather than through the public network. The thought of most of Victoria PSB area being worked through the telephone exchanges of South London makes my blood run cold.

" said:

There have been wiring degradation issues; it seems to me they've affected 1980s installations far more than the earlier ones (Cambridge relay room is full of notices on the racks warning against disturbing the wiring, while the 1960s panels on the WCML continued without significant problems until replaced under the recent resignalling). That would have happened equally in a locally-controlled small installation, or indeed in the electrical portion of a mechanical installation.

It was explained to me elsewhere that the 1960's Panels were being replaced ahead of many century old mechanical frames because parts were no longer available to maintain the panels and they had become unreliable. Mechanical frames on the other hand could be repaired with basic tools.

Assuming that has any truth to it at all, is it not likely related to the fact that mechanical boxes have been closing continually for decades, providing a stream of spare parts, whilst also reducing the number of complicated layouts controlled or rare bits of equipment still in use? Even then, it's a rarity to see mechanical signalling replaced like-for-like with spares - there's plenty of precedent for signals and trackwork controlled by mechanical boxes being removed or modernised when parts could not be found to replace them.

" said:

Assuming that has any truth to it at all, is it not likely related to the fact that mechanical boxes have been closing continually for decades, providing a stream of spare parts, whilst also reducing the number of complicated layouts controlled or rare bits of equipment still in use? Even then, it's a rarity to see mechanical signalling replaced like-for-like with spares - there's plenty of precedent for signals and trackwork controlled by mechanical boxes being removed or modernised when parts could not be found to replace them.

All that mechanical stuff can be fabricated or machined. Many electronic components from the 1960's and 70's require complex manufacturing processes that no longer exist. (I am reminded of the propulsion group of a recently retired class of electric MU's that relied on the analogue properties of specific germanium semiconductors.) At least this is what explained to me on another signaling forum.

The argument against 1960s electronics applies equally to the mechanical stuff. It gets hideously expensive to manufacture small quantities of specialist components, especially when what you want is in imperial sizes that no-one makes any more.