Our new ticketing site is now live! Using either this or the original site (both powered by TrainSplit) helps support the running of the forum with every ticket purchase! Find out more and ask any questions/give us feedback in this thread!
I was surprised to see on Friday that TfW Journeyplanner was already advising of train cancellations on Monday due to "over-running engineering works". I was surprised that they knew so far in advance that engineering works would over-run, but having discovered this thread I now know why.
When I started on the Western, I was astonished to find the works staff at Reading building replacement relay racks for existing interlockings. I was told they did this when they had no other work to do. As MarkyT says, the new rack could then be swapped in to replace an existing rack, by simply disconnecting the rack-to-rack multicore cables on the terminals at the top of the rack, lifting out the old rack and putting in the new rack, and reconnecting the multicores. By this means, an interlocking could effectively be renewed rack by rack. It is a pity this option doesn't appear to have been used for this work at Gloucester, with new racks ready-wired and basic testing ready-done prior to the shut-down.
== Doublepost prevention - post automatically merged: ==
Westbury and Exeter didn't reuse any equipment from elsewhere as far as I am aware. Whereas traditional WR interlockings used what are generically termed "Post Office relays" (as used in traditional telephone exchanges) for non-safety functions, Westbury and Exeter used a more modern form of non-safety relay, with IIRC a number of such relays inside a case similar to a BR930 case, but smaller. These relays proved troublesome almost from the start. In addition, Exeter used a "panel processor" computer system for many of the non-safety functions. This was developed by GEC, but they soon lost interest even before it was commissioned when it became apparent that it was going to be a one-off, and they weren't going to be selling any more as it was effectively bespoke to the WR. In addition, both Westbury and Exeter were built to electrification standards, with significant circuit differences to other WR interlockings.
Westbury definitely had the small relays to begin with. I thought they were individual plug-in units with separate bases on the racks but I could be wrong. They disappeared quickly and were replaced with Keyswitch Varley relays, well proven PO3000 series units like the earlier standard WR panel relays, but individually packaged in a transparent case with a matching plug base, allowing each relay to be easily removed, unlike earlier implementations where several were housed in a telephone exchange-like 'can' with the wiring soldered onto the rear tags of the relays themselves.
I'm not sure if any of the original tiny relays were actually installed at the slightly later Exeter panel, though I don't doubt they were originally intended. It was clear by then from Westbury that they were troublesome. When I visited the new Exeter box for a period as a trainee in 1984, while the early stages were being tested, all the panel relays were the Keyswitch Varley type. The local station interlocking and the directly wired remote interlockings out to Cowley Bridge and Exeter Central were not covered by the panel processor system, so had the full complement of panel relays as used on earlier WR systems, though modified for push-push route setting rather than the traditional turn-push method.
Presumably a 'Gloucester' type job is going to have to be done on Exeter at some point if a 'mini-ROC' (which I thought was the latest plan) can't be afforded? We now have the slightly odd situation of Liskeard box in Cornwall interfacing to Exeter in the west and Plymouth in the east!
Westbury definitely had the small relays to begin with. I thought they were individual plug-in units with separate bases on the racks but I could be wrong. They disappeared quickly and were replaced with Keyswitch Varley relays, well proven PO3000 series units like the earlier standard WR panel relays, but individually packaged in a transparent case with a matching plug base, allowing each relay to be easily removed, unlike earlier implementations where several were housed in a telephone exchange-like 'can' with the wiring soldered onto the rear tags of the relays themselves.
The small "continental"-type relays were used extensively by Westinghouse for indication processing with their TDM69 (e,g. at London Bridge), and by the Southern with its Electronic Route Setting Equipment (ERSE) (also used by the LM), without significant problems. I can only assume that the version used at Westbury must have been a cheap and not particularly reliable version. The Keyswitch Varley relays took up considerably more rack space than the continental relays they replaced, and as I recall, more space for an equivalent number of relays than the old can-mounted relays.
I had a look through my records of my time on the WR, and found details of the Keyswitch Varley relays, but nothing about the relays used for Westbury. But then for the Westbury and Exeter schemes, I was mainly involved in the remote interlocking areas, while a dedicated team dealt with the panel circuitry.
There's plenty of room for new equipment in the cavernous relay room beneath the operating floor. It was sized to accommodate panel relays for all the remote interlockings, most of which weren't needed when the panel processor system was selected. There's probably room to replicate the entire station interlocking on new racks if desired.
They disappeared quickly and were replaced with Keyswitch Varley relays, well proven PO3000 series units like the earlier standard WR panel relays, but individually packaged in a transparent case with a matching plug base, allowing each relay to be easily removed, unlike earlier implementations where several were housed in a telephone exchange-like 'can' with the wiring soldered onto the rear tags of the relays themselves.
The PO relay sets used in the 1970s WR PSBs like for example Bristol PSB (including remote interlockings / relay rooms) were mounted "in a can" with five relays per "can". The cans fitted into plug boards (bases) (for all electrical connections) but were secured to the rack via two nuts on threaded studs (part of the rack). So they could be changed fairly easily. The wiring connections on the plug board / base were soldered. So if a plug board / base had to be changed it was a nightmare.
Even in the Westinghouse F1 TDM (Time Division Multiplex), the rack mounted relays were on plug boards / bases. Only the output relays were soldered, but these were on PCB cards, so if one became faulty, you just changed the card.
The late 1980s alterations and extensions used Keyswitch Varley relays instead of PO relay sets. A load are in use at St. Andrews Junction Interlocking for example (dating from 1988).
But 930 type relays were used for non-safety functions when Stoke Gifford Interlocking was completely replaced (in the same relay room as the existing interlocking) in connection with the Royal Mail terminal and for the new Filton Abbey Wood Interlocking. Both since decommissioned due to TVSC taking over the areas they covered.
In my travels, the only relay cases used for signalling that had multiple smaller relays in were the sets used with the GEC (General Electric Company, the British one, not the American one) FDM (Frequency Multiplex) Reed systems (these were 12V relays IIRC) and relay sequence timers (or guess the time delay sets, as they were not very accurate), also 12V IIRC. I'm not including the 7.5 second point operation timer or ELDs (Earth Leakage Detectors).
Even in the Westinghouse F1 TDM (Time Division Multiplex), the rack mounted relays were on plug boards / bases. Only the output relays were soldered, but these were on PCB cards, so if one became faulty, you just changed the card.
The late 1980s alterations and extensions used Keyswitch Varley relays instead of PO relay sets. A load are in use at St. Andrews Junction Interlocking for example (dating from 1988).
Even in the Westinghouse F1 TDM (Time Division Multiplex), the rack mounted relays were on plug boards / bases. Only the output relays were soldered, but these were on PCB cards, so if one became faulty, you just changed the card.
TDM69 used a similar system, with small PCBs that plugged into a rack. Each PCB had a number of small "continental" relays soldered onto it. As I recall, there were ones with a red plastic front handle with LED indicators used for signals, and ones with a black handle used for points. The contacts on the relays were connected together on the PCB to combine the TDM outputs to give the required panel indications, saving a lot of free wiring.
In my travels, the only relay cases used for signalling that had multiple smaller relays in were the sets used with the GEC (General Electric Company, the British one, not the American one) FDM (Frequency Multiplex) Reed systems (these were 12V relays IIRC) and relay sequence timers (or guess the time delay sets, as they were not very accurate), also 12V IIRC. I'm not including the 7.5 second point operation timer or ELDs (Earth Leakage Detectors).
I have seen BR930 cases containing a PCB with multiple small "continental" relays used for non safety functions, produced by Westinghouse. And as previously mentioned, ERSE used similar things, produced by the relay shop at Wimbledon.
We are aware of an issue with emails from the Forum to Microsoft-based email accounts (hotmail/outlook/live.com email addresses). This is being looked into currently, thanks for your patience meanwhile.