Those traction reliability stats all (ok some…) of you have been eagerly awaiting… apologies the post is a bit long, but a fair bit of info to cover.
For 100% accurate answers to these questions, you’d probably need to be whoever manages the sleeper loco for GBRf and they’re not going to share on here. This is from my own observations/records taken from publically available sources, but I’m confident it’s a fair reflection. Happy for anyone to add further information.
Goes without saying, but info about what actually ran is obviously more available than issues at depots “behind closed doors” – but to be honest if a working loco from the overall fleet appears from the depot on time I don’t think it really matters – I’m sure there are switches of allocated loco/stock many times every day across the rail industry for a myriad of reasons.
All info covers the last 12 calendar weeks (excluding tonight).
Split of traction
The four “main” runs (1M11, 1M16, 1S25, 1S26) have been:
79% Class 92
20% Class 90
1% Class 86/87
If the Lowlander Edinburgh portions are added in as a fifth run, the split changes a bit as these have been more heavily worked by the 90s:
72% Class 92
23% Class 90
4% Diesel (Colas Class 47s / Class 73s)
1% Class 86/87
Delays/cancellations attributable to Traction
In terms of ANY delay due to traction, I don’t have this sort of info. However, I’m fairly certain that the list below covers all significant delays (including those >1 hour). I cannot recall any other sizeable delays due to traction in the last 12 weeks. Where there’s been relatively small delays (say 10-30 mins) it’s usually due to other operational/network issues.
Failures resulting in significant delay/cancellation:
Class 92s (3/228 workings = 1%):
1) 92038 on northbound Lowlander on 8 May – replaced by 87002 and IIRC arrived c.45 mins late at Glasgow
2) 92044 on southbound Highlander on Sunday night (11 Jun) – 1M16 cancelled from EDB
3) 92033 on northbound Highlander on Wed night (14 Jun) – replaced by 90046 – 85 Late at EBD / 98 Late at Inverness
Class 90s (1/81 workings = 1%):
1) 90046 on the southbound Highlander last night, replaced by 92010 – 97 Late arriving at Euston.
Class 87 (1/1 booked working):
1) 87002 failed prior to departure on the “special” on Mon 5th June – replaced by Class 90
Other known failures not significantly impacting services:
I’m aware of a few other failures prior to departure/at depots which had no impact on the services involved – as above, I’m sure there’s others that have happened within the depots. But I also think these are the sort of things that happen to all fleets on a pretty regular basis and dealt with as a matter of course.
Class 92s
- 92023 due on 1M11 failed on night of 27 Mar, but a loco reshuffle involving 92010 taken from 1M16 and 86401 taking over 1M16 meant services ran to time.
- Reports of a Class 92 failing at Wembley on 11 Apr and being replaced by a 90 – didn’t impact service.
- 92010 was due to work the northbound Highlander on 2 May but was pulled well before departure and replaced with 90049. Again, no impact on service.
Class 90s
- 90047 failed at Polmadie on 22 May, stock was turned and 86401 on the other end originally for the ECS duties worked the service South instead. No impact on departure time.
Couple of overall observations (based on last 12 weeks)…
- The level of 92 failures causing significant delay/cancellation is low (1%) and there’s only been one outright cancellation.
- The level of 90 failures causing significant delays etc. is proportionately the same taking into account the split of workings. It is also low.
- Three of the four "significant impact" failures (2x 92s; 1x 90) have occurred in the last week – hopefully just “one of those weeks”; but can give the impression of there being traction/Class 92 issues, yet this longer term view suggests otherwise.
- Although the info is more limited, there’s been a handful of times locos have been failed (well) before departure (and possibly some of these have been erring on the side of caution). But again, they appear to be occurring about once/twice a month and are proportionately split across 92s/90s. They do not appear to be indicative of widespread/above average issues with the 92s.
So, on the basis of the above from the last 12 weeks, the 92s and 90s have both been very reliable in the main (particularly considering they’re c.25 year old locos) and would also appear to be about as reliable as each other.