LU Fleet strike (Train Inspections)

This is the question I asked some few posts ago, and in many places it has been answered (it's all about statistics).

For instance, I'm fairly sure that most railways know just how long a brake block/shoe will last on given equipment & line; let's call that 48-50 weeks for discussion. With the rate of wear, if you check them every week (or after certain events, such as emergency application), you will always find the replace-now ones before they're down to the nubs and know that daily checks are not necessary. Is this acceptable? I'll suggest that it is because a given train will have multiple brakes. Contrast with "do the brakes operate?" which is different from the block wear. Likewise you do want all the doors to close (and open) correctly and they're prone to failure and abuse; test them daily.

Tripcock? There's only one active on a train and it's essential safety gear; test daily.
Destination signs? They're not always right anyway.....

What evidence do I have of all this? None at my fingertips, however a perusal of existing procedures, failure and discovery rates, and RAIB reports will probably point the way.

I think you're missing the point. My question is whether all of the things on the "24-hour" check list need to be on it. Simply, I posit it's likely that not every one of the tasks need to be done daily, either because they aren't as safety-related as other parts or because their failure rate is quite low. Or, because things change (i.e. LED indicator lamps do not "burn out" as often as incandescent ones, most industries acknowledge that and don't check them as often). Check lists are reevaluated all the time.

That said, I'm not advocating that a daily check be abandoned, only that it check the right things.

BTW, this is nothing like the 373MAX debacle, which was poor system design (a single point of failure; had two sensors, ignored one of them).

Regardless what the manufacturer states on maintenance frequencies LU has a own rarely likes to change processes believe it or not.

Trouble is with all the cuts back that need to be done they are cutting back on some items which I think especially in signalling to there maximum and still getting bit on the backside with maintenance related failures

Fair comment, as was the rest of your post; but I'd like to try and answer the bit above if I can.

Now I'm not a train maintainer, but I am a driver that see's what train maintainers do on train preps when I stable trains at night so my understanding of the general process is something like this:

1) Check drivers defect log book for new defects.
2) Check public address system is working.
3) Does the whistle work?
4) Is screen wash topped up and are wipers clearing windscreens?
5) Are all motors engaging?
6) Check passenger door operation - do all doors open?
7) Do all doors open and close within time tolerance?
8) Do the motors engage* when passenger doors are open (they should not - this is safety critical)
9) Operate Passenger Emergency Alarm - does this apply the emergeny brake? (it should - this is safety critical)
10) Do motors engage* when Passenger Emergency Alarm is operated? (they should not - this is safety critical)
11) Do motors engage* when Passenger Emergency Alarm is operated and brake overide device is operated? (they should)
12) Do motors engage* when Passenger Emergency Alarm is operated and emergency brake cut out is operated? (they should)
13) Does the emergency brake apply when the tripcock is operated? (it should - this is safety critical)
14) Does the tripcock cut out release the emergency brake? (it should)

* All motor checks are done on both full & slow speed / tripcock & restricted mode circuits (or equivalent depending on particular rolling stock).

This list isn't exhaustaive and I'm sure a train technician will add to it or correct any errors but these are the sorts of checks that are done to all trains every 24 hours (ie, when they stable in the evening).

Hope that clarifies what's being discussed.

I can see the unions objecting to additional jobs being given to train operators!

But every single driver's duty that brings a train into service would need to be extended to allow time to carry out this activity. That's less efficient than having a few dedicated people to get the task done as you will need take driving time off the duties to compensate.

And what if a defect is found right before the train enters service? Isn't it better to check trains when they arrive in the depot? That way you have to time to rectify the problem before the train leaves again.


This graphic taken from an offial LU document explains it very well:

In theory it seems a train can enter service 23 hours after being checked and then run in service for (in theory) five traffic days without another check (checked one hour after the start of traffic Thursday morning, enter service and run until close of traffic, then enter service at start of traffic on Friday without a new check (still within 24 hours of the most recent), run continuously in service Friday (day and night), Saturday (day and night) and Sunday (day) before leaving service as the last train to stable on Sunday night. (I know that several reforms would be needed for this to happen in practice). With a 96 hour window that's theoretically allowing nearly a week of service without a check...

As for transferring the job to train operators - presumably if a train technician finds a fault in a check they have the skills etc to fix them then and there if possible. I have confidence that all train operators could top up the screen wash, but could they fix a wiper motor?

Chris M has made the point I was hoping to get to, but I wanted the facts before I made it!

My point about some tests being transferred to train operators was only the simple tests, ie those which I would expect to be within their current competency. Things like "whistle test".

I don't think anybody in this thread has previously mentioned the 'service clock' and crucially how long it runs for (if they recently have I've missed it). The graphic certainly doesn't explain that.

Am I correct that a train is prepped, then can sit for 23¾ hours, then run in service continually for a further 23¾ hours (assuming not stabled) before finally being taken out of service? Thus towards the end it's 47½ hours since the passenger alarm was last checked?

Colin Ah ok, I understood the wording of the graphic to mean that 24-hours after it is checked, a line is drawn. If at that time the train is stabled the check expires there and then, if the train is in service at that time then the check expires the next time it stables - however long the interval between the line and the stabling is. I didn't realise that there was a 24-hour limit to the in-service time as well.

I thought these were a simple pair of diagrams that explain it all visually well, but its thrown up just as many questions as it answers

I'm not entirely sure why the "+26h" bit is on there unless it's perhaps indicating that the 24 hour "service clock" began 2 hours after the "train prep clock" thus showing how a train is able to remain in service even though its more than 24 hours after the train was prepped by depot staff.

In my defence, it's not my diagram - I did say on the previous page that it was lifted from an LU document.

No.

As per my post higher up this page, the 24 hour service clock cannot be extended.

Firstly let's take an example on a day with no night tube:

Let's say a train is stabled by a late duty driver just after midnight, having done its days work on the railway. It's taken through its checks by depot staff and this is completed at 1am. So the depot prep clock starts at 01.00

Having completed their checks an early duty driver takes the train out of the depot at 06.00. The depot prep clock is now overidden and becomes spent. The 24 hour service clock now takes over.

On non night tube lines, all trains are stabled by about 01.45 - given that the service clock will expire on our train at 06.00, a new depot prep will be carried out overnight and so the process repeats.

Now onto a night tube example:

Theoretically, a night tube line could keep our train out in service until 6am by use of reforms but a timetable wouldn't schedule a train out to its limits like that. As I've previously said, night tube lines will usually put freshly depot prepped trains out, say ones that have been depot prepped around 20.00 or 21.00 - these will enter service around 23.00 so their service clocks run until 23.00 the next evening. Once stabled they'll be depot prepped again and drop back into the routine of the earlier "normal" example above.

Even with night tube in the mix, the 24 hour service clock cannot be extended so service control and depot staff have monitor their trains and take care when doing reforms to ensure none are out in service longer than 24 hours.

I'm sorry, but this confuses me. In the example shown the train is in service for 26 hours, or does the stabling time during the day not count.

Colin, I understood your post where you mentioned about operators doing checks but drivers at one time carried out 4 walks of the train testing the deadmans device and having a brake test, they also checked to see that the tool boxes were in order. The same thing happened on the Victoria line being OPO.