Rolling backwards for 2 miles

[martin_wynne]

This occurrence reported by RAIB is surely very odd?

 

http://www.raib.gov.uk/publications/current_investigations_register/100817_shap.cfm

 

Martin.

[James]

This occurrence reported by RAIB is surely very odd?

 

http://www.raib.gov....100817_shap.cfm

 

Martin.

 

It definately is

[beast66606]

James - they can do crew changes, the departure time is not necessarily the time the driver started

 

As usual - unnecessary speculation should be avoided

[Kris]

That's a different one, not an occurrence that you hear of every day.

[martin_wynne]

Without speculating about the cause, it would be interesting to know what was happening between the locomotive wheels and the rail for such a sustained period of time. They could be:

 

1. rolling backwards, or

 

2. stationary and sliding backwards, or

 

3. rotating forwards and slipping backwards.

 

Whichever, there is surely some equipment on the locomotive to detect such a situation and do something about it? Even the mouse in my hand can detect which way it is moving without touching the desk.

 

Martin.

[The Stationmaster]

Without speculating about the cause, it would be interesting to know what was happening between the locomotive wheels and the rail for such a sustained period of time. They could be:

 

1. rolling backwards, or

 

2. stationary and sliding backwards, or

 

3. rotating forwards and slipping backwards.

 

Whichever, there is surely some equipment on the locomotive to detect such a situation and do something about it? Even the mouse in my hand can detect which way it is moving without touching the desk.

 

Martin.

 

The normal detection method is the Driver. The whole thing sounds rather odd and there are a number of likely reasons for it but to select any one of them would be to speculate on the cause. All we know thus far is that for some reason (unknown) the train began to go backwards and continued to do so with an implication (solely from the wording of the RAIB item) that no action was taken by the Driver until a radio broadcast had been made.

 

We now await the RAIB's report, no doubt sometime during next year.

[Leicester Thumper]

It is Autumn..... that's all im saying.

 

Yeah it would be interesting about the wheel movements....... got my physics memeory working now that has!

[JOHNMCDRAGON]

I seem to remember reading years ago of this happening to a passenger train. I think it was a boat train from Liverpool sliding backwards in the Lime St tunnel or thereabouts and the crew were unaware they were going backwards. It was due to greasy rail ,train too heavy for loco etc. And it resulted in a minor collision. I think it happened on the real railway more than we modellers realise.

[martin_wynne]

It is Autumn..... that's all im saying.

The incident occurred on 17th August.

[roythebus]

Quite odd I must admit. A lot could depend on the type of wheel slip protection on the loco and train, if indeed there is any. I can't speculate as I'm not familiar with that type of loco.

 

It happened to me once in about 1977 when trying to lift a 900 ton ballast train out of Kings Cross one Sunday with a class31. We managed to get up Holloway bank, after that the wet rail conditions meant forward movemtn was impossible, and the unfitted train decided to go backwards. The loco brake was insufficient to stop the train. Luckily the guard realised what was happening and applied a few wagon handbrakes which averted a more serious incident.

 

The WCML is luckily no longer fitted with catch point which would have derailed such a runaway.

[James]

James - they can do crew changes, the departure time is not necessarily the time the driver started

 

I'm well aware of operating practices; but my comments are based on the research I undertook for my former safety director on ROGS 25; though my work was based around OTM crews, the fundamentals are the same.

[beast66606]

I'm well aware of operating practices; but my comments are based on the research I undertook for my former safety director on ROGS 25; though my work was based around OTM crews, the fundamentals are the same.

 

The specifics of this case are what matters though

[34theletterbetweenB&D]

Look at the time the event happened, just before 2 am. Anyone who has worked night shifts knows when errors, incidents and accidents peak, and why; there is a large body of evidence from a range of industries around the world, and the data is very consistent.

[Guest D.J.K.]

These things are complicated, and very serious.

 

I've done quite a bit of railway incident investigation in my time, and things like this are never as straight forward as they seem......................

 

Wait for the The RAIB report.........................

[martin_blog_search]

Report published today:

 

http://www.raib.gov.uk/cms_resources/110815_R152011_Shap_Summit.pdf

 

Martin.

[trisonic]

Very interesting read. Many thanks for posting the link.

 

Best, Pete.

[Ohmisterporter]

A very interesting read as usual from the RAIB. As a shift worker on 12 hour day and night shifts I can concur with the findings related to fatigue in the report. However, I am left confused by the chart figure 8, which seems to show "fighting sleep" at the same point as maximum "alertness". Either they have the graph's description wrong or I am completely missing the point. Clarification gratefully accepted.

 

Geoff.

[roythebus]

A good argument to use the secondman in a loco?

[Ragtag]

However, I am left confused by the chart figure 8, which seems to show "fighting sleep" at the same point as maximum "alertness". Either they have the graph's description wrong or I am completely missing the point. Clarification gratefully accepted.

 

Geoff.

 

 

Hi Geoff

 

The graph is plotted against 2 different y axes, so we can assume (from this data, ignoring common sense) there is no direct causal relationship between sleepiness and alertness and we should be looking at the effect of # hours sleep on each. It's a peculiar graph regardless, in that the axes have been inverted for some reason. Since it doesn't refer to a source I assume it's been created by the RAIB.

 

 

If we look at their explanation of results:

 

b. Time spent asleep: most people need between seven and eight hours sleep each night; sleeping less leads to a slow build up of sleep loss which leads to reductions in alertness and performance on test.

 

they are saying that sleeping for four hours a night leads to a build up of sleep loss (sleepiness) [which the graph shows], and a loss of alertness [which the graph doesn't show]. I suspect they have accidentally inverted the second y axis, and the arrow labelled "more" should be pointing upwards. It's an easy mistake to make, since they've deliberately plotted sleepiness on an inverted axis to get a nice set of matching lines.

 

 

 

HTH,

 

Stu

 

 

EDIT: Here's a quick rework of the data. I've done away with the inversion jiggery-pokery on the sleepiness scale (no idea why they did it that way, the data is far more intuitively read like this) and inverted the alertness data. This graph now fits the description the RAIB give in the report.

 

[meil]

 

 

It's an easy mistake to make

 

Especialy if you haven't been sleeping well.

[Ohmisterporter]

Thanks Stu,

 

This confirms my suspicion that this would have been better done by having two separate graphs. Not to detract in any way from the usual RAIB reports that get to the bottom of things far better than public enquiries that IMHO end up playing the blame game and all round back covering exercises.

 

EDIT : I have just seen your new graph. Plotting "Sleepiness" and "Alertness" this way clarifies things.

 

Geoff.

[Ragtag]

I'd guess they've fallen foul of the classic mistake - they've written the explanations first and then made the graphs to fit those. Had they made the graphs first they'd have realised the error when they tried to write the explanation. All in all the report does make sense, and they've gotten the important bits right - just a botched graphic it would seem.

[Oldddudders]

I'm pleased to record that I haven't worked regular night shifts since 1973. I did 5 years of round-the-clock work in Control, with a week of nights every 4th week. The first night - Monday, was usually the worst, as stated in the report, but after a few nights - and assuming an undisturbed day in bed - the body adjusted. I occasionally worked a fortnight of nights, and while that knackered your social life, it did become a very tolerable routine in other respects. In those days "doubling back" off nights was normal - you finished Monday morning,and were back 8 hours later for late shift, which could be hard work, not least because Monday would be the day when problems often arose, due to that week's rosters for traincrews etc being new.

 

Other grades had it much worse. Signalmen finishing a week of nights would regularly double back Sunday, from nights to late turn, and then double back again to do early turn Monday - perhaps they still do. Thus in 40 hours from 2200 Saturday to 1400 Monday they would work 24 - which is a lot. Is their work "Safety Critical" - to use a modern term -? You and I might think so.

 

It has been suggested that having a secondman in the cab might have promoted greater alertness. True, but human nature suggests in many cases the secondman would be sent off to get his head down in the back cab!

[Glorious NSE]

It has been suggested that having a secondman in the cab might have promoted greater alertness. True, but human nature suggests in many cases the secondman would be sent off to get his head down in the back cab!

 

If you compare with the US where 2-person crews are still the norm then fatigue related collisions, (often fatal to both crewmembers,) are not at all unknown. Whilst there's lots of differences in how the industries appear to work that would suggest having two people in the cab is not a guaranteed way of preventing problems.

[The Stationmaster]

It has been suggested that having a secondman in the cab might have promoted greater alertness. True, but human nature suggests in many cases the secondman would be sent off to get his head down in the back cab!

All too true, or both would be nodding off together. In the modern traction era the presence of a second person in the cab doesn't seem to be much of a mitigation against things like this and indeed has on occasions been the cause - one spectacular (but fortunately casualty free) SPAD/collision on the Western when an HST ran up the back of the Malago Vans between Hayes & Southall one afternoon was a direct result of the two Drivers paying more attention to their conversation than a succession of adverse signals which ended with the red one.

 

Night shifts and shift patterns, especially for traincrew, have been the source of many studies over the years (including some which I think are far more relevant than the stuff quoted by the RAIB) and they have usually shown that changeover patterns are by far the biggest factor in subsequent drowsiness/loss of attention. One factor which is rarely mentioned however is increasingly extended travel to work times (something which my final 'big railway' employer rigorously applied with traincrew and was pretty keen on for the rest of us) plus time on what are bundled up as 'ancillary duties' Another thing - which Ian highlighted well - is that it has long been known than longer periods of night turns in a roster help to adjust people to working at nights, I for instance could never understand why people wanted a Rest Day in the middle of a week of nights (a feature which seem now to have vanished from modern rostering, thank goodness). And it is of course worth remembering that many people, albeit not in the mainline rail industry, work permanent nights for years on end with no apparent ill-effects on their concentration levels etc.

 

As for the RAIB Report - no comment