Collision and derailment near Salisbury (Fisherton Tunnel) 31/10/21

[pompeyfan]

it wouldnt be a rear cab the train is travelling on the left hand rail which is the rails it would be driving on in the direction. Just a thought.

exactly, which is why I think the live broadcast of a unit with its hazards on is not the incident SWR service.

 

[MadMac]

It’s certainly technically possible to design signalling such that track circuits must be operated in a logical order, and for signals to be turned to danger if an irregularity is detected.

”Out of sequence“ alarm is generally programmed in at the “control centre“ end nowadays. Unlikely that it exists here in 40 year old relay based systems operated from a “hard” panel.

 

[Wilts Wanderer]

I suspect there is a physical explanation for why both the units appear to be side by side in the tunnel but were apparently travelling in the same direction over the same track, if the rear car of the GWR service had derailed and the following SWR service struck it, as was reported upthread.

 

[ScotRail158725]

do they have any reflective stuff on them to help a driver see a unit signal. maybe the number plate?

I’m pretty sure the plate with the signal number on it works in the same way as roadworks signs at the side of the road for example by reflecting in the headlights

 

[redbutton]

do they have any reflective stuff on them to help a driver see a unit signal. maybe the number plate?

Technically the number plates are reflective but in my experience they always have crud on them. They're only required to be legible from about 20 metres away afaik.

 

[Gloster]

I don’t know if the junction has a switch diamond or not and that is in a way irrelevant. The second train shouldn’t have been anywhere near the junction before the first had cleared the section through the tunnel. (This is based on what little we know at the moment.)

 

[Spirit555]

SWR was 3 x 159 (159014, 159017 & 159018).

Pics of 159 show car 52803 which is from 159102. Still another illustration of how fragile the network is.

Mark

 

[bramling]

It is possible and newer systems than that at Salisbury (the control system rather than the interlocking as I understand it) detect a track circuit occupying or clearing out of sequence. I don't know whether it puts signals back to danger though.

The Met Line has such a system on the Amersham end, primarily as a mitigation against a wrong side track circuit failure occurring should a train lose detection as a result of leaf mulch in autumn. The Central Line certainly has it too. I believe in these instances it would set signals to danger, and is “reset” by the passage of a train having to run through under failure conditions.

Interesting though this topic is, it’s a bit of a digression. The likelihood of a derailed train disappearing from track circuits altogether is very remote indeed, as the action of a derailment is highly likely to damage something, which being safety-critical equipment will then fail in a safe state.

 

[Parallel]

Glad to hear that everyone involved is OK. This could've ended so much worse. Thoughts to everyone involved in what must've been a scary incident.

 

[Horizon22]

Politics For All has leaked the incident report on Twitter

That isn’t the incident report. That will be significantly longer & detailed racking up to (probably) hundreds of log items.

 

[richw]

The second train shouldn’t have been anywhere near the junction before the first had cleared the section through the tunnel.

The photos circulating appear to show that the SWR was on the other track. That would normally be ok on a bidirectional line.

 

[Crossover]

I think we can count ourselves lucky we haven't had any especially when you look at the damage to the Swr unit.

My stomach lurched when I saw it come up on BBC news. Thoughts with all the train crew

Agreed. That cab looks to have taken something of a battering

 

[David Stewart]

I don’t know if the junction has a switch diamond or not and that is in a way irrelevant. The second train shouldn’t have been anywhere near the junction before the first had cleared the section through the tunnel. (This is based on what little we know at the moment.)

Yes, I accept that it shouldn't have been near the junction but it was. It would not be the first time that for whatever reason a train has passed a protecting signal (wheelslip, equipment failure etc) and taken switch diamonds as a facing crossover (Greenhill lower many years ago is one).

 

[LUYMun]

Terrible news. I send my prayers to those affected in the incident.

 

[westcoaster]

Thats what I was thinking, the image show in the tunnel makes no sense

I think the image from the bbc shows the front 2 cars of the GWR unit. The picture from in the tunnel shows the rear 2 cars, and the front of the SWR unit.

 

[Helvellyn]

The fact 159102 is pictured, but that RTT says 159014 and 159017 were working 1L53 suggests either RTT is wrong, or possibly an ECS train (159102) in the tunnel?

The most important thing is that no fatalities and fortunately it seems only twelve injuries, none of which were serious.

 

[Surreytraveller]

Agreed. I don’t have a problem with sensible and informed discussion, which for the most part is what we see here.

Exactly. Its an internet forum. That's what its for

 

[alxndr]

I’m pretty sure the plate with the signal number on it works in the same way as roadworks signs at the side of the road for example by reflecting in the headlights

Depends on the age. The older ones are not retroreflective at all, just black and white paint. ID plates are often mainly black too, which won't help visibility. It's more down to route knowledge than spotting plates.

 

[RPI]

I think the image from the bbc shows the front 2 cars of the GWR unit. The picture from in the tunnel shows the rear 2 cars, and the front of the SWR unit.

So the unit split?

 

[D365]

I guess the guard has switched the hazard lights on the rear of the SWR unit.

Yes. The daytime running lights are always used as hazards, as they are the brightest.

 

[bramling]

The fact 159102 is pictured, but that RTT says 159014 and 159017 were working 1L53 suggests RTT is wrong, or possibly an ECS train (159102) in the tunnel?

I wouldn’t place too much reliance on the unit numbers shown on RTTT, especially on a day when the railway was already disrupted as a result of the weather.

 

[bishdunster]

Coastguard helecopter now returning to St athen, South Wales, and Dorset and Somerset air ambulance en route to Southampton on adsbexchange radar.

 

[323 Class]

The fact 159102 is pictured, but that RTT says 159014 and 159017 were working 1L53 suggests either RTT is wrong, or possibly an ECS train (159102) in the tunnel?

The most important thing is that no fatalities and fortunately it seems a only twelve injuries, none of which were serious.

Not unusual. I was at Stalybridge a couple of days ago and RTT said 150101 was coupled to a 156. When it entered P5, it was 156423 & 156428 so proof that its not totally accurate.

 

[Horizon22]

So the unit split?

That does appear to have been the case, yes.

 

[irish_rail]

As a driver, I can confirm that it is NOT easy to see an outed signal in darkness, whether it has a plate on it or not.

 

[158756]

I’m confused, I don’t think that train is actually involved in the incident. If it’s the rear cab of the incident train it’s likely to be on the other track surely?

It is on the right track to be the rear cab I think, with the hi-viz guys standing on the other line from that direction we can't see.

 

[pompeyfan]

That does appear to have been the case, yes.

The unit with its flashing lights on is an SWR train and I believe is at Laverstock north.

Edit: having just seen it on TV I suspect it’s the rear cab on the SWR unit. The rear bogie does not appear to be on the track.

 

[Ken H]

Technically the number plates are reflective but in my experience they always have crud on them. They're only required to be legible from about 20 metres away afaik.

so a driver is expected to see a signal that is unlit on a very dark night, and treat it as a stop signal? Hmm

 

[43096]

As a driver, I can confirm that it is NOT easy to see an outed signal in darkness, whether it has a plate on it or not.

I thought that was what route knowledge was for? Given that 1L53 should have been held at the signal protecting the junction to allow 1F30 to pass, the driver would have had a single yellow at the previous signal, so would have expected to stop at the protecting signal. Plenty of questions on this, and given the time of year, railhead conditions may also be a consideration.

 

[Gloster]

If, as it think was mentioned earlier, a train from Salisbury towards Romsey had just passed, then the points and diamond might still be set for that route, which could (the conditional is important) divert a down train from the Andover line to the Up line. It still shouldn’t have been there. I wonder if we have two bits of very bad luck combining.