Devonport derailment

 
  Myrtone Chief Commissioner

Location: North Carlton, Melbourne, Victoria
I have been obsessed with vigilance control before. Obsessed with the fact that drivers of these trains have to press a button just to keep the emergency brake off. These are absent from any road vehicles, and it makes think that drivers of these trains need to be more disciplined than road vehicle drivers.
The fact that rail vehicles have things like this but road vehicles don't and that rail vehicles don't crash as often may make one tempted to suppose that this is part of the reason why.

Sponsored advertisement

  RTT_Rules Dr Beeching

Location: Dubai UAE
I have been obsessed with vigilance control before. Obsessed with the fact that drivers of these trains have to press a button just to keep the emergency brake off. These are absent from any road vehicles, and it makes think that drivers of these trains need to be more disciplined than road vehicle drivers.
The fact that rail vehicles have things like this but road vehicles don't and that rail vehicles don't crash as often may make one tempted to suppose that this is part of the reason why.
Myrtone
Vigilance control on road vehicles has limited benefit, within a few fractions of a second of their eyes drooping, they have crossed the centre line and killed a family, long before the brake pads have even touched the disks.

With a train, the death toll can be in the hundreds, overall cost $10m's just for the lost rolling stock, plus medical costs, fairly straightforward to implement and it actually works.

However, car/truck automation is literally just around the corner and now we have things like auto braking in modern trucks which is proving to reduce accidents and lane discipline detection with steering shake to alert the driver.
  RTT_Rules Dr Beeching

Location: Dubai UAE
Easily my favourite part of that report:

Spencer Street Signal Box – I haven’t got, I haven’t got a platform to put it in. Metrol – You got no platform to put it in?
Spencer Street Signal Box– Nope!
Metrol – (INAUDIBLE)
Spencer Street Signal Box – The only place I’ll have is platform one when the Adelaide goes. And then it’ll run down and hit the bottom anyway.
Metrol – Yeah.....yeah, that’s the thing.
Spencer Street Signal Box – And it’ll run out of wire and then of course nothing moves. Metrol – Mmm-hmmm.
Spencer Street Signal Box – And that’s what they want to do?
Metrol – Yep, that’s right.
Spencer Street Signal Box – Righto
Metrol – OK?
Spencer Street Signal Box – No worries.
Metrol – No worries, seeya.
Spencer Street Signal Box – Yeah, hey listen.....(cut off)

In fact I’m chuckling copying it across right now.

A very unfunny situation, but picturing the unfortunate No 1 signalman as he hangs up is priceless...

The whole episode is an interesting juxtaposition of a number of human errors and some brilliant human creativity under stress.
potatoinmymouth
A train is running on only gravity at speed with no body on board and they question it will run into an unwired area!
  ngarner Junior Train Controller

Location: Seville
VC, in the 1980s, was questionable in its effectiveness in keeping crew awake on diesels, from my experience.

I can cite an example, in the early 1980s, where I was the loco assistant on a down Seymour goods out of Melbourne about 9pm. My driver had obviously not slept well during the day (understandable in my opinion, as I had that problem regularly) and got me to take the controls after going through Broadmeadows. He quickly fell asleep and I watched his arm rise to punch the roof mounted VC every time the air hiss began prior to the VC going off properly. We stopped at Wallan for some time waiting line clear on the goods ahead of us, where I had a (solo) meal break. We got line clear and I drove as far as Broadford where we changed over to our, return trip, up goods.
My mate took the drivers seat after the changeover and soon after leaving Broadford queried me whether we'd had 'tucker' to which I replied "yep, at Wallan". He had slept the whole way north, after I took the controls, and kept the VC from stopping the train the entire time, to the point of not being aware of anything that had happened during that time.
I know that I did likewise, more than once; the most memorable being on an up Bendigo goods where I lost track of time between Kyneton and Riiddells Creek and obviously hit the VC as needed to keep the brakes off.


If they've improved VCs since then that would be good to know, but I have my doubts about the likelihood of that happening. I know there was talk at the time of the Germans having invented glasses that had a light beam that had to be broken by the wearers eyelashes regularly breaking the beam by blinking or the brakes would be applied. It would certainly be a better method than the one the VR used, which assumed that a sleeping person couldn't/wouldn't raise their arm and press a roof mounted button. The only thing it really did was to prevent any of the crew from leaving the cab for more than 90 seconds; a reaction to the obvious absence of the fireman in the cab of the Southern Aurora for a prolonged period of time in 1969, since prior to that accident, to my understanding, there was no VC at all. A more positive reaction was to monitor senior driver's state of health a lot more closely and restrict them to pilots or other jobs that wouldn't have such disastrous consequences if they had a health problem whilst on the job.

Having said that, either Tasrail don't have even that basic level of protection, or it wasn't working at all, because flawed as the VR's VC was it should have stopped the train long before the siding diversion was necessary in Devonport.

Neil
  YM-Mundrabilla Minister for Railways

Location: Mundrabilla but I'd rather be in Narvik
VC, in the 1980s, was questionable in its effectiveness in keeping crew awake on diesels, from my experience.

I can cite an example, in the early 1980s, where I was the loco assistant on a down Seymour goods out of Melbourne about 9pm. My driver had obviously not slept well during the day (understandable in my opinion, as I had that problem regularly) and got me to take the controls after going through Broadmeadows. He quickly fell asleep and I watched his arm rise to punch the roof mounted VC every time the air hiss began prior to the VC going off properly. We stopped at Wallan for some time waiting line clear on the goods ahead of us, where I had a (solo) meal break. We got line clear and I drove as far as Broadford where we changed over to our, return trip, up goods.
My mate took the drivers seat after the changeover and soon after leaving Broadford queried me whether we'd had 'tucker' to which I replied "yep, at Wallan". He had slept the whole way north, after I took the controls, and kept the VC from stopping the train the entire time, to the point of not being aware of anything that had happened during that time.
I know that I did likewise, more than once; the most memorable being on an up Bendigo goods where I lost track of time between Kyneton and Riiddells Creek and obviously hit the VC as needed to keep the brakes off.


If they've improved VCs since then that would be good to know, but I have my doubts about the likelihood of that happening. I know there was talk at the time of the Germans having invented glasses that had a light beam that had to be broken by the wearers eyelashes regularly breaking the beam by blinking or the brakes would be applied. It would certainly be a better method than the one the VR used, which assumed that a sleeping person couldn't/wouldn't raise their arm and press a roof mounted button. The only thing it really did was to prevent any of the crew from leaving the cab for more than 90 seconds; a reaction to the obvious absence of the fireman in the cab of the Southern Aurora for a prolonged period of time in 1969, since prior to that accident, to my understanding, there was no VC at all. A more positive reaction was to monitor senior driver's state of health a lot more closely and restrict them to pilots or other jobs that wouldn't have such disastrous consequences if they had a health problem whilst on the job.

Having said that, either Tasrail don't have even that basic level of protection, or it wasn't working at all, because flawed as the VR's VC was it should have stopped the train long before the siding diversion was necessary in Devonport.

Neil
ngarner
Would the VC application (or not) be dependent upon whether the TasRail train 'powered' away from Railton or just 'ran away'. Does VC operation still depend on the throttle setting being above notch 0?
IIRC there was a means of deactivating the original VC in Victoria by means of loosening an airline union somewhere many years ago? Trains were allegedly stopped mid section so that Loco Inspectors could check the VC operation unannounced?
Was the Mainline at Railton not protected from a Siding runaway by catchpoints or other derail mechanism?
  lkernan Assistant Commissioner

Location: Melbourne
A loco under remote control is basically being driven as a Multiple Unit setup where the lead "loco" is the remote box.

Correct me if i'm wrong but i'd assume a second unit's VC does nothing in that situation.
  YM-Mundrabilla Minister for Railways

Location: Mundrabilla but I'd rather be in Narvik
A loco under remote control is basically being driven as a Multiple Unit setup where the lead "loco" is the remote box.

Correct me if i'm wrong but i'd assume a second unit's VC does nothing in that situation.
lkernan
If that is the case (I am not doubting you) it should not be allowed without other safeguard(s).
  RTT_Rules Dr Beeching

Location: Dubai UAE
Would the VC application (or not) be dependent upon whether the TasRail train 'powered' away from Railton or just 'ran away'. Does VC operation still depend on the throttle setting being above notch 0?
IIRC there was a means of deactivating the original VC in Victoria by means of loosening an airline union somewhere many years ago? Trains were allegedly stopped mid section so that Loco Inspectors could check the VC operation unannounced?
Was the Mainline at Railton not protected from a Siding runaway by catchpoints or other derail mechanism?
YM-Mundrabilla
Railton is not protected by catchpoints (Google location on Google Earth) and if I recall my frequent visits around Tas in 90's when it had two separate railway operations under two different organisations, neither is any other loop, siding or junction.  Now while you can count the number of train movements each day in Tassie on 3 hands, maybe 4 and there hasn't been RPT pax for 40+ years, in Qld, things are not much different. For example, none of the passing loops on NCL have catch points.
  YM-Mundrabilla Minister for Railways

Location: Mundrabilla but I'd rather be in Narvik
Would the VC application (or not) be dependent upon whether the TasRail train 'powered' away from Railton or just 'ran away'. Does VC operation still depend on the throttle setting being above notch 0?
IIRC there was a means of deactivating the original VC in Victoria by means of loosening an airline union somewhere many years ago? Trains were allegedly stopped mid section so that Loco Inspectors could check the VC operation unannounced?
Was the Mainline at Railton not protected from a Siding runaway by catchpoints or other derail mechanism?
Railton is not protected by catchpoints (Google location on Google Earth) and if I recall my frequent visits around Tas in 90's when it had two separate railway operations under two different organisations, neither is any other loop, siding or junction.  Now while you can count the number of train movements each day in Tassie on 3 hands, maybe 4 and there hasn't been RPT pax for 40+ years, in Qld, things are not much different. For example, none of the passing loops on NCL have catch points.
Might help if we knew what RPT meant and whose NCL you are talking about, please (NSW or Bananaland)?

Crossing loops (ie running lines) where trains are crewed are a different case from shunting/marshalling yards especially where movements are, in effect, remote shunting. If there is one thing worse than remote shunting it is remote shunting complete trains.
  ngarner Junior Train Controller

Location: Seville
In Victoria both the driver and loco assistants VC were functional when the loco reverser was set for a direction and the brake handle was in the running position. The position of the throttle was irrelevant as an experienced driver could maintain his trains speed without using the throttle in some locations, e.g. a goods on the down Seymour line between Broadford and Tallarook (although I wouldn't necessarily want to been the guard at the back with the coupler slack running in and out on the way).

Yes, the VC could be cut out by closing some air cocks in the nose or short end of mainline locos (excluding the E, F, Y and, probably, W classes) but these cocks were secured open by wire ties which ran through a hole in the cock handle itself and around the pipe, a bit like a cable tie. To close the cock meant breaking the tie(s) and if a loco came into a fuelling point with damaged or missing ties the last crew to use the loco on a road trip would be getting a "please explain" from Head Office.
Unless you have an extremely good reason there would be penalties applied. Probably the only reason accepted would be a faulty VC but, I'm fairly, sure they would have tested the loco to confirm that before accepting the drivers word.
In 5 years I never saw anyone dare to tamper with the VC for these reasons.
Yes, in MU, only the lead unit would have its VC operational as the brakes would be cut out on the trailing units.
You might note the H class did have VC fitted, which was an absolute pain on the hump yard pilots when all you did was push a rake up a slight hill at 2.5mph and then roll back to the other end of the arrivals yard and couple up to the next rake to be humped.

It's possible trains were stopped to check on the VC but, again, I never experienced that. I can only speculate that soon after VCs were fitted crews worked out how to avoid having to respond to them but Head Office got wise and had the ties fitted to stop it from happening anymore.

From what I've read about this accident  the TR was the lead (only powered) loco but the driver rode in DV1 so VC in the TR would have been pointless. Does DV1 have any controls or is it the remote unit the only way of controlling the train? If the remote does not have a VC fitted then that would explain the ability to run away and not be stopped. With the brakes off and the throttle set for the Devonport direction a (Victorian) VC should have stopped it, if fitted.

If the remote doesn't have a VC I agree with YM-M that there need to be other safeguards in place.

Neil
  RTT_Rules Dr Beeching

Location: Dubai UAE
Would the VC application (or not) be dependent upon whether the TasRail train 'powered' away from Railton or just 'ran away'. Does VC operation still depend on the throttle setting being above notch 0?
IIRC there was a means of deactivating the original VC in Victoria by means of loosening an airline union somewhere many years ago? Trains were allegedly stopped mid section so that Loco Inspectors could check the VC operation unannounced?
Was the Mainline at Railton not protected from a Siding runaway by catchpoints or other derail mechanism?
Railton is not protected by catchpoints (Google location on Google Earth) and if I recall my frequent visits around Tas in 90's when it had two separate railway operations under two different organisations, neither is any other loop, siding or junction.  Now while you can count the number of train movements each day in Tassie on 3 hands, maybe 4 and there hasn't been RPT pax for 40+ years, in Qld, things are not much different. For example, none of the passing loops on NCL have catch points.
Might help if we knew what RPT meant and whose NCL you are talking about, please (NSW or Bananaland)?

Crossing loops (ie running lines) where trains are crewed are a different case from shunting/marshalling yards especially where movements are, in effect, remote shunting. If there is one thing worse than remote shunting it is remote shunting complete trains.
YM-Mundrabilla
RPT - Regular Public Transport

NCL - North Coast Line (Qld)

True, but I calculated previously that they are up to around 40,000 movements since Remote operation was introduced, so they must have got something right more than once.
  Z1NorthernProgress2110 Chief Commissioner

Location: Burnie, Tasmania


From what I've read about this accident  the TR was the lead (only powered) loco but the driver rode in DV1 so VC in the TR would have been pointless. Does DV1 have any controls or is it the remote unit the only way of controlling the train? If the remote does not have a VC fitted then that would explain the ability to run away and not be stopped. With the brakes off and the throttle set for the Devonport direction a (Victorian) VC should have stopped it, if fitted.

Neil
ngarner
Woah there. The Driver uses the remote control box while leading out of Devonport in DV1. On the return to Devonport, the driver would be using the normal controls in TR11.
The driver would be in the cement solo control room to help load the cement into the wagons, using the remote to load two cement wagons at a time.

Link here is the cab of DV1, the hand held remote control box sits in the spot near horn controls. Perhaps the green button is the VC.

http://railtasmania.com/photogal/showimage.php?id=633
  Valvegear Dr Beeching

Location: Norda Fittazroy
For all of the interesting discussion here, the fact remains that a train got away and people were injured. I would be very interested to (eventually) find out what is being done to ensure that it cannot happen again.
  YM-Mundrabilla Minister for Railways

Location: Mundrabilla but I'd rather be in Narvik
For all of the interesting discussion here, the fact remains that a train got away and people were injured. I would be very interested to (eventually) find out what is being done to ensure that it cannot happen again.
Valvegear
It will be equally interesting to find out, not only how this particular train got away but also the thinking/logic/safety case behind the original remote control concept and implementation.
  ngarner Junior Train Controller

Location: Seville
Thanks for the information Z1N-P; it clears up my misconceptions of the setup.
The photo, unfortunately, doesn't show any labels but you could be correct about the green button, as it isn't obvious what it is there for; unless it is for power on? The large lever would be the throttle and the small black switch would be the reverser. Amusing they use a recycled speedo (on near left). The other gauge is for the brake pressure as I can just make out Kpa below the needle. The rest of the switches would be the usual mix of lights, generator field switch and such. one of the two push buttons in the centre of the console might be VC but they're awkwardly placed for that function.

As the remote was in use, the VC, if any, on the TR would have been cut out and once the train was out of range of the remote any VC function on it (the remote) would have been unable to interact with the TR so the runaway continued; but that begs the question of secondary safeguards when the loco loses communication with the remote, as questioned earlier in this thread.
Either way, the results of the investigation will be very interesting reading indeed and what is proposed to prevent a repeat.

Neil
  YM-Mundrabilla Minister for Railways

Location: Mundrabilla but I'd rather be in Narvik
Rightly or wrongly (more likely) I assumed that the stick with the black knob would have been the whistle. Again, without any knowledge, the 'recessed' green button seems a strange design for the VC button to me.
Apart from the emergency dump valve on the cab wall is there no other brake valve?
  ngarner Junior Train Controller

Location: Seville
That's a really good question! I should have picked that up too. Nothing else obvious in the picture looks brake related so unless there is a brake stand out of view, your guess is as good as mine.
In my experience VCs were not surrounded so as to be easy to push so your point about the button being 'recessed' is a valid one IMO.

Neil
  Z1NorthernProgress2110 Chief Commissioner

Location: Burnie, Tasmania
There is no brake stand in the cab of DV1.

The raised bit between the horn control and ATN track warrant you sit the remote control box, use it like a desktop

http://www.cattron.com/dnn/CANTRAC/tabid/153/language/en-US/Default.aspx

See the yellow box? that sits on there instead of being strapped to the driver.


When the train is loading or unloading, no one is in the loco or DV van, they are in the silo's helping load and unload the wagons.

All the braking etc all comes from TR11. DV1 is just a converted engineless loco to have someone to ride in on there way back to Railton.
  Z1NorthernProgress2110 Chief Commissioner

Location: Burnie, Tasmania
Rightly or wrongly (more likely) I assumed that the stick with the black knob would have been the whistle. Again, without any knowledge, the 'recessed' green button seems a strange design for the VC button to me.
Apart from the emergency dump valve on the cab wall is there no other brake valve?
YM-Mundrabilla
Yes, not sure what the green button is for. But the stick with the black knob is the horn control.

I've been on this very early in its life, the driver often pressed the green button and automatically thought it was a VC.
  ngarner Junior Train Controller

Location: Seville
Well, there you go.
My speculation was somewhat off, to say the least.
The last two posts clear up a lot. The green button, if pushed regularly for no apparent reason, would be a VC.
YM-M was correct about the whistle/horn lever. The picture of the (missing) remote made things a lot more understandable.
Now for the report on what went wrong...

Edit: This train doesn't seem to have been having a very good time of it recently; just found another post with the same title as this one where TR03 (I think) appeared to have split the points at Devonport with same cement load.


Neil
  YM-Mundrabilla Minister for Railways

Location: Mundrabilla but I'd rather be in Narvik
Thanks Neil and Z1..... for your contributions.
As has been said we will await the results of the ATSB investigation with interest.
You will all have gathered that I am not a fan of some of these gee whizz radio controlled computer based thingys. Smile
  ngarner Junior Train Controller

Location: Seville
ATSB website indicates that the report on this is probably going to be released in the 3rd quarter of 2019.
Nearly a year to wait to find out what they think caused it. Don't think I have an alarm that I can set for 12 months duration

Neil
  KRviator Moderator

Location: Up the front
ATSB website indicates that the report on this is probably going to be released in the 3rd quarter of 2019.
Nearly a year to wait to find out what they think caused it. Don't think I have an alarm that I can set for 12 months duration

Neil
ngarner
Have a look at some of the OTSI reports! Cityrail puts one car inthe dirt after SPADing at Hornsby in 2015 and coming up to 3 years later there's still no report.

It's taken them nearly 2 years to release the report into the NP23 debacle at Muswellbrook, and the ATSB 2.5 years into a phuckup at Kilbride where the crew detrained passengers onto a live line...

Bottom line - don't hold your breath for any meaningful information to come out.
  BrentonGolding Chief Commissioner

Location: Maldon Junction
https://www.atsb.gov.au/newsroom/news-items/2018/freight-train-runaway-and-derailment/?fbclid=IwAR0P6UhqCxeFsy1KvpV257_rRg5R-YlQR9nBDoRLvXtlwP904sjZPG9DgcI
Some news on this issue.
Z1NorthernProgress2110
Oops! Sounds like a list of recommendations as long as your arm will follow this incident!

BG

Sponsored advertisement

Display from: