Video: BHP’s twisted train wreck

Straight out of the movies this:

Trains to be running again within a week.

David Llewellyn-Smith

David Llewellyn-Smith is Chief Strategist at the MB Fund and MB Super. David is the founding publisher and editor of MacroBusiness and was the founding publisher and global economy editor of The Diplomat, the Asia Pacific’s leading geo-politics and economics portal.

He is also a former gold trader and economic commentator at The Sydney Morning Herald, The Age, the ABC and Business Spectator. He is the co-author of The Great Crash of 2008 with Ross Garnaut and was the editor of the second Garnaut Climate Change Review.

Latest posts by David Llewellyn-Smith (see all)


  1. Maybe BHP could come to Sydney to help out with the new light rail / tram thing, if they can get that 1.5 km done in a week and these muppets can’t get it done in 2 years I am sure they could show them a thing or two.

    • Yeah…because building a bit of track in the desert is directly comparable to building in the CBD.

    • It’s easy – and anyone can do it if it costs them a lot more than having it working.

      Good, fast, cheap.

      You can guess which two they picked.

    • Not with multi-million revenue lost each day when the heavy-haul railway line is shut down. Paying demurrage charges for the bulk carriers on standby outside the port of Port Hedland and rerouting bulk carrier ships at short notice. Adds up quite quickly. If you impose that kind of LD charges the “muppets” will work really quickly.

      • Hey, you’ve just spelled ACCOUNTABILITY!

        it’s a long word, that’s why nobody in NSW government can spell it. Also they don’t have a concept of CONSEQUENCES…. might be related.

    • Cost IndexMEMBER

      Crashes like this are frequently blamed on pilot/driver error, never mind fatigue, preposterous time targets etc, where the systemic fault actually was created higher up the chain. But with out someone on the frontline to take the blame, who will? The middle level manager that oversaw it all? Answer: “It was the system” faceless blame on a computer can be plausibly assigned and everyone will nod in agreement, heck even bonuses will be handed out on this

      • darklydrawlMEMBER

        Yes – the real failure and issue here is the lack of a dead man’s brake or ability to remotely stop the engines. You can be confident that those system will be in place soon. This incident is likely to be costly to BHP on several fronts.

      • DD, I had forgotten about the deadmans brake until you mentioned it. I thought all trains had those things?

    • had that ac in Indonesia been left on auto
      it wouldn’t have crashed
      pilot error.
      HUGE pilot error
      Huge lawsuit coming up

      • That must be why Boeing is busily releasing warnings about electronic stabilisers.
        Because it was all the pilot’s fault.

      • And had the aircraft been pilotless, the manufacturer bears all responsibility for every crash that ever happens, no possible doubt at all.
        This is why autonomouos cars are a lot further away than anyone thinks. Autopilots have been capable of completely autonomous flight for 40+ years, yet no autonomous passenger aircraft have been produced.

      • WW, rubbish. Have you seen the EOP’s for a new 737? Unless you have, you are talking rubbish. I also suggest you youtube “children of magenta” if you think that autopilots are all seeing and knowing. Sophisticated autopilots bring in a totally new raft of issues that need to be worked around. You’ve got no idea how quickly two pilots can be task saturated during a busy flight phase such as takeoff or landing, especially in busy airspace and a modern sophisticated aircraft.

      • t
        the instrument was an aoa indicator
        no influence at all after to.
        you are taking through yr beanie

      • well good thing the data recorders have been recovered
        we will soon know, and i will let you know
        in case u miss it.

      • Tell me what an autopilot will do if the ASI is misreading and what a human will do when he notices the plane isn’t acting like it should. I know you seem to think AI will take over the world, but in the 35 million euro helicopter I used to fly, that shares the autopilot with the A380, the processors were 486’s. The Augusta 139 FMS runs DOS FFS! You have no fvcking clue wolfy. Commercial aviation technology runs decades behind due to certification issues.

        Read the Air France crash report for the one out of Brazil if you want to see what an autopilot does with the most basic of instruments failing. It disconnects itself. The rest of what followed was a tragic case of the pilot’s not understanding what was going on. Again, I suggest you watch this. because the levels of automation vastly decrease pilot skills and increase workload if they don’t keep a check on it. I’ve watched it happen myself.

      • WTF does dead reckoning in the GPS have to do with an ASI malfunction? The FMS uses the GPS for tracking. The AMC uses ASI for the most basic of control inputs. You really are clueless.

        Tell me, if you are at the height envelope top and you are flying with a 200kt jetstream tailwind and relying on GPS ground speed, if the ASI fails, (actually, it’s usually pitot blockage that makes these scenarios) how fast in the air do you know you are going? Heard of high speed stall or coffin corner?

      • @Timmeh
        “how quickly two pilots can be task saturated”
        That is the advantage of an autonomous system, they simply don’t task saturate.
        If you can tell me what to do in advance of it happening the system can be programmed to deal with it, and 486 class processing power is plenty sufficient for an autonomous aircraft given it requires none of the complex visual environmental processing for obstacle avoidance that a car requires. If you can fly the plane in 0 visibility so can an autopilot.
        Of course, they still aren’t and won’t be implemented any time soon for non technical reasons.

    • It had a driver,but the driver stepped out
      “to look at a wagon”
      there is the hole in the story.
      what feedback would the driver had received of anything amiss with a wagon???
      What was the driver doing out of the cab??
      where was the remote isolator for the engines??
      how come the brakes, were released??

    • darklydrawlMEMBER

      This train had a driver. The driver stopped the train and exited the cabin to check a wagon problem en-route. It seems like the brakes were not applied correctly (or at all – although we don’t know the exact details at this early stage). However, as others have pointed out, the real failure is the lack of a ‘dead man’s’ brake or switch to stop the train remotely from the BHP control rooms. The only option was to derail it by remotely changing the points further down the line and by then the train was doing approx 200 km/hr. This will now impact BHP’s ability to get IO downstream for processing and export.

      • DD
        no chance the train was ever doing 200 clicks
        nor that the driver was inspecting a wagon
        so what happened??

      • darklydrawlMEMBER

        Hi WW, yes I agree with your thoughts. The initial report I read said the train had reached 200 km/hr – which seem a stretch and I was doubtful it was correct. It seems that circa 120 km/hr is more accurate.

      • WW: 500m elevation with a steady gradient and a shitload of tonnes = yes, can get to 200. Apparently.

        “500m above sea level and 28000 tons gives 137 Gj of energy to dissipate.
        Rolling resistance over 92kms would only account for 5.97 joules and would largely be negligible…
        Assuming 60 cars had locked brakes, coefficient of sliding friction of 0.3, that’s 187 joules of braking.
        Net result = big crash when it derailed. Would have been incredible to witness.”

      • myne
        those curves arent banked for 200
        it would have derailed much earlier

        how about if the train was a fair dinkum runaway
        and the driver jumped clear

      • What are ya
        ya cant tell from a road map
        those trains normally run at about 60k and the track is designed for that speed
        especially the cross track tilt of the curves, and the support in those sleepers on the curves
        the radial forces are huge.
        tracks which use transverse sleepers are especially vulnerable to excess radial loads in curves from rolling stock either over mass or moving too quickly.

      • i cant get satellite from that link
        what are the co ords
        for where the driver exited
        and where it derailed

      • Mining BoganMEMBER

        80k running. Rumour is reached 160 but who knows, information not coming my way. Where it took off from is steep so wouldn’t surprise me.

      • MB
        so you recon a 28 thousand tonne train can roll away from a standing start faster than the driver can return to the cab?
        wiht 60 cars having locked brakes???

      • Mining BoganMEMBER

        Without doubt. If the brakes failed or kicked off or whatever happened it would go quickly. Very steep grade there. No chance to run it down if you were back up the rake.

      • WW it derailed at Turners siding (northern part)

        Since you seem not to have used google maps before (wtf?!) click the thing that says satellite and zoom in on the end. You’ll see the rail line not far from the road. Follow it back. It started rolling somewhere about 90km back – near Mulga Downs somewhere.

      • Mining BoganMEMBER

        Can’t edit…

        That’s an area if a train became disabled it’s 100% handbrakes applied before trying to recover the system. If the driver only had 60 handbrakes on before the train brakes failed or kicked off or whatever it would roll. No way could you run over 600 metres on broken ground to get on board…and probably a suicide mission if you could make it.

      • Looked it up on google earth pro
        the track is gently curved
        but essentially level at 262m amsl
        especially the last 30 or so km
        so any speed you have heard about is rubbish
        in fact the last few km are up hill??

      • so now yr telling me the train dragged 60 cars, brakes on for 90km
        the flanges woulda worn off the wheels
        thus derailing those wagons???
        and the rail will have to be re profiled.

      • How do you know it was 600m
        and google shows the formation to be in good condition
        a man could easily run on that surface
        why would the driver be inspecting a wagon 600m from his cab
        do they have wheel monitoring??
        was he checking the vacuum hoses?
        or did he jump off??

      • Mining BoganMEMBER

        Around the 210-211 the track is through cuttings. Very uneven ground in parts on the handbrake side. 60 handbrakes is going by the link myne has above. I didn’t know that but I do know why the driver got out and it makes sense.

        If the link is right he would have been going to do 260+ handbrakes but only got to 60 before it all went to sh!t.

      • Mining BoganMEMBER

        Oh, there are hot wheel detectors everywhere. When detected it sends a message to the driver over the radio to pull up and have a look. Unfortunately this time the driver was still back in the hills worrying about wild dogs so not much could be done…

      • so burning question
        why did the driver get out??
        what are the co ords of where he jumped off?
        and if he only made 60, was he working from the locomotives back

      • Mining BoganMEMBER

        That one I’m not going to say. Might get someone in trouble for spreading stories. BHP hates social media.

        Yep, locos back. Each car is 10 metres so if it was 60 cars=600m.

  2. We had a train derailment and the line was operational again very quickly – they’ll build a parallel temporary diversion around the crash – a few km’s of rail. It is a flat, dry area – hardly challenge construction conditions. The actual wagon clean up might take a while – but that is kind of irrelevant when you’re not using that track. The temporary diversion won’t be built to a high standard but will work fine until no longer needed.

    • Mining BoganMEMBER

      They’ll bulldoze it all to one side and lay the replacement track. Might be a bigger delay finding out why so many failsafe systems failed. They won’t run until then.

      • You have to ask if it was a driverless train then why did it even have a driver on it? If he was just monitoring it then there should have been a fail safe system where the driver must acknowledge he is on-board periodically otherwise the train stops.

      • Isn’t that why there are drivers on trains? So it can be someone’s fault other than either the manufacturer of the train or the company operating it?

        If they introduce driverless trains, they’ll only have themselves to blame next time this happens.

  3. Reminds me of what is coming for the Australian housing market with all that brown fine dirt being infestors.

    • APRA is the train driver – world class monitoring of multiple failsafe systems.


      • The Traveling Wilbur

        APRA couldn’t regulate anything train related, only monorails.

        They’ve got a one-track mind (MOAR debt for Aus is good).