Australia: Rail crash inquiry reveals serious mechanical flaws left uncorrected

By Barry Jobson and Terry Cook
2 July 2003

Evidence presented to the judicial inquiry into the Waterfall rail disaster in New South Wales, Australia has pointed to serious long-term mechanical faults with the 15-year-old double-deck Tangara passenger fleet that have been ignored by rail authorities and the state government.

On January 31, a Tangara train—labelled G7—derailed at high-speed outside Waterfall station, about 40 kilometres south of Sydney, resulting in the death of the 53-year-old driver Herman Zeides and six passengers. Simulated tests showed that the train was travelling at 117 kph as it attempted to negotiate the bends in a zone with a 60-kph speed limit.

In the aftermath of the disaster the NSW Labor government, already under fire over the deteriorating state of the rail network, sought to avoid responsibility by declaring the train mechanically sound and attempting to direct blame toward the driver. Evidence, however, points to the contrary.

In the early stages of the inquiry, several train drivers testified that Tangara trains were prone to “surging”—gathering speed unexpectedly—and braking faults, which accentuated the problem. Their testimony is consistent with statements by passengers indicating that the train had speeded up alarmingly as it entered the bends south of the station.

One driver described the particular Tangara—G7—as “unruly” and “having a mind of its own”. Another, Timothy Sharp, said he had experienced a “pushing sensation” after applying the brakes when driving the train on January 10. “It was surging,” he said. He only managed to control the train by applying maximum brake pressure but the “surging” returned as soon as he eased back on the brakes. Previously non-operating motors at the rear of the train suddenly came to life, propelling the Tangara forward.

Driver Randell Goddard testified to similar experiences with the Tangaras, while another driver Mark Facey reported trains surging as he attempted to bring them into stations. Evidence was presented that a train had rushed forward, overshooting a station and going through a red light, after the brakes were applied. The inquiry was told that drivers reached an “understanding” to avoid applying the emergency brake because the train sped up for a few seconds before the brake became operational.

The most damning fact to emerge was that rail authorities ignored repeated reports from drivers about these problems. Driver David Johnson testified that he and fellow driver Don Millar reported about 25 defects on the G7 over a three-year period. He said that he had been “disappointed” by the response of State Rail. Johnson said there was no feedback on any measures taken. Often drivers who contacted the defects branch “were abused”.

The testimony from the drivers certainly warrants further investigation into what was done about the reported faults. Such inquiries would raise serious questions about the culpability, not only of State Rail, but the Labor government, which has corporatised the public rail system and subjected it to severe financial constraints.

Instead, the inquiry headed by Acting Justice Peter McInerney has begun examining unsubstantiated allegations that Herman Zeides, and rail drivers in general, engaged in unsafe practices.

The accusations emerged alongside the claim that Zeides may have suffered a heart attack just prior to the accident causing the train to run out of control. The issue was raised again even though a post-mortem examination carried out immediately after the crash failed to establish exactly what killed Zeides. Its findings were declared by the Glebe Coroners Court to be “inconclusive”.

Dr Graham Peel testified in June that the driver’s cardiovascular system was diseased. Under questioning, he declared that Zeides “was most likely incapacitated by cardiac arrest” before the crash. But the possibility that Zeides suffered a heart attack raises more questions than it answers.

The Tangara trains are equipped with a number of fail-safe devices, including one known as a deadman’s brake. These are meant to activate the braking system and bring the train to a standstill if the driver becomes incapacitated. The deadman's pedal, for example, activates the brakes if the driver becomes unconscious and his foot lifts from the pedal.

State Rail authorities attempted to show that drivers deliberately circumvent these safety devices. On June 19, safety expert John Gusselli claimed that drivers may disable the deadman’s brake by jamming a flagstick under the console. His only evidence, however, was that, in the course of his inspections, he had had found black marks under the console of 29 Tangara trains, including G7.

Gusselli suggested that in each case a gum-like substance may have been used to jam a flagstick to hold down the deadman’s pedal. Without any corroborating evidence, he intimated that “possibly for the 15-year history of the Tangara fleet, some trains were operating without a safeguard against an incapacitated driver”. He admitted, however, that he did not know if it was “one person who did it, or every driver in the fleet”.

The media picked up on the Gusselli’s statements. In particular, Murdoch’s tabloid, the Daily Telegraph, embellished the story by making outlandish accusations that drivers read while operating trains and are generally inattentive. No evidence was presented—only photographs of printed material lying on the consoles of some trains.

The rail authorities, and no doubt the Labor government, had hoped that Gusselli’s testimony would point to the conclusion that Zeides—who was praised by fellow workers as a competent and careful driver—had disabled the deadman’s pedal on the day of the disaster. However, all three flagstaffs on the crushed Tangara were found intact on the floor of driver’s cabin. Gusselli admitted under cross-examination that if a flagstaff had been jammed under the console at the time of the crash, it would have been smashed.

But having eliminated the possibility that the deadman’s brake had been deliberately tampered with, the issue remained: if Zeides had suffered a heart attack, why didn’t the safety brake work? Three reports by risk engineer Kevin Anderson, presented to the inquiry on June 24, pointed to another possible mechanical defect serious enough to cause the crash. Anderson had been retained in 1992 to examine the braking system on the Explorer train, which is similar to the Tangara. He had concluded that there was a danger that a deadman’s pedal could fail due to the passive weight of the leg of an incapacitated driver.

Anderson testified to raising the same concerns in 1994 and 1999 in relation to the Tangaras and recommending the addition of a “vigilance control device” to monitor if the driver was alert. He stated that the train posed a “significant risk” unless the back-up system was installed. His proposal was dismissed by a rail steering committee, which insisted that the motion of a moving train would be sufficient to dislodge an unconscious driver’s foot.

Driver John Kennedy told the Waterfall inquiry in late June that he had refused to drive a train with a failed deadman’s brake, even though he feared being “stitched up” by Cityrail for taking a stand. He said he reported the fault and refused to continue driving, despite being ordered to do so. The train was carrying 500 passengers at the time, including scores of school children.

The evidence so far presented to the Waterfall inquiry points to the opposite conclusion to that being pushed by the Carr government and the media. Rather than an irresponsible driver being at fault, everything points to mechanical defects that were well known, had been reported to State Rail on a number of occasions and had been ignored.

This is not just the responsibility of rail management but of the Labor government which, in slashing costs, is compromising public safety. The crash at Waterfall, like the Glenbrook rail disaster in 1999, which claimed seven lives, is symptomatic of a systematic rundown of maintenance, infrastructure and safety standards.