» Views

It all began with one faulty fastener

Publication Date : 05-07-2012


Blame it on a piece of metal no bigger than a mobile phone.

An inherently defective metal fastener was found to be the trigger point of an unprecedented breakdown of the MRT system on December 15 last year, causing grief to thousands of commuters.

The fastener is part of a support assembly that holds up the third rail, which supplies power to trains.

Coincidentally, two adjacent third rail supports were also found to be defective. There were cracks in their insulators.

These flaws contributed to a collapse of a section of the third rail, disrupting power supply to trains.

Undetected damages sustained by trains that ran into the collapsed third rail are believed to have triggered the second breakdown on December 17, infuriating thousands more commuters.

But it all started with that one faulty fastener.

The Committee of Inquiry (COI), which questioned 116 witnesses over six weeks and which sent damaged components for forensic tests, could not ascertain when the defect occurred or when that first support assembly failed.

It concluded, however, that both breakdowns were preventable.

And they probably were, although the cost involved would have been high.

First, there are roughly 30,000 of these assemblies in the North-South and East-West lines, the oldest of Singapore's metro network.

Inspecting an assembly entails removing an unwieldy plastic cover and scrutinisng no fewer than four individual parts.

SMRT, with its current resources, takes about a week to complete inspecting the entire network - without removing the plastic covers.

Cracks on the insulator may not be visible unless their bolts are removed. Inherent material flaws in the fastener will not be detectable unless high-tech equipment such as ultra-sound and X-ray scanners are employed.

Carrying out such 'non-destructive tests' - especially along vulnerable stretches of the network - is one of several recommendations the COI made in its comprehensive 358-page report made public yesterday.

These tests are now being done at third rail joints, which are relatively flat structures. It may or may not be technically feasible to do so for a complex component such as the third rail support assembly, which is also made up of different types of materials.

Assuming it is feasible, these tests would have to be done during hours when the trains are not running.

If third rail support failures had occurred during the hours when trains were running, what then?

Another suggestion by the committee addresses this: installing a high-speed camera system on trains to detect third rail sags, which occur when a support fails.

Currently, such a system is used for detecting flaws on the rail which the trains run on - that is, the running rail.

It is understood that no similar system exists for detecting flaws in the third rail, although it is technically feasible to devise one.

This is because third rails rarely collapse, if ever. The inquiry heard no evidence of similar failures in other metros. Also, the rail industry naturally pays more attention to the running rail, as flaws here can cause derailment, with dire consequences.

It is for the same reason that SMRT's maintenance crew continued to check the temperature of the axle box - a part of the train connected to the wheels - after the December 15 breakdown.

The Committee of Inquiry questioned this move in its report. But this is a crucial process, as an overheated axle box can lead to bearings seizing and the train stalling. In extreme cases, it could also lead to derailment.

Be that as it may, SMRT should have put in more resources to check for damages sustained by trains that plied the collapsed section of third rail on December 15.

If these defects had been detected and corrected, the December 17 incident could have been averted.

SMRT could also have done more to tighten its entire maintenance framework - something the inquiry report minced no words in pointing out, and which the operator accepts.

Still, questions remain about the defective metal fastener, and the adjacent support assemblies which also had defective insulators. The inquiry also noted that the third rail claws had a weak design that made them prone to dropping.

That does not mean SMRT had a raw deal and could have done nothing about it. It could have been more proactive in getting the Land Transport Authority (LTA) - which oversees rail projects and regulates the rail industry - to replace the ageing and poorly designed third rail supports.

In fact, SMRT had brought up the issue of dropping claws in 1995, but did not pursue the matter further.

The LTA, on its part, did not follow up on the dropping claw issue after 1995, as the COI noted.

The whole issue of defective or poorly designed third rail support components is water under the bridge though, now that it has been decided that all claws should be replaced by the latest fifth-generation model, which can be bolted to the third rail.

These new claws, which come with new insulators, are far more secure, but can also be dislodged. But being made of stainless steel, they are more detectable when dropped than the old black cast-iron claws.

Other methods of detecting anomalies on the third rail - such as the high-speed camera system mounted on trains - will also help prevent a repeat of December's incidents.

But breakdowns of other nature can and will still occur. That is an undeniable fact of all mechanical systems.

The inquiry's report has also put in a comprehensive list of recommendations to mitigate the impact breakdowns have on commuters.

Like its engineering recommendations, many are overlapping. And so they should be.

A system that transports millions of people a day should likewise have built-in redundancies so that the failure of one component or process does not jeopardise the integrity of the entire network.

Built-in redundancies aside, the design and build quality of each component must be robust.

As far as the third rail support assemblies on the North-South and East-West lines go, that has clearly proven to be not the case.

Between maintenance and design, especially in an ageing system, there is no running away from the fact that accountability must lie with all players - operator and regulator.

Commuters, for sure, will be demanding that.


Mobile Apps Newsletters ANN on You Tube