Manual Labor, Buy The Book

Air Safety Week, April 11, 2005

The Pakistani Civil Aviation Authority (CAA) has found that "human error" has been to blame throughout the incidents involving wheel-brake fires on Pakistan International Airlines' (PIA) aircraft, previously examined in this publication (see ASW, March 7).

A PIA spokesperson has acknowledged the report and stated that disciplinary action has been taken against those responsible. "This is the fifth incident with our Boeing [BA] B777 and the matter was reported to the manufacturer. They changed the lubricant quality and quantity," but even then, the fires continued to happen, PIA maintenance chief, Mukhatar Qazi, told the Asian News March 30 (www.theasiannews.co.uk/news/index/11629.html). By "they" he would've been actually, but obscurely, referring to his underlings.

The larger question is why this situation was allowed to prevail for so long. It would seem to relate more to entrenched attitudes among maintenance staff and the difficulty in absorbing new methods and technologies in a short time. During the investigation, it came to light that the staff deployed for the B777's introduction was not properly trained and, in particular, that there was no record of any training for assembly in the B777 wheel-shop.

Following the PIA induction of the B777 in March 2004, there were six incidents, all of the same type. Fire broke out in the brake assemblies after landing. Four incidents took place at Mancheste,r UK, one at Lahore in Pakistan, and another in Toronto, Canada. The sixth and last Manchester incident was March 1. The CAA concluded that "as all the incidents have occurred on newly installed wheel assemblies, it was easily established that there was a flaw in shop maintenance procedures." One has to wonder why PIA safety sleuths could not have concluded this before reaching incident number six.

It's sobering to reflect that Manchester was the scene of one of the UK's worst aviation disasters when a British Airtours B737 caught fire on a taxiway in August 1985 and more than 50 persons perished. Due to the PIA wheel-fires being directly beneath the wing fuel tanks, the potential for a disaster was significant.

The original loophole discovered by the UK Air Accidents Investigation Branch (AAIB) after the June 8, 2004, incident was that PIA had been improperly using NYCO 22 grease. The Aircraft Maintenance Manual (AMM) clearly specifies the use of general purpose wide temperature grease with the specification MIL- PRF-81322 in the wheel sub-assembly. Three "to spec" types were available: Aeroshell 22, Mobil 28 and Mobil Aviation Grease SHC 100. PIA took the UK AAIB information onboard and switched grease - yet a further incident occurred. The two further loopholes, yet to be discovered, were actually more directly related to the heated wheel-brake's flammability. But the AAIB investigators would've had to visit the PIA maintenance base in Karachi to have uncovered those.

"There were two main reasons for the fires," the Pakistani CAA report said. The more serious one was using an inappropriate solvent while servicing brake assemblies and, even then, using it in an improper manner. The solvent being used for the cleaning of the bearing and heat shield of the carbon brake assemblies was carrier-2. The correct (AMM manual demanded) solvent was minimally P-D-680. The shop foreman decided unilaterally that he would accept carrier-2 as its equivalent. The CAA remarked that P-D-680 was a category I type solvent but that the recommended solvent type for the B777ER was a category II and III type. "The justification for using the carrier-2 solvent as provided by the Chief Engineer Power Plants was the non-availability of the P-D-680 and his also considering carrier-2 as a sufficient equivalent," the report said. In addition, the PIA supervisory staff disregarded that while the cleaning process at the manufacturer is automated, PIA had provided for no such facility. The mechanics directly involved in the process were conventionally dipping the parts into solvents and kerosene-based oils. Due to this immersion short-cut, the solvent becomes entrapped inside the heat-shield covers (between the two assemblies) and the glass-wool sheets inside the covers become soaked and impregnated internally with solvents. When subsequently heated during the braking process, the solvents vaporize and easily ignite. The clue was there all along in Boeing's Maintenance Tipsheet (777 MT 32-021, dated May 29, 2001): "Residual cleaning fluids can be retained by some wheel heat-shields after being saturated with flammable solvents during maintenance." The Pakistani CAA also found that approval documents provided to them for "authority to maintain" had been back-dated.

In March 2004, another AAIB report had blamed the standard of aircraft maintenance at PIA when a large B747 engine panel sheared off shortly after it landed at Manchester from New York. Rivets had been replaced incorrectly.

The PIA spokesperson, Samina Pervez, responding to the Pakistani CAA report, said, "They did not dry out the solvents in the proper way." She observed and added dryly that the basic cause of these accidents had now been removed. Stating that the responsible person had been replaced, she also denied that there was any chance of a fire in the fuel-tanks or that passengers' lives had been at risk. Obviously, there is a perception problem for discriminating between the relatively common scenario of hot brakes and the potential of wheel fires. It's worth remarking here that the B777 has had a very clean run in the safety arena and no doubt Boeing would like to keep it that way. What more could be done to avoid these sorts of techno-phobic situations arising?