Industry:
Commercial Aviation
Baggage Handling Conveyors
Problem:
A luggage conveyor at a busy international airport stopped working
while
it was fully loaded with bags. Everything had been working fine up to
that moment.
Injuries:
None.
Damages:
None, but the customers had a tough time getting their bags and
this
incident would temporarily put the carousel out of service.
Observations:
After routine electrical problems were ruled out, the large drive
motor
was sent to the manufacturer for repairs. At the plant the motor was disassembled.
It was soon discovered that broken fasteners prevented
the motor from
working correctly. The fasteners breaking were M6 x 45mm (1.8") socket
cap screws, Property Class 12.9. These were being used in a circular pattern
of eight equally spaced screws in what is called the
drive drum. The remains
of all screws were still within the tapped holes.
Investigation:
We began with a careful visual observation under a stereomicroscope.
This
showed that the fractured surfaces were clearly fatigued in nature, with
the accompanying striations (beach marks). The screws broke about 6 to
8 mm (0.236" to 0.315") from the end, and all screws had a
nearly
identical fracture appearance. The point end of the screws were immobilized
with what looked like a chemical locking product. The hardened residue
was apparent. At the end of the threads on the long section of the screws,
the major diameter appeared slightly necked down. We placed some of the
pieces on a video measuring machine at a medium high magnification. This
showed that distention clearly was present. A study of the thread engagement
indicated there were barely enough threads to prevent
stripping. And,
a discussion with the installers on the assembly line showed that they
didn't have a good understanding of torque or bolt tension. They put the
chemical on so vibration wouldn't loosen the screw, but had no idea what
frictional properties the chemical had on the assembly. However, they
did point out the 45mm (1.8") length screw was "just temporary
until the new screws came in," as a 50mm (2") length screw
was
normally used.
Conclusion:
In a bolted connection there is a balance between the force created
by
tightening the fastener (clampforce) and the force that the machine will
put out. Generally, when the clampforce is greater than the force made
by the machine, the bolt lives. If we assume that a good design practice
was used on this motor, then the torque was unreasonable for the screw's
strength with the chemical locking product. Yes, they would have been
better off with the longer screws, but that didn't cause the failures.
The fact that the screw had "necked out" slightly reduced the
stress area, and thus made an imbalance. The machine force now was greater
than that provided by the screw, therefore an opportunity for a fatigue
failure.
Recommendations:
While a tensile-to-fatigue failure is relatively rare, you can expect
any number of things to go wrong if each aspect of the
joint and assembly
aren't carefully understood, and successfully addressed. It pays to take
the time to do so.
As appeared in Fastener Technology
International magazine. |
