Over the holidays, I happened on a documentary that told the story of American Airlines flight #191 – to date, the single worst air crash in our country’s history. The story covered the aftermath of that accident and the resulting FAA investigation into the causes. It was the conclusions of that investigation that really raised my interest, and as you’ll see as I share what I learned, this tragic occurrence has lessons to share with all of us in our roles as professional technicians.
On May 25, 1979, 271 passengers boarded AA 191 at Chicago’s O’Hare airport. Shortly after the big jet rotated on takeoff, the left engine separated from the wing and literally fell off of the aircraft. It also ripped through vital hydraulic and electrical lines, including the “shaker” that warns the pilots of an impending stall by vibrating the primary controls.
The pilots knew they had lost an engine’s power, but didn’t realize they had lost the engine structurally. They followed emergency procedure for a loss of power as the plane continued to climb, attempting to gain enough altitude for a safe return back to the airport. When the last of the hydraulic fluid leaked out, the front wing slats on the left wing closed, reducing the amount of lift on that side of the DC-10-model aircraft, causing the plane to dip over into a hard left turn until the wings were nearly vertical to the horizon and stalled. With the wings no longer producing lift, the plane crashed shortly after, killing everyone on board.
Why?
The aircraft had just been in for maintenance eight weeks earlier. Part of the maintenance was an inspection of the engines and their mounts. On the DC-10, the engines are attached to the wing through a highly engineered part called the pylon. This mount supports the weight of the engine and is attached at two points, fore and aft. Accessing the areas of the mount for inspection is supposed to be performed after the engine has been removed from the pylon but the procedure to remove the engine was involved and time-consuming.
And here is where the story really caught my attention.
Rather than remove the engine first, someone decided it would be so much easier if the engine and pylon were removed as an assembly. An engine stand was mounted on a forklift and the lift was used to support the weight of the two during removal. Once the work was completed, the lift was used to raise the assembly into place while the pylon was reattached to the wing. And here is where the problem occurs. During the reinstallation, proper positioning of the pylon mounts is critical to protecting them from damage. That kind of precision placement is kind of hard to do with a forklift positioned perpendicular to the mount. During the job, the pylon aft mounting plate was cracked, ever so slightly, and the damage went unnoticed by the maintenance team. That’s easy enough to understand, since they just inspected that component and found it to be serviceable.
During the next eight weeks of takeoffs and landings, additional stresses caused the crack to grow until the aft mount finally failed. It failed on the takeoff roll of AA #191, causing the full weight of the engine to pivot down on the front mount until it too sheared, allowing the engine to complete its rotation over the front edge of the wing and finally coming to rest on the runway.
Sound familiar?
There isn’t a commissioned technician I know of that hasn’t come up with ways to make a job go faster or easier. I’ll never forget an instance back in the early days of the Dodge Neon. When first introduced, they were notorious for leaking oil from the original head gasket design and we repaired dozens of them under factory warranty (I worked for the local dealer at the time) using the new MLS-style gasket. Some of the techs where I worked loved the jobs and could complete them in just a few hours. How, you ask? By cutting some major corners, that’s how.
Rather than remove the head and clean the gasket surfaces, these “techs” would remove the head bolts, leaving the cam and timing belt intact, and simply lift the head just enough to slide the old gasket out and the new one in. They weren’t concerned if the gasket failed, or some other problem arose, arguing that they would just fix it again “under warranty”.
And how many of us love to tackle A/C evaporator cores by peeling the dash back and laying it on the seat, rather than removing it completely from the car? I loved the Chrysler minivans when the book time was still something like 12 hours or so. Heck, I could have one done in four hours, and that included sucking out the refrigerant and charging it back up!
The difference between these two examples is significant though. In one, there was no regard for the consequences of the shortcuts taken while the other was a generally acceptable way of approaching the repair (so much so that now flat rate times have been adjusted accordingly). The yardstick, for me anyway, has always been based on industry acceptable “best practices” – in other words, “First Do No Harm”.
That can be a major no-no, like the gasket replacement example I just shared. The surfaces of the head and engine block were never cleaned, the timing belt tension and alignment were never verified and the head bolts were installed with air tools rather than a torque wrench. How much potential for harm is there in that scenario? I guess it’s a good thing that we don’t have to sign our names to the maintenance logs for each car we service, isn’t it? But if it was required, would you be willing to sign yours?
Avoid harm when performing electrical checks
On a related note, let’s talk about “doing no harm” when performing electrical tests on your customer’s car. The most effective tests are performed with the system operating and that means we have to access the circuit wiring while it remains intact. Two ways we commonly do that; backprobe the connector or pierce the insulation of the target wire. Both can be effective and efficient – both can create future problems if done improperly.
Backprobing promises the opportunity to avoid damaging the wiring but only if done carefully. It is imperative that your backprobe be designed for the purpose. Many techs use the “T” pins you can find at any store that sells sewing supplies, but it is too possible to short two of these pins together during testing and that may be just enough to let all the smoke out of the ECU you’re connected to! It is also not uncommon to place your probe in one cavity yet come out on the other side on one of its neighbors, again resulting in a potentially component damaging short. Last but not least, if the probe passes anywhere other than the space between the weather seal and the wire’s insulation, damage to the weather seal may be enough to allow moisture in with the resulting corrosion that accompanies it.
Piercing will certainly cause a hole in the wire’s insulation so be sure to seal that hole when your done with some liquid electrical tape. When clamping down on the wire with your piercing probe, use only enough tension to make contact with the wire strands inside to avoid damaging too many of them in the process.