Aircraft Accidents and Lessons Unlearned XXXII: Northwest Airlink Flight 5719

On December 1, 1993, on approach into Chisolm-Hibbing airport in Minnesota, an Express II Airlines, Inc. Jetstream BA-1300, doing business as (dba) Northwest Airlink, tail registration number N334PX, impacted terrain. The accident number, DCA94MA022, was researched for information through the National Transportation Safety Board (NTSB) archives, but no archived reports were found, except for those that were comprised in the final accident report AAR-94/05.

The problem with a report like AAR-94/05, is that it is based on an emotional argument, not factual. It preceded other emotionally based reports, like ValuJet 592 and Colgan 3407. The facts of the investigation, often straightforward, get lost in the tragedy’s victim numbers or circumstances. An investigator’s job is not to get distracted but to remain focused on the accident’s facts.

The Probable Causes of the accident were as follows [numbering added for clarity]: “The National Transportation Safety Board determines that the probable causes of this accident were [1] the captain’s actions that led to a breakdown in crew coordination and the loss of altitude awareness by the flight crew during an unstabilized approach in night instrumental meteorological conditions. Contributing to the accident were: [2] The failure of the company management to adequately address the previously identified deficiencies in airmanship and crew resource management of the captain; [3] the failure of the company to identify and correct a widespread, unapproved practice during instrument approach procedures; and [4] the Federal Aviation Administration’s inadequate surveillance and oversight of the air carrier.”

What is gained by looking at an accident that occurred twenty-six years prior? What is to be learned from a small commuter accident? That is precisely the point, that each accident, from a passenger-packed jumbo jet to a single-engine private plane, is important because each accident, if incorrectly investigated, leads to another. Each has a lesson – or lessons – to teach us and, as in this case, lessons to be unlearned that were taught wrong to begin with.

Express Airlines II, dba Northwest Airlink, flight 5719 was a Title 14 Code of Federal Regulations Part 135, ten or more passenger air commuter, contracted to Northwest Airlines as a regional server. The investigation was a major accident investigation – accident number DCA94MA022 – because it involved certification requirements for its operations and maintenance, exceeding those of a general aviation aircraft and other smaller certificates. The timeframe, from accident (December 1, 1993) to final report (May 24, 1994) – no hearing – was just under six months; there were nine investigatory groups formed that contributed to the accident investigation. However, six months is extremely quick and did not lend itself to much time for testing, detail or completeness.

The root or actual causes of the accident were not run to ground; instead the investigators settled on probable cause. Beginning with the fourth probable cause: “… the Federal Aviation Administration’s inadequate surveillance and oversight of the air carrier;” was ambiguous. Imagine a picture of a square mile of the mid-Pacific Ocean; an arrow is placed on a wave, stating: ‘You Are Here’. The probable cause had no point of reference, no usable information. What did the NTSB, with its limited certificate holder experience, find that the Federal Aviation Administration was inadequate in? A finding about Air Carrier Operations Bulletins failed to clarify the collapse of trust or what constituted a failure of surveillance and oversight. No lesson was learned; indeed, no lesson was generated at all.

Probable cause number one: “… the captain’s actions that led to a breakdown in crew coordination and the loss of altitude awareness by the flight crew during an unstabilized approach in night instrumental meteorological conditions.” The first ten report pages, ironically titled: 1. FACTUAL INFORMATION, described a captain who was hard to work with, following him from the previous day to the day of the accident. His every growl and ill-tempered action from 24 hours before the accident flight was analyzed and commented on by gate agents and cleaners. Were these persons experts in human behavior? Did they even know the captain? This was hearsay. From the cockpit transcript, crew conversation leading up to the accident was casual dialog; no disagreements. There was no arguing or interrupting associated with poor communications. Where was the breakdown in crew coordination?

The captain was criticized by a cleaner (no one else was onboard) for dressing down the first officer for an improper preflight because the first officer missed the broken landing lights. It was the captain’s job to require the first officer to do his job. Crew coordination takes two; each pilot is responsible for his/her breakdown of the communication. However, from the transcript, there were no miscommunications or problems. The captain was tough, but for an investigator to base a report’s probable cause on the word of people the captain rarely had interactions with, let alone conversations with, was unprofessional. The NTSB Board Members should have pushed back on the hearsay; instead, they were disengaged.

The second probable cause: “The failure of the company management to adequately address the previously identified deficiencies in airmanship and crew resource management of the captain.” Was this probable cause correctly focused? The captain’s training history demonstrated several problems, including several failed proficiency checks, yet the captain maintained his rating after retraining. The investigators did not succeed in their pursuit of training information. Serious training issues should have raised alarms with the Operations investigators. Why did the investigators not question the instructors’ training techniques, pursued the instructors and their quality? The instructor information was scarce.

The third probable cause: “… the failure of the company to identify and correct a widespread, unapproved practice during instrument approach procedures.” What ‘widespread, unapproved practice’ did the company fail to identify? The report did not say. The investigators had access to the trainers who gave instruction to the pilots. Were they teaching unapproved practices and what were those practices? The problem with interviewing persons who have much to lose (the instructors) is that they will shine attention on the deceased crew, the ones who cannot argue in their own defense. The investigators should have had better interviews with other Express II pilots to see if they were receiving different training.

The Findings in this report did not represent factual information; the use of terms, such as ‘suggested’ or ‘should have’, diverted from the facts of the investigation, allowing speculation to be reported as fact. The nonexistent amount of time dedicated to testing, the lack of a hearing and the unnecessary analysis in factual matters, signified that the investigators’ efforts were inadequate. The report never made clear why the captain’s actions were unsafe, or why the first officer’s responses were in question.

The investigators dropped the ball and focused attention in the wrong direction. What, then, could have been the cause of the accident? Two questions: What about icing and was the aircraft deiced in Minneapolis airport before it departed? The investigators spent a lot of the report questioning the pilots’ integrity; what efforts were given to prove mechanical integrity? Too little was done to show if icing or frozen water was a contributor.

Deice fluid is made of 50% deice fluid and 50% water. During deicing, water has been known to get trapped in fairings surrounding the elevator pushrods. At altitude, the deice water freezes at below freezing temperatures. At cruise, small inputs are the norm, while at approach speeds the larger inputs are critical. What if they were hampered by ice? If the aircraft was deiced in Minneapolis, trapped ice could have made elevator movements sluggish or non-existent. The aircraft would have been more difficult to control at slow speeds or to recover from a steeper descent. Speculation? Perhaps, but the investigation failed to explore other effects upon the aircraft’s flight control integrity.

This accident investigation showed no lesson learned; nothing of value came out of the report. The investigators’ rush to close the accident report found nothing that would increase safety and ignored the obvious. This questionable investigation itself should be a lesson unlearned.