Following a lengthy and expensive annual inspection, the pilot-owner of a Piper PA-31P — the pressurized version of the twin-engine Navajo — is preparing to depart Myrtle Beach Airport (KMYR) in South Carolina for his home base of Grand Strand Airport (KCRE), a mere 12 miles up the coast.
It’s a lovely Friday evening in late May of 2021. The sky is clear, it’s 75°, and there’s a 10-knot breeze off the ocean.
But the airplane, and its pilot, will never see their destination — because two small pieces of the airplane’s flight controls, after having been removed for service, have been re-installed incorrectly.
The Flight & The Final Moments
Right after takeoff, the first time the big twin has flown since the annual, the pilot radios the tower that he needs to return to the field. He’s cleared for right closed traffic at 1,500 feet msl (the airport’s elevation is 25 feet msl) but he can’t seem to maintain altitude.
The tower asks if he needs assistance, and the pilot transmits, “Yes, we’re in trouble.”
The plane drops to 450 feet msl, bounds up to 700 feet, then drops to 475 feet. Then radar contact is lost, only a little past mid-field on the right downwind.
The entire flight lasts two minutes.
The Accident
The crash is… ferocious. The NTSB’s Wreckage Examination Summary reports that the fuselage hit the ground in a near 90° nose-down attitude.
The engines had to be dug out of the earth. The right one was five feet deep.
The left was six feet under.
A passing airport police officer witnesses the crash and stops to render assistance, but as he runs toward the aircraft it explodes into a fireball, injuring the officer. The airplane, recently fueled with 168 gallons of 100LL, is quickly consumed by the post-impact fire.
The Pilot
The final report from the NTSB doesn’t tell us much about the pilot. He was a 60-year-old male with a first class medical. He held an ATP certificate and was a CFI/MEI. The NTSB listed him as an “occupational pilot.”
But an online obituary gives us more detail. He received a discovery flight as a gift for high school graduation and fell in love with flying. He went to Embry-Riddle Aeronautical University, and then worked as a freight dog, followed by a 30-year career with US Air and American Airlines. He capped off his career as an A321 captain, with over 35,000 hours in his logbooks.
The NTSB
In their inspection of the shattered and charred wreckage, the NTSB investigators lucked out, as the tail section survived relatively intact, allowing the discovery that both elevator trim tabs had been installed upside down and “reversed.”
The right-hand trim tab was installed on the left elevator and vice versa.
The NTSB noted that this “resulted in a reversal of the direction of the trim tabs’ movements; a command of nose-up trim from the cockpit controls would incorrectly move the trim tabs in the nose-down direction, and vice versa.”
The NTSB believed this error happened at the recent annual, listing the probable cause of the accident as “the mechanic’s inadvertent installation of the elevator trim tabs in reverse, which resulted in the pitch trim system operating opposite of the pilot’s input and the pilot’s subsequent loss of control.”
So how could such a maintenance error have occurred? And why was it missed by everyone?
Analysis & Discussion
In the course of the twin’s four-month, $42,951.65 annual inspection, corrosion was discovered on all of the flight control surfaces. They were removed, repaired, repainted, and re-installed — including, of course, the trim tabs. They were part of Discrepancy #38 on the 13-page work order, with a logbook entry stating that the work was done in accordance with the Piper maintenance manual.
But the NTSB seems to cast doubt on the use of the maintenance manual, as they included both the parts catalog and the service manual in its report.
Interestingly, the parts catalog actually shows the elevator trim tab control horn in the wrong place, which could lead to upside-down installation, but the NTSB calls attention to the fact that the preface matter of the catalog clearly states that the book shouldn’t be used for maintenance, and that the control horns are properly illustrated in the service manual.
But oddly, the summary of their interview of the director of maintenance (DOM) for the Part 43 shop never mentions the parts catalog. So why was it included? Did one of the employees say something off the record?
And while the DOM emphasized to the NTSB that he “tries to drill it into the guys” that they always need to reference the maintenance manuals, that wasn’t so easy for the mechanics to do. The maintenance manuals were on computers that, at the time of the twin’s annual, were not located in the maintenance hangar, requiring mechanics to leave the job, go to a computer in a nearby office, check whatever info they needed, then go back to the hangar.
Why?
Apparently, at one time, the shop had computers in the hangar, but “the mechanics weren’t getting as much work done” as before they had the computers, so the management decided the computers were “a distraction” and moved them to the office space.
An additional potential confounder noted by the NTSB was that the shop also maintained a non-pressurized version of the Navajo, and that this version features a bottom-mounted control horn on the single elevator trim tab, unlike the pressurized version, which has the control horns located on the tops of the twin elevator trim tabs.
The IA on point for the annual stated that he examined the primary flight controls for proper movement after they were reinstalled, but admitted that he did not verify proper movement of the elevator trim tabs, so the shop missed the fact that they had been re-installed incorrectly.
Now, elevator trim tabs are like ants: Crazy powerful for their size. If you don’t believe that, I urge you to fly to a safe altitude, set up for cruise flight, then — hands off — rapidly deploy your elevator trim tab full forward and see what happens. After you recover, and let your stomach settle back down, do the same thing with full aft trim.
But could improperly rigged elevator trim tabs really have caused the loss of control? Well, as noted, trim tabs are powerful. But unlike say, improperly rigged ailerons undiscovered until liftoff — where you have mere seconds to sort out what is happening — this is a slightly slower motion emergency, and trim tabs can generally be overpowered. (Disclaimer: I’ve never flown a Navajo, and the Navajo’s trim tabs look pretty beefy, making up a larger percentage of the elevator’s surface area than airplanes I’m more familiar with so Navajo fliers, please chime in on this.)
Shedding more light on the loss of control from the tabs, however, is the FAA inspector’s statement. He talked to family members at the crash site the day after the accident. The pilot’s son reportedly told the inspector that the airplane “took both hands and full nose up trim” to get “off the ground,” and that this had been true “ever since they first got the airplane.”
This suggests that the pilot was in the habit of setting full up trim and hauling back. But with the tabs reversed, the plane is now trying to nose down, even while he’s got the yoke waaaaay back. It doesn’t take much imagination to picture the wild see-saw ride that ensued.
Additionally, with the plane fighting him to dive, you can see why attempting to trim the nose “down” (which in this case would have raised it) might have seemed a suicidal choice to the pilot — rather than a logical trouble-shooting step.
The Takeaway
The mistake with the trim tabs is mechanical malpractice, but what about the pilot? Could he have — should he have — caught the mistake?
Well, in case you don’t know, post-maintenance is the time to preflight like your life depends on it — because it does.
And while preflighting trim can be difficult, especially unassisted, it’s hardly impossible. You can deploy it fully one direction, get out of the plane, verify its position, get back in the plane, deploy it fully the other direction and once again, clamber out and check where it is.
Should the pilot have done that? Given that he must have known the tail feathers were removed, worked on, and re-installed: Yes. Absolutely. And for all we know, he did.
Because for that “inspection” to work, he would have needed to understand how the trim tab should look for a given setting, and frankly, a lot of pilots don’t. There’s no shame in this. External inspection of trim tab deployment angles just isn’t one of those things we deal with during routine operations — I don’t think I’ve ever seen a trim tab position check on a preflight checklist.
So a good takeaway for all of us is to review this smallest of secondary controls. A properly-installed and rigged elevator trim tab points in the opposite direction from the commanded setting. Dialing full nose-up trim in the cockpit will result in a downward-angled trim tab, and vice versa.
For my student pilot readers, this is because trim tabs move control surfaces in the opposite direction. Sticking with the nose-up command example, the down-angled trim tab forces the elevator upward in response. The up elevator, in turn, forces the tail of the airplane downward, which, in turn, raises the nose.
Is it any wonder folks sometimes get confused staring at the trim tab at the back of the airplane, trying to visualize all the various motions and forces at play?
And, finally, there’s one more takeaway to consider from this accident: Could that long airline career have worked against the pilot, rather than giving him the increased wisdom we usually expect with increased experience?
He spent decades in an environment in which maintainers were specialists in a limited number of airframes and the airplanes were sufficiently complicated that — while flight crews still did walk arounds — there’s a lot they simply couldn’t see, so in airline operations, the needle on the trust/verify meter shifts heavily toward trust.
But GA is different. Our maintenance folks work on a lot of different airplanes, and in many cases the pilots know the systems better than the mechanics. Returning-to-GA retired airline folks should take note of this takeaway: In GA, you need to move the trust/verify needle closer to verify.
The Numbers
Want to read more? Download the NTSB’s final report here or view the items on docket here.
I owned a P Navajo N57Jk. The elevator needed a ladder to examine. Full trim application for takeoff seems odd to me.
JJ
Non-pressurized Navajos use a servo tab on the elevator. The P-Navajo is fitted with anti-servo tabs, the precursor to the Stability Augmentation System (SAS) used in the PA31T turbine versions of the Navajo. Precise rigging is required, none of which should be committed to memory or rote.
There appears to be an accumulation of errors here. I think the plane may have been flying in an Un airworthy condition for some time according to the son. Then possibly the mechanic put the trim back just like he found it not realizing that it was not correct. If I flew a plane that handled as the son described I’m pretty sure I would be looking to see what was so different about why my “ new” plane was so nose heavy. Trying to find one good thing about this, is that it is a learning moment. The lesson might be that after major maintenance test fly the plane, think how much worse if the pilot had his children with grand children on the first flight. Not much but a small consulation.
I have been caught in the same type of situation on the first flight after maintenance. As I lifted off it rapidly became clear that there was a control problem. By using very strong arm tactics I avoided making my acquaintance with the local undertaker – I had one arm hauling back on the control yoke and assisted with leg pressure on the rudder pedals leaving me with the other arm free to trouble shoot so immediately returned the trim to neutral and regained sufficient control to get to circuit height and return to the airfield with some questions for the maintenance facility, the most pressing being WHAT HAPPENED TO THE DUAL CONTROL INSPECTION AT THE FINAL CHECK AT THE END OF THE ANNUAL MAINTENANCE . ? THE TRIM TAB IS A CONTROL SURFACE ! This occurred many years ago (I have now been flying for 60+ years.) and I have since this incident included a trim check with my preflight checks. Whether other pilots think it is useful or not is their decision but I intend to continue using it in my preflight check list.
Paul NZ.
This part seemed a little odd to me. “He talked to family members at the crash site the day after the accident. The pilot’s son reportedly told the inspector that the airplane “took both hands and full nose up trim” to get “off the ground,” and that this had been true “ever since they first got the airplane.”
That just does not seem right to me, no I have never flown a PA-31P. PA-31P pilots please tell us if this is “normal”.
Could there have been other issues that made it that hard to get off the ground even before the mess up with the trim tab?
Yes, NEVER trust the mechanic (or anyone else) to know how to do ANY job correctly. Plenty of crashes confirm this. FAA rarely pulls the license of the incompetent, negligent, or careless mechanic. The “human factor” part of this crash was the pilot not considering and realizing the faulty trim tab in climb out and putting in the reverse input. It was not computing as he was fighting the yoke. AQP might have helped rescue the situation. Something to consider on the check list: “What do I do if the trim fails?” Could include not checking out the work, if that was not done by the pilot. Really have to AQP every scenario you can think could happen. On the ground is the bast place to do that.