Galloping Ghost: NTSB Nails It
I took an hour and a half to attend Monday's NTSB hearing on the Reno Galloping Ghost accident. To say it was an eye opener is an understatement. It's not much of an exaggeration, in my estimation, to assert that before it even took off, Galloping Ghost's survival in the race was in question. The NTSB found that it was flying at the very edge of its structural and performance envelope, if not well beyond it.
How could this have happened? Perhaps the easiest way to answer that is to suggest that it occurred because of an uncertain confluence of an owner and team willing to press the limits, a racing association with weak technical oversight and a regulatory agency—the FAA—that simply wasn't in the loop. The NTSB found that the aircraft was significantly modified, to include the removal of the belly scoop, the addition of a new canopy, structural mods to the fuselage and tail and a boil-off unit that's popular among Reno races as a supplemental cooling system.
These are hardly uncommon mods for this class of airplane, but the FAA had no records on any of this except the boiler. The owners hadn't reported the rest. Given that Reno racers are experimental, I'm not sure they're required to, so that's not necessarily unusual. Yet had they done so or sought additional consultation, opined the board, more flight trials to prove the mods might have resulted. Would a structures guy have seen serious issues with the scoop removal just by inspecting it? Maybe. The NTSB said that much of the data it reviewed was unique to Galloping Ghost, even though other unlimited Reno racers are similarly modified. In other words, it wasn't necessarily a par-for-the-course airplane.
Galloping Ghost was clearly built to win. It was flying faster than it ever had by at least 35 knots and the engine was delivering more power than was ever asked of it, said the NTSB. It rounded its last turn at more than 400 knots, rolled sharply left, then pitched up violently into a 17-G uncommanded pull that the NTSB said no human could endure. Yet even before the moment the accident scenario began, there was some evidence that the airplane was coming apart. Like most aviation events, Reno is widely filmed and the NTSB had an unusual amount of good imagery to investigate this accident. It did a superb job of analyzing it.
In lap two of the accident race, the imagery showed deformation in the aircraft's skin, indicative of overloading, and a visible crack or gap opened in the canopy. Wouldn't the latter have been evident to the pilot? The conclusion seems to be that it should have been. Why it wasn't is a mystery. Again, the NTSB said these indications were unique to the Ghost, which I take to mean it's not normal to see that much deformation nor to see canopies parting.
On the accident lap, the final failure mechanism was loose or fatigued screws holding the left trim tab in place. (The Mustang has a pair of trim tabs, for redundancy, but the right one on Galloping Ghost was fixed in place.) The investigation revealed that self-locking nuts were re-used on the left tab and old paint on the fasteners suggested they were last installed 26 years ago. The screws were incapable of being properly torqued.
In the final turn, something excited flutter in the loose tab. Was it wake turbulence from the proceeding airplane or sympathetic vibration with a structure that might have already been buzzing? We may never know. But we know the result. Flutter is as relentless and unforgiving a phenomenon as anything in aviation and it can destroy robust structures in mere seconds. In Galloping Ghost, the fluttering tab failed the trim actuator rod, rendering the trim useless. It didn't help that the P-51's elevator bob weights and balance had been significantly modified. Jimmy Leeward was doomed the instant the tab buzzed. It didn't actually depart the elevator until well into the uncommanded pitch up.
In my view, as surely as the technical explanation for this accident was a structural failure, the reason for it was a cultural failure. In her opening remarks, NTSB Chairman Deborah Hersman showed an acute understanding of why Reno pilots are willing to assume risk, but she also observed that exposing spectators to risk is quite another thing and an out-of-control airplane is a risk to everyone.
As pilots, we tend to dismiss the concerns of non-aviators as the paranoia of people who live uninspired lives in a cocoon, unwilling or unable reach out for the thrill that animates the rest of us. But there's a degree of cynicism in that dismissiveness and it can get people killed. In my view, the Galloping Ghost accident doesn't appear to be the result of willful ignorance, rather, perhaps, a failure to fairly consider all the ramifications in maintaining an airplane to be flow to very edge of its performance envelope, if not beyond. It also seems reasonable to assume, based on the NTSB's findings, that it could have been avoided if the owners had merely examined the risks critically and conducted more flight testing. As the NTSB suggested, the Reno Air Racing Association needs better technical oversight of aircraft flying there and it has agreed to do that. RARA members certainly have the expertise.
A word here about the NTSB, which a friend of mine once described as "government done right." I'm not easily wowed, but watching this hearing, I couldn't help but be impressed with the thoroughness and speed of the NTSB's probe into this accident. Moreover, the board members questioning of the investigators showed deep technical grasp of the issues. They asked what I'd ask. And then some. Deborah Hersman's queries and closing remarks were respectful and set just the right tone; firm, no-nonsense, but not overbearing. Positive changes have already come in the wake of this accident. Let's hope they stick.
A video of the full hearing will be available on the NTSB Web site in a few days. It's worth the time to watch. I'll add a link when it's available.
In the meantime, you can download the NTSB's statement of probable cause here (PDF).