The flight was going fine right up until the moment everything got quiet.
The passenger, startled, asked the pilot what had just happened.
He turned to her and said, “We’re out of fuel.”
At least, that’s what she related to the NTSB investigator from her hospital room telephone four days after the crash. And that does seem to be the case.
At the accident site, an Amazon warehouse parking lot in Akron, Ohio, investigators found no fuel in the 1973 Cessna 172M’s tanks, lines, or carburetor bowl — and only trace amounts in the fuel filter.
The Flight
The pilot and his passenger were returning to home base at Weltzien Skypark (15G), near Wadsworth, Ohio, after a trip to Connecticut. They stopped roughly midway for fuel in Tunkhannock, Pennsylvania, took off again, climbed to 4,500 feet MSL and — pushing through a mild headwind — began their descent after two and a quarter hours of flight.
At that point, the pilot glanced at his fuel gauges and “had never seen them so low.”
Despite being only 20 miles from home, he decided to divert to Akron Fulton International Airport (KAKR) for fuel. He almost made it. The crash site is nine-tenths of a mile from the threshold of KAKR’s Runway 25.
The Pilot
The 71-year-old male pilot had a casual, but pragmatic, approach to fuel planning.
From experience, he told investigators, with full tanks his airplane could fly for “just under four hours,” so he rarely flew more than three hours at a time — giving him double the minimum legal fuel reserve for day VFR flight.
When he fueled, his standard operating procedure was to fully fill his tanks, unless he was carrying more than one passenger, which was rare.
On the accident flight he had topped off at his midway stop, so there should have been plenty of fuel for the two-and-a-half hour flight home.
Or was there enough fuel?
The NTSB spent more time — and math — on that problem than the pilot did.
Math Class at the NTSB
The 172 was just nine hours out of annual at the time of the crash, most of that time likely racked up during the flight to and from Connecticut.
Of interest to the NTSB was the fact that the fuel system was worked on during the inspection, raising the question as to whether it was a fuel leak or poor fuel planning that led to fuel exhaustion.
As luck would have it, before leaving Connecticut, a line person overfilled the airplane’s tanks, spilling fuel over the wing. We all hate that when it happens, but it gave the NTSB investigators a solid gold base line. They knew the aircraft had a full load of 38 gallons of usable fuel at the start of the day. And they knew, from fuel receipts, that he purchased another 11.23 gallons at his fuel stop.
So the question was: Did that 11.23 gallons really top off the tanks or was the pilot short?
To answer that question, Air Safety Investigator Lynn Spencer set out to reconstruct the fuel burn for the first leg of the trip. I won’t bore you with the details of her computations, but they’re impressive.
In the end, she came to the conclusion that the pilot departed his fuel stop with only 28.66 gallons in the wings, about 10 gallons short of a full load. But even so, as she wrote in her report, “at the time of the accident, the aircraft should have had 5.74 gallons of fuel remaining.”
That’s not a lot, but it would have gotten the pilot to his diversion with the legal minimum — or all the way home with precious little reserve.
Spencer next looked at the numbers another way. She took the known total fuel, made some adjustments for taxi, takeoffs, and a go-around the pilot executed at the refueling stop, and then divided that by the total flight time of the two legs: The first one from Connecticut to the midway point and the midway point to the moment the engine stopped. That gave her a fuel burn of 9.29 gallons per hour to, as she put it, “reach fuel exhaustion.”
That’s nearly 14% higher than the book value of 8.1 gallons per hour. As that seemed like a lot to account for, she next did something interesting. She ran the math a second time, but this time she used fuel burn figures from the engine manufacturer, rather than from the airplane’s manufacturer.
While Cessna uses an 8.1 gallons per hour baseline, the operator’s manual for the Lycoming O-320 that powered the Cessna — right up until the Amazon warehouse — uses a baseline of 8.8 gallons per hour.
Re-crunching the numbers, Spencer came up with not over five-and-a-half gallons that should have been remaining, but only 2.38 which, she states in her report, could have “easily” been used up during start, taxi, run-up, or even through leaning procedure variations.
In the end, the NTSB’s statement of probable cause was the pilot’s improper preflight fuel planning and in-flight fuel management, which resulted in a total loss of engine power due to fuel exhaustion.
Analysis & Discussion
Well, we can’t really argue about the “charge” of improper preflight fuel planning, as there was none. Not in the traditional sense, anyway.
Although, I for one, am willing to forgive a pilot for figuring a four hour fuel load will do for a two-and-a-half hour flight without doing any deeper math. Three times the minimum reserves seems more than adequate for a margin of error to me. And it’s not like the guy stretched it — flight-time wise — as we see in so many of these cases of fuel exhaustion.
Of course, he wasn’t as topped off as he thought he was. Why was that? The 172, as a high-wing airplane, does require a bit of gymnastic skill to fill-by-view.
At 71, was our pilot less spry than he used to be? Perhaps, but perhaps not. I know some elders who are in better shape than most young punks.
And what about the fact he first looked at his fuel gauges right at the end of the flight? We all know that fuel gauges can’t be relied on, but they at least deserve some monitoring.
The Takeaways
One obvious takeaway is a reminder that — whether you calculate or SWAG your fuel burn — you need to be 100% sure how much you have to burn. That, ultimately, was the real cause of this crash.
So get in a position to see into the tank, whatever that takes. Consider carrying a super-bright flashlight to overcome sunny ramps or use a fuel straw to “stick” the tanks for final verification of fuel levels.
Another obvious takeaway is a reminder of the importance of monitoring all your instruments and gauges throughout the flight.
In this case had there been a leak, which post-accident testing revealed there was not, he could have been alerted to the loss. As it was, he could have at least suspected he was less fueled than he thought earlier in the flight, when there was more time to deal with it.
But the most disturbing takeaway from this accident report was the finding that the Cessna POH and the Lycoming engine operator’s manual don’t agree with each other when it comes to fuel consumption. I’m still unpacking that in my mind.
There are many variables that make predicting fuel burn challenging, but we assume that we at least have a solid foundation from which to start — and, as pilots, we are taught to use the data in the airplane’s manual as the “bible.”
But what if that data in the “bible” is wrong?
The Numbers
Want to read more? Download the NTSB’s final report here or view the items on docket here.
Aircraft fuel gauges, as noted in comments) are generally estimates of actual fuel. When I built my RV-7 I was scrupulous about fuel senders and the like. My Garmin G3X and fuel pump combine to provide reasonably accurate accounting- but- I still use a dip style fuel measuring rod and always top off my tanks. I measure after flights and have found I actually burn less than indicated, but I wish I had a spot-on accurate set of senders. “Fuel estimators”might be the best term in small aircraft.
Comparing fuel gauge readings before and after fueling yields an opportunity to check gauge accuracy. In the case of filling a tank, the gauge reading before fueling will be a guide to the quantity (in gallons) that will need to be added. A significant shortfall may be a ‘tip off’ that the tank was not, in fact, fully fueled.
Did the NTSB check the fuel selector valve? My 69 C172K had a leaking valve that only occurred in the “selected left tank” position. I discovered it after shutdown after I failed to switch to BOTH for landing. This would not be discovered in a pre-engine start checklist.
I simply never leave an airport that provides fuel without filling the tanks (R172K) and waiting a bit to sump for water. This aviation business is a fun personal pursuit so always plan for extra fuel and extra time!!
LOOKS LIKE A STALL SPIN IN TOO . LIKE BOB HOOVER ALWAYS SAID FLY IT TO THE CRASH SITE. .THEY ARE LUCKY TO BE ALIVE.
I do not agree with the NTSB’s determination of inadequate fuel planning. The planning could have been perfect, but the best plan in the world is a worthless piece of paper if it is not executed properly. It doesn’t matter what the POH or Lycoming fuel burn rates show, those are only guidelines of what can be expected under a certain set of conditions. Your airplane may, or may not perform to those guidelines for any number or reasons. A person needs to know what the fuel burn rates are for that airplane, the way they fly it, to develop a good plan. Then the plan execution must be continuously monitored throughout the flight to insure the flight is conforming to the plan.
This reminds me of one of an old saying;
“NOTHING is foolproof to a sufficiently talented fool.”
In other words, assume Murphy is out to get you. What if a gas cap was left loose and comes off in flight? What if a fuel drain starts leaking mid-flight? What if one fuel tank feed line gets plugged preventing any fuel from that tank from getting to the engine? Without good plan execution, i.e. monitoring the flights progress against the plan, sooner or later Murphy will find a way to get you.
GOTTA” be more to thin story! After over-fueling one tank, did the line guy forget to top off the other tank? I ALWAYS top off and check for WET FINGER before attaching gas cap on a 172. No gas stick required.
The pilot stated that he didn’t always lean in cruise. My Cessna POH lists the lean fuel burn at 67% power as 9.6 gph, but at a ‘rich’ mixture, it’s 10.8 gph.!! A big difference if flying for 3 hours.!!
I’ve found the POH is accurate, if not a bit too high. But My Cessna has flap gap seals, full wheel pants, and polished paint.
After topping one tank of a 172, fuel will flow from the full tank to the other tank through the vent line connecting the two tanks, especially if the tank is over filled. So by the time the second tank is filled the first tank will be less than full. Additionally, if the fuel selector is on both, the full tank will transfer to the lower tank until both tanks are equal given enough time. If the aircraft is off level, the higher tank will transfer to the lower.
Vollkommen richtig!
Leanen ist auf einem Reiseflug immer angesagt. Dadurch hätte die Treibstoffknappheit eindeutig vermieden werden können. (Absolutely right!
Leaning is always popular on a cruise flight. This would clearly have avoided the fuel shortage.)
“But the most disturbing takeaway from this accident report was the finding that the Cessna POH and the Lycoming engine operator’s manual don’t agree with each other when it comes to fuel consumption.”
I cannot speak to this airplane and engine combination but I do know that the data from a Turbo Saratoga and Lycoming’s TIO-540 Operator’s Manual is operationally divergent in the case of cruise performance and leaning recommendations. Interested parties from way back indicate that range profiles had to meet marketing specs one way or another.
Wow! Been using the aircraft flight manuals for years. Not accurate? Maybe time to rethink that. Thanks for alerting us.
I, personally, don’t blindly trust or trust not the manuals. May be they’re right or wrong. But, may be facing this mine high degree of uncertainty, I always put my the minimums of fuel calculations around 25%/30% more fuel than that expressed in the manual that expresses the higher consumption.