The pilot reported that, two days before the accident, the multiengine airplane’s fuel tanks were filled (150 gallons).
During the taxi to the runway, the Cessna T337’s right main tire blew.
The right side of the plane was placed on a dolly to support the gear so that the airplane could be towed. According to the pilot, due to the airplane’s fuel system design, when one side of the airplane was raised, all the fuel could be transferred to the opposite tank, which then forced the fuel to be released out of the air vent line.
On the day of the accident, the pilot completed his preflight inspection and visually confirmed the fuel quantity by checking both fuel gauges, which were “green.” However, he did not verify the fuel onboard by checking the tanks.
About three hours into the flight, the rear engine lost power. Before he attempted to restart the rear engine and after he verified the correct engine to feather, the front engine also lost power.
When he realized the airplane would be unable to reach the nearest airport, he landed it in a grass marsh near Homosassa, Florida, with the landing gear retracted.
During the landing, the wing hit grass and then veered right about 90°, which caused the left wing to dip and hit terrain. The airplane sustained substantial damage to the left aileron and empennage.
The pilot reported that there were no preaccident mechanical failures or malfunctions with the airplane that would have precluded normal operation.
During his preflight inspection, the pilot should have verified the fuel quantity in the fuel tanks to ensure there was sufficient fuel onboard for the flight, and his failure to do so led to fuel exhaustion and the subsequent total loss of power in both engines.
Probable cause: The pilot’s inadequate preflight inspection, which resulted in fuel exhaustion and the subsequent total loss of power in both engines.
NTSB Identification: GAA19CA070
This November 2018 accident report is provided by the National Transportation Safety Board. Published as an educational tool, it is intended to help pilots learn from the misfortunes of others.
I fly a 74 year-old aircraft. The original “main” fuel system has a float gauge (not electric). The STC’d aux tanks added in 2000 have a fairly accurate electrical gauge. But it’s a tail-wheel plane, and the gauges only read correctly in a level flight attitude. And I’ve been around long enough to never place much faith in a GA aircraft fuel indicator. My solution to avoid running out of gas is bone simple if you have a notebook, pen, and can do simple math.
I know the conservative average fuel burn per hour, and keep a small spiral notebook in which I record the tach time of each flight and estimated gallons burned. I also record the gallon amount each time I add fuel to keep track of how much fuel is onboard. Pretty simple…For instance, if I take off with full tanks (5 hours usable) and fly 4 hours, I will land with roughly one hour of fuel on board. And I always plan to land with an hour of fuel. So far, after 22 years of ownership, I find that each time I refuel, the amount needed to fill the tanks is always within one gallon of what I estimated would be needed. And even if I make several refills of less than a top-off, inevitably when I do top off, the amount added is always within a gallon of my estimate. So simple a cave man can do it.
And BTW, planning to land with an hour reserve may sound overly conservative, but there’s been more than one time when having planned that much reserve came in handy.
As mentioned, this aircraft didn’t dump 80-90 gallons of fuel because one main gear was on a dolly. Figure out the angle of tilt that would require.
Didn’t visually confirm quantity on board.
Gauges ” in the green”…yup, showing 3 hours worth of fuel and guess what happened 3 hours later ?
No surprise and adding anything other than intelligence won’t fix the ….self inflicted problem.
I imagine the FAA knows the guy is distorting the truth but not worth their time to investigate.
It seems a bit over the top to believe the plane vented off 50 – 70 gallons of fuel and nobody would have noticed. That’s a lot of flammable liquid to spill either on the ramp or in a hangar.
Methinks that the fuel venting system has a problem. On the C77R the left tank vents out of the right wing “tip” trailing edge and the right tank vents out the left. IN this way, when in a bank one is not dumping fuel.
And if the T337 is like the C337 that I’ve flown in (not pilot, not multi rated), it has inboard tanks (left and right) and outboard tanks (left and right). At 150 gallons (full tanks) and only getting ~3 hrs flight, a lot of fuel was dumped on that taxiway and parking area, etc.
So to me, with my limited experience with a friend’s 337, I think something was significantly wrong with the fuel vent system.
As mentioned, visual verification and time are the operatives.
Gauges being there are a backup to time to inform you there’s excessive consumption, running rich or a leak.
And any vehicle has it’s peculiarities, all my GM trucks have a third when showing half….know your equipment.
However there are more accurate systems, for a price.
It’s a bit ridiculous that fuel gauges can’t be trusted. If fuel gauges are inaccurate, there should be an AD to fix that. If the altimeter says you are at 1000 feet and you are really at 50 feet, is it a “pilots failure” if he hits a 100 foot tower on an instrument approach? “Check the fuel tanks” is obviously on the checklist so we should always do it, however, we should be asking why that necessary. A fuel gauge isn’t some complicated concept and yet they are almost completely worthless as this accident would indicate.
Agreed. Imagine if the only way you could get an accurate measure of how much gas was in your car was to dip the gas tank. How long do you suppose people would put up with that?? Yet an accurate measurement of gas in the tank is far more critical in an aircraft.
Aircraft fuel gauges were never designed to be accurate, with the FAA requirement to only be accurate at empty, [ no usable fuel remaining ]. They operate on the buss voltage that varies from 12.5 volts to 14.5 volts. Then the gauges are so small, they are difficult to read. Not to mention that there are no baffles in the tanks, so the fuel sloshing causes the gauge needle to swing wildly.
By contrast, auto gauges run on a 1% regulated 5 volts, and the gauge is readable and with baffles, there is no fuel sloshing..
But, the actual useful measure of fuel remaining is time in flight; in hours… The POH is quite accurate in specifying the fuel use in GPH for various engine rpm , or rpm and manifold pressure.
After engine start , I set my clock to 12-noon. Then a glance at the clock will let me see my flight time. If it gets to 3:00, I land and refuel….simple and accurate to within less than 1 gph, or 3 gallons in my case.
‘We don’t need no stinkin’ gauges.’
there are STCs to add a digital engine monitor which includes a fuel flow sensor, which will indicate fuel used. BUT, it’s still up to the pilot to input an accurate number for fuel in the tanks….Are they really full when he inputs the value for ‘full fuel’ ?
The idea that fuel gauges only have to be accurate when empty is a myth and needs to die. I got berated for saying this on my PPL checkride 4 years ago. The DPE was beyond frustrated that people are still spreading this misconception and chided my CFI for teaching me as such. It’s nowhere in the FARs,
What is in the FAR is this:
14 CFR 91.205 (b)
(9) Fuel gauge indicating the quantity of fuel in each tank.
I have read it in the FARs requiring gauge to accurately display empty.
Me Too. By the way, the most accurate fuel gauges in aircraft are in the Piper J-3, Taylorcraft, etc.
“Aircraft certification rules require accuracy in fuel gauges only when they read “empty.” Any reading other than “empty” should be verified.” PHAK pg 7-26. Let your DPE know so he won’t have an attack. https://www.faa.gov/regulations_policies/handbooks_manuals/aviation/phak/media/09_phak_ch7.pdf
91.205[a] specifies that the instruments are to be ..’..in operable condition’, and in 91.205[b], there is no accuracy requirement specified for any of the 9 required instruments, for VFR flight, just that they have to be there and operating.
My old Cessna fuel gauges will show about 1/4 tank difference between the engine running with 14 volts , vs 12.7 volts with the engine not running.
And, depending on temperature, the buss voltage varies from 14.7 to 13.8 volts
Without a regulated voltage to the gauges, they cannot display an accurate reading of fuel quantity.
As pointed out elsewhere the fuel gauge system currently used in automobiles is far more accurate then what is in almost all GA aircraft.The real problem is the cost to replace those old tech systems that came out of the factory with that new technology. The FAA made some good progress when they started allowing lower cost avionics from the LSA/EAB world into certified aircraft, now that needs to be applied to the less glamorous but obviously more critical area of fuel gauges. Lets start taking out those rarely accurate legacy systems and putting in systems that a pilot can have some faith in. Until that is done this is going to keep showing up in accident reports.