Very similar accident in 2013 near Bordeaux, B90C, but no casualties:
Link
Took off on fumes and landed in the vineyards !
Look at the picture:
Link
Abstract from BEA’s report (French NTSB):
“Flight from Bordeaux Leognan Saucats airport (LFCS) to Sarlat airport (LFDS). 15mn after take-off, at an altitude of 2000ft, pilot notices fluctuating Fuel Flow on right engine. He contemplates diversion to Sainte Foye La Grande (LFDF). Few moments later both engines die. Pilot extends landing gear using emergency procedure and lands amidst vineyards, sideways from rows direction. Undercarriage breaks upon reaching ground. Aircraft, slowed down by wires, stops in 50 meters. On the ground, it is noticed that all 4 tanks are empty.”
Aircraft owner is a company: Air Chateau ;-)
Link
As I read on this blog:
Link
The morale of this story could be “better a fully loaded pilot than an empty tank”
By way of clarrification that was why I specifically said some sort of visual check. In other words on most types there is a visual level which will give you a pretty good idea of the physical quantity but I agree on some types this can be literally full or not possible at all but they are few and far between. Even the 42 has a special gauge to measure the fuel physically.
Capacitive gauges have been around “for ever” industrially but in aviation they are very expensive – of the order of 10k-20k.
They are not at all expensive to make, but as with so much in aviation the technology is a victim of “product differentiation” whereby absolute crap is 1k, relative crap is 5k, and stuff that works is 10k 
The gauges in most of low end GA are of the “absolute crap” grade but in the heyday of GA (1960s and 1970s) people were much more accepting of absolute crap, in all areas of life, and in GA they got it in copious doses. A Mk 1 Cortina was pure luxury and even in my time at university (1975-78) if you had one you had absolutely guaranteed “female company” 
Needless to say, I had one of these… (but made up for it in later years!).
Then GA production and innovation almost died so we are still flying the same crap, mostly.
The TB20 has had capacitive gauges since the GT (year 2000) and probably earlier. They are very accurate – I’d say to within 3% of full scale deflection. But they still can’t be used for accurate fuel estimation low down the scale because the indicators are cheap moving coil meters and anyway Socata didn’t even bother to put numbers on the scales. Nothing comes close to a fuel totaliser – used correctly of course. Those can be in the 1-2% range and it’s hard to improve on that over the wide surface temp range encountered (fuel expansion 0.1%/degC).
I don’t know what the paperwork involves for retrofitting a capacitive fuel gauging system onto a certified aircraft. Some stuff at that level requires an STC (N-reg) i.e. it is a Major Alteration but normal approved data is not good enough. And gawd knows what under EASA. I would not spend money on it when a totaliser is about 2k, is a dead easy installation, and is more accurate than anything.
what next
I have a 600 litre ferry tank for doing some of my sillier trips, and with that full and a bit of equipment and spares, the aircraft very rapidly gets up to 120% MTOW.
Obviously, being that much overweight leads to concerns, so I have done some fairly extensive test pilot work on asymmetric behaviour at different altitudes and temperatures, and my experience is that you get a good ROC at ISA up to 600’ airfield elevation.
So my personal view is that I would rather have more fuel and be overweight than have less and stay within the envelope.
Peter,
my FS-450 was 40% off after installation … because it was installed wrong. I then downloaded the installation manual and showed the Avionics Company how to install it correctly. They swore that the instrument was “bad” – but after the installed it correctly it was only 1 % off … until today
Just for the record …
Is there any technical reason one can’t fit capacitance gauging systems to light aircraft?
The DA42 has it too.
Is there any technical reason one can’t fit capacitance gauging systems to light aircraft?
No some have it – PA46 and I think even a TB20.
If you flew a twin like most fly a single (solo, 2, or very rarely 3-up) you could depart with full tanks almost every time. It’s the culture…
But there’s another culture as well: Safest possible single-engine handling. As we all know, light twins can have marginal performance on one engine. So in order to have some margin in case of an early engine failure, especially in IMC and/or difficult terrain, you want it to be as light as possible. Therefore many pilots tend – for a perfectly valid reason – to only take as much fuel as is leagally or operationally required.
Is there any technical reason one can’t fit capacitance gauging systems to light aircraft?
On my TB20, and a lot of other SEPs, you can’t visually check if below about 50%.
But then a “reasonable payload” plane like that doesn’t normally need to depart with less than 50%. In 12 years of ownership I have only once had to depart with less than full tanks, and that was a flight to the IOM with 3 massive blokes, which was cancelled due to fog.
Twins are different. Google G-OMAR AAIB report for a classic one. You still have the 50% (or whatever) lowest visually checkable fuel issue. But on top of that you have the extra problem that a “6-seater” used commercially might actually have to carry 6 if 6 turn up, and then you have a problem if the last renter had a brain and didn’t return with empty tanks. “Non visible fuel” departures appear to be commonplace. I have no personal twin experience but quite a bit of exposure and the whole culture seems different. Again, read the G-OMAR report. That was an AOC operation (that actual flight wasn’t) and it is a total eye-opener. You might smile if I told you the operator was my 170A examiner (and the result of that is not in my JAA IR writeup)
If you flew a twin like most fly a single (solo, 2, or very rarely 3-up) you could depart with full tanks almost every time. It’s the culture…
