[Username Protected]

It seems odd to me that the Garmin G3X has lots of data channels coming from the Rotax engine, but not fuel flow. I understand that the fuel injection design of the Rotax is more complex than legacy aviation piston engines, but I find it hard to believe the problem can't be solved. Further, with the Icon weight limited and often needing partial fuel loads, it's fairly crucial to me that the fuel state be accurately tracked, particularly when flying cross country. My local Rotax agent says there is a fix available using known components, but I also understand that Icon then needs Garmin to modify the Icon software for the G3X before that option can be explored.

Has anyone had any luck with this issue?

[Username Protected]

My understanding from Brett (who led the G3X project at Icon) was that the fuel flow number coming from the Rotax ECU wasn't accurate enough (maybe because of return line?) - although I would argue it is better than looking up performance tables in the POH. I think it has to be enabled on the ECU side, if you set FF to "Rotax FADEC" on the G3X in setup mode it still has no data to my recollection.

https://static.garmin.com/pumac/190-01115-01_aw.pdf

Rotax 912iS/915iS: For the GEA 24, configure the FUEL FLOW input to "Rotax FADEC" to use fuel flow data from the Rotax 912 FADEC interface.

I think to do it properly you would need two fuel flow readers, one on the intake and one on the return line.

The GEA 24 and GSU 73 also have provisions for a second fuel flow input (FUEL FLOW 2) to use in aircraft that require a second fuel flow sensor for differential fuel flow measurement. If both fuel flow inputs are configured, the displayed fuel flow will be FUEL FLOW 1 (feed) minus FUEL FLOW 2 (return).

[Username Protected]

The ECU fuel flow is estimated based on assumption of the fuel pressure and the opening timing of the injectors. We and many others (not icon specific) find its lower than the fuel flow actually is. That number has nothing todo with the return line.

What Marcus is referencing is correct. You install a fuel flow sender or in this case two 1 on the feed 1 on the return. They do the math and subtract the two so you only are displaying the consumed fuel. Here's the rub the paddle wheel style fuel flow sensors are very susceptible to inaccuracies do to installation location and air bubbles. The way the Rotax system works is it pressurizes the fuel to the fuel rail (3 bar) and thus the areas where the injectors get fuel that fuel pressure is regulated by a mechanical fuel pressure regulator that is vented to the intake manifold. The nature of taking pressurized fluid and depressurizing it to the local atmosphere can cause an air bubbles to form causing the return fuel flow sensor to read wrong. Its a complete PITA to get calibrated correctly it can be wildly off. Ask me how I know.

my cousin has a legend cub with a Rotax 916iS and Garmin G3X he just did a big trip from Texas to Northern Nevada to Alaska and back to NorNev and shortly after one to Oshkosh and back. He put over 100 hrs on the plane right out of the factory. He was complaining that the fuel flow was fairly wrong when he did fuel ups and not in a conservative way. I told him we noticed the same thing at ICON during development. I brought this to Rotax's attention but they didn't care enough to do anything about it.

I spent a stupid amount of time convincing the NTSB that the fuel flow on the Rotax was wrong during the fuel starvation accident in Chicago in 2019 cause they were looking at the Flight data recorder data. We did multiple data gathers to prove to them the performance charts were accurate and the Rotax fuel flow was wrong (the performance charts were developed NOT directly from fuel flow values off the ECU) I have a photo somewhere where I filled a fuel tank using graduated paint mixing cups 1/2 gallon at a time, then drained it after we did the flight to prove the total consumption during the flight. I don't think any of that made it into the report because the pilot didn't have access the bad rotax data.

That all said you could enable it to show up. In the original configuration some 4 years ago I chose not to enable it because of that previous accident I didn't want people pushing it like that guy did thinking that data was correct which is certainly is not and will not match the perf charts.

[Username Protected]

Sorry everyone to ask maybe a dumb question but I'm curious: if there was a way to know the exact amount of fuel in the tank (which of course is undermined by constantly changing body acceleration - both directional and rotational), would the ongoing differential be an accurate reading of fuel flow? i.e. is this an interest in the engine itself (how much fuel is it burning) or in the rate at which fuel is being depleted overall (the rate at which the "plane" is consuming fuel - even if it's for ex. leaking out the tank into the wind)?