Fuel Calculator inaccuracy

Glenn Martin wrote:

There seems to be some misunderstanding of EFI systems in these discussions.

Engine developers use injectors that are tested for fule flow accurately (less than 1% error). They do this by checking flows at different opening times and durations referenced to temp, and fuel pressure. So a variable fuel pressure is compensated for by the ECU adjusting the opening times.

This is done because fuel flow needs to be very accurate in an ICE engine to get efficient burn and power.

Therefore the ECU counting up the fuel injected is a very accurate way of measuring fuel used, much more accurate than almost all fuel flow transducers.

Remember you need at least two fuel flow transducers to measure used fuel in an EFI system.

Most people I have seen making fuel gauges for their aircraft, eg making sticks and marking them are not very acurate. We do not take into account temperature and pressure that can make additional errors.

For instance the difference of say 20 degrees C can change the volume by 24ml/Litre 2.4%.

Then when we fly fuel vapour pressure and vent tubes means we loose some, particularily when high and fast.

Being a boring old scientist I spent a full day with calibrated measuring cylinders a plane on a flat surface checked by a spirit level and as close to a 15 degree C day as I could get

I am planning a few "long flights" where I will burn about 200L. I have practiced these and have found the G3X to agree with my dip stick within 3-4%.

I suspect the error is MINE not the G3X.

Therefore to go against the flow , pardon the pun. I think the first step is for anyone having significant errors to accurately recheck your own fuel quantity measuring system, your vents etc.

Sorry to disagree !

Glenn

We're not comparing it against fuel tank quantity measuring systems. We're topping off the tanks, going flying, topping it off again, then comparing the amount pumped to the Fuel Used displayed on the G3X. The Fuel Used displayed on G3X is 10 - 15% (typically 3 to 5 gallons) below the amount shown on the fuel pump meter. This is multiple different airplanes and different types. The only things in common are Rotax iS and Garmin G3X. So the question is, is the error on the Rotax end or the Garmin end? I've emailed Garmin - awaiting a response. 

My current aircraft has a 912S. I’ve only had it for a few months, so it’s preliminary info but the fuel quantity based on the fuel flow transducer (so measuring actual fuel flow) seems to correspond very well to the actual fuel in the tanks. Garmin G3X avionics.

My previous plane used Thielert/Conti Diesel(s) which uses the same method of calculating fuel flow as in the 912iS/915iS. I did a thorough comparison between calculated fuel flow vs true flow and while at 60% power the data corresponded, at higher power settings, 75%, the calculated fuel flow underread by about 10%. I checked with another owner with the same type of aircraft; same there.. Garmin 1000 avionics.

So in the end maybe the old tech FF meter is more accurate? Will do more measurements in the months to come to verify accuracy of the FF meter.

I know it’s a little off-topic because we are talking about the 915 and its particular way of measuring fuel flow, but here are my measurements over the last few months of the ‘old fashioned’ FF meter/G3X combo: 774 litres refuelled (corrected for drained fuel), G3X says 757 litres used. So the FF meter/G3X combo underreads 2.2%. Not exact, and ‘on the wrong side’ for safety, but not all that bad I would say. 

Well, maybe the missing fuel is not being burned in the engine. Maybe it escapes somewhere else. For example via not well sealed tank caps on the upper side of the wings. The effect is more visible when the tanks are toped up. The fuel can be in such a situation easily sucked away by the low pressure area that generates the lift during the flight.

If anyone comes across these issues with fuel calculator inaccuracy, It is important that you create a CSIR - https://www.rotax-owner.com/en/support-topmenu/csir-service-information-report