Odd readings from fuel capacitance plates

[david.perl]

Plane has just passed the 200 hour mark.

I noticed a couple of months ago whilst refueling the tanks, they were overeading about 20 ltrs (4-5 gallons) per side when about 1/3 full. This is on both sides so that would eliminate a bad electrical join. I've only used Avgas so am pretty sure its not a fuel mismatch issue.

When full, the readings are accurate so the capacitance plates are putting out the correct voltage at that point, however below that, the misreading somewhere starts and then continues.

Im keeping track of fuel burn with my red cube which is pretty accurate.

I plan on draining both tanks and undertaking a recalibration on both tanks.

Before doing that I wondered if anyone might shed some light on this odd behaviour?

[david.perl]

I drained both tanks and recalibrated the sensors.

The problem hasn't gone away which i dont understand.

Anyone got any other ideas as to why the readings should be different?

I plan to take the voltage readings from the capacitance plates inflight in 10 ltr incremental drops on each side and then adjust the voltage readings as per each drop on each side to try and get more accurate readings from the plates

[Capt]

Mine are erratic but I never trust the Vans units anyway. I have digital FF indicator which is accurate to within 1 Ltr per full tanks (pretty good) that plus an accurate wooden calibrated dip stick I know my fuel Qty very accurately at anytime, comforting👍

[jcaplins]

I'm sure my setup is different; I have the Vans plates, dynon converters and GRT eis4000, mine is only accurate at full and empty. I can only calibrate an offset and scaling factor, not specific points in between.
I've had fuel leaking through the bnc connector cause a shift in the fuel readings. I don't remember the exact number, but I think it was reading 5 to 10 gallons high.

I'm not sure if it was the fuel wicking through the connector or fuel getting into the converter that was the problem.

[longranger]

Originally Posted by jcaplins

...
I've had fuel leaking through the bnc connector cause a shift in the fuel readings. I don't remember the exact number, but I think it was reading 5 to 10 gallons high.

I'm not sure if it was the fuel wicking through the connector or fuel getting into the converter that was the problem.

If you didn?t seal BOTH ends of the wire from the inboard capacitance plate to the center terminal of the BNC connector, it WILL leak. Fuel has a very low surface tension and will wick between the conductor and insulation inside the wire.

[jcaplins]

Originally Posted by longranger

If you didn?t seal BOTH ends of the wire from the inboard capacitance plate to the center terminal of the BNC connector, it WILL leak. Fuel has a very low surface tension and will wick between the conductor and insulation inside the wire.

Agree 100%. For 1 tank I either forgot to seal the wires or I just didn't do a good enough job.

I put a tiny o-ring in the bnc connection, and a year later seams to still be working great.

[Snowflake]

I haven't flown an airplane yet that had an accurate fuel level reading in flight using a method that measures the level directly. Floats stick, or don't read the full tank. Capacitive senders can be confused by different fuels or additives.

By far the most accurate is the dip-stick before flight. Equally accurate has been the red cube fuel flow sender installed firewall forward. If I were building new, I wouldn't put any senders in the tanks. I would put a red cube in each wing root (or on each side of the fuel selector) and measure flow out of each tank directly. No senders in the tanks = less places for leaks, and more volume in the tank for fuel.

[jcaplins]

Originally Posted by Snowflake

I haven't flown an airplane yet that had an accurate fuel level reading in flight using a method that measures the level directly. Floats stick, or don't read the full tank. Capacitive senders can be confused by different fuels or additives.

By far the most accurate is the dip-stick before flight. Equally accurate has been the red cube fuel flow sender installed firewall forward. If I were building new, I wouldn't put any senders in the tanks. I would put a red cube in each wing root (or on each side of the fuel selector) and measure flow out of each tank directly. No senders in the tanks = less places for leaks, and more volume in the tank for fuel.

Agree with using a dipstick. I time my flights and the fuel computer is a backup telling me how much is remaining. I never look at the fuel gauges other than to note they are there and working.

Unfortunately FAR 91.205 says you will need fuel gauges regardless if you want them or not (for the US anyway).

[Mark33]

I?m going to give these units from Belite a try. http://www.aircraftspruce.com/catalo...ecfer=15727683 They?re ?kind of? like a capitative system but somewhat of a different technology. They?re crazy expensive, but if they work as advertised, they?ll hopefully provide me with many years of trouble free and accurate fuel level readings.

[Waiex-guy]

I really like the idea behind the Belite sensors, but unfortunately they don?t seem to work well with vented fuel systems. They measure tiny pressure changes and equate that to the fuel remaining, but the vent system changes the pressure above the fuel which completely disrupts the process. There are other threads in vaf addressing this problem.

Gordon