Fuel tank calibration

[microflying]

Calibrated the right tank OK and the display reads 50lts. which is the full value. The left tank appeared to calibrate ok with all the right figures and messages throughout calibration but then right at the end message came up “tank calibration unsuccessful. Do you want to use the settings anyway” answered yes and the display came up with a full value of 44lts. Thought I had maybe miscounted the number of 5lt additions so calibrated again only to get the same message and result. The unit was told a full tank was 50lts and I definitely added 50lts in 5lt increments as requested so why does it come up with an error message and tell me there is only 44lts in?
5lts. was added before calibration started as the sender doesn’t register the first 5lts.

 

[dynonsupport]

Did you happen to jot down the sensor values that are displayed at the end of the calibration? Generally, that error means that there was something abnormal about the way the voltage/resistance went up and down as you calibrated. For example, if you plot the liters/sensor value, the resulting graph should only go one way. If there are "humps" in the chart, say if 5 liters was 3.4V, then 10 liters was 3.5V, then 15 liters was 3.6V, then that creates a situation that the calibration process has trouble with, since a reading of 3.4V is is non-deterministic. I'd probably see how the calibration you have performs as you drain and fill a tank of fuel in order to test it.

 

[microflying]

Calibration results were as follows.
1    2929    0.0
     (dif 419)
2    2510    1.32
     (217)
3    2293    2.64
     (374)
4    1919    3.96
     (330)
5    1589    5.28
     (269)
6    1320    6.60
     (161)
7    1159    7.93
     (233)
8     926     9.25
      (180)
9     746     10.57
      (456)
10    290    11.89

Not sure what all the figures mean and how they should be. Seems to be a lot of variation between lines on the middle column. Can you offer any advice?

 

[dynonsupport]

Those number looks sane - I'm having one of our developers take a look to make sure there's nothing I'm missing.

 

[microflying]

Checked the calibration on my left tank by filling in 5litre increments.  The overall calibration seems fairly accurate after all. Just not happy with the results at either end of the tank capacity though. The first 5lts I added gave no reading as expected. The next 5lts didn't register the full amount either. The display only recorded 2lts. So from there upwards any amount of fuel added is displayed as 8lts less than what is actually in there. Maybe that's not too bad a thing as it gives an inbuilt safety margin but on top of that the sensor stops recording at 44lts(indicated) so any fuel added above this amount is not sensed either. So if I have a full tank of 60lts, as measured, I have only 44lts recorded. For a sophisticated bit of kit like the Dynon is I don't think that is very satisfactory to have a sensor that only reads 70% of the fuel in the tank.

Is there anything that can be done to improve the range on this?

The right hand tank seems to have a lot larger range. 5lts to 50lts. Which is maybe not too bad.

 

[PhantomPholly]

Capacitance sensors historically are poor at measuring the "top" and "bottom" of the can. Some people install them diagonally hoping for a bit more voltage variation.

Your best bet for a more accurate measurement comes from a couple of tactics.
1. Install high/medium/low sensors for which you have a good idea what level of fuel they represent
2. Use the fuel flow sensor from Dynon and calibrate it. Most folks say that gives you a reading that is spot-on.