Fuel Probes (again)

            Could the pressure increase inside the tank going up with speed
            via the venting cause a capacitance variation thus a
    different DC
            reading.

            It could be verified on the ground by increasing lightly the
            pressure in the
            tank.
            Or in flight taking a reading in slow flight and another one at
        cruising
            speed.

            Just guessing have no logical explanation for such a phenomena.

            JP

        JP,
        I paid close attention to the fuel level readings on my last
    flight and
        I have to say that they appeared to be pretty steady in level
    flight. I
        may have caused some of the change in readings via uncoordinated
    turns.
        Not that the ball was all the way left or right, but just a little,
        possibly enough to cause the fuel inside the capacitive probes
    to move
        up the tube a bit, which would cause the fuel readings to vary a
    bit.

        The big surprise was when the right gauge, the one with the
    Dynon CTV
        converter installed, went to zero as soon as I keyed the mic. Up to
        that point, it had been the more reliable of the two. I'll have to
        check again, but the left gauge, the one with the stock EI circuitry
        (square wave a/c) appeared to drop by a few gallons while the
    mic was
        keyed. An undesirable feature, but not as drastic as the right
    side.

        I plan on doing the following:
        1) re-route the antenna cable away from the efis and other wiring
        2) mount the CTV in a grounded metal box
        3) run a separate ground from the capacitive probe to the a/c
        ground bus
        4) Purchase an EI fuel gauge and hook them up and forget about
    reading
        fuel levels on the EFIS. ;-)

        Mark S.

    It's somewhat likely the noise is radiated as opposed to conducted to
    the sensor circuit.

    Do you have access to a handheld transceiver (with a "rubber ducky"
    antenna)? While transmitting with it (on a locally inactive frequency),
    you may be able to identify a portion of the circuit that is vulnerable
    to the RF radiation. If you don't find any part of the circuit
    responding to the handheld then grounding properly likely will improve
    things.

    There are instruments sold by Van's which are vulnerable to RF. The
    only solutions found have been to shield the instrument (in your case,
    probably the CTV box) with metal - heavy foil may work, but a metal box
    might be better. Or ignore it. When the mic is un-keyed, do the
    readings return to normal immediately?

    Regards,

    Matt-

Matt,
I hadn't considered using a hand-held radio to trouble-shoot this
problem. I'll give that a try. Once the mic is unkeyed, the readings
will return to normal, but not immediately since they have a built in
damping factor.

For now, my most reliable method of tracking remaining fuel is with the
Real World Solutions EM-2 engine monitor. This unit calculates the fuel
used based on the duty-cycle of the injectors. I enter the number of
gallons added when I refuel, and it counts down (or up if you wish) from
there. Even though it doesn't track fuel levels in each individual
tank, it is still a pretty nice backup system.

I also have a 7-gallon low fuel warning light for each tank, driven by a
GEMS sensor. Again, nice to have a backup system when the primary is a
bit quirky.

Mark

Hi Mark,
Matt may be on the right track with the handheld idea. If its possible
align the antenna of the handheld as parallel to the installed cap. tube
as you can then key the handheld while someone else watches the gage.
Get as close as you can to get the greatest and quickest effect. If you
get no response try aligning perpendicular to the tube and key the
handheld again. It's possible that you were lucky (?) enough to get your
installed tube and installed antenna exactly in parallel
geometric planes and the radiated energy is the culprit.     Dave M

Mark,
I think you have a diode junction somewhere that is sensitive to the RF
and is charging something that upsets the bias on your capacitance
sensor. This link: http://rbi.ims.ca/5698-100 from a recent issue
of Design magazine will take you to the Analog Devices web site and an
article about capacitive sensors. From there you can probably find some
application notes on recommended RF bypassing techniques for the Analog
Devices IC's and capacitive sensors. Ferrite toroids on your wiring
near the termination points may be all you need to cure the problems or
maybe a 1Meg resistor somewhere to bleed charge.

Aubrey Thompson

The Rotary Engine NewsLetter. Powered by Linux.
ACRE NL web site. http://www.rotaryeng.net
Copyright 1998-2007 All world wide rights reserved.