paparazzi-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Paparazzi-devel] Re: Tuning Suggestions Part II (TurboTas)


From: Gautier Hattenberger
Subject: Re: [Paparazzi-devel] Re: Tuning Suggestions Part II (TurboTas)
Date: Mon, 02 Feb 2009 09:35:05 +0100
User-agent: Thunderbird 2.0.0.19 (X11/20090105)

Hi,

I can add that when you are using an ADXRS 150, the last value in the telemetry is not the pith rate but the temperature compensation. And as Christophe said, the "estimator_q" value is not used anyway. Your problem probably comes from IR sensors. Have you checked in the makefile section that you add all the needed flags -DUSE_ADC_x ?

GH

Christophe De Wagter a écrit :
Toby,

Have you done the following?

-put your funjet on your desk
-turn on everything (gcs, rc (manual mode throtlle off or even remove propellor), funjet on)
-look at the PFD on the GCS (http://paparazzi.enac.fr/wiki/GCS#PFD)
-put one (warm) hand underneath the vertical infrared sensor (otherwise the contrast calibration goes crazy)
-switch to auto1 (throttle off)

-move your second hand in front of the horizontal infrared eyes (aircraft nose side)-> you should see the elevators go up and the PFD should show you are diving -move your hand behind of the horizontal infrared eyes (tail side) -> you should see the elevators go down and the PFD should show you are climbing
-hand right -> right aileron drops and vice versa

Make sure you read this: http://paparazzi.enac.fr/wiki_images/Users_manual.pdf

Whether or not you have no pitch gyro, the pitch angle is estimated from the infrared sensors, and the pitch rate is computed as the filtered derivative of the pitch angle. [BTW: even if you have a pitch gyro it is not used yet in pprz (cvs code) as the pitch gyro does NOT measure the pitch rate whenever the plane turns]

If you hesitate about any measurements (infrared, gyro or GPS), don't try it out while flying but try them om the ground. If your gyro is calibrated in the wrong direction for instance and your gain is low you might only find out when the plane encounters a big gust. Before every launch its good to check in auto1 if the controls work fine by holding the aircraft over your head and checking if the elevators go up when the nose goes down, it rolls back when not held straight, and the ailerons counteract a fast roll motion. So if by accident your temperature sensor is connected to your infrared sensors, you won't find out while flying! Best of luck,

Christophe


    ---------- Forwarded message ----------
    From: TurboTas <address@hidden <mailto:address@hidden>>
    To: address@hidden <mailto:address@hidden>
    Date: Sun, 1 Feb 2009 03:51:08 -0800 (PST)
    Subject: [Paparazzi-devel] Tuning Suggestions Part II
    Hi All,

    I posted a couple of weeks ago about a tuning problem I was
    having.  To refresh your memories, my funjet plunges groundward
    every time I try AUTO1.

    I have made some progress but would be great for one of you
    techies to validate my steps!

    What I did was to use the Log plotter and go through all the
    available variables plot by plot to see if I could see anything odd.

    I think I found something which may be the culprit, but I don't
    know how to fix it - or even if my diagnosis is valid.

    My airframe is a Funjet and I have a pretty standard setup: Horiz
    and Vertical IR sensor, Roll Gyro, Xbee etc.

    For the Roll Gyro, I use the ADXRS 150 and have both the Temp and
    Rate connected, so I have a ADC section like this:

     <section name="adc" prefix="ADC_CHANNEL_">

       <define name="IR1" value="ADC_1"/>

       <define name="IR2" value="ADC_2"/>

       <define name="IR_TOP" value="ADC_0"/>

       <define name="IR_NB_SAMPLES"        value="16"/>


       <define name="GYRO_ROLL" value="ADC_5"/>

       <define name="GYRO_TEMP" value="ADC_6"/>

       <define name="GYRO_NB_SAMPLES"      value="16"/>

     </section>

    Look good?
    Then I hve a Gyro Section which looks like this:

     <section name="GYRO" prefix="GYRO_">

       <define name="ADC_ROLL_COEFF" value="1"/>

       <define name="ADC_ROLL_NEUTRAL" value="495.464"/>

       <define name="ADC_TEMP_NEUTRAL" value="476"/>

       <define name="ADC_TEMP_SLOPE" value="0"/>

       <define name="DYNAMIC_RANGE" value="150" unit="deg/s"/>

       <define name="ADXRS150_RESISTOR_BRIDGE" value="(3.3/(3.3+1.8))"/>

       <define name="ADXRS150_SENSITIVITY" value="5" unit="mV/(deg/s)"/>

       <define name="ROLL_SCALE"
    
value="3.3*1000./1024./(GYRO_ADXRS150_SENSITIVITY*GYRO_ADXRS150_RESISTOR_BRIDGE)"
    unit="deg/s/adc_unit"/>

       <define name="ROLL_DIRECTION" value="1."/>

     </section>

    Okay, to me this looks correct but I get an odd result in the Log
    plotter.  In the Plotter under the Gyro _Rates Section, I only get
    three variables: Pitch, Roll & Roll_ADC.

    If I plot these in the log plotter, I get pretty much what I
    expect from the ROLL and ROLL_ADC.  Good numbers that tie up with
    the flight.

    However, If I plot the PITCH param, I would expect to get nothing
    as I don't have a pitch Gyro, But what I actually get is a number
    that starts off around 25 for the few minutes before flight.
     After takeoff, the number gradually drops to about 5.  After
    landing, the value gradually starts to climb agian.

    My guess is that I'm seeing the gyro temperature in the PITCH
    position on the plotter.

    The obvious question is therefore: Does this mean that the AP is
    seeing the temp as pitch as this may explain the lack of pitch
    controll?

    Any comments would be great!

    Toby.




------------------------------------------------------------------------

_______________________________________________
Paparazzi-devel mailing list
address@hidden
http://lists.nongnu.org/mailman/listinfo/paparazzi-devel




reply via email to

[Prev in Thread] Current Thread [Next in Thread]