Re: [Paparazzi-devel] ArduIMU

[Hector Garcia de Marina]

About the vibration tolerance, it strongly depends on your sample frequency. I would check this out before purchasing more expensive sensors.

I see in many UAVs projects that the sample frequency is still lower than two times the maximum propeller rpm (main fuselage vibration source), thus

most of the spectrum power from high frequencies is aliased and moved to low frequencies, distorting for example the gravity in case of accelerometers.

A typical 1000kV motor with 3S LiPo can achieve 6000rpm (1Hz is 60rpm) at half-throttle signal, that means you need to sample @ 200Hz minimum if you don't want an aliased signal.

The same happens when the telemetry is stored at the ground station with a low frequency data-packet.

It is funny seeing a ~1Hz signal (very common, just check it out) in Y-Axis accelerometer when the fuselage is in resonance with the motor and the sampling frequency is "low".

Héctor

Cheers

On Thu, Jul 28, 2011 at 5:17 PM, Prof. Dr.-Ing. Heinrich Warmers <address@hidden> wrote:

Hi,
as you know we use the DCM port of the dydrohnes project  inside the tiny and twog by using the razzor pcb.
We always make a drift compensation  for each power on. Therefore we glue a bubble on 
the aircraft amd put the aircraft  in the calibration attitude before power on. This work well but need 10s.
The zero level of the gyros is critical. If there is an error of one or two LSB the DCM will fail and you can see the pfd rotating.
On the other side can you fly circles for an halve hour without any problems. It seems this is not true for the paparazzi AHS.
The only problem with the low cost sensors is that there deliver a bad  vibration tolerance.
Regards

Heinrich


Reto Büttner schrieb:

Hi all,

I got some feedback from the guys of ZHAW directely:

- The throttle slew can be ommitted.
- The newest code is arduino_basic.pde (cleaned up by Gautier Hattenberger)
- ENABLE_AIR_START 1 should be set to avoid recalibration according to
current random aircraft position when plugging in the battery.

I experience the following drift behaviour:

arduino.pde and ENABLE_AIR_START 0: No drift.
arduino.pde and ENABLE_AIR_START 1: A lot of drift.
arduino_basic.pde and ENABLE_AIR_START 0: A lot of drift.
arduino_basic.pde and ENABLE_AIR_START 1: A lot of drift.

I thought I could avoid the annoying drift by going outside to get
gps. Unfortunately that didn't change anything. Therefore the only
configuration looking good to me is the first one.

Can someone explain the drift depending on different configurations?

Cheers,
Reto

2011/7/23 Reto Büttner <address@hidden>:
  

Hi guys,

I followed the wiki (http://paparazzi.enac.fr/wiki/ArduIMU) to put my
ArduIMU into service. Thanks for the great instructions! I updated the
wiki with some minor points according to my experience.

I am using the most current git master repo (no branches). I had to
comment out the following two lines in the file ins_arduimu.c for the
code to compile:

float throttle_slew;
throttle_slew = V_CTL_THROTTLE_SLEW;

Has anyone experienced the same?
Should it be THROTTLE_SLEW_LIMITER instead of V_CTL_THROTTLE_SLEW?
Are these lines (the throttle slew) needed?

Then the ArduIMU works just fine. At least in the lab, the field
experience is to come :-)

Where can I get the most recent, proven, stable and sophisticated code
for the ArduIMU?

Cheers,
Reto

    

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

  

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

-- 
Héctor