AutoQuad Forum

Johannes

I see in your log that your RC channel 6 starts out at -700 and then goes to 350 micro seconds. If your using channel 6 for flight mode control then you may be flying in APH and not manual control. The only difference in APH, without a GPS 3D lock, is that you can't arm in APH and you have altitude control with the barometer.

If you are in APH when you get a 3D lock you should expect to have control problems like you described. After a minute or two the GPS gets more accurate and then APH is very precise if you don't have some of the GPS signals blocked by trees.

Larry

Hello,
today i did some testing and i think i got a bit further^^.
I did a hover in 10cm height indoors and stable, then i went outside and held it over my head and then i did a short hover.
Test 1 and 2 was successful. Though test 1 was a indoor test i had no gps signal. When i was holding the copter over my head i had a stable 3d fix and the copter behaved normally. This motivated me to test a outdoor freeflight. This was test 3 in my logfile. First everything was as it should be, but right after the takeoff, the blue gps light went off and motor number 3 decelerated and the quad started to flip. My reaction was to set throttle to 0 and it landed without damage:-). I think the gps antenna is damaged somehow, could that be? Perhaps i should order a spare one for testing.
I attached a screenshot of the qGroundControl where u can see the stick positions and flap positions of the radio i configured for manual flight. The gui also says manual.
@lpr i think i switched to set home position last time and then back, just to be sure this was not the fault.
I configured the limits for channel 6 to -350 to 350. The gui says -350 for manual 0 for pos hold and 350 for mission mode.
Does that mean i was in mission mode?

picture of my normal take off configuration

Hi Johannes

Good you could land your craft without damage. Instead of closing the throttle, you can always switch back to manual (throttle at center) and land more graceful.

Flight modes:

<-350 = manual
>=-350 to <=350 = Pos Hold
>350 = mission mode

So you where in manual mode. But RC can jump 2 or 3 points easily - I would set the channel to -700 to + 700 as we suggest.

Check your GPS connection. You should always see a steady Hacc GPS value of 0.4 - 0.8 for good performance. I usually do not use GPS modes (PH and NAV) with Hacc above 1.0.


Cheers
Christof

Actually the switch thresholds (in the fw code) are +/- 250, so 350 gives a safe overhead. < or > 350 is also fine, of course. The important part is that middle switch value should be > -250 and < 250 (which isn't obvious from the hints in QGC).

-Max

the strange thing is i never tried a mode other than manual, what gives me headaches is, that even when i am in manual mode, i can fly normally and stable, but as soon as i get a gps fix it goes mad, so i sadly cant switch to manual mode to land softly:-(. This also prevents me from testing in open airspace.
The compass stands still and doesnt move a bit if i dont move the aircraft.
I always go testing with live telemetry and do a preflight check with the groundcontrol.
I think even magnetic interference should not have any influence here since the dimu firmware doesent use the mag after arming, or after the init phase. I saw in the logs, that the mag is a flatline after arming the aircraft.
One more point that gives me headaches is that if i test the aircraft above my had and holding it in my hands, i cant reproduce that behaviour, it behaves normal (with 3d fix) and counters the movements of my hand. Taking of from the ground is impossible. If i disconnect the gps antenna, i can easily take off from the ground to and hover in place for minutes.

Man, this is giving us headaches, too!

Ok, replace the antenna. While waiting on a new one you could do further tests with a simple Dipole antenna:
http://autoquad.org/wiki/wiki/m4-microc ... PS-Antenna

Now that you know how to set it up you could also start from scratch again. That´s sometimes the only way to find out.
Especially check the plugs of the GPS port, both on the wire and the board.

If this doesnt help, you might have a rare case of bad GPS module or something else in the way producing abnormal behaviour....

jap;-)
I will also try to disconnect all my peripherals like the wireless telemetry and the teensy for the taranis telemetry and do a bare metal test. Also i will build a gps shield.
But there is still one question that drives me mad, where is the difference between manual mode without gps antenna and manual mode with gps 3d fix?
Because in any case something goes wrong, i want to have the ability to switch to manual and have the rudimentary gyro acc stabilization to land savely land the craft.

Of course this is the way it should work. Manual is manual, no matter what the GPS says.
This is the first case I´ve ever heard of, where a 3D fix makes it freak out.

"Something" is triggering that odd behaviour and so far no real hint of what it could be.

Could be both (corrupted) firmware or bad hardware or a fatal combination of both.
Atm I`m tending to a hardware problem...

btw: what does the MCU load and tx/rx losses say in the status details? Any significant change when the 3D fix occurs?