AQ flip over after lift off

[Lindbaergh]

Hi , today i want to test different battery setups. directly after the first start my octo flips and crashes. one arm was broken, so i replaced it. after the repair i flew 4 battery packs without a problem. i made a video (sorry for the bad quality, it was recorded with an old digicam). also i have a logfile, i post the links here. maybe someone who is more experienced then i can have a look at it.
i was flying with this setup several days before, and never experienced a behaviour like this.

my setup is:

AQ6 firmware r65
flyduino HL-FCP
MT 4006 740Kv motor
13x4 xoar wood propeller
ESC32
5000 4S turnigy battery
graupner GR12 receiver with uBEC (turnigy 5A / 6V)
cinestar 2 achs gimbal, servos powered by another uBEC (turnigy 5A / 6V)

the video:
the logdatei:

[bluuu]

are you shure that mixing table is ok and you have right motor connection ?
on this video i see that FC want to stable platform but reaction is oposite to right direction
check twice motor connections and compare them with mixing table (i do the same mistake at first tryies)

[LPR]

Lind

I have seen that behaviour when I was having trouble with the ESC32 disarming. I had calibrated the ESC32 with too high of calibration amp setting and also set the max amps too high. Then on a rare occasion like when the batteries were just charged and the weather was cold a motor of two would disarm and my octo would flip.
Did you calibrate your ESC32? How many amps do the motor draw?

Larry

[Lindbaergh]

Hi, i use the mixingtable Larry posted a while a go here:

Motor thro___ pitch___ roll __yaw
1______100____70____0__-100
2______100____50___-50___100
3______100_____0___-70__-100
4______100___-50___-50___100
5______100___-70____0___-100
6______100___-50___50___100
7______100_____0___70__-100
8______100____50___50___100

also i changed the prop size a week ago and therefore recalibrated the esc. the motors are taking 24 amp's max so i set the limit to 20 amp's. the table and motor assingment should be fine, i had several flights now that went pretty ok. and after repairing the broken arm yesterday it fly's quite well again.
anyhow terramax and jussi also give some advice that something with my motors/esc is going wrong. so i will do the esc calibration again now and also do some extra measurement's and test's.

[LPR]

Lind

If you're using the QGCS method of calibrating ESC32 here is some more information to make it a little easier to use.

You do not usually have to calibrate the ESC32 if your motor and prop combination takes less than 20 amps. To have maximum performance calibration is necessary

After many tries to calibrate with the windows QGCs I found it does work but a few things have to be done a certain way.

1. Flash the ESC32 with the newest firmware.

2. Set the calibrate amps a little lower or at the maximum amps your motor will use with your prop and battery.
Don't bother setting the Max amps in the upper right hand corner till after you calibrate.

3. Now run the calibrations. If the motor stops while making an odd noise or loud noise you should run the calibration again with a lower calibration amp setting.

4. After calibrating both the RPM and the current limit you must recycle the power to the ESC32. Make sure you unplug the main power to the ESC32 first and then the MC if it is powered separately.

5. I close the QGC, then plug in the USB, then connect main power and then restart the QGC.

6. Now set the Max amps at the same setting as you had for calibrate amps or less, but not higher.

7. I find you can test the calibration by starting the motor and then increase the RPM by moving the slider and then click on "Set RPM".

8. If the motor will jump from start to full throttle without disarming the ESC32 has been calibrated with a low enough amp setting to not disarm in flight.

9. If the motor stops when testing, recalibrate at a lower calibrate amps and test again.

Your motor will draw more current if you fly in very cold and dry air. If you don't fly in cold weather using an amp setting the same as the motor takes at maximum throttle should work ok and give the fastest change in RPMs.

This is what I found doing a few different motors.

The motor mix your using I'm still using and it seems good.

Larry

[Lindbaergh]

i recalibrated 4 times using different motors, the way Larry subscribed above. i transferred one of the results and did some test flights. the copter flies as usual, there is no difference to the way it acts before the flip. i also use a watt meter while calibration and flying, the motors wont take more then 18 amps in calibration, and around 10 amps while hoovering.
the more i test and watch the video the more i think it was a one-time problem. maybe the balance was not perfect, i don't know. i run the video with VLC player in super slomo, all motors are turning everything looks great, an then it is simply lowering the nose and raising the tail...

i append the result of the calibration and links to the log file. maybe someone can look at it, if there is something i haven't found. on the weekend i will have some more flights with a real camera and will report back next week.

recalibration for 13x4 xoar props 13.03.02 before the flip, used this setup for several flights before the crash.
FF1 +7.29164e-08
FF2 +1.45185e-03
CL1 +3.82231e-01
CL2 +1.28078e-03
CL3 +5.28156e-02
CL4 +5.58613e-06
CL5 +7.80967e-01

log's, all made under full load with gimbal etc.
http://www.flugbaerchen.de/lnkimg/aq6/049-AQL.LOG
http://www.flugbaerchen.de/lnkimg/aq6/050-AQL.LOG
http://www.flugbaerchen.de/lnkimg/aq6/051-AQL.LOG

M1/M8 13.06.16 after flip (i use the M1 result, for a test flight's)
FF1 +6.56015e-08/+6.70636e-08
FF2 +1.42875e-03/+1.44939e-03
CL1 -3.25420e-02/-1.83710e-02
CL2 +1.30697e-03/+1.22782e-03
CL3 +4.06860e-02/+3.50490e-03
CL4 +2.62383e-05/+5.94519e-05
CL5 +7.68039e-01/+8.61940e-01

M6/M7 13.06.18 (haven't used this til now)
+6.64616e-08/+7.12749e-08
+1.44803e-03/+1.45233e-03
+1.70081e-01/+1.04426e-01
+1.36862e-03/+1.35328e-03
+4.02465e-02/+3.63988e-02
+1.63737e-05/+3.09393e-05
+7.88361e-01/+7.47196e-01


Michael

[LPR]

Michael

It sounds like you have done all that is possible to avoid having a ESC32 disarm.
I haven't seen a control problem like the one in your video other then if I fly with a very low battery. I use A123 batteries so taking them down to a very low voltage is not a problem. I only do it when I'm flying very close to the ground. It should not have been caused by a Tx glitch so a momentary loss of power is all I can think of.

Are you finding like me that the octo is a very smooth flying craft?

Larry

[chschmid]

Michael
I was checking your LOG file and the motoroutput does not look as it should.
Usually the motoroutput overlapps @ hover more or less. Yours do not.

I think something is not correct with the motor order or mixingtable or direction of motors. Do you have all quadrants right? Please tripplecheck.

When you lift the octo off, did you comand the yaw to the left?

Best
Christof

[Lindbaergh]

Christof

because it was the first take of with the full load, i was a bit 'shy' on the gas. so the octo starts moving on the ground and i have to correct it a bit, its not in the video but in the log you can see it . after it leaves the ground i cant remember if i use yaw, but i dont think so because the octo is moving only a bit forward, not rotating.
i just checked directions again and think they are ok. because i number my props im pretty sure that i have them placed the same way even if i had removed them, but i haven't removed them that days at all. i put only one esc at a time to the FC to make sure the connections are right. here is my layout

upperside / lower side
M03 (Motor 3) / M6 (Motor 6)
M02 (Motor 2) / M5 (Motor 5)
M01 (Motor 1) / M4 (Motor 4)

M10 (Gimbal) / M14 (PPM input)
M09 (Gimbal) / M13 (NC)
M08 (Motor 8) /M12 (NC)
M07 (Motor 7) /M11 (NC)

my copter props and directions

directions

20130320_092456-prop-direction-cut.jpg (569.34 KiB) Viewed 13295 times

my motormix table (taken from Larry)

motormix table

QGC-DBR02-motormixing.jpg (63.74 KiB) Viewed 13295 times

and my parameter file (til now i modified onle throttle factor and gimbal settings)

parameters-130320.txt

parameters

(18.2 KiB) Downloaded 648 times

Larry

the battery maybe a thing i use 5000 turnigy , but i think when it was a weak battery all motors will go down not only one. i missed to put the beeper on once and when the battery was empty the octo just moves down slowly.
aside this little problem i was very pleased with my octo, i made some videos with a contour action cam to test the gimbal settings. the octo is moving like it is on rails, very smooth even on windy conditions i love it .
http://www.youtube.com/watch?v=CxvWewpLnaM
http://www.youtube.com/watch?v=Q8ISbP8zG6g

[chschmid]

Your timers are correct.
Your motordirections are correct.
Your mixing table is correct.
I can not understand, why your motor outputs are so much different. Your CG looks good.
Could it be that one motor is turning the wrong way around? Or a motor is connected to the wrong port? (My hexa even flew this way, although not very good).
I would really investigate that problem, even if she flies well.

Best
Christof