Page 3 of 5

Re: DIMU fail or FW bug ?

PostPosted: Wed Apr 22, 2015 11:48 am
by Crizz
nobody any idea ? :(

Re: DIMU fail or FW bug ?

PostPosted: Wed Apr 22, 2015 10:34 pm
by teramax
Just to be sure which part of the system makes troubles could you remove the DIMU, flash some AIMU firmware and test this once again.

Re: DIMU fail or FW bug ?

PostPosted: Thu Apr 23, 2015 7:56 am
by Crizz
like i already wrote, this AQ6 never had a calibso and i would like to spare all the work and let my pc run all night long on calculations just to see if its the DIMU itself - thats the reason why i bought a DIMU.
Normally the proferssionals should be able to see if the files contains abnormal values.

To complete, i attach the onboard params as file here also.

Re: DIMU fail or FW bug ?

PostPosted: Thu Apr 23, 2015 8:21 am
by chschmid
Hi Crizz

Your params show that no Mag cal (onboard ) had been done.
Could you do that and check again.

Cheers
Christof

Re: DIMU fail or FW bug ?

PostPosted: Thu Apr 23, 2015 8:44 am
by teramax
Crizz, theres no need for Calypso and freezing, i did not whant you to fly and go for a mission with the AIMU, just a short hover and see how it behaves. The point of the whole Calypso thing is to get the sensors propper alligned but for a quick flight theres absolutly no need for.
Just wanted to be sure if the problem is on the DIMU or an the AQ6 board.

I did a quick LOG lookup yesterday with Tilman and we discovered also strange behaviour in MOT_Pitch and IMU_RateY graphs. With the attached params file theres nothing wrong beside the missing MAG Cal but thats not the reason for this strange behaviour.

Re: DIMU fail or FW bug ?

PostPosted: Thu Apr 23, 2015 9:38 am
by Crizz
okay, understand - will give it a try at the weekend and see if there´s a difference. Be back l8r with AIMU-logs. Maybe i get a little time in the evening after work.

Re: DIMU fail or FW bug ?

PostPosted: Thu Apr 23, 2015 7:41 pm
by Crizz
okay, took a little time and installed the same FW for A-IMU. Checked Radiosetup and tried reaction of controller.

File AIMU_Disarmed : check in disarmed mode for sensor reaktion and HUD

File AIMU_MotRunning : same with running motors ( little bit over iddle ). Reaction in HUD seemd okay and it feeled better in hand, like controller tries compensate offset

File Handcheck : with higher rpm for checking sterering commands. Felt like there was NO reaction onroll left (or was it right ? Should make note next time...) and no reaction on nick-backwards (nick forward comes as fast as the stick moves, no problem ). On yaw i didnt felt anything but heared changing rpm on motors, maybe it was to blame on the 5x3" i mounted for the "handheld-test".

Now i´m a little confused - is there any basic-calibration i have forgotten that caused the fail on one side of roll and nick (pitch) ? Or is my AQ6 wrecked ?

Re: DIMU fail or FW bug ?

PostPosted: Thu Apr 23, 2015 9:14 pm
by Crizz
update :

tested again in basement because i was pretty unsure about reaction on steering commands. Seems all axles are correct, will give it a try in the field tomorrow, but one thing is sure :

compensation of movements are fine on all axles, even with running motors - and HUD shows in time, but not with D-IMU. So guess D-IMU is to be blamed for the problem. Can´t load the logfile up here, size is over 10 mb as ZIP, but guess you professionals will see the same in the smaller logs i posted before.

Re: DIMU fail or FW bug ?

PostPosted: Thu Apr 23, 2015 9:40 pm
by Crizz
Question :

i´m right back from the basement, mounted the props for testing tomorrow. As i pulled down the D-IMU i saw that "R3"-component is not tipped (!) .

Is this required and a tipping-failure or isnt it required ? May this causes my problems when DIMU is used ?

Forgive me for all that questions, but to be honest : meanwhile it is a 2-years-project and still no maiden, so maybe you guys can understand that i´m sort of pissed (sorry) at the moment, season starts and still no proper running AQ in sight.

Hope we can get rid of my problems instead of trashing the hardware, guess there even is no waranty at the AQ6, maybe on the DIMU if any luck......

Re: DIMU fail or FW bug ?

PostPosted: Thu Apr 23, 2015 10:33 pm
by teramax
R3 is empty also on my boards, should be ok so.
Be aware that flying without calibration could and will result in drifting and some strange behaviour so fly close and not to high ;-)

LOG seems OK to me, nothing wrong for an uncalibrated AIMU.
Could you take a magnifying glass and look for the solder job on the AQ6 board where the DIMU socket is located.
Beside that without a second DIMU for testing i´m running out on ideas, sorry.