Desperately need help analysing log files!!

Info and discussion about the autonomous Micro FC, suitable for nanos, micros and mini multirotors

Re: Desperately need help analysing log files!!

Postby LPR » Tue Aug 30, 2016 9:31 pm

Patscherpofel

Like Adam says the problem with the extra weight above the M4 + N4 is that it will only cause the M4 to be moved out of its normal position when the frame is being rapidly tilted.

Anything that may cause the M4 to move a very small amount out of its position relative to the quads frame may cause problems.

Larry
LPR
 
Posts: 1323
Joined: Tue Jun 26, 2012 1:29 pm
Location: MN, USA

Re: Desperately need help analysing log files!!

Postby chschmid » Wed Aug 31, 2016 5:00 am

Hi Patscherpofel

I had a look at your crash log.
It seems that at 97200 the system is fine. At 97210 the motors 5&8 start to spool up, 6&7 to slow down. After that the system starts to tilt. That is where the disaster begins. The FC does nothing unexpected as far as I can see.

It seems that you have been in a steady hover at crash time. Your alt is a bit jumpy but fine. Why are your radio sticks not centered? What flight mode have you been in? Please describe your radio channels and functions.

Cheers
Christof
chschmid
 
Posts: 1800
Joined: Wed Jul 11, 2012 7:41 pm
Location: Herrliberg, Switzerland

Re: Desperately need help analysing log files!!

Postby Patscherpofel » Wed Aug 31, 2016 6:45 am

the channels are in order of what AQ suggests in default
this crash happened not in a steady hover ...I was doing a some slow flying I think thats why the sticks are not centered
but the problem occurs on all flight modes mode ( but not rate mode so far but I havent tested this much to be sure about this)

a similarity in all thes "scenarioas"I had (must be around 50 times now)

-its allways easy flying or hover
-its mosty yaw that freaks out with short pulses...you can mostly hear 4-8 time the motor jump rapidly..then it dissapears
-50% of the time the quad will loose its level
-if I keep on flying there is allways at least 1-2min gap till it happens again...never closer together
it feels a bit like there is some sort of filter closing out and then resets somehow

what about electronical outer influences from esc or the rx? could this lead to a problem like this?
Patscherpofel
 
Posts: 91
Joined: Fri Feb 12, 2016 10:03 pm

Re: Desperately need help analysing log files!!

Postby chschmid » Wed Aug 31, 2016 11:50 am

I was thinking about this as well.
I had this issue once using "ungeschirmte" PWM lines.
You can use a microfone cable with shield and put one side to PWM gnd.
Usually a PWM malfunction is seen past a certain current if the PWM line is close by this powerline..

I am not sure what we see here. What ESC's do you use and is your cap capacity big enough?
Witch motor Port goes to witch motor ?

Your log does not look like flying arround except your sticks. And I see quite a stick deflection.
Can you post a hover log in PH.
chschmid
 
Posts: 1800
Joined: Wed Jul 11, 2012 7:41 pm
Location: Herrliberg, Switzerland

Re: Desperately need help analysing log files!!

Postby Patscherpofel » Wed Aug 31, 2016 1:14 pm

there is some position hold in both logs..really solid...as uausal with AQ
I was testing also aggresive flying to test pids and see if I get clipping of motors ,so yes this is not only easy hovering but there is some in there as well
but I am 99% sure these problems are not related to fast movements or stick input , they appear randomly but mostly when i do easy stuff I would say

motor port 5-8 are used, 5 is front/left and so on

its an afro hv 20A esc, flashed with blheli, damped light activated...huge caps on them
but I remember this happened on differnet seutps allready...also 4s and smaller motors
and this frame and motor combo flies rock solid with a different FC
Patscherpofel
 
Posts: 91
Joined: Fri Feb 12, 2016 10:03 pm

Re: Desperately need help analysing log files!!

Postby Patscherpofel » Wed Aug 31, 2016 6:38 pm

so I did a test
I dismounted the gps antenna and shield completely
I realized I put the fairly heavy taohglas gps on a double sided tape which has a small surface only, meaning the gps flexes a lot...not good, yes

I did a 10min flight and no problemes you can camper the log if you want
hard to say for me...is it cleaner? I mean besides those 3-4 "freak outs" the crash logs has
https://we.tl/7y94gRTODx


I took these screenshots also
are these shifts on the accZ normal?...they are linear to motor throttle...quite aggressive on the sticks yes




vibes.png
Attachments
moutor output.png
Patscherpofel
 
Posts: 91
Joined: Fri Feb 12, 2016 10:03 pm

Re: Desperately need help analysing log files!!

Postby sandmen » Wed Aug 31, 2016 10:21 pm

It looks really not too bad, but without any scaling in the pic's... I didn't know.

Looks like, I need to register for "wetransfer", sorry didn't want to do that.
sandmen
 
Posts: 997
Joined: Fri Jun 22, 2012 7:25 am

Re: Desperately need help analysing log files!!

Postby Patscherpofel » Wed Aug 31, 2016 10:46 pm

you dont need to register...its a simple download normally
Patscherpofel
 
Posts: 91
Joined: Fri Feb 12, 2016 10:03 pm

Re: Desperately need help analysing log files!!

Postby Patscherpofel » Thu Sep 01, 2016 1:32 pm

second flight without problems...this time with gps, but no fixed to the m4 tower
could this be it? just in theory, can something this small make a frame flip mid air without stick input?
Patscherpofel
 
Posts: 91
Joined: Fri Feb 12, 2016 10:03 pm

Re: Desperately need help analysing log files!!

Postby Patscherpofel » Thu Sep 01, 2016 2:38 pm

and one more flight
some screenshots again

vibes 2.png
vibes

10min flight.png
total flight

pos hold.png
pos hold
Patscherpofel
 
Posts: 91
Joined: Fri Feb 12, 2016 10:03 pm

PreviousNext

Return to AutoQuad M4 Micro Controller

Who is online

Users browsing this forum: No registered users and 8 guests

cron