Desperately need help analysing log files!!

[whizz101]

[whizz101]

Fw 1881 also works ok. No spikes.

I did have a very aggressive yaw movement that i guess was it correcting from gps, but i have never seen it do it so aggressively!

Anyway's i hope that you can now pinpoint the problem. You have been a big help to me in understanding the problem and fixing it! Thanks very much...

What firmware would you recommend running. I like the extras you have added in you next branch, but am a bit weary of new FW's since this issue with the hott FW. I guess the FW's 449 or 442 have been flown and tested a lot more.

Or perhaps i'll just stick with 1881 since it's already loaded..

Thanks again,

Tristan

[Max]

That's good news about the spikes being gone. But if they're gone in 1881, I don't know why they would be back in the later versions. 1881 includes a fix which doesn't init the 2nd serial port when there is no comm protocol defined on it (unlike the original 1873). So that's still a mystery.

I appreciate your patience in tracking this down.

Don't know about the aggressive yaw... I see your setting for max. yaw rate is/was only 15 */sec. Maybe it got changed in params? That should be pretty slow.

I guess the big question now is if you get the same out-of-control problem or not.

Cheers,
-Max

[whizz101]

well, just to confirm that they existed, i went back to FW 1897. And they are exactly as before. loads of spikes!

[Max]

Yea that's a bit strange. If you're up for it, here's a version of 1897 but with the same change which I applied to the last TEST version (removed support for 2nd serial port).

You could also try the very latest version... The only reason I suggest that is that it does have some stability improvements in terms of memory management, and also some fixes for Nav with PID (which didn't occur with Quatos).

ftp://ftp.autoquad.org/3/334693_22529/a ... .6-quatos/
https://github.com/mpaperno/aq_flight_c ... CHANGES.md


-Max

[whizz101]

Fw 1897 TEST is ok....1 spike!

log here

https://www.dropbox.com/s/yckaml32ho7n4 ... t.LOG?dl=0
So how would you rate this firmware....would you "bet" on it so to speak!

Thanks again

Tristan

[Max]

I guess I'd fly this version some more and see how it goes. The spikes might be due to some other variable at this point... it's not conclusive.

I would not bet any more than I could afford to loose. On pretty much anything.

But seriously, there is no way I can promise anything and I'm not going to speculate. There are way too many variables involved and there's no way I, or anyone, can test all possibilities.

Cheers,
-Max

[whizz101]

Disaster!!

Whilst doing a second flight with FW 1897TEST, I had exacly the same problem as i original started this thread..........

I was doing a test flight, just keeping it static, 50cm above ground in manual. Was flying perfect when suddenly it tipped to the left and tryed to fly off. I immeadetly cut throttle and had a small crash landing.

you can see the motor outputs go crazy again....

log here
https://www.dropbox.com/s/9rcrb3raki01i ... h.LOG?dl=0

[whizz101]

Well, I'm not sure how to continue! I have now had the exact same problem with and without the hott extension FW, so we can rule that out. It happened in FW 445 hott and in Max's new builds. It has also happened on two different M4's and two different adapter boards.

I guess this all points to the problem lying else where, but what should i be looking for. All these problems started when i changed the FC from AQ6 to M4. I have done hundreds of flights with this exact set up with a AQ6 and only ever had one problem, which was associated with vibrations, i have gone back and looked at the log of that incident and it is not the same as i am suffering now.

Could this be associated with the "hacking" of ports 1-4 on the M4. There does seem to be a difference between the ports 5-8 on the expansion board and 1-4 that i have hacked. For example when i power up the system the two esc's connected to the ports 3 and 4 initialise a few seconds before the esc's connected to ports 5-8 on the adapter board. In fact say 20% of the time when i connect power to the system, the esc's connected to ports 5-8 don't initialise at all and i have to power down and try again....

Or, could this be caused by a loose connection or bad wiring? I am tempted to take the whole machine to parts, inspect all the cabling and rebuild from scratch. But that is a big job! And i really doubt that the problem lies there.

Any one with any ideas i am all ears!!

[whizz101]

Well, I was going to order a new AQ6, but they are not available! not in any of the 3 distributors!

@ Jussi- Are there any of the big M4 prototype adapter boards available, the ones with osd ect and 12pmw outputs? perhaps this could help in my problem!

I am going to load FW 442 and see if the problem still exists.