AQ board after repair going crazy, dangerous accident

[LordPikaChu]

Okay, tested with older firmware (r440, was working good with this firmware before) and the same happens - the board randomly hangs mid-flight.

[Florian75]

I think in this case you have to contact Jussi. If the old (tested) Software shows the same problem it can be a Hardware Problem. I have no more ideas.

Florian

[JussiH]

I will send you a new board to test with. Pls hang on to the current board for now.

I am still not convinced that this is not a Radio problem, but another board will confirm or rule it out. Pls for now post some logs of this phenomenon.

[LordPikaChu]

I will send you a new board to test with. Pls hang on to the current board for now.

I am still not convinced that this is not a Radio problem, but another board will confirm or rule it out. Pls for now post some logs of this phenomenon.

Thanks for your reply Jussi, I will be waiting for the new board, then. You already have my address.

Regarding logs - the thing is that there are no logs. I mean that the board is logging fine, until it hangs. Please watch my video again and focus on the telemetry on my computer's screen. Once the telemetry stops - this is the moment when the board freezes and stops logging. The log is just cut off at this very moment. You can also see that my radio is working fine - sometimes the green LED responds to arm/disarm commands, but this is the only thing that seems to work. This seems pretty random, as sometimes the board will not respond at all.

[JussiH]

I still cannot shake the thought that this is something to do with the radio somehow. In the last clip it seems to happen exactly at the time you lower the stick.

In the first clip the board is logging. Are you saying you got no data at all or that the data clipped when the board freezed? If there is any data, I would like to see it.

[LordPikaChu]

Yes, the log is clipped when the board freezes.

I will send you an example log later today.

In the last clip it was really a coincidence, that's the only time that it happened. I just attached it because it was a good demonstration of the board hanging without being subject to vibrations. I have done the test like 30 or 40 times already and the board freezes without me touching the radio every time, except that one in the video.

I can also test with Turnigy 9X if you insist, I just don't yet know how to setup this radio with AQ. Any suggestions?

[LordPikaChu]

Okay, so, it looks like I cannot test with Turnigy 9X since it doesn't output PPM signal...

Today it was a bit difficult to cause the board to freeze. It only bugged out one or two times out of ten. I was able to record a log where it bugs out really quickly after powering up - please see the attachment.

Also, how would it even be possible for my radio to cause the board to freeze?

Now, I have another proof that my radio indeed works fine. When I was trying to record a log for you, a weird thing happened that has never happened before - the board bugged out, that is it didn't respond to external inputs (moving the copter), but the telemetry was still working! I could also turn the motors on and off, but they wouldn't go above idle RPM, no matter how high the throttle was set or how I would move the copter / any other radio stick. I have recorded this incident with my phone and you can clearly see in the QGC that my radio is working fine! Please listen carefully to the motor sound - they don't react to anything at all. They could only be off or run at idle speed.



If this doesn't prove to you that everything's okay with my radio - I don't know what does.

EDIT: Forgot to attach the LOG file:

AQ malfinction.LOG

(4.63 MiB) Downloaded 83 times

[bluuu]

its a full log from video ?

this shows half throttle (to the end)

[Max]

OK you're gonna hate having to do this again, but if no one finds anything in the logs...

Can you get it to crash while you have the full Diagnostic Telemetry running? Set it to 25Hz or so, and enable Default and Stacks checkboxes (and Gimbal if you're using any gimbal/passthrough features). Then watch the Values Grid screen.

I'm specifically looking for "NaN" values after the "freeze" event, or stack(s) size going to zero (if it starts out at zero to begin with, that's fine). But post the whole screen. The fact that you still have comms during/after the crash should help a lot in diagnosis.

Thanks,
-Max

[LordPikaChu]

its a full log from video ?

this shows half throttle (to the end)

aq_thr.jpg

No, it is NOT the log from the video.

The log that is attached to the previous post was the shortest I was able to get today. I don't have a log from the video accident.

This log shows the standard behavior when the board is freezing - please read all my previous posts. The logging just stopped when the board froze - I was moving throttle and other sticks after the board has crashed, but as you can see it is not present in the log.

OK you're gonna hate having to do this again, but if no one finds anything in the logs...

Can you get it to crash while you have the full Diagnostic Telemetry running? Set it to 25Hz or so, and enable Default and Stacks checkboxes (and Gimbal if you're using any gimbal/passthrough features). Then watch the Values Grid screen.

I'm specifically looking for "NaN" values after the "freeze" event, or stack(s) size going to zero (if it starts out at zero to begin with, that's fine). But post the whole screen. The fact that you still have comms during/after the crash should help a lot in diagnosis.

Thanks,
-Max

Thanks Max, I will post this tomorrow. Do you want me to record a video of the screen? I have been unlucky in getting the board to freeze today (on previous days I could reproduce the behavior every time, but not today) - it only bugged out 2/10 times on average, so it might take some time to catch it on video. I will try my best, though.