Bad crash after radio loss.

[Johan_vV]

Was out flying to day and suffered from a really bad crash.

This was my 5/6 time flying with the AutoQuad and i only had one minor crash during the first flight when entering Postition hold (I had not done any PID tuning at all and it over adjusted and fell to the ground from a low altitude and nothing was damaged).

After adjusting the PID a bit it was quite stable and RTH etc worked without any issues.

Yesterday i tried to tweak the P-value a bit and changed it from 20 to 35 since it seemed to react a bit better to stick input when having it attached to a rope in my apartment.

Today when i went out and flying it was quite windy. I powered it up and Telemetry and all was working. I was able to arm the FC and perform a lift of. Directly after the lift of it felt like i had 0 control over it and it was about to fly away from me and then it stopped and stayed in a hover position. After some time i realised that it must have lost the Radio contact since it didn't react to anything i was doing on my TX. I tried flipping the swith to manual/position hold but i was not able to control it in any way. I then walked up to it and grabbed it with my hand while it was flying and then unplugged the battery.

I was then scrathing my head a bit and could not understand why the PPM didn't work any longer. I was not able to arm the FC anylonger and AQGC didn't show any RC input either.

I unplugged the battery and plugged it back once more and then all of a suddenly it started working. I could no longer reproduce the problem even if i unplugged the battery and connected it again so I was thinking that i should take try flying once more.

The second flight the behavior was the same. I was able to luanch but then it lost contact again a few seconds after the lift off and this time it didn't initiate the fail safe and it crashed really really hard into the ground and broke allmost everything. All props, three motor mounts, the GPS Sheild etc =(

The radio loss is probably related to my flashed Hitec optima 6 which stopped working for some odd reason which i don't know at the moment and i don't think that it was the AQ:s fault. But i have a couple of questions regarding the fail safe and behaviour of the FC in these scenarios.

1: During my initial PPM tests, it took around two seconds for the AQ to detect the radio loss and initiate the Fail safe. Is this normal behaviour and can this be adjusted so that fail safe occurs faster? In my case the lift of and crash probably occured within two seconds so the Fail safe was probably not initated at all.

2: During all my crashes I noticed that the AQ tries to spin the motors really fast even after it has hit the ground in Position Hold/fail safe mode. The first time i noticed this was during my first flight with the AQ where it over adjusted it self when entering Position Hold and fell into the high grass. I the lowered the throttle to zero and the enignes/ESC was still screaming until i flipped the switch to Manual.

Similar behaviour today when the motors was running with a really high throttle after it crashed up side down and since i had no radio control i wasn't able to turn them of. However, i suppose that the Fail Safe should have kicked in since much more that two seconds had passed since it lost radio link until i managed to run to it and unplug the battery.

I must also say that previusly when hovering the quad in position hold and putting the throttle to minimum the props almost stopped spinning as soon as the quad touched the ground and "landed". - but this spin-down does not occur when crashing.

I did have my SD card plugged in during both flights so I do have the logs if someone is interested.

Thanks!

[kinderkram]

Dammit - sorry for your loss, Johan!

Sounds like the failsafe kicked in properly on the first radio loss so it performed a hover in PH.
Second time sounds really strange...
I don't know of anyone else flying with a flashed Hitec Rx in PPM mode.
I'd investigate on that matter. Can you point me to where to download the new firmware for the Optima6?
I have one here and could try to use it on my AQ.

P.S.: please post your log files for an inspection.

[GoFaster]

Johan -

Sorry about the crash

Please do post your logs. How do you have failsafe set for the Optima RX?

- Felix

[Johan_vV]

Thanks for your reply.

I found a guy (Libor) who built a PPM firmware for the Optima RX. He was very helpful and answered on all my questions. His site can be found here: http://tomsik.eu/optima_ppm

Since he at that time couldn't provide a HEX file and only the Source code i later on found a thread on RC groups where another person also had found Libors web site and then decided to make his own PPM firmware which is the one i ended up using: http://www.rcgroups.com/forums/showthre ... st22674881

This firmware has been working really good and when i during my initial tests on the workbench powered of my TX then i could see that the AQ notices radio loss and initiates fail safe which it also logged in the Mav Link and once i turned on my TX then it was working again. This means that i have not set up the fail-safe in the Optima it self since it is programmed to stop sending the PPM signal when the TX-RX link goes down. I have successfully flown the AQ with this firmware several times.

At the moment I can't get my PPM output working again so i suspect a issue with my Optima and not with AQ. But im not sure yet since i haven't got anything else then the AQ that can handle PPM input which i could use for testing.

However, now to my AQ concerns:
1. Would it be possible to speed up the fail-safe event "triggering" so that it falls in to this mode faster compared to now?

2. would it be possible to have the AQ stop spinning the motors when it crashes? Bump detection works when i previusly tested landing it in position hold mode, but it does not work when it "lands" up side down =)

I gladly upload the Logs. I think that the crash log is the most interesting for you experts so I attach it to here.

[GoFaster]

Johan -

Can you please clarify:
1. The custom Optima RX PPM firmware stops sending PPM frames including PPM guard pulse upon radio loss.
2. What AQ version are you running and is it self compiled or downloaded from here? I'm curious because as far as I can tell, AQ does not know about radio signal quality, hence trigger the internal failsafe/signal loss routine when in PPM mode.
3. Is the log from your second flight today - sequence as described in the first post?

[Lyagukh]

[Lyagukh]

[Johan_vV]

[GoFaster]

@Igor -
Thanks for the explanation, I see it now that I took a closer look at radio.c and supervisor.c. If there is no update from the radio for 50ms, a value of -1 is set as input to the Signal Quality moving-average algorithm. With +1 being 100% signal quality and -1 being 0%. Something like that?

Personally I would prefer Radio_Loss_Stage_2 to rise to a settable altitude (if it's not already at or above), then RTH.
I would like Battery_Low_Stage_2 to be AutoLand.

@Johan -

I am no expert but I loaded up your log and here is my interpretation:

In the first part of the log file,
You power up everything.
Arm, then immediately disarm.
Then you shut off RX or TX. (then move the copter?)

Zooming in to the last part:
You turn on RX or TX and signal quality ramps back up.
Arm and then apply some throttle (light blue). The red line is MOT_THROTTLE.
Almost immediately, you lose signal. However AQ is still using the last throttle value.
Signal comes back for a little bit and sends the new/higher throttle setting.
Just before signal goes away again, it appears that AUX2 was set to RTH. Did you do that?
Signal goes away again and AQ continues to use the higher throttle setting.
Now what's interesting is after signal is lost, it appears the AQ saw simultaneous RC commands to switch from Manual to PosHold AND throttle to max setting. I know all this happened very quickly but perhaps you jammed the throttle to full and flipped the switch while signal came back for a split second?
The motors go to full power because the throttle setting is at max.
Then if I am correct, we see from IMU_ACC_Z that it's flipping upside down and bounced a few times

To me it appears that your RX rebooted as the motors were spooling up. Either a lose connection, or more likely a brownout from voltage sag? How is it being powered?

[Lyagukh]