Hi Guys,

In two occasions I have experienced an issue when increasing throttle. I listened to a pop sound and all of a sudden the whole quad came down.  When I turned it on back on my bench, all motors were working.

Is this a desync?  I have noticed in the two times it has happened, the humidity and heat were very high where I fly. I also noticed that when varying the throttle, the quad wobbled a little. Like there was a voltage spike.  

The motors are Iflight Xing2 2506 1650kv
FC is a Mamba Pro with 65 amps ESC
Props are Genfam Flash 60433
I have put a capacitor and I am using Caddx Nebula Pro.

Any suggestion on what could be happening is greatly appreciated.

Thanks.

Did you hear the FC/ESC reboot sounds?

I would not have thought that a desync would have a popping sound.

A popping sound would indicate an electrical fault or short. Perhaps a voltage spike.

Did you inspect the board to see if any of the FET's on the ESC may have caused this.

Perhaps on the next flight you are better off recording some DVR (even though we wont get any sound) and Blackbox logs.

probably a bell bended? are there differences in gap between stator and magnets?
the only thing i can mention is an esc issue or a physical blocked motor.

probably a video can bring some more infos?

Hi Guys,  The issue happened Today again when flying a QAVS with a TMOTOR F7 mini and TMOTOR motors Black Bird 2207 1950KV.   I hope you can see what the issue is.  I haven't seen any fets burned or bent motor bells.  This is happening also with my QAVR2 as I reported before. The only thing in common between the two copters is CROSSFIRE.  It seems the forum doesn't let me attach .BBL files.  Do I need to convert them to something else?  I am attaching my cli diff all though.

Any help will be greatly appreciated.

Thank you!

Crossfire will simply pass your stick commands. This is probably not the cause.

Even if you can’t attach a .bbl file you could share a dvr video so we can see what happens and when.

I don’t see anything wrong with your dump. Although you do have airmode off but that should not cause motors to pop.

What firmware are you running on your ESC? What version?

Also did you try the stock Betaflight pid tune?

(19-Sep-2021, 11:35 PM)kafie1980 Wrote: Crossfire will simply pass your stick commands. This is probably not the cause.

Even if you can’t attach a .bbl file you could share a dvr video so we can see what happens and when.

I don’t see anything wrong with your dump. Although you do have airmode off but that should not cause motors to pop.

What firmware are you running on your ESC? What version?

Also did you try the stock Betaflight pid tune?

Hi Kafie1980,

I am running both quads with stock PIDS.  Didn't touch them. Lamentably I haven't record the DVR for this occurrence. I will make sure it is enabled for next time. Is there a way I can hang the BBL files in this forum?

The ESC firmware version is 32.8 for both quads. I did notice I have BAD frames when I plug the ESCs to BLHELI_32.  Is this a signal they are damaged?

BLHeli_32 has had some issues lately on 32.8. There were issues caused by the variable PWM function recently added which several users had reported: https://github.com/bitdump/BLHeli/issues/546

Perhaps you can test 32.7. There is also a pre-release 32.8.2 out but is not officially released yet.

You can upload the .bbl log files to https://filebin.net/and paste the link here.

Hi Kafie1980,

Here is the link for the uploaded files:

https://filebin.net/aqiwi4ja7bhrnz3d

Let me know if you notice anything.

Thanks a lot for your help!

You didn't say which of the two quads the blackbox log is for but at 02:15 motor 3 instantaneously rises to 100% throttle and motor immediately 2 drops to 0% throttle while your stick commands are relatively static. Looking at the gyro trace the quad starts to flip out diagonally with motor 3 arm dipping down (even though it is at 100% throttle) and motor 1 arm rising up (even though it is at 0% throttle). Keep in mind that this is the flight controller commanding 100% throttle on motor 3, not what that motor is necessarily doing (there is no feedback from the motor to the FC), so the fact that the fight controller is commanding 100% throttle on motor 3 but that arm is still continuing to dip down means that the motor is no longer spinning and the FC is trying to command 100% throttle to get that arm raise back up but it's not working, thus the eventual flip out / death roll. This sort of behaviour indicates a desync on motor 3 of this particular quad.

The problem can't be an issue with Crossfire otherwise it would show up in the RC command graph which it doesn't as the graph lines for those are relatively smooth and you only see the throttle command drop as you drop the throttle stick yourself when you obviously then realise that your quad is flipping out.

I can't comment on the other quad without also seeing a backbox log from that. Maybe you have a similar (coincidental) motor desync on that quad as well.

EDIT: I just noticed that this log is for the QAVS.

I just noticed from the header information in your blackbox log that the one you posted is for your QAVS which you say is running T-Motor Blackbird motors. Be aware that there is a known issue with some batches of the Blackbird V2 motors whereby they have mismatched magnets, so if your motors are from an affected batch then that is likely what your issue is on the QAVS quad. See the following thread for more details...

https://intofpv.com/t-t-motor-users-beware

As for the issue on your QAVR2, that one is using different motors and a different stack, so we would also need to see a blackbox log of the QAVR2 in order to try and figure out what is up with that other quad.

I just plugged in my personal laptop was I am unable to run Blackbox Explorer on my office laptop and Snow has done the investigative work.

Add to what he has pointed out and suggested so far, can you also provide a snapshot of your BLHeli_32 settings. What I am interested in the Rampup power, Motor timing and Demag compensation settings.

Hi Snow,

I took a deeper look at motor 3 in the QAVS. See the attached picture.  It feels like the magnets are burned, but not sure.  In regards to the QAVR2, I don't have any logs of the problem yet. I will check also its motors to see if something similar is happening.

(20-Sep-2021, 04:12 PM)lleston2 Wrote: I took a deeper look at motor 3 in the QAVS. See the attached picture.  It feels like the magnets are burned, but not sure.

That is just balancing "glue". They apply that where needed to balance the motor, just like a tyre fitting shop who put a new tyre on your car fit small balancing weights to prevent wobbles and vibrations when the wheel is spinning.

Hi Kafie,

I am attaching the ESC Overview.  Let me know if you can see it OK.

Some suggestions I have with regards to your BLHeli_32 settings. 

First, you seem to have setup Variable PWM frequency. Your PWM Frequency Low is fairly low at 16kHz. I will advise you to set both PWM Frequency Low and High to 24kHz or 48kHz to disable variable pwm frequency as there is a bug in 32.8 as I mentioned earlier.

Usually for motor desyncs the recommendation is also to try Demag Compensation set to High to see if this helps. 

Finally, if the above still causes motor desyncs then try increasing the Motor Timing to 23 and go no further.

You might want to experiment with these one at a time which can be a pain but if you have an android phone or a laptop then you can experiment with these in the field.

But for the most part do the variable pwm frequency and demag compensation first and if that doesn't help then try motor timing. 

Also take short test flights and keep checking your motors for heat.