Drone Crashed - Help required to investigate the cause of the crash

Hi
We need help in finding the issue that caused the crash of our quadcopter. The dimensions and configuration of the Quadcopter are as follows:
Dimensions: 1250mm motor to motor
Flight Controller: Cube Black
Firmware: ArduCopter Stable 4.0.7
Batteries: 12S
Motors: U8lite KV 85
ESCs: T-Motors Alpha 60 A HV
Prop Size: 29 inch
Current Sensor: Mauch PL-200
PMU: Mauch 053 BEC

The log Files are as under: https://www.dropbox.com/sh/d7kywxu48iyni2o/AABIeXpYCd33PIfGYMmdmXAaa?dl=0

We have done a lot of flights (more than 20-30 hrs) using the same configuration and no such issue has been experienced before.
In this case the quadcopter took off without showing any sign of instability or any warnings but as soon as it reached a height of 25m it got unstable rolling right at 80-90 degrees thereby falling down. Looking at the logs we see RCout.C4 peaking and RCout.C3 dropping to the bottom whereas bird yaw can also be seen. The ESC and motors have been recovered and have been tested on the ground showing no abnormal behavior. The arms Motor 1 and 2 are broken apart with connections intact whereas Motor 3 and 4 are severely damaged with intact connections; yet, we are able to power on the motors. The cube black is damaged and is not able to get power from the board.
We have tried hard but are unable to make the linkage of the probable reasons causing the crash. As per our deliberations/analysis, following could be the possible reasons:

  1. Motor 4 shut down of Quad in Air
  2. Mechanical failure of Quad in Air i.e. Motor 1 Arm broken
  3. In air, the current sense is observed to drop to zero, can this be related with motor shut down or motors taking no loads or escs failure
  4. Reqd PWMs not been published by Hardware (Possible malfunction of Motor 1-4 PWM Pins of Autopilot)

@Leonardthall @Eosbandi @mboland @hosein_gh @rmackay9 @Harald, @xfacta, @andyp1per, @dkemxr … your expert review will be very much valuable in this regard

Sorry to see your crash.

All I can tell you is you lost motor 4:

You have pretty much covered all the options: PWM output, wiring to ESC signal and power, ESC failure, Motor failure, propeller failure.

3 Likes

The attitude graph shows that drone was flying well until the actual roll is far from desired roll. That suggest external reason that lead to the failure. In the same time RCOU.C3 is low and RCOU.C4 is high which means loss of RCOU.C4

2 Likes

Hi @Leonardthall
Thank you for replying.
Can you kindly give us your input on the behavior of the current as well.
A current surge of 10A can be seen just before the quadcopter becomes unstable. Moreover, there are also currents sinks at three different instances. At 10:57:35.859 the current drops down to 205mA, at 10:57:37.384 it falls down to 142mA and at 10:57:38.566 it sinks to 900mA.
Analyzing the current and PWM plots we can say that the current drops are due to the slowing down of the motors 1 and 2 which can be seen from the lowering of PWM RCOU.C1 and RCOU.C2 at these instances. However, we are unable to link the surge current to any specific event.
Your expertise in this regard will be very much appreciated
Thanks

There is a slight increase in average throttle but you are right it doesn’t look enough to justify the increase in current. This may suggest an increase in load of one of the motors.

Hi.
A second drone of ours got crashed. The configuration was the same. The propulsion system wass comprised of T-motor U8-lite KV 85 motors with Alpha 60A HV ESC. The prop size is 29 inches as was in the previous case and was operating on 12S battery. The log files are attached as:
https://drive.google.com/file/d/1vE_HvoUXalnJBg6adIrdg-wNQBGeGHWi/view?usp=sharing

In this case we lost motor 2 as can be seen by RCOU.C2 peaking and RCOU.C1 dropping to minimum throttle. The drone had completed 10 hours of flight and no abnormality had been observed as such. We are trying our best to find the reason behind the crash but have been unable to link motor failure to a substantial cause.
As per our detailed analysis, following could be the probable reasons:

  1. ESC desync
  2. High ripple voltage causing the ESC to malfunction.
    Finding the root cause is very critical as two of our drones have crashed in almost the same manner.
    Please provide us with your valuable input that can help us in resolving this issue.

@Leonardthall @Eosbandi @mboland @hosein_gh @rmackay9 @Harald @xfacta @andyp1per @dkemxr

1 Like

@Asher @rickyg32 @CBwintertime @Corrado_Steri @ThePara @fnoop @davidae @MHefny

My recommendation is change to a different ESC entirely. APD or something reputable, not another T-Motor ESC.
Keep the ESCs inboard, shorter power wires, longer motor wires. Use extra capacitors if you really need to mount the ESCs close to the motors.

1 Like

Thank you for your swift reply @xfacta we really appreciate it.
We have already adapted the approach of shorter power wires and long phase wires. The ESCs are mounted in the hub.
Have you had any experience with alpha escs? I have heard about Flame ESC that they are quite prone to malfunction. What about the alpha hv?
Also there is a slight ambiguity that concerns me which if you can address.
If you analyze the logs you can see that the drone goes into counterclockwise yaw and undergoes 18 to 19 rotations despite the failure of motor 2 and motor 1 attaining min rpm (both counterclockwise motors) while motor 3 and 4 are performing in normal range. Is it possible for the drone to rotate in this manner with motors 3 and motors 4?

Motors 1 & 2 rotates counterclockwise… that means their yaw effect is clockwise.
It seems that Motor 2 was down because RCOU.C2 is high and because RCOU.C1 is very low as a response to bad M2 so torque in CW is lower compared to torque in CCW of Motors 3 & 4. so quad should rotate in CCW.
I am not sure if this is your case or not. But for me I used bad motors and once they get hot they loose torque, it does not need to completely stop, or disconnect.

1 Like

Yours is just the latest in a history of issues with regards to T-Motor ESC’s. Take Shawn’s advice and acquire some others.

1 Like

Check your motor wires. How are they connected to the ESC? do you use a connector or directly soldered to ESC? We’ve had 3 U8 lites fail on us in flight, thankfully no major crashes. Root cause was traced back to motor wire fracture, specifically at the point where the enameled portion ends and the factory tinned portion of the wire begins. Our motor wires were not secured down and stuck out past the arm, meaning they were susceptible to being knocked during transport.

@1afridi Please do not spam people like this.

Thanks for your reply Dave.
Can you please tell me why would the ESC suddenly stop sending signals. Could it be hardware failure, firware issue of the ESC, ESC desyncing or ripple voltage related problem?

Thank you @MHefny for calrifying my concept about the yaw.
The motors were new as i have mentioned they had done 10 hours of flight time and they were in good condition having no such observations about them.

I am disappointed that you feel you are entitled to tag 17 people on your post. The discuss forum is where people get FREE help from members of the community. It is obvious that you do not respect the time and effort of those people.

2 Likes

Crash all your fleet so that you’ll be very sure.

I do not feel entitled. I am just asking for help regarding the issue we are facing.
I respect all members of this community and i am very grateful to everyone that have given their input.
All the people i have tagged i consider experts and seek their guidance in this issue.
I am sorry if you feel disappointed.

In future show that respect by not spamming the communities’ “experts”.

Have you tested the C2 ESC with a different motor after the crash ? Is it still working ?

I’ve hit the sweet-spot with Alpha HVs at 24 inch and 135 kV. Bigger props are harder to handle. Spooling them up needs more torque and thus higher currents.