Hi community,
Sadly my quad crashed after 21 minutes of flying, everything seemed normal before quad suddenly fell from the sky. I have been flying the quad for a total of 8 - 10 hours with more or less the same setup without any issues before. I was flying with altitude hold around 40m in more or less at one position when the quad dropped from the sky. From what I see in the log a brownout occurred at altitude 11.73m, but from what I can understand the issue happen before the brownout, battery status was at 32%. Battery did detach from the power module when the quad hit the ground. I have newer seen Roll failure in any log before this crash (Roll (120.92, line 302416) > maximum lean angle (45.00)) I suspect mechanical failure after reviewing the log. Weather condition was a windy from time to time.
Known issues, vibration level has always been is a little high, I have reviewed some older logs looking at vibration level, more or less the same as in this flight. After adding a secondary GPS I started to receive GPS glitches, but the GPS-Sats (15 - 20) and GPS-Hdop (1.08 - 1.63) is always acceptable. Slow loop lines has always been a problem.
Video of the crash, it makes a strange sound right before it goes down:
https://youtu.be/vr7nOH0is2Q
Log files (dataflash log):
https://drive.google.com/folderview?id=0BzWoDXwJ2f-kfjRJcXlPekxvX1FKUTV4emxPVzQwNE1ORHFqV0wtM0g2RlBQN0xQWG9mTHc&usp=sharing
Hardware information:
- Pixhawk PX4 2.4.6 (Firmware Version V3.2.1)
- UBLOX MIni NEO-M8N GPS GNSS + HMC5983 Compass Triple Band ANT (Primary GPS)
- UBLOX Micro MAX-M8Q GPS GNSS + HMC5983 +SAW +LNA+ Triple Band ANT (Secondary GPS)
- Power Module v5 - 90A
- Tattu 16000mAh 22.2V 15~30C 6S 1P Lipo Battery Pack (separated battery pack/circuit for fpv gear)
See the included txt for full specification list.
Log auto analyses output:
Test: Autotune = UNKNOWN - No ATUN log data
Test: Balance/Twist = GOOD -
Test: Brownout = FAIL - Truncated Log? Ends while armed at altitude 11.73m
Test: Compass = GOOD - mag_field interference within limits (9.07%) Max mag field length (597.11) > recommended (550.00)
Test: Dupe Log Data = GOOD -
Test: Empty = GOOD -
Test: Event/Failsafe = FAIL - ERR found: GPS_GLITCH
Test: GPS = FAIL - GPS glitch errors found (1) Min satellites: 15, Max HDop: 1.53
Test: IMU Mismatch = GOOD - (Mismatch: 0.21, WARN: 0.75, FAIL: 1.50)
Test: Parameters = GOOD -
Test: PM = FAIL - 126 slow loop lines found, max 16.00% on line 730
Test: Pitch/Roll = FAIL - Roll (120.92, line 302416) > maximum lean angle (45.00)
Test: Thrust = GOOD -
Test: VCC = GOOD -
I would truly appreciate any help that could shed some light on what caused the quad to drop from the sky. As explained, this quad has been reliable for many hours of flying before the crash, so I see no point in starting any rebuild since I do not know what caused the crash.
//Seiern
Replies
Hi Ivan, thanks for taking the time to view my flight log, may i know is that possible that my battery had failed me, it seem like the amp suddenly drop to zero.
this one is simple, mechanical issue likely motor/esc failure, look the the outputs going crazy after 6:42:30 as FC tries to compensate for a motor failure, also note the large divertion between requested roll and actual roll, most definitely looks like intermittent motor failure so you either lose a magnet on the motor, loose wires, ESC....thats where the problem is....the Pix did its job best it could nothing wrong anywhere else
Can anyone help me to analyze my flight log, it crashed suddenly while hovering with position hold mode. Flight data as below:
https://logs.px4.io/plot_app?log=88bfc93e-6092-4dbc-8e58-e74d74615a7c
yea I had same thing happen to mine like yours, this is definitely a motor or ESC issue....I didnt see any brownout on your log file except when it hit the ground, didnt you say battery came off?
Hi Ivan, thanks for responding.
Yes, battery came off when it hit the ground. The auto analyze stated: Test: Brownout = FAIL - Truncated Log? Ends while armed at altitude 11.73m. Look at my included image which shows Vcc and alt.
Never less I found the root to the problem. The wires between the motor and ESC are soldered to banana connectors, glued heat shrink was used. The wire inside was cut directly at the connectors backend when I removed the heat shrink, how this happen I am not sure. They were firmly tighten with cable ties. Should have looked at hardware in more details in the first place, also my friend had this happen to him twice with Tarot motor (TAROT 5008/340KV). But his cables was not firmly secured, so we assumed that vibration caused them to break. Not a good cable core for such usage. Should have been delivered with proper silicone wire (flexible stranded copper cables).
What do you guys think of reusing the Pixhawk (3D ACC / Gyro / MAG / Baro), gps/magnetometer after such a crash. Are the sensors affected?
//Seiern
yea just looks like poor soldering job, glad you at least found the root cause......i have had my share of crashes (hard and soft) and Pix has always managed to survived...just to make sure you test it will in the ground and then in a "safe" flight, take a close look at your log values for the sensors and make sure its all good....
Hi Ivan, thanks for your reply.
It's not due to poor soldering, what you see at the backend of connector is flux, and the rest of the cable is still deeply into the connector. The cable wire does not loosen because of poor soldering, it's cut off. As for my friend it has even been cut of inside the cable insulator. When such cable types are soldered, they get too stiff and break easily. So if you dismantle your setup sometimes, you would eventually weaken the cable wire from the Tarot motor. I have other motors for my regular copters, cables from those motors cannot be compared to the ones that Tarot uses, same banana connectors. Those inexpensive copter motors uses good quality silicone cables.
We did look at the possibilities to replace the cables coming from the motor windings, but it's a lot of work. We have dismantle a few of the Tarot motors, and it's shocking, cables has several insulation tears inside the motors (between the winding connections, and outwards). Buying another motor brand.
Hopefully this post would warn others about this problem.
//Seiern
ah ok good clarification, yea I wouldnt get into messing around with replacing the cables close to the windings, I just bought 6 of these and I love them so far, great quality, power and price....will see how it goes....but am on 4s setup and I think you are on 6s...
my next copter will be 6s based, and for that one am going with tmotor all the way for sure....
http://www.hobbyking.com/hobbyking/store/__63643__Multistar_Elite_4...
Looks like a mechanical failure of motor #1 (front right). It's got the classic signs of this:
1. the vehicle is doing fine, maintaining desired roll and pitch and then the action vs desired suddenly separate.
2. It tilts forward right (indicating the front right motor failed)
3. the output to the front right motor (#1) goes to full while it's opposite motor (#2) goes to it's minimum
I can't tell you while that motor failed but that looks like the cause. Certainly doesn't look like a software problem.
Thanks Randy for taking time to analyzing my log and for providing a detailed explanation.
Any clue on what caused the brownout afterwards? As you see in the video, led lights are still on while the quad falls to the ground.
//Seiern