P3 crashed and now roll motor will not stabilize

Joined
Mar 24, 2022
Messages
7
Reaction score
0
Age
22
Hello everyone, I'm new here so if I'm posting in the wrong place or this has been discussed else where I apologize however I can't seem to find this issue anywhere. I have a P3 4k here that someone brought me to fix. The P3 suffered a fall from at least 30feet said the guy. The drone works except the gimble. When you turn it on at first it would just grind and not move. I ordered a factory ribbon as well as a roll motor. After installing now everything moves however the roll motor is smashing back and fourth against it's stop. I have watched a few videos and the flat on the shaft is aligned correctly as well as the set screw on the yaw shaft. If any one has any other ideas I'd be glad to hear them. I did try a calibration and an IMU calibration but neither would take.
 
Joined
Jun 30, 2016
Messages
1,442
Reaction score
637
Location
PL
Check if potentiometers in roll/pitch arms are ok (in kOhms and increase gradually without bumps as you rotate), the carbon resistive sheet often gets broken.
Also check yaw arm - it uses a magnetic ring encoder and magnetic field reader instead of potentiometers, make sure the ring is there.
Check if magnets within motors are not crushed.
Check if gimbal arm didn't crushed a quartz oscillator on gimbal top board.
 
Joined
Mar 24, 2022
Messages
7
Reaction score
0
Age
22
Check if potentiometers in roll/pitch arms are ok (in kOhms and increase gradually without bumps as you rotate), the carbon resistive sheet often gets broken.
Also check yaw arm - it uses a magnetic ring encoder and magnetic field reader instead of potentiometers, make sure the ring is there.
Check if magnets within motors are not crushed.
Check if gimbal arm didn't crushed a quartz oscillator on gimbal top board.
Thank you for the reply, checked everything this morning both the roll and pitch pots test good (0-8k) no bouncing. The crystal which I assume is the small black rectangle on the main board looks intact. As for the motors I'm not sure how to get at the magnets, the motor dosent seem to want to come apart. They all feel like the rotate fine no binding. I'm not sure where the magnet for the yaw arm is either I can't seem to find that or it's encoder.
 
Joined
Jun 30, 2016
Messages
1,442
Reaction score
637
Location
PL
Magnet encoder - see the photos here:

Crystal osc - yes that's it - the usual metal tube is inside the black plastic to make it easier for machines which place the parts.

Magnets inside motors - if you can't hear grinding and don't perceive an uneven resistance when rotating, they must be ok.
 
Joined
Mar 24, 2022
Messages
7
Reaction score
0
Age
22
Magnet encoder - see the photos here:

Crystal osc - yes that's it - the usual metal tube is inside the black plastic to make it easier for machines which place the parts.

Magnets inside motors - if you can't hear grinding and don't perceive an uneven resistance when rotating, they must be ok.
Yep the magnetic ring is still on the Yaw motor shaft. Any other checks or tests I could do? For some reason it seems to think it's position is way off I can try to take a video later if that would help diagnose the issue
 
Joined
Jun 30, 2016
Messages
1,442
Reaction score
637
Location
PL
Since we don't really know how to interpret error codes form gimbal MCU (STM32 microcontroller), at this point it is best to start replacing parts with known working ones.

The messages from gimbal are in serial packets, they are stored on a special sd-card near FC, and you can also see them fly live by connecting to the front USB of the drone - in case you still want to investigate that route. But it would take you a lot of research to get an understanding there. We have several parsers for the packets though, ie. DATCon, or Wireshark dissectors.

It is very hard to narrow down the issue based on gimbal motion, unless one arm is just clearly limp. Otherwise, all arms will be moving erratically as the gimbal tries to stabilize position based on pots and magnetometer from each arm and IMU placed on the board with image sensor.
 
Joined
Mar 24, 2022
Messages
7
Reaction score
0
Age
22
I guess at this point I will pull both the roll and pitch boards and take a look under the microscope and look for cracked caps or resistors, order up both the pitch and yaw motors and see what happens. Otherwise I guess I will need to find a way to secure the camera in a fixed position. I know the yaw arm does have a slight bend in it but I didn't figure this would make it go wild like it is probably just effect the resting position. I will get that on order as well
 
Joined
Mar 24, 2022
Messages
7
Reaction score
0
Age
22
Since we don't really know how to interpret error codes form gimbal MCU (STM32 microcontroller), at this point it is best to start replacing parts with known working ones.

The messages from gimbal are in serial packets, they are stored on a special sd-card near FC, and you can also see them fly live by connecting to the front USB of the drone - in case you still want to investigate that route. But it would take you a lot of research to get an understanding there. We have several parsers for the packets though, ie. DATCon, or Wireshark dissectors.

It is very hard to narrow down the issue based on gimbal motion, unless one arm is just clearly limp. Otherwise, all arms will be moving erratically as the gimbal tries to stabilize position based on pots and magnetometer from each arm and IMU placed on the board with image sensor.
Just for kicks I pulled the roll axis control board back out and remeasured resistance between the single upper leg and the bottom left leg while the pot is in it's horizontal position I get 8k on left leg but on right leg I'm getting 180k no matter where the pot is turned im assuming that shouldn't be the case?
 
Joined
Mar 24, 2022
Messages
7
Reaction score
0
Age
22
The pot is identified here:

It says that's 6-10 kOhm pot, definitely not 188k.

There's also a link to datasheet if you're confused by 4 legs instead of 3.
Okay that cleared it up then we have identified most likely our issue! I will order it up and recheck the pitch potentiometer and retry. Thank you so much!
 
Joined
Mar 24, 2022
Messages
7
Reaction score
0
Age
22
The pot is identified here:

It says that's 6-10 kOhm pot, definitely not 188k.

There's also a link to datasheet if you're confused by 4 legs instead of 3.
New pot installed test good now, drone now take IMU calibration but still goes crazy like before, what else could cause this? Both pots and incoder ring seem fine mabye ribbon got damaged during installation? Video works find and it seems to read all xyz sensor data
 

Members online

Forum statistics

Threads
142,178
Messages
1,461,230
Members
103,960
Latest member
TECNOPADO