Does the same thing with litchi and other 3rd party apps. Judging from looking at log files it appears to be some kind of problem between the IMUs and compasses. This problem has seem to come about from mostly new firmware or new drones. I’m sure you are right will it being a buggy code. All these problems have seem to come about ,about the same time dji was installing new computer systems for the engineering department. That’s something they do not want the public to know but they have a major problem that is affecting more and more drones every day. It’s also probably part of the reason why they canceled the announcement on the 18th . Fix the problem before they release more problemed drones into the market.The common denominator is GO 4. Must be some buggy code that was recently introduced with the latest update. That's my hypothesis.
P4P 2.0 slow drifting in hover
- Thread starter brantel
- Start date
GadgetGuy
Premium Pilot
- Joined
- Jun 18, 2015
- Messages
- 6,881
- Reaction score
- 2,199
Must be in the SDK or the FW updates then, as you point out. Spreading like a virus! If it ain't broke yet, don't update anything!
The common denominator is GO 4. Must be some buggy code that was recently introduced with the latest update. That's my hypothesis.
Except the app plays no part in stabilizing the aircraft, so that seems improbable.
In the logs that I've looked at from aircraft showing this issue, the common factor has been systematic disagreement between the rate gyros and the compass. It could be hardware-related.
GadgetGuy
Premium Pilot
- Joined
- Jun 18, 2015
- Messages
- 6,881
- Reaction score
- 2,199
Yet people are reporting that it only occurred after a FW and/or app update, and was working fine before, which should eliminate hardware as the problem, unless the FW update is causing a hardware failure.
If you are correct that means that the p4pv1 and 2 along with the Mavic Air and Pro all share the same IMUs and compasses, which may be a possibility,that I do not know because I’ve never had the chance to tear them apart to find out. My p4pv1 was on the factory firmware for the longest time. Then it was updated to 04.0602. Left it there for 20 flights and everything was the way it should be,no problems what so ever. Then updated to 05.0600. Skipped the 300 version because I read it was causing motor surging and shaking when landing and the 600 version supposedly fixed that. At this update is when the yaw drifting and motor surging/ shaking started. I downgraded back to 04.0602 and the problem was still there. Updated a friends p4pv1 the same day and his flys perfect. His also went from the 04.0602 to the 05.0600 firmware skipping the 300 version. So what exactly is the problem? Is it faulty hardware, faulty firmware or a combination of both, or is the newer firmware versions damaging certain hardware components?
Yet people are reporting that it only occurred after a FW and/or app update, and was working fine before, which should eliminate hardware as the problem, unless the FW update is causing a hardware failure.
Agreed - it's certainly possible that it is firmware rather than hardware. The point I was elaborating on is that it's not going to be software (DJI Go app).
My first guess would be firmware interaction with components from different manufacturers, which could explain why some aircraft are affected but other are not. Total speculation, of course.
Noticed one thing about the yaw drift, the gimbal was yawing in the other direction to compensate for the drift! That means that the live feed you saw on the screen was not moving but the P4 was drifting away.
Either the gimbal knows the correct yaw for some reason or it uses object recognition in the picture to keep it in focus. If it hovers with no stick input they might do this.
The gimbal continue to compensate until it was getting close to mechanical stop and after that the picture started drifting.
One of the drawbacks of this is that if you start flying the craft with the gimbal off center it will move in one direction and the camera will point in another direction, really hard to make nice shoots with that!
I also think that this is a hardware/software problem.
/hakan
Either the gimbal knows the correct yaw for some reason or it uses object recognition in the picture to keep it in focus. If it hovers with no stick input they might do this.
The gimbal continue to compensate until it was getting close to mechanical stop and after that the picture started drifting.
One of the drawbacks of this is that if you start flying the craft with the gimbal off center it will move in one direction and the camera will point in another direction, really hard to make nice shoots with that!
I also think that this is a hardware/software problem.
/hakan
GadgetGuy
Premium Pilot
- Joined
- Jun 18, 2015
- Messages
- 6,881
- Reaction score
- 2,199
Yes, that hypothesis was only postulated because it was the only known common denominator. If multiple FW versions for different aircraft were also pushed out at the same time, new FW updates would be a far better hypothesis for instability, as the aircraft doesn't need the app to fly.
GadgetGuy
Premium Pilot
- Joined
- Jun 18, 2015
- Messages
- 6,881
- Reaction score
- 2,199
Makes sense to me. Components are constantly being outsourced from different suppliers, as needed, such that two apparently identical aircraft could have several different internal components, and react differently to FW updates, much like a patient may respond better to a named drug than the generic version. Subtle differences in manufacturing can make all the difference.
Similar threads
