This is a new one for me. Have a new P4Pro I picked up, running the latest firmware on ipad mini 2, etc. When I sit and hover, the drone ever so slowly rotates towards the west until it's facing directly west. Doesn't matter which direction I point it, after about 30-45 seconds it will rotate until it gets to west. If I face west, it doesn't rotate. I tried a compass calibration and IMU calibration, but it still does the same thing. Is it worth trying them a few more times maybe? I might pull the DAT and look at the mag data on there too just to see, but very strange behavior! I could also take it back to best buy and exchange it I suppose?
Phantom 4 Pro - always wants to face west
- Thread starter F3honda4me
- Start date
Maybe it's a 'Mecca' thing? Have you tried doing a RC calibration as one of sticks (the left one in mode 2) may not be perfectly centred when in neutral.
- Joined
- Oct 19, 2017
- Messages
- 19
- Reaction score
- 21
- Age
- 68
- Joined
- Jun 14, 2016
- Messages
- 1,744
- Reaction score
- 791
- Age
- 66
Ah yes. But why would it stop when it gets to west and why would it not rotate if it is already facing west when it starts? I suspect there’s something nefarious and evil at play here.
I didn't do it because on the screen it shows centered, plus it wouldn't stop once it gets to facing west.
@BudWalker
Starting at 160s, I took the drone up and just sat there. Notice how magyaw constantly moves even though I'm giving zero RC Rudder input. I did several tests (where relative height is at 33)
Starting at 160s, I took the drone up and just sat there. Notice how magyaw constantly moves even though I'm giving zero RC Rudder input. I did several tests (where relative height is at 33)
Here is my DAT file if anyone else wants to take a shot at it.
Dropbox - DJI_ASSISTANT_EXPORT_FILE[2017-11-08 20-07-52].FLY009.DAT
Dropbox - DJI_ASSISTANT_EXPORT_FILE[2017-11-08 20-07-52].FLY009.DAT
So the question is, is this being caused by compass or IMU... Probably going to box it up and exchange it at best buy tomorrow. I want a gremlin free drone, not one that always wants to point west lol.
N017RW
Premium Pilot
If there was an offset due to a ‘mis-cal’ it would never stop rotating.
i.e. No different than an intentional input or stick deflection.
Good point, mine used to rotate very slightly in hover and I fixed it with a recal.
Cheers.
The reason that the P4Pro rotates on it's own is that it's trying to eliminate a Yaw/magYaw separation. Seems to be a problem with compass #0 which is the one that it's using. I don't think it's associated with pointing west. At time 170 there was a Yaw/magYaw separation that caused the P4P to fix by rotating while holding Yaw fixed. At about 220 the separation was eliminated and the P4P stopped rotating at the heading -27 degrees.
The same thing happens at 223 secs but stopping at -150 degrees.
There is an indication that the compass isn't calibrated as can be seen while the AC is rotating from 28 secs to 78 secs
There are several points in the flight where this doesn't happen, it's only after 45 degrees or more that the Yaw/magYaw separation happens followed by the uncommanded rotating to fix the separation.
Got all that ?
Can you please do two tests. 1) At altitude turn it pointing North and then wait a couple of minutes to see if it really does stop at west.
2) Cause the P4P to switch to compass #1 (it's normally on #0). Then do your test. Compass #1 seems to be better behaved. The numbering scheme that you see in the App is probably 1 and 2 whereas in the .DAT it's 0 and 1. So just get it to switch to the higher numbered compass. I don't know where the compasses are but moving a screwdriver close to a compass will cause it to switch.
Don't know why it seems to have this problem. I always ask is there some external equipment attached?
No external equipment. This is a brand new p4p I got for myself. My wife has one too but hers is fine, she's had it for 6 months.The reason that the P4Pro rotates on it's own is that it's trying to eliminate a Yaw/magYaw separation. Seems to be a problem with compass #0 which is the one that it's using. I don't think it's associated with pointing west. At time 170 there was a Yaw/magYaw separation that caused the P4P to fix by rotating while holding Yaw fixed. At about 220 the separation was eliminated and the P4P stopped rotating at the heading -27 degrees.
View attachment 90557
The same thing happens at 223 secs but stopping at -150 degrees.
View attachment 90558
There is an indication that the compass isn't calibrated as can be seen while the AC is rotating from 28 secs to 78 secs
View attachment 90559
There are several points in the flight where this doesn't happen, it's only after 45 degrees or more that the Yaw/magYaw separation happens followed by the uncommanded rotating to fix the separation.
Got all that ?
1) At altitude turn it pointing North and then wait a couple of minutes to see if it really does stop at west.
2) Cause the P4P to switch to compass #1 (it's normally on #0). Then do your test. Compass #1 seems to be better behaved. The numbering scheme that you see in the App is probably 1 and 2 whereas in the .DAT it's 0 and 1. So just get it to switch to the higher numbered compass. I don't know where the compasses are but moving a screwdriver close to a compass will cause it to switch.
Don't know why it seems to have this problem. I always ask is there some external equipment attached?
I'll try your tests along with another compass calibration or two.
Thanks Bud!
Ok so I did the test. It stops eventually but before hitting due west. I then tried the screwdriver trick and got it to flip to the other compass, but soon as I take the screwdriver away it flips back to the better working one.
Then I shut it off, powered up, did another compass calibration, and THIS time it seems that it has fixed it. Or at least made it not as bad. Maybe the first time the calibration was bad or something. But I did the test again, this is the 2nd DAT you'll see. This time it did not seem to yaw on its own.
Dropbox - DJI_ASSISTANT_EXPORT_FILE[2017-11-09 10-28-53].DAT
Then I shut it off, powered up, did another compass calibration, and THIS time it seems that it has fixed it. Or at least made it not as bad. Maybe the first time the calibration was bad or something. But I did the test again, this is the 2nd DAT you'll see. This time it did not seem to yaw on its own.
Dropbox - DJI_ASSISTANT_EXPORT_FILE[2017-11-09 10-28-53].DAT
Last edited:
Looking at the data now, the yaw and mag yaw are still way off even on the 2nd test flight. You can see it still doing the correction, but it is slower than the first flight test. Granted it is windy today so that made it harder to tell I think. But, probably going to return this bird and exchange it for another.
Got a new P4P, works flawlessly. Does none of the uncommanded yaw that the previous one does. Is rock solid, very stable even in the crazy high wind here today. That other bird definitely had something going on. I told best buy what was wrong with it so hopefully it gets back to DJI for repair.
Similar threads
