P4P 2.0 slow drifting in hover

[brantel]

@BudWalker and others....

Unfortunately the slow turning in yaw has returned for some reason. I ran over 4 packs thru the unit after the firmware upgrade and the slow turning was almost not detectable. Yesterday and today it has returned. This is driving me nuts!

Here is the last flight's dat off of the drone...

If anyone wants to take a look and let me know if there is anything I can do to help minimize this, please let me know and thanks for your help!

DJI_ASSISTANT_EXPORT_FILE_2018-06-29_06-51-13.DAT


PS...Every video I have viewed from a P4P 2.0 online where they were hovering at altitude, I have noticed this same slow turn in yaw. Am I asking too much from these drones? Is this basically normal and I have to live with it? Some people claim it is the gimbal but the logs prove it is not. The drone is turning and sometimes it is upwards of 20° if you let it hover long enough.

 

[GadgetGuy]

@BudWalker and others....

Unfortunately the slow turning in yaw has returned for some reason. I ran over 4 packs thru the unit after the firmware upgrade and the slow turning was almost not detectable. Yesterday and today it has returned. This is driving me nuts!

Here is the last flight's dat off of the drone...

If anyone wants to take a look and let me know if there is anything I can do to help minimize this, please let me know and thanks for your help!

DJI_ASSISTANT_EXPORT_FILE_2018-06-29_06-51-13.DAT


PS...Every video I have viewed from a P4P 2.0 online where they were hovering at altitude, I have noticed this same slow turn in yaw. Am I asking too much from these drones? Is this basically normal and I have to live with it? Some people claim it is the gimbal but the logs prove it is not. The drone is turning and sometimes it is upwards of 20° if you let it hover long enough.

P4P original does not exhibit this behavior, so it is not "normal" for these drones, but the V2 is still in its infancy, going through the same growth pains that took almost 12 months for the P4P to get all worked out. Hopefully, they can get it sorted out with a mere FW fix, and it isn't a hardware issue, requiring a recall! The problem is that won't be known, until it actually gets fixed, one way, or another.

 

[Paul2660]

The P4 vr2.0 plus I have shoes the drift in hover for sure.

Today just flying it around my house it was quite apparent.

In take off during the hover right after launch mine went hard left about 2 to 3 feet then stabilized. After climbing straight up to 200 feet and hovering mine goes forward a few feet.

Strange thing is it doesn’t happen all the time. About 50 % or so.

Does make a tight landing a bit harder as you have to watch for the drift.

This is with new firmware.

Paul C

 

[GadgetGuy]

The P4 vr2.0 plus I have shoes the drift in hover for sure.

Today just flying it around my house it was quite apparent.

In take off during the hover right after launch mine went hard left about 2 to 3 feet then stabilized. After climbing straight up to 200 feet and hovering mine goes forward a few feet.

Strange thing is it doesn’t happen all the time. About 50 % or so.

Does make a tight landing a bit harder as you have to watch for the drift.

This is with new firmware.

Paul C

Update to the latest FW and GO 4, and then do a cold IMU and gimbal calibration on a level surface. That should address your concerns, according to the most recent DJI posts by Ken/blade strike.

 

[Paul2660]

I did try that.

Firmware is current on drone. It’s a pro plus so I assume the djigo4 app is current.

Re did an IMU calibration and the problem did go away for 1 flight. Once I tired the controller and drone off and then tried to fly the problem came back.

Paul C

 

[GadgetGuy]

I did try that.

Firmware is current on drone. It’s a pro plus so I assume the djigo4 app is current.

Re did an IMU calibration and the problem did go away for 1 flight. Once I tired the controller and drone off and then tried to fly the problem came back.

Paul C

I feel your pain. Unfortunately, the P4PV2+ is an entirely different animal, running a buggy customized Android version of GO 4, written by 3rd string Android programmers, instead of DJI's crack iOS crew. Would highly recommend exchanging the V2+ version for the standard V2 version and an iOS tablet, or buy a separate $319 standard V2 RC, as the aircraft are identical.

 

[Paul2660]

It’s headed back today.

I wonder if those same Android programmer do the coding for the Crystalsky?

Paul C

 

[brantel]

It’s headed back today.

I wonder if those same Android programmer do the coding for the Crystalsky?

Paul C

I have tried 3 of them... they all did it. Ken Booth from DJI says it is a known issue and does not recommend sending them in. Likely to get the same thing back.... He keeps saying wait on the firmware fix.

 

[Paul2660]

I understand, but this one also had a off centered camera. I flew a P4 Pro Vr 2.0 non pro yesterday, one that shipped during the first week a Best Buy order. This machine has no forward movement issues in hover like my Pro +, and yaws only during a pan. So as you pan all is fine but as you stop the pan even with the smoothness control the drone will just ever so slightly continue in the previous direction of the pan. If you push back on the stick just a slight amount this can be defeated.

Back to my Pro P+, to be honest after working with a iPad or CS, the monitor on the Plus is not for me.

Controls are not as convenient to me as on the CS or standard iPad layout.

Surprising to me that since the GPS, compass etc on the Vr 2.0 should be the same as on the P4 Pro, that such issues are happening, and that they were not caught during testing. Surely a lot of these were put out to beta testers. The Occusync vs Lightbridge should not make a difference in flight characteristics should it?

I realize the firmwares are different but the drones should be sharing pretty much the same components.

Paul C

 

[brantel]

I understand, but this one also had a off centered camera. I flew a P4 Pro Vr 2.0 non pro yesterday, one that shipped during the first week a Best Buy order. This machine has no forward movement issues in hover like my Pro +, and yaws only during a pan. So as you pan all is fine but as you stop the pan even with the smoothness control the drone will just ever so slightly continue in the previous direction of the pan. If you push back on the stick just a slight amount this can be defeated.

Back to my Pro P+, to be honest after working with a iPad or CS, the monitor on the Plus is not for me.

Controls are not as convenient to me as on the CS or standard iPad layout.

Surprising to me that since the GPS, compass etc on the Vr 2.0 should be the same as on the P4 Pro, that such issues are happening, and that they were not caught during testing. Surely a lot of these were put out to beta testers. The Occusync vs Lightbridge should not make a difference in flight characteristics should it?

I realize the firmwares are different but the drones should be sharing pretty much the same components.

Paul C

It is my understanding that much more was changed with the P4P 2.0 than just changing to Occusync and new ESC's. It appears pretty much the whole control package was apparently changed to the newer DJI standard. This seems to be the reason for the drastically different firmware and the new version of DJI Assistant 2.0 just for the P4P 2.0 and for the need to update the SDK. Ken Booth even said there must be more differences inside with the hardware and definitely with the code.

My comments in the prior post were strickly related to the slow yaw drifting issue. I have no experience or knowledge of any issues with the Plus version. I also don't have an off center camera issue. Based on your issues and distaste for the Plus model, I would return it as well.

 

[GadgetGuy]

It’s headed back today.

I wonder if those same Android programmer do the coding for the Crystalsky?

Paul C

The answer is a definite "Yes!" Same lousy Android programmers that write the customized GO 4 for the Plus model RC's write the buggy Android code for the CS tablets. Both require an Android version of GO 4 specifically written for their own known hardware! They still can't get it right! Had DJI licensed iOS for their own branded tablets, the Dream Team iOS coders would have produced a winning product. Alas!

 

[$gambino$]

I will say I don't not have any problems with my Galaxy S9 plus and v2.... as you would with a cheaper Android device or Crystal Sky/+ model

 

[Paul2660]

The version of DJIgo4 and android OS on non CS or Plus models is different. It's at least 2 versions behind the main android version. This is stated by DJI themselves on the Crystalsky FAQ. I believe the Android version is a custom piece of code.

Updates appear to happen in this order, Mac iOS, Android, Crystalsky/plus Android.

The current state of Crystalsky and Phantom 4 connectivity is pretty darn bad IMO. I have long since given up on the CS for P4, only use it on my Mavics, as it works pretty flawlessly on them.

Paul C

 

[Mlebout]

It is my understanding that much more was changed with the P4P 2.0 than just changing to Occusync and new ESC's. It appears pretty much the whole control package was apparently changed to the newer DJI standard. This seems to be the reason for the drastically different firmware and the new version of DJI Assistant 2.0 just for the P4P 2.0 and for the need to update the SDK. Ken Booth even said there must be more differences inside with the hardware and definitely with the code.

My comments in the prior post were strickly related to the slow yaw drifting issue. I have no experience or knowledge of any issues with the Plus version. I also don't have an off center camera issue. Based on your issues and distaste for the Plus model, I would return it as well.

I had the Compass Redundancy errors also. I never noticed the drifting as I only took night shots (they looked good to me 6-8 second shutter). I did the latest firmware update and I no longer get the compass errors. However, I do get the drifting you've described. I have also done IMU and compass calibrations before and after the FW update. Yes, you can stop the drifting if you counter it. But, this is obviously not ideal.

 

[GadgetGuy]

The version of DJIgo4 and android OS on non CS or Plus models is different. It's at least 2 versions behind the main android version. This is stated by DJI themselves on the Crystalsky FAQ. I believe the Android version is a custom piece of code.

Updates appear to happen in this order, Mac iOS, Android, Crystalsky/plus Android.

The current state of Crystalsky and Phantom 4 connectivity is pretty darn bad IMO. I have long since given up on the CS for P4, only use it on my Mavics, as it works pretty flawlessly on them.

Paul C

Exactly! The generic Android version for all Android devices (with the exception of DJI devices) is vastly better than the custom Android GO 4 version written for the known hardware of DJI's creation. Go figure!

 

[Scoobywrx05]

To bring this back up I’m going through the same thing with my p4pro v1. Ran it on factory firmware for the longest time. Updated to 04.0602 and things where fine. Went to 05.0600 and it acquired the slight rotational drift just like a number of the Mavic Airs have, plus the death shake when landing seconds before the motors shut off. That shake doesn’t happen when I hand catch it in the air tho. I downgraded back to the 04.0602 and the yaw was still there and the shake was also but the shake was not as violent. So both versions of the p4ps have the same problem along with the Mavic Air and heard of the Mavic pros having the same issues. Question is, what is the common factor between all of them that is causing the drift issues.

 

[herkam]

Hi!
Same issue here with my P4 Pro V2, drifting in yaw and not flying straight.
Done all calibrations, latest software and this yaw problem is really BAD!
It is strange that is sometimes hovers without yaw drift and flies straight...haven't found any pattern when it drifts or not yet

My P4 was NOT having any problems with yaw drift so they have software that handles the data fusing between gyro and compass (kalman filter) properly.
My fear is that the IMU hardware is bad (may due to cost reductions or change of supplier) and now they tell the software guys to fix it....good luck!

I was planning to use the P4 Pro V2 for surveying but did one test today using Litchi and a mission with constant heading but the yaw drift caused the pictures to have different rotation which made it a mess!

/hakan

 

[Scoobywrx05]

Hi!
Same issue here with my P4 Pro V2, drifting in yaw and not flying straight.
Done all calibrations, latest software and this yaw problem is really BAD!
It is strange that is sometimes hovers without yaw drift and flies straight...haven't found any pattern when it drifts or not yet

My P4 was NOT having any problems with yaw drift so they have software that handles the data fusing between gyro and compass (kalman filter) properly.
My fear is that the IMU hardware is bad (may due to cost reductions or change of supplier) and now they tell the software guys to fix it....good luck!

I was planning to use the P4 Pro V2 for surveying but did one test today using Litchi and a mission with constant heading but the yaw drift caused the pictures to have different rotation which made it a mess!

/hakan

I’m not so sure it is a hardware problem yet. If it was it wouldn’t be affecting other lines of dji products. Have seen this issue on recent firmware’s for the p4pv2,p4pv1, Mavic Pro,Mavic Air so far.

 

[GadgetGuy]

I’m not so sure it is a hardware problem yet. If it was it wouldn’t be affecting other lines of dji products. Have seen this issue on recent firmware’s for the p4pv2,p4pv1, Mavic Pro,Mavic Air so far.

The common denominator is GO 4. Must be some buggy code that was recently introduced with the latest update. That's my hypothesis.

 

[herkam]

Hi!
Did a simple hover test today, had the P4 Pro V2 hovering for a minute, turned it 180 and took the hands off the controller and it started drifting in yaw.....

This is a very nice situation for the IMU, no accelerations, turns or other stuff just a gently rotation of the craft.

/hakan