Neither P4P will maintain altitude

[spirytus]

Damnit guys, 4 pages and no one asked him to pull the FC logs?

1. Fly a test flight with one of your faulty P4Ps; Nothing too fancy: take off, hover, climb, descend, yaw left, yaw right, move fwd, move back, bank left, bank right, land -- in that order testing all the axis responses individually.

2. Plug her in and download the .DAT log for this one flight, and while you're at it download the controller .txt log from you iPad.

3. Upload them to Google drive or Dropbox and share them on here.




Sent from my Pixel XL using PhantomPilots mobile app

 

[spirytus]

The only common denominator at this particular juncture is you, maybe you've miscalibrated some setting on all 3 drones.

Run through a stick calibration again as shown below:

Lastly, if all else fails do a factory reset with the Assistant 2 software.


Sent from my Pixel XL using PhantomPilots mobile app

 

[jpk1080]

I will give this a shot today as well as getting some logs posted. I agree that I am the common denominator. I have had probably 6 drones in the last 2 years though and have never had this issue. I'm not saying it isn't something I'm doing, but I find it unlikely that it would be affecting all three aircraft. I did do the factory reset through the DJI assistant app as well as the controller calibration. I will do it again to see if it makes a difference, then try to get the logs posted. Thanks for the help.

 

[jpk1080]

Here is a link to the flight dat. files i uploaded today. The first one was me just testing to see if the drone was still doing the same thing after calibrations, resets, etc. The second was what you asked for which consisted of hovering, climb, descend, and going in each direction. The drone would descend in all directions by about .4 to .7 fps. I hope it indicates that on the logs. Thanks for the help. Let me know if I can get anything else for you. I was unsure of how to pull the controller logs off the ipad.

Dropbox - DataLogs

 

[spirytus]

FLY18.DAT is the aircraft sitting in one spot at relativeHeight of 0 the entire recording with motors off. Looks like you were messing around with the settings.

FLY17.DAT is interesting. Can you pull the txt log from your ipad for that flight?

Please see https://www.phantomhelp.com/LogViewer/upload/

 

[jpk1080]

I will try to get the text logs tomorrow. I was up and flying both times

 

[D'sAdvanced]

Glad I found this thread as I am having the exact same issues as the OP on my P3A. Took is out a few days ago and just wanted to play around at low altitudes. I changed the Braking setting a few days prior and wanted to see how the new setting worked. I would go full speed starting at say 20' altitude and little by little, the drone kept getting closer and closer to the ground. First thought it was a fluke but the more I did it, the more certain I became there was a problem. Altitude would drop around 20' in a distance of around 100-150 yards. Hope a resolution can be found.

 

[spirytus]

Well I reviewed the only good FC log you uploaded: FLY17.DAT and the data is too messy to see anything concrete.

At several points of the flight it looks like you're trying to go full positive then later full negative with your aileron, and elevator inputs (right control stick [mode 2]) but you're also peppering the data with additional inputs from the throttle / rudder.

Worst off all I don't see any considerable loss of altitude on any of the 3 primary altitude datapoints: gpsAltitude, baroAlt(m), or relativeHeight(m) when I actually have clean full right stick inputs (either aileron or elevator).

Can you fly this thing in an open field (football filed would be good) and follow a script we can later analyze in the log?

0. Start video recording (nothing fancy, 1080 at 30fps).
1. Take off and hover at a height of your 30m for 1 minute.
2. Aileron: roll left for aprox 3 seconds (right stick - full left).
3. Pause for 10 seconds.
4. Aileron: roll right for aprox 3 seconds (right stick - full right).
5. Pause for 10 seconds. Readjust height back to 30m if you lost altitude.
6. Elevator: pitch down for aprox 3 seconds (right stick - full forward).
7. Pause for 10 seconds.
8. Elevator: pitch up for aprox 3 seconds (right stick - full back).
9. Pause for 10 seconds. Readjust height back to 30m if you lost altitude.
10. Yaw 360 degrees to the left.
11. Pause for 10 seconds.
12. Yaw 360 degrees to the right.
13. Land and end the flight.

Go home and download the .DAT log from the quad, download the .txt log from the ipad, and download the cached video file saved on your ipad also. Don't mess around after as this may make it harder for you to upload the correct log.

here's mode 2 controller configuration for your reference:

 

[BuickGS]

Since these are new, I think I would open a support case up with DJI. They will more than likely have you send one in for repair and that 'should' answer the question. I am very interested in hearing what the fix is once one is eventually found.

 

[spirytus]

Since these are new, I think I would open a support case up with DJI. They will more than likely have you send one in for repair and that 'should' answer the question. I am very interested in hearing what the fix is once one is eventually found.

This is true, however DJI support can be a hit or miss so it would help to have undisputed data to back up the claim - I'm willing to vest time to get him just that.

It could also be helpful information to the P4P community if we disclosed findings (and maybe a fix) before loosing the opportunity to do so -- Support is not know for publishing fault findings. They'll just ship him a new one (or a refurb) if they can duplicate it at the repair station.

Sent from my Pixel XL using PhantomPilots mobile app

 

[Russ43Phantom]

Does the setting (On/Off) of the Downward Vision Positioning System influence minimum altitude above the ground (vs. above sea level)? The barometric reading only works on altitude above sea level. If the ground level is rising relative to your initial launching position it would seem, if you are depending on the barometer, that the behavior you are seeing would be expected. GPS also measures x-y-z but has no idea if the ground level is rising.

I don't know the answer to this question, but logically, it seems your investigation should start with how the Downward Vision Positioning System addresses your problem.

 

[Fat City]

This may be kind a longshot, but I think a possible cause of the altitude loss could be a strong local updraft.The pressure is lower in an updraft, and hot dry areas are prone to strong localized updrafts, especially at boundaries between different surfaces types such as asphalt, grass, irrigated field or water. It's common to see dust devils in the morning in the warmer parts of California's Central Valley due to the differential warming by the sun on these surfaces.
The A/C relies on air pressure to sense and maintain altitude (unless using VPS), so an updraft's lower pressure makes the A/C think it has risen and to compensate it will go down.

 

[jpk1080]

Thanks all for the replies. I will try to get a clean file today for you guys to look at. It seems to be much worse at lower altitudes, but I will try the 30 meters and see how it goes. VPS being on or off doesn't seem to make any kind of dfference. As I mentioned, this is happening on all 3 drones, in multiple locations and states. Strange that it isn't showing altitude loss but it's possible the video will. I will get on that today. Thanks again.

 

[spirytus]

Thanks all for the replies. I will try to get a clean file today for you guys to look at. It seems to be much worse at lower altitudes, but I will try the 30 meters and see how it goes. VPS being on or off doesn't seem to make any kind of dfference. As I mentioned, this is happening on all 3 drones, in multiple locations and states. Strange that it isn't showing altitude loss but it's possible the video will. I will get on that today. Thanks again.

30m test height is arbitrary, do 6m (20ft you reported in first post) if you prefer. Just do it in an open field away from buildings or other possible sources of interference, and *try* to follow the flight script please.

Sent from my Pixel XL using PhantomPilots mobile app

 

[jpk1080]

I just took it out but messed up the test and did other stuff before landing so I will redo it. At 100 feet it does much better. It basically levels out in al directions. However low to the ground especially moving forward it dives. It just will not maintain. I'm going to head to a field so I can get a bit lower and hopefully show the data along with video. Sure yiur drones have no issues descending when 20 feet or so above the ground?

 

[jpk1080]

Okay I went out and did a few different test on a field. The first few, are what you asked for. The last one, is just a simple hover, followed by forward, and back. As you can see, there is a drop off on the field a little ways. After going forward, and letting it hover, you will see it climb. That is not me adjusting the height. It is climbing itself almost as if it knows it descended. At these heights, the gps and vps will be significantly off from one another until it regains its altitude. You can tell from the video that the drone is descending both forward, and backwards. Up at altitude, it doesn't seem to be as much of an issue. Thanks for taking a look.

One more thing, the other P4P and Inspire 1 do the exact same thing to the T.

Dropbox - FlightLogsUpdated

 

[pomonabill220]

Like the cartwheels your daughter was doing.... funny little addition to your videos.
Very interesting problem!

 

[jpk1080]

Ha yeah sorry about that. She likes to be on camera! I know. It's been frustrating to say the least. Another user said he was having the same issue. No one else has a similar issue?

 

[pomonabill220]

No need to apologize for your daughter's acting. It was fun!
I have had my P3A for almost a year, and have YET to take it out! I know... "why do I even have it??" Time and weather.
I know it isn't a P4, but I will keep your problem in mind. I just downgraded from 3.1.2 to 3.1.1 for my GO app since so many others have found a serious HP problem, so I want to try it out soon.

 

[Doctorcobani]

Stupid question but maybe helps:

In your 3 drones did you use the same device?

You have DJI GO for Inspire and DJI GO4 for P4P in the same device?

Which type of device do you have? Android? iOS?

Enviado desde mi Nexus 6P mediante Tapatalk