Near disaster- warning on running Waypoints w/o a device barometer

Joined
Jan 14, 2015
Messages
907
Reaction score
233
Location
Orange County, California
Was testing the waypoint feature out today with my iPad mini3. Flew to 4 points at slightly varying altitudes; none less than 40ft from ground. Set the system to start. It flew to first wp fine, then to the second wp about 20ft above trees its elevation was too low, so had to abort by flipping the mode switch to 'P'. You would think I would have clued in after a month of testing with Autopilot and their app constantly popping up windows to verify current altitude reading....
Nonetheless, I just figured I did something wrong, so flew 4-wp path all same altitude and saved to favorites this time in case I wanted to redo. Started the system, all was well...wp1, wp2... on way to wp3 it started losing altitude [emoji37][emoji37][emoji37]
I managed to disable about 3ft from the ground...whew! [emoji37]
Then it hit me, my iPad doesn't have barometer so it's relying on the bird. The bird isn't very accurate, especially when path elevation changes.

Just a watchout for those either flying close to objects or locations where elevation changes. If you use iOS, Iphone6/6+ and iPad Air2 all have barometers. iPad minis do not.
Edit: adding video
 
Last edited:
Was testing the waypoint feature out today with my iPad mini3. Flew to 4 points at slightly varying altitudes; none less than 40ft from ground. Set the system to start. It flew to first wp fine, then to the second wp about 20ft above trees its elevation was too low, so had to abort by flipping the mode switch to 'P'. You would think I would have clued in after a month of testing with Autopilot and their app constantly popping up windows to verify current altitude reading....
Nonetheless, I just figured I did something wrong, so flew 4-wp path all same altitude and saved to favorites this time in case I wanted to redo. Started the system, all was well...wp1, wp2... on way to wp3 it started losing altitude [emoji37][emoji37][emoji37]
I managed to disable about 3ft from the ground...whew! [emoji37]
Then it hit me, my iPad doesn't have barometer so it's relying on the bird. The bird isn't very accurate, especially when path elevation changes.

Just a watchout for those either flying close to objects or locations where elevation changes. If you use iOS, Iphone6/6+ and iPad Air2 all have barometers. iPad minis do not.
You've lost me. I thought the only way the bird knew elevation -and changes in its elevation- was via the bird's barometer (and no other). How is a barometer in any smart device used?
 
Yeah, I guess I didn't read your post close enough. What does a barometer on the device do for the bird's altitude?
 
You've lost me. I thought the only way the bird knew elevation -and changes in its elevation- was via the bird's barometer (and no other). How is a barometer in any smart device used?
Ya... I think this has nothing to do with anything controller based. There is no way that a the altitude could be tablet based when the P3 is sending ITS altitude constantly back to the tablet.
 
You've lost me. I thought the only way the bird knew elevation -and changes in its elevation- was via the bird's barometer (and no other). How is a barometer in any smart device used?
I think it uses the bird by default. This is known to be limited in accuracy. It's more exposed as an issue with wp mode versus flying manually. DJI doesn't share documentation on the system, so can only assume they would utilize a devices barometer that's more accurate if available. Same as is done with GPS for home point/etc. It's in the fw to allow and 3rd party apps explicitly give you the option. Maybe DJI didn't even consider this!
 
Added video of 2nd failure- near crash into ground
 
My Phantom 3 was consistently showing it's height as 45 to 46 feet on landing after 16 minutes or so of flying. Actually when I was hovering about 1 foot above the takeoff point. I did an IMU calibration after last month's firmware update and that appeared to fix it. My P2V+ had a similar issue but IMU cal didn't help with that one.

What does the app show as the Phantom's height when you land right before shutting down the motors?
 
The P3 doesn't use anything other than it's internal barometer. As far as it is concerned, the take off point is 0ft. All other heights are relative. Barometers are not perfect. They can fluctuate by as much as 10ft.
 
The only utilization of a device barometer I've seen is in Litchi's Magic Leash app, which is used to allow Follow-me to follow something other than the device hooked to the rc. The concept is that if the device being followed goes up or down then you may want the uav to follow suit (or risk plowing into a hill as it follow its target). Not sure whether DJI's implementation does the same, but something to think about before doing a follow-me up a hill with the p3 at low altitude.
 
I think it uses the bird by default. This is known to be limited in accuracy. It's more exposed as an issue with wp mode versus flying manually. DJI doesn't share documentation on the system, so can only assume they would utilize a devices barometer that's more accurate if available. Same as is done with GPS for home point/etc. It's in the fw to allow and 3rd party apps explicitly give you the option. Maybe DJI didn't even consider this!
Whether your tablet has a barometer and whether it is accurate or not is completely irrelevant.
Your tablet could only report the altitude of the tablet. It knows nothing of the altitude of your Phantom and the altitude of your tablet is useless information for waypoint flying.
The barometer in the Phantom is much more accurate than you think and your problem will be something else.

Not sure what you mean about GPS for home point etc.
Your Phantom only uses barometer for altitude data because GPS is very inaccurate in the vertical axis (commonly +/-100-200 ft)
 
[snip]

Not sure what you mean about GPS for home point etc.
Your Phantom only uses barometer for altitude data because GPS is very inaccurate in the vertical axis (commonly +/-100-200 ft)
I have read posts to that effect several times on this forum and have to question it. I have half a dozen GPS's for use in my vehicles. They're from differing mfrs. (Garmin, TomTom, etc.). At least two of them show elevation above sea level and are extremely accurate and repeatable. Are they not getting that info extrapolated from Satellites? Or is some software looking up the Lat./Long. coordinates and reading elevation from a table?
 
I have read posts to that effect several times on this forum and have to question it. I have half a dozen GPS's for use in my vehicles. They're from differing mfrs. (Garmin, TomTom, etc.). At least two of them show elevation above sea level and are extremely accurate and repeatable. Are they not getting that info extrapolated from Satellites? Or is some software looking up the Lat./Long. coordinates and reading elevation from a table?
I would have to guess a table, because the stated accuracy of gps altitude is something like 30-50m (100-150 feet).

Here is what Garmin says:
How accurate is the GPS elevation reading?
 
That video doesn't show anything relevant without flight data. For all I know you could have manually flown it.
As far as your explanation about barometer and device barometer goes, nothing holds water. The bird doesn't give a **** about if you have or don't have a device with barometer.
Your logic is very very flawed.
 
I have read posts to that effect several times on this forum and have to question it. I have half a dozen GPS's for use in my vehicles. They're from differing mfrs. (Garmin, TomTom, etc.). At least two of them show elevation above sea level and are extremely accurate and repeatable. Are they not getting that info extrapolated from Satellites? Or is some software looking up the Lat./Long. coordinates and reading elevation from a table?
GPS is much less accurate for altitude than for horizontal position.
Nowdays many consumer units also have a barometer because raw GPS altitude data is next to useless.
To save typing it all out again, read this post (and post #43 following it), particularly the explanation from Garmin.
Height from where? | Page 3 | DJI Phantom Forum
 
The P3 doesn't use anything other than it's internal barometer. As far as it is concerned, the take off point is 0ft. All other heights are relative. Barometers are not perfect. They can fluctuate by as much as 10ft.
It can be off more than 10 ft. As I mentioned above, mine was consistently reporting 45 feet higher than it actually was at the end of a battery. I've got screen shots somewhere showing this. I suggest the OP fly his Phantom for 15 or so minutes and bring it back to a hover one foot above the takeoff point and look at what the height reading is. Then he'll know whether that's the problem.
 
It can be off more than 10 ft. As I mentioned above, mine was consistently reporting 45 feet higher than it actually was at the end of a battery. I've got screen shots somewhere showing this. I suggest the OP fly his Phantom for 15 or so minutes and bring it back to a hover one foot above the takeoff point and look at what the height reading is. Then he'll know whether that's the problem.
That's confusion between calibration and precision.
Your unit was out of calibration but it could still tell altitude differences of 1 or 2 feet.
 
The internal barometer is stupid accurate, hence the **** near perfect hover in a windless situation.
 

Recent Posts

Members online

No members online now.

Forum statistics

Threads
143,086
Messages
1,467,525
Members
104,965
Latest member
cokersean20