Barometer calibration

Joined
Sep 3, 2016
Messages
5
Reaction score
2
Age
52
Any advice / experiences to share on calibrating the barometer on a P4 ?- after flying at high altitude (around 800 m asl) I now get very big errors in altitude estimates in the image data when flying nearer sea level.....
Could this be affected by temperature (flying at < 5 degs celcius / 41 fahrenheit)...
 
Joined
Apr 2, 2016
Messages
1,873
Reaction score
986
Location
Canada
The barometer (non calibratable) takes a reading and sets your take off point as 0 feet. (Altitude is now 0, not the actual elevation from sea level) As you climb with your craft the atmospheric pressure will be less and the barometer will read and the differential pressure from take off to present will be caculated and your altitude will be shown as calculated. Any variation in atmospheric pressure caused by weather systems comming in or temperature change, humidity will vary your readings. This is why you should use these values as estimates because even on a good day there usually is a variance in altitude readings. It dosn't take mutch variance, 0.266 psi is about 500 feet, even the velocity of the air going by your drone will reduce air pressure to the barometer somewhat.
 
Joined
Sep 3, 2016
Messages
5
Reaction score
2
Age
52
Hi Erised - and thanks for your response - yes, this was quite an issue. Actually it is the altitude estimated in the image exif file that seemed to be causing the problems when generating a surface model (using Agisoft photoscan) -
Here the estimated altitude of the images was around 175 m (I believe the EXIF altitude is the altitude above sea level - in the mountains it was around 800 m, while the drone only flew about 30 m above the ground). Modelling software like Agisoft uses the references in the EXIF data.

So, after reflying the same (very small) area again a week later (under calm conditions) the estimated altitude was suddenly 55 m and not 175 m, and the model generated from the images fitted very precisely (nett overall error of 6 mm, not > 2m as it had been).

In the meantime, I had done an IMU calibration as suggested by Mark the Droner (above).
 
Joined
Apr 2, 2016
Messages
1,873
Reaction score
986
Location
Canada
Hi Bruce, a 6mm overall error is very respectable, In my EXIF files the altitude is AGL. The only place that I can figure out you will have actual take off elevation recorded from the drone is in flight records in GO, top altitude(android) or top take off altitude(ios), on the left hand side under the growth value thing and your totals. The value shown as far as makes sense to me is that this value has to be from sea level. To test this I synced all my flight records (in my computer also) then deleted all the records in GO then took off from a point way higher than I flew and the number in top altitude was about 13 feet off from my take off point of the elevation I could get from google earth (I know not the best data). To bad this number is only the max of all our flights because I think the craft has the capacity to track elevation above sea level.
 
Last edited:
Joined
Jul 26, 2016
Messages
1,539
Reaction score
480
Yeah, mine isn't so good at altitude numbers either. I sort of worry about using it with some autonomus flight app in that it drifts by 50 feet at times on the screen from take-off at zero. I think the Autopilot app can take the barometer reading off the tablet or phone and incorporate the two somehow since the ground based one may be more stable than the one in the drone and the turbulence it may be getting into. I dunno though, and maybe the P4P might be a bit tighter in the altitude reporting as the hovering drift specs do seem better than the P4 specs and the Inspire 2 even more. I know from watching their Flight School videos they are not too keen on primary setup menu distances under 33 foot tolerances for Autoflight missions.
 
Joined
Apr 2, 2016
Messages
1,873
Reaction score
986
Location
Canada
Yeah, mine isn't so good at altitude numbers either. I sort of worry about using it with some autonomus flight app in that it drifts by 50 feet at times on the screen from take-off at zero. I think the Autopilot app can take the barometer reading off the tablet or phone and incorporate the two somehow since the ground based one may be more stable than the one in the drone and the turbulence it may be getting into. I dunno though, and maybe the P4P might be a bit tighter in the altitude reporting as the hovering drift specs do seem better than the P4 specs and the Inspire 2 even more. I know from watching their Flight School videos they are not too keen on primary setup menu distances under 33 foot tolerances for Autoflight missions.
Inspire2 has 2 barometers and maybe you will get the calculated average. A 33 feet tolerance is a lot, GPS elevation is beter than that if we could see it because im pretty sure the craft knows it. The reason I say that it also uses gps elevation is that a few months ago I was flying purpously towards a on-comming storm front to get a flight in before the storm arrived. I was 365 feet and a ways out when I got a critical barometer error, red screen and all and a use caution when landing. GO maintaind the critical message all the way back home. Pucker factor was high, all the way back because the warning is that i had barometer error or no barometer and my maximum altitude was 5333 feet in GO flight records(checked after). All this happening the craft maintained 365 feet altitude for the whole flight home (365 feet live in GO and the same in flight records). It maintained perfect altitude cruising at 70+mph in a hell of a tail wind all the way home with possibly no barometer, thats why I think it also uses gps for elevation as a back up. I wish dji would tell us or let us use it because tracking 17 satellites you can't tell me that I don't have enough data for elevation.
 

New Threads

Forum statistics

Threads
139,514
Messages
1,439,328
Members
101,314
Latest member
GI Joe