On the GO4 small "radar" indicator (the red AC symbol) is pointing about 3 degrees less than the actual aircraft direction.The Spark does not show any indication that it needs calibration. NO interference is indicated.

Please keep the answer a simple YES or NO to these two questions:

1. Can compass calibration affect this?

2. Can IMU calibration affect this?

YES,  NO    short enough ?

You do understand that it is the compass of your device in question, here, right?

JJBspark Posted at 2018-10-8 06:59
YES,  NO    short enough ?

Nice answer, JJB

JJBspark Posted at 2018-10-8 06:59
YES,  NO    short enough ?

You blew it.
1. You failed to comprehend the OR qualification.
2. You also missed the TWO distinct response areas.
3. However, you are totally in line with the quality of answers so typical of DJI.

Thanks

S-e-ven Posted at 2018-10-8 07:02
You do understand that it is the compass of your device in question, here, right?

So typical. More questions. Not answers.
How about a GO4 error OR a Spark calibration problem?
I am so accustomed to convoluted answers from DJI.
There seems to be no way around them no matter how hard I try.
Thanks anyway.

Calibrate the bloody compass in your phone.
Where do you think get the app the compass parameter?

S-e-ven Posted at 2018-10-8 07:02
You do understand that it is the compass of your device in question, here, right?

Hi 7,

50% ok.   

The red arrow gets it heading information from the drone compass.
The outer compass ring gets te information from the mobile device.
If the drone compass is OK then you see a red arrow pointing the same direction as the drone, if the device is calibrated OK than the North sign is pointing North.

So if the drone heading is set to the mobile device heading and both are calibrated OK than its 1 nice view....  ;-) if not [ and the drone compass is OK ] than pointing the red arrow towards the center brings the drone home (it looks only weird....)



cheers
JJB

Chuck101 Posted at 2018-10-8 07:13
You blew it.
1. You failed to comprehend the OR qualification.
2. You also missed the TWO distinct response areas.

OMG i blew it.  

1. nah, there was no OR option in your question
2. do not understand
3. uhmm, normally i give better quality answers....( i hope )

But, need any help ?

If your 3 degrees is offsett the direction of your drone that its a drone compass thing, if 3 degrees is related to the outer compass scale than its a mob device thing.

cheers
JJB

Hi, Sorry for the troubles that it caused. I would recommend calibrating the compass to see if there will be changes. Please let us know on the latest progress.

DJI Tony Posted at 2018-10-8 16:04
Hi, Sorry for the troubles that it caused. I would recommend calibrating the compass to see if there will be changes. Please let us know on the latest progress.

It has not really been a trouble. Mostly just a curiosity about this very minor  discrepancy.
Actually I pretty much assumed the compass calibration would affect this. I am going to recalibrate it anyway.
I was really looking for  a YES or NO on the IMU calibration affecting this. But, as usual, simple, straight forward answers are just not part of the DJI support culture.

Chuck101 Posted at 2018-10-8 16:27
It has not really been a trouble. Mostly just a curiosity about this very minor  discrepancy.
Actually I pretty much assumed the compass calibration would affect this. I am going to recalibrate it anyway.
I was really looking for  a YES or NO on the IMU calibration affecting this. But, as usual, simple, straight forward answers are just not part of the DJI support culture.

Hi, Thanks for your response. The IMU will take care of the stability of your aircraft and measure the acceleration of your aircraft whilst flying the drone. Please let us know on the latest progress.

DJI Tony Posted at 2018-10-9 10:22
Hi, Thanks for your response. The IMU will take care of the stability of your aircraft and measure the acceleration of your aircraft whilst flying the drone. Please let us know on the latest progress.

I did a compass re-calibration. I made no difference. The "arrow" still points a little to the left of where the aircraft is actually pointing. As I previously stated it is NOT a serious problem. I was merely curious.

Since a simple YES or NO answer about the IMU effect is not possible I will just have to make the assumption based on your response that it has no effect on this.

I am getting the same thing after updating 2 sparks with firmware .1000

they flew great with .701 but after .1000 upgrade and calibration .. several in fact.. the aircraft is pointing 9 degrees to the right vs where it thinks and reports its pointing..

on RTH the home point is in the far left of the screen when it use to be centered..

the aircraft flys RTH in a straight line.. just flying crabbed off angle..

it doesn't land in the direction it was pointing at takeoff.. you guessed it.. off 9 degrees... oh .. and my declination here is .. wait for it.. 9degrees so perhaps the compass algo was messed with on the latest firmware..

i also note that when the aircraft is powered up and motors started with the sticks.. that the camera picture will yaw / pan right.. without the aircraft moving or even being off the ground.. now as i understand it.. its a 4k sensor that DJI motion stabilizes yaw with cropping.. so the aircraft thinks its turning when its not even off the ground...

also when flying in sport.. with .701 the horizon moved correctly .. now left turns are mostly normal but right turns and the horizon dips into a left turn first and snaps into a right turn very jerky .. like compass dip where the compass will turn opposite to a turn initially before catching up.. again.. compass algorithm..>?

DeuceDriv3r Posted at 2018-10-9 11:38
I am getting the same thing after updating 2 sparks with firmware .1000

they flew great with .701 but after .1000 upgrade and calibration .. several in fact.. the aircraft is pointing 9 degrees to the right vs where it thinks and reports its pointing..

Did you downgrade back to .701 and the problem go away?

downgrading is more frought with peril than upgrading.. I am waiting out what DJI is going to do about it first.. both birds are under waranty still and if they dont have a fix in 10 days they are both going in and they can fix them on thier dime

It is usually the mobile device that is the issue.
There was a GO 4 app version out there that iPhones needed to be re-calibrated.

For those that may not remember...you get a message every once in a while to "tumble" your phone to calibrate the compass app.

pmshop Posted at 2018-10-9 14:26
It is usually the mobile device that is the issue.
There was a GO 4 app version out there that iPhones needed to be re-calibrated.

Can you elaborate further?
The North on the phone is dead-on.
It's the AC and the red arrow in GO4 that don't agree.
The red arrow is always slightly left of the true Spark direction.

Chuck101 Posted at 2018-10-9 14:32
Can you elaborate further?
The North on the phone is dead-on.
It's the AC and the red arrow in GO4 that don't agree.

Yes, try to "tumble" your phone while in the DJI GO 4 app, regardless of iOS or Android.
Flip it around like you see that 3 axis "flat spin" trainer for astronauts and at carnivals.

This syncs the red arrow with the craft orientation.
Don't ask me why it works, but it does.
Everything will tell you you have to do something with the craft to fix it...but not in this case.

Its not the phone that is pure bunk.

The magnetometer in the spark determines how it points to the n marker on the app. Now if your n marker on the app or the maps are not aligned then yes that is the phone.

But the app an the phone have nothing to do with how the bird flies. I can send it 5000 ft out and turn the remote and app off and the bird will fly home and SHOULD have the home point centered in the video the whole way back. Not in the left hand edge of the screen

Ok, I just write, like I see it.
(Think my experience with the radar, bird is not where it shows, calibrate compass of phone, bird is where it shows, more or less accurate is supporting this)

The radar has 2 functions:
1. Telling the pilot, in which direction from his position the bird is
2. Telling the pilot, what orientation the bird has right now.

Function one is totally depending by the devices on board compass.
Every time the difference is not helpfull, doing "the 8" with a compass app is correcting that to just some degree difference.


To function 2 I have just not really seen there bad (unuseful) info, yet.
But I have not much experience with it, I have to admit
(I just use "1" for some seconds, in case I lost the bird b/c display watching or such out of my eyes, to know where I have to search)
For getting the bird back, if really out of sight, since I am flying "mobil data on", almost always, I use the map view, often in sat mode.
there you have this nice path in green, which shows the shortest way. Even in Atti!
I just would like to have google maps in the app, or nokia here ;-)

Question:
As he bird did not shown right in the radar, has anyone used the RTH button, to let the bird turn its nose direct to the homepoint?
Perhaps to confirm the wrong readings?

That would be an IMU thing, methinks

pmshop Posted at 2018-10-9 16:06
Yes, try to "tumble" your phone while in the DJI GO 4 app, regardless of iOS or Android.
Flip it around like you see that 3 axis "flat spin" trainer for astronauts and at carnivals.

I'll be damned! It did affect this indication.
While it did not correct it 100% it DID improve it tremendously.
After several iterations of this procedure observing its effect it ended up with the red arrow pointing about 1 degree to the right (it was about 5 degrees to the left) of the true AC direction.

but did it change the blue triangles orientation to N ... not necessarily to the center of the indicator...

the rotation of the whole 'radar' is determinant on the phones understanding of orientation... I suppose its possible that a grossly out of calibration of the phone would also skew its values for magnetic deviation and the placement of the N marker on the 'radar' ring... but the rotation of the triangle (drones heading) is mearly plotted agains the phones understanding of N .   its position relative to the center is GPS position based...

generally as part of preflight checks .. I pull up the full screen map and while holding the controller in the 'in flight' orientation and with the bird in the hover .. I check the map for correct orientation to geographical features.. i.e. are the roads on the map lining up with reality.. or other features... if this alignment is good.. my phones understanding of N is good.. and thus my radar N will be good...

if the blue triangle represents the correct orientation of the bird in the hover.. than I know that its understanding of N (compass cal) is in sync with my phone.  

what I have noticed after updating to .1000 is that the spark is generally not orienting correctly durning this preflight...

for example I will on a football field line up myself and the craft on some chalk lines.. immediately after getting into a hover the bird will slightly drift heading but it wont match what the triangle is doing... thus there is a decoupling of what the physical craft is doing vs what the gyros think is going on and is reported in telemetry

Chuck101 Posted at 2018-10-10 05:23
I'll be damned! It did affect this indication.
While it did not correct it 100% it DID improve it tremendously.
After several iterations of this procedure observing its effect it ended up with the red arrow pointing about 1 degree to the right (it was about 5 degrees to the left) of the true AC direction.

Like I said...don't ask me why it works but it does

The nay-sayers can continue on but I just pass on real experiences that I have been through.
I agree, it makes no sense...but it does work.

And Chuck, you may have to do that again every once in a while.

Great discussion guys, even allowing for the angst. I got a lot from it about the function of the compass ring on the Go4 app.

One question though. If the arrow on Go4 compass ring shows the direction that the Spark compass is indicating, is there any Go4 indication of where the gyro in IMU is indicating? (Might be useful when compass error forces an Atti mode.)

if you can see 3 degrees on that indicator you have better eyes than me.. that would be less than 1% of the dials circumference.... if my spark or any dji drone for that matter had a compass that was accurate to 3 degrees I would be ecstatic  

I have taken out a magnetic compass

aligned a landing mat H to magnetic north..

put an aircraft on the mat lined up with the H  i.e. facing mag north

calibrated the phone/ac so that launching go 4 produces a radar that when facing mag north everything more or less lines up. N at the top of the radar and the red aircraft more or less lined up with that N

after several minutes of flying around .. the iPhone's idea of north generally goes nuts.. and can be way off... like a lot..

when I bring the A/C back for a precision landing almost every time the spark/mavic has its nose pointing several degrees right BUT yet the radar shows the red chevron pointing at N so it thinks it landed the same way it took off.. lately this angle has not been too great I would say in the area of about 10 degrees which is just about what my mag deviation is here but I would have to look and see if its facing true north (i.e. correct deviation) or if its off in the wrong direction for the declination angle.. and dji still never got back to me if their flight controller does its headings in true or mag...


edit

ok so went and check and I am at 9ish west declination so that would put mag north to the left of true north.. so the aircraft seems to more align itself with true.. I think I have my iphone prefences to set the compass to show mag north.. might have to change that to true north to see if it makes much difference. perhaps the dji FC is programmed to fly with all calcs converted to true north.

as an assisde.. interesting trivia.. but the magnetic strength on record for my area is about 48000 nanoteslas and it matches what the drone reports in its csv data ... doing more research dji's mag interference calculations are basically measuring the difference between expected earth gravitational contstant vs measured .. so it does not detect if the heading it being pulled off per-se .. but whether the field strength is greater than earth constant for a geographical area

so theoretically if a magnetic or metalic inteference forces was in proximity to the drone such that it was affecting its heading but was close to overall expected gravitational constant. your heading would be off but your compass interference bar would show no interfernce ...

DeuceDriv3r Posted at 12-5 14:59
if you can see 3 degrees on that indicator you have better eyes than me.. that would be less than 1% of the dials circumference.... if my spark or any dji drone for that matter had a compass that was accurate to 3 degrees I would be ecstatic

Same here!

1/no
2/no
Just try lifting your craft while turned on turn left turn right if the red triangle moves the same as aircraft then everything is ok to fly.
Compass Calibration craft won’t help, if the attitude of your aircraft is good and aircraft is flying well and you have good values shown in your app then your IMU is fine.

1/no
2/no
Just try lifting your craft while turned on turn left turn right if the red triangle moves the same as aircraft then everything is ok to fly.
Compass Calibration craft won’t help, if the attitude of your aircraft is good and aircraft is flying well and you have good values shown in your app then your IMU is fine.

DeuceDriv3r Posted at 12-5 15:09
I have taken out a magnetic compass

aligned a landing mat H to magnetic north..

Thread you might find of interest on Magnetic Field, Compasses, Interference.

Another Thread you might find of interest on Magnetic Field, Compasses, Interference.

And one more Thread you might find of interest on Magnetic Field, Compasses, Interference.

Or Not!  

#1-Yes

#2-No

From what I have been able to dig out, 3-degrees is not an issue.  Somewhere around 15-degrees off is when problems start.

DeuceDriv3r Posted at 12-5 14:59
if you can see 3 degrees on that indicator you have better eyes than me.. that would be less than 1% of the dials circumference.... if my spark or any dji drone for that matter had a compass that was accurate to 3 degrees I would be ecstatic

Has been my point for a long time.   

Nearly impossible to discern 3-degrees of accuracy via GO-4 Indicator.  Really need an option to enable a digital compass heading readout on display.  Similar to option to enable digital battery voltage on display.

Adding to impossible is issue of discerning exactly which way Drone is pointing.  Almost need a built-in Laser-pointer, or Peep sights to determine what direction Drone is pointing.   There isn't even a Line on top like Weather stations have for wind direction.

Chuck101 Posted at 10-10 05:23
I'll be damned! It did affect this indication.
While it did not correct it 100% it DID improve it tremendously.
After several iterations of this procedure observing its effect it ended up with the red arrow pointing about 1 degree to the right (it was about 5 degrees to the left) of the true AC direction.

hi chuck, unrelated question but would you like an awesome and high quality garfield profile picture??

ManOfSand Posted at 12-6 13:10
hi chuck, unrelated question but would you like an awesome and high quality garfield profile picture??

Sure. I'll try it.

JJBspark Posted at 2018-10-8 06:59
YES,  NO    short enough ?

i liked ur answer. good humore