Please select Into the mobile phone version | Continue to access the computer ver.
[New pilot] Need tips not to lose the sparky during my voyage
12Next >
49420 55 2018-4-10
Uploading and Loding Picture ...(0/1)
o(^-^)o
ashikaumanga
lvl.2
Flight distance : 19821 ft
Japan
Offline

I bought my first drone a week ago just for my upcoming Peru adventure.
I will be travelling in high plains of the Andes to deep dark corners in Amazon jungle.
During the last couple of days, I got myself familiar with all the features.
After going through a dozen of threads in the forum, I want to make sure that I won't lose my Sparky..Especially due to the weird "flyaways"...

Here is a list of checks/steps I am going to follow :

1) Calibrate the compass before flying: Since I will be travelling to Peru from Japan
2) Set Home Point every time before flying.
3) Do not fly above 4000m: Most of the places in Cusco area are higher than 3000m, but some places, especially higher mountains can be more than 4000m sea-level
4) Do not attach peripherals: No gimbel covers, camera covers, stands..etc.. Maybe only properllar covers.
5) Do not fly above 100m of height eventhough I have RC.
6) Do not fly beyond 200m of distance eventhough I have RC.
7) Check for magnetic interference before flying.
8) Make sure I have enough juice : Own 3 batteries

Is there anything else I can add to this list?

2018-4-10
Use props
msinger
Captain
Flight distance : 183647 ft
  • >>>
United States
Offline

Here are some modifications I'd make to that list:

1) Don't calibrate the compass before flying. There's no need to in most cases. See more details here.

2) The home point will normally automatically be set by the time DJI GO shows your Spark is ready to fly. Check the map in DJI GO to ensure you see the "H" marker and it's in the correct location.

And some additions:

9) Make sure there are no trees or other obstacles between the Spark and remote controller.

10) Be mindful of the wind. If you fly high and the wind is strong, your Spark might not be able to make it back to the home point in time. It could be tough or impossible to retrieve it if lost in a jungle.

2018-4-10
Use props
DJI Elektra
Administrator
China
Offline

Thanks for your hard work. Fly in a good environment is also important.
2018-4-10
Use props
InTheReeds
Second Officer
Flight distance : 347014 ft
United States
Offline

Also, disable bluetooth on any devices near the remote. Phone, watch, etc.
2018-4-10
Use props
hallmark007
Captain
Flight distance : 4151791 ft
  • >>>
Ireland
Offline

Use this exercise it will help.


https://forum.dji.com/forum.php? ... &fromuid=260008
2018-4-10
Use props
S-e-ven
Captain
Flight distance : 3541229 ft
Thailand
Offline

Since not every Spark seem to set the homepoint just in the beginning of the P-GPS Mode and Start: Make sure, yours did it before leaving the homepoint.
Can be done in 5 or 10 m height. It is just a bit 'uncomfortable', if  'Homepoint set, please check on map' happens 50-100m away, b/c the drone wasn't happy with 10 Sats for the homepoint, but allowed to fly GPS-Mode ;-)
2018-4-10
Use props
Gunship9
Captain
United States
Offline

I would check the channel page on the DJI Go 4 app if flying in a new area.  You need to check the radio spectrum for heavy use before sending your Spark up in it.  Like general aviation aircraft such as the Cessna, it is not always safe to fly.  In fact it is not safe to fly as often as it is safe to fly.  
2018-4-10
Use props
ChrisJG
Captain
Flight distance : 455061 ft
United Kingdom
Offline

Good luck with your trip, it sounds as if it will be an epic adventure and I'm sure we all are looking forward to seeing your footage. How are you transferring the files from your memory card whilst you are in Peru? Have you some back up method in place?
2018-4-10
Use props
Fer García
lvl.2
Flight distance : 3054 ft
Mexico
Offline

Check that there are no large birds flying over the area; it would be nasty if your Spark were eaten by a condor from the Andes!

Have a nice trip and don't forget to share your footage with us!

2018-4-10
Use props
ashikaumanga
lvl.2
Flight distance : 19821 ft
Japan
Offline

ChrisJG Posted at 2018-4-10 11:26
Good luck with your trip, it sounds as if it will be an epic adventure and I'm sure we all are looking forward to seeing your footage. How are you transferring the files from your memory card whilst you are in Peru? Have you some back up method in place?

only can backup to my laptop when I get back to the hostel at the end of the day.
2018-4-10
Use props
ashikaumanga
lvl.2
Flight distance : 19821 ft
Japan
Offline

Fer García Posted at 2018-4-10 11:38
Check that there are no large birds flying over the area; it would be nasty if your Spark were eaten by a condor from the Andes!

Have a nice trip and don't forget to share your footage with us!

will definitely do
2018-4-10
Use props
sSkyPilot
First Officer
Flight distance : 290782 ft
United States
Offline

Fly with the skills you'v learned...  Smalls steps... Get the basics ....  understand the software, wifi, software, AC and RC.
2018-4-10
Use props
Ram-UK
First Officer
Flight distance : 405103 ft
United Kingdom
Offline

Keep your AC within VLOS at times. Also learned to fly with in ATTI mode. It will help if you encounter trouble with GPS.
2018-4-11
Use props
RjMllr
Second Officer
Flight distance : 17909 ft
Philippines
Offline

Add this -> Always check the battery locks if they are properly installed
2018-4-11
Use props
CSmithKev
lvl.1

Kenya
Offline

You could consider travelling with a small juicer as well for your flying journey. Imagine the possibilities of making Peruvian Chicha Morada juice on your own with some pineapples, apples, and lemons! By doing so, you will not only have enough juice for your gadget but for your body as well.
2018-7-16
Use props
Antipaxi
First Officer

Romania
Offline

Here's my correspondence with a peruvian official, you might find it useful. I let go of the idea of bringing my Spark to Peru.


Mr. Adrian:



ADUANA -AIRPORT

About your questions,   when you will arrive in our country, the entity ADUANA (in the airport) will verify and determine whether internal or not the drone, it is jurisdiction of them.



You don´t need a permission only bring your one (01) drone to Perú. (Attached RESOLUCIÓN DIRECTORAL N 479-2016-MTC-27 31 10 2016.pdf), please you will show document to ADUANA, and you will say them, to use drone is only transitory



If ADUANA confiscates the drone, indicate the owner to approach the Ministerio de Transportes y Comunicaciones - Dirección General de Concesiones , where they will give you a document that will allow you to make the process of getting your drone in the airport.



Concerning the placement of equipment Drones, you can make consultation in the email  avasquez@mtc.gob.pe and ymaguina@mtc.gob.pe They will guide you all the procedures related ADUANAS- Airport.



FLIGHT IN PERU(RPAS)

Only you will can fly drone in rural zone in Peru with according Norma Técnica Complementaria NTC:001-2015, you must apply Point (8) Regulación(Regulation) (e) Limitaciones de Operación (Operating limitations in RPAS)



http://www.mtc.gob.pe/transporte ... /normas/normas.html



In Urban Zone is necessary an authorization of the Dirección General de Aeronaútica Civil – DGAC. Please you give the local tour company about ruler and procedures (attached) ,they can write me or call me for more information.



ABOUT MACHUPICHU:

you can make question to Entity Servicio Nacional de Áreas Naturales Protegidas por el Estado



http://www.sernanp.gob.pe/

sernanp@sernanp.gob.pe

https://www.facebook.com/Sernanp ... 211888321/timeline/



http://www.culturacusco.gob.pe/i ... formulario-contacto
2018-7-16
Use props
Northwood
First Officer
Flight distance : 266949 ft
Canada
Offline

Wow, sounds like the greater risk is the authorities, not the technology or pilot skills...
2018-7-16
Use props
N2QLT
Second Officer
Flight distance : 29222 ft
United States
Offline

Practice practice practice.
2018-7-16
Use props
Mirek6
Second Officer
Flight distance : 239319 ft
Canada
Offline

msinger Posted at 2018-4-10 03:09
Here are some modifications I'd make to that list:

1) Don't calibrate the compass before flying. There's no need to in most cases. See more details here.

msinger,

I know it is already past the fact - however, since the thread is active again.

Who is maintaining the descriptions you are referring to regarding compass calibration?

There is at least one thing - important one - missing. Compass should be calibrated if you travel long distances. Ashikaumanga was travelling from Japan to Peru. It is almost certain that ge-magnetic field properties in Peru are different than in Japan. His compass should be calibrated in Peru no matter if DJI GO prompts for it or not. I have seen multiple cases where DJI GO showed normal yet Spark experienced issues during flight which went away after calibration.

As a matter of fact DJI does recommend to re-calibrate compass if you move 50 km or more from the location you calibrated it before. This recommendation is not in DJI Spark manual but it is in DJI Mavic Air manual, which, for all intents and purposes, does not differ in magnetometer or recommended procedures.

Mirek

  
2018-7-16
Use props
msinger
Captain
Flight distance : 183647 ft
  • >>>
United States
Offline

Mirek6 Posted at 2018-7-16 08:22
msinger,

I know it is already past the fact - however, since the thread is active again.

DJI GO does not always show a warning when the compass is being negatively affected by some type of external magnetic force. Before taking off, you should always do the following:

1) Make sure the Spark is never near magnetic metal objects after it's powered on.

2) When the Spark is on the ground at the takeoff point, look at the map in DJI GO and verify that the aircraft symbol is pointing in the same direction as the Spark.

3) Make sure the aircraft symbol in DJI GO is not slowly rotating as the Spark is sitting on the ground.

I cannot comment on the multiple cases you're claiming you saw without links that provide a complete explanation of those incidents. Feel free to share them if you want to dig in deeper.
2018-7-17
Use props
Mirek6
Second Officer
Flight distance : 239319 ft
Canada
Offline

msinger Posted at 2018-7-17 03:56
DJI GO does not always show a warning when the compass is being negatively affected by some type of external magnetic force. Before taking off, you should always do the following:

1) Make sure the Spark is never near magnetic metal objects after it's powered on.

msinger,

You misunderstood me and quoted something which is quote obvious.
Everything you claim is true and everything the page you referred to is true.
I just commented that one very important part is missing form the instructions which are given on the page you referred to - need for re-calibration of compass when you travel long distances.

No need to dig deeper into specific cases (which, as I said, abound on this forum).
It is just physics about geo-magnetic field and how magnetometer works. Magnetometer measures geo-magnetic field (which is very weak hence all the precautions you are talking about are very valid) in three axis. Based on these measurements it establishes exact direction of magnetic North. Geo-magnetic field properties differ in different places on Earth. They are affected by geographical location, presence of various geological formations and magnetic disturbances with various sources. This is why you need to calibrate when you travel long distances. And you need to calibrate even if DJI GO shows that everything is normal (BTW - the page you referred to does confirm this as well - do not believe DJO GO app).

Besides - check Mavic Air manual. It is right there. Re-calibrate compass if travelling more than 50 km or not using your drone for more than 30 days.

In summary - I believe we are in agreement - I was just pointing out missing information which is important.

Mirek
2018-7-17
Use props
Tyr76
lvl.3
Flight distance : 28503 ft
Netherlands
Offline

Ehmmm... also a [new pilot] so forgive me if this is a dumb question...Why NOT use gimbl cover,or anything to protect the drones fragile parts?especially when travelling I assume people tend to wanna protect their expensive 'toys'...
(Me,I throw my stuff in a backpack which never leaves my side.I know ppl ll think I'm crazy but the spark fits cozily in the goggles headband, pretty much protected by the pgytech prop guard-with-top.)
2018-7-17
Use props
Mirek6
Second Officer
Flight distance : 239319 ft
Canada
Offline

Tyr76 Posted at 2018-7-17 07:55
Ehmmm... also a [new pilot] so forgive me if this is a dumb question...Why NOT use gimbl cover,or anything to protect the drones fragile parts?especially when travelling I assume people tend to wanna protect their expensive 'toys'...
(Me,I throw my stuff in a backpack which never leaves my side.I know ppl ll think I'm crazy but the spark fits cozily in the goggles headband, pretty much protected by the pgytech prop guard-with-top.)

Tyr76,

Does you question have anything to do with what was described above? I do not see any connection.

Mirek
2018-7-17
Use props
msinger
Captain
Flight distance : 183647 ft
  • >>>
United States
Offline

Mirek6 Posted at 2018-7-17 05:05
msinger,

You misunderstood me and quoted something which is quote obvious.

You're suggesting the compass needs to be calibrated after traveling. I'm suggesting that is does not need to be calibrated in most cases. While some DJI manuals recommend calibrating after traveling and other DJI documentation recommends calibrating before every flight, it's really not necessary most of the time. Will calibrating the compass before every flight hurt anything? No. You'll likely just waste a lot of time.

Here are times when you should definitely calibrate the compass:

  • Compass interference values are out of whack or DJI GO is displaying a compass error (check area for magnetic metal objects before calibrating)
  • The Spark is circling in flight while hovering in place (also check for other possible causes)
  • New metallic equipment has been attached or removed from the Spark (e.g. GPS tracker)
  • If you just degaussed your compass (don't degauss the compass unless instructed)

It appeared that you were suggesting many incidents have been caused by not calibrating the compass after traveling. If that's what you meant, then, yes, I'd definitely like to delve into those cases some more. If we do take the time to do that, I think we'll find the following causes in a majority (or all) of those cases:

  • The issue was caused by powering on the Spark near a magnetic metal object
  • The issue was caused by taking off near a magnetic metal object
  • The compass was not calibrated after some type of magnetic metal accessory was attached to the Spark
  • The Spark was stored near a strong magnetic source (e.g. a subwoofer in a car)
  • The compass was defective
  • The OP did not provide enough information in order to determine what caused the compass problem



2018-7-17
Use props
Kingram
First Officer
Flight distance : 84063 ft
United States
Offline

Even though these are for a Mavic there is very good info in them.




2018-7-17
Use props
Mirek6
Second Officer
Flight distance : 239319 ft
Canada
Offline

msinger Posted at 2018-7-17 09:07
You're suggesting the compass needs to be calibrated after traveling. I'm suggesting that is does not need to be calibrated in most cases. While some DJI manuals recommend calibrating after traveling and other DJI documentation recommends calibrating before every flight, it's really not necessary most of the time. Will calibrating the compass before every flight hurt anything? No. You'll likely just waste a lot of time.

Here are times when you should definitely calibrate the compass:

msinger,

Yes. All true.
I just do not understand why people are so defensive regarding calibrating compass after moving long distances. It is additional insurance policy and it is recommended by both scientific explanation how magnetometers work and by what DJI recommends.

I do agree that the list of causes you provided is probably much more important and detrimental to proper operation of compass than re-calibration due to changes in the geo-magnetic properties in different locations. Ge-magnetic field deviations have definitely weaker impact than many of the other things you listed. But, this does not change the fact, that if you want to protect yourself to the fullest, just do it as well. And no, I do not recommend re-calibrating before each flight.

Yes - I have seen many cases on this forum where person flew on a vacation and than had trouble with compass. Can I definitely say that it was caused by differences in geo-magnetic field rather than, let's say, storing his Spark on a plane in a vicinity of somebody's hard-drive? No, I cannot. And, frankly, it is impossible to do. I just asked these people - did you re-calibrate your compass? The answer was usually - no, I did not.

Latest case was from South Africa from, perhaps three weeks ago. Read entire discussion I had with Nilesh here: https://forum.dji.com/forum.php?mod=viewthread&tid=154710. And check my question and his response in posts 52 and 53. He was 400km from the location he re-calibrated his compass last time. No other interference sources since he was in the middle of nowhere.

Does this conclusively prove that Nilesh's problem was caused by travel and geo-magnetic field differences? No, it does not. Is it likely? Perhaps, it is. Certainly, it is possible.

We will never find a definite proof by analyzing cases from this forum. There is not enough volume and there is not enough 100% conclusive data. Due to the nature of such research, this can only be done by scientific experiments in a controlled conditions. However, such research was done by scientists which developed magnetometers and there are plenty of scientific articles about how and when to calibrate the magnetometer - including when moving long distances.

Again msinger, I really do believe that we both understand and agree on what is a good practices and how it should be done.

Mirek
2018-7-17
Use props
msinger
Captain
Flight distance : 183647 ft
  • >>>
United States
Offline

Mirek6 Posted at 2018-7-17 10:16
msinger,

Yes. All true.

According to the flight log, the Spark was not flying in the direction shown by the GPS coordinates at the start of the flight. The heading looks like it was off by about 25 degrees. That's normally caused by some type of magnetic metal source at the takeoff point.
2018-7-17
Use props
BudWalker
First Officer
Flight distance : 150877 ft
  • >>>
United States
Offline

Mirek6 Posted at 2018-7-17 05:05
msinger,

You misunderstood me and quoted something which is quote obvious.


It is just physics about geo-magnetic field and how magnetometer works. Magnetometer measures geo-magnetic field (which is very weak hence all the precautions you are talking about are very valid) in three axis. Based on these measurements it establishes exact direction of magnetic North. Geo-magnetic field properties differ in different places on Earth. They are affected by geographical location, presence of various geological formations and magnetic disturbances with various sources. This is why you need to calibrate when you travel long distances.


This is all true but a compass calibration doesn't measure or determine geomagnetic field properties. A calibration can only measure and then compensate for magnetic properties that rotate with the AC. The data arising from external magnetic effects are effectively ambiguos and, therefore, can not be used to compensate the magnetometer data.


Since the P3 the geomagnetic declination has been determined by using a model based on GPS coords. This can be seen by noticing that at batteryOn the Yaw and magYaw (heading determined by the magnetometers) are equal. When gpsHealth gets above 3 the Yaw is then adjusted by adding the GeoDeclination and magYaw.



2018-7-17
Use props
Mirek6
Second Officer
Flight distance : 239319 ft
Canada
Offline

BudWalker Posted at 2018-7-17 13:49
It is just physics about geo-magnetic field and how magnetometer works. Magnetometer measures geo-magnetic field (which is very weak hence all the precautions you are talking about are very valid) in three axis. Based on these measurements it establishes exact direction of magnetic North. Geo-magnetic field properties differ in different places on Earth. They are affected by geographical location, presence of various geological formations and magnetic disturbances with various sources. This is why you need to calibrate when you travel long distances.

Bud Walker,

Yes - perhaps I choose imprecise words to describe what magnetometer does but the final result and final conclusion and recommendation for Spark users is identical :-)

Where do you see metrics you are talking about (e.g. magYaw?). For my analysis I use csv file which is generated from a flight record of Spark. I do not see the data you are talking about in this file. Perhaps you are using different decoder of the DJI flight record? I use TXTLogToCSVTool published on phantom site.

Mirek
2018-7-17
Use props
Mirek6
Second Officer
Flight distance : 239319 ft
Canada
Offline

msinger Posted at 2018-7-17 12:09
According to the flight log, the Spark was not flying in the direction shown by the GPS coordinates at the start of the flight. The heading looks like it was off by about 25 degrees. That's normally caused by some type of magnetic metal source at the takeoff point.

msinger,

I did suspect compass interference in my first analysis of that case (see post 21) and I asked Nilesh about it. He was vehement that there was nothing around which could have caused magnetic interference. He was in the middle of the forest.

Out of curiosity, how did you determine mismatch of about 25 degrees between the direction Spark was flying and GPS co-ordinates during start?

As far as I can see in the first 12 seconds pilot is moving rudder back and forth adjusting yaw and, at the same time, he is using elevator to propel Spark forward. This is very dynamic situation and measuring any possible deviations are difficult and are bound to be imprecise.

At about 12 seconds he did direct Spark to a final destination yaw of about 10 degrees and started flying straight. It's path is not exactly in line with the direction where it is flying (based on GPS) but it does compensate for breeze from the Westerly direction so any conclusive findings here are difficult.  However, this small mis-alignement does not change while Spark is gaining height. Which led me to suspect that Nilesh might have not re-calibrated his compass. I asked and he did confirm.

I would be interested to see where you got your 25 degree mis-alignement at the beginning of the flight from. My point here is not to argue minutia of flight analysis (which is often speculative given plethora of various inputs), but rather to learn something.

Mirek
2018-7-17
Use props
msinger
Captain
Flight distance : 183647 ft
  • >>>
United States
Offline

Mirek6 Posted at 2018-7-17 15:31
msinger,

I did suspect compass interference in my first analysis of that case (see post 21) and I asked Nilesh about it. He was vehement that there was nothing around which could have caused magnetic interference. He was in the middle of the forest.

Compare the heading in the TXT flight log to the direction the aircraft is actually heading (according to the GPS data) when the pilot has only the elevator in the up position.
2018-7-17
Use props
Mirek6
Second Officer
Flight distance : 239319 ft
Canada
Offline

msinger Posted at 2018-7-17 16:21
Compare the heading in the TXT flight log to the direction the aircraft is actually heading (according to the GPS data) when the pilot has only the elevator in the up position.

Yes - this is what I explained in my post above. My conclusion is different than yours although slight difference between Spark's yaw and flight direction is there.

Mirek
2018-7-17
Use props
BudWalker
First Officer
Flight distance : 150877 ft
  • >>>
United States
Offline

Mirek6 Posted at 2018-7-17 15:04
Bud Walker,

Yes - perhaps I choose imprecise words to describe what magnetometer does but the final result and final conclusion and recommendation for Spark users is identical :-)

Yes - perhaps I choose imprecise words to describe what magnetometer does but the final result and final conclusion and recommendation for Spark users is identical :-)

No. A compass calibration can not determine and compensate for geomagnetic properties because they are external to the AC. It's mathematically impossible. It's incorrect to say that flying in a new location requires that the compass be calibrated.


Where do you see metrics you are talking about (e.g. magYaw?). For my analysis I use csv file which is generated from a flight record of Spark. I do not see the data you are talking about in this file. Perhaps you are using different decoder of the DJI flight record? I use TXTLogToCSVTool published on phantom site.

To see this information you'll need to look at the contents of the .DAT file which resides on the AC itself. Here are instructions for retrieving the .DAT
http://www.datfile.net/DatCon/retrieveV3Dat.html


You can use CsvView or DatCon to process the .DAT file
http://www.datfile.net/index.html
2018-7-18
Use props
Mirek6
Second Officer
Flight distance : 239319 ft
Canada
Offline

BudWalker Posted at 2018-7-18 04:54
Yes - perhaps I choose imprecise words to describe what magnetometer does but the final result and final conclusion and recommendation for Spark users is identical :-)

No. A compass calibration can not determine and compensate for geomagnetic properties because they are external to the AC. It's mathematically impossible. It's incorrect to say that flying in a new location requires that the compass be calibrated.

BudWalker,

You say:” No. A compass calibration can not determine and compensate for geomagnetic properties because they are external to the AC. It's mathematically impossible. It's incorrect to say that flying in a new location requires that the compass be calibrated.”

I am not sure whether continuing this discussion makes sense. Let’s agree to disagree. Here is why:

Your sentence is entirely incorrect. It implies that magnetometer does not measure geo-magnetic filed “because it is external to AC”. ???
If it does not measure geo-magnetic filed than what does it measure and how does it figure out magnetic north? From inside of AC (as you seem to imply)?

Why would you turn Spark in two axis during calibration? This is not IMU calibration which does not care about geo-field and only cares about what being flat means in multiple axis.
You turn Spark so its magnetometer can carefully measure geo-magnetic field around it in different positions (360 degrees horizontally and 360 degrees vertically). This is why you cannot have anything which disturbs geo-magnetic field in the vicinity where you calibrate Spark. Geo-magnetic field is very weak and any distortions (which are outside of AC) would be detrimental. Distortion from inside of AC are taken into account during measurements because they do not change – they fly with Spark.

Please refer to this article from Teslabs explaining exactly how magnetometer reads external geo-magnetical field, how field properties differ in different places and what is involved in calculating magnetic north. This article has three maps of earth field magnetic models (intensity, inclination and declination). https://teslabs.com/articles/magnetometer-calibration/
Of course this is just an example – you can find myriad similar scientific articles.

Thank you for tips on .dat. I’ll check them out.

Mirek
2018-7-18
Use props
BudWalker
First Officer
Flight distance : 150877 ft
  • >>>
United States
Offline

Mirek6 Posted at 2018-7-18 07:53
BudWalker,

You say:” No. A compass calibration can not determine and compensate for geomagnetic properties because they are external to the AC. It's mathematically impossible. It's incorrect to say that flying in a new location requires that the compass be calibrated.”

Your sentence is entirely incorrect. It implies that magnetometer does not measure geo-magnetic filed “because it is external to AC”. ???

You're mistaken. What I said is that a calibration can not determine the properties of the geomagnetic field because it is external to the AC. In particular, a calibration can not determine the geodeclination. As I said earlier geodeclination is determined from a model based on GPS coords.


I read that paper. If you looked at it closely you would see that calibration detects and compensates for both hard iron and soft iron effects. Both of these are on the AC itself; they are not external to the AC. Nothing in that paper even remotely claims that calibration detects and compensates for geomagnetic properties external to the AC.


From your questions regarding the reason for the compass dance it's apparent you don't understand how a calibration works. This paper describes the compass calibration without all the math
http://www.sensorsmag.com/components/compensating-for-tilt-hard-iron-and-soft-iron-effects

and it makes the point that
In contrast, it is much more difficult—if not impossible—to compensate for distorting effects exhibited by material external to the aircraft/sensor platform.


If you think you know how a calibration can detect and compensate for geomagnetic properties you should publish your idea. You'll be famous.








2018-7-18
Use props
BudWalker
First Officer
Flight distance : 150877 ft
  • >>>
United States
Offline

Mirek6 Posted at 2018-7-18 07:53
BudWalker,

You say:” No. A compass calibration can not determine and compensate for geomagnetic properties because they are external to the AC. It's mathematically impossible. It's incorrect to say that flying in a new location requires that the compass be calibrated.”

Maybe you'll find this interesting. [size=14.6667px]I normally fly at home where DatCon computes the geoDeclination to be 12.72 degrees. On start up the Yaw is set to a value computed from the magnetometers. Then, when gpsHealth gets to 4 (out of 5) that Yaw value is then adjusted to reflect the geoDeclination
P1.jpg
[size=14.6667px]Without re-calibrating I flew at another location 1200 miles distant where the GeoDeclination is 8.61. It can be seen that the Yaw adjustment reflects that of the new location.
P2.jpg
2018-7-18
Use props
Mirek6
Second Officer
Flight distance : 239319 ft
Canada
Offline

BudWalker Posted at 2018-7-18 13:58
Your sentence is entirely incorrect. It implies that magnetometer does not measure geo-magnetic filed “because it is external to AC”. ???

You're mistaken. What I said is that a calibration can not determine the properties of the geomagnetic field because it is external to the AC. In particular, a calibration can not determine the geodeclination. As I said earlier geodeclination is determined from a model based on GPS coords.

BudWalker,

I am scratching my head.

I previously said that we are in agreement - I just used words which were imprecise. Than you said we are not in agreement - misinterpreting my words and stating obvious (at least for me). Than I misinterpreted your words and so on. Let's stop please because we are confusing people on this forum. And let's not judge what is "apparent" and who knows what - you or I may be sorely mistaken - let's avoid that and let's keep it civil so we can learn from each other.

It is quite obvious to me that we both know very well how magnetometer works - we just cannot use proper words and we are just confusing each other. Perhaps language. I described soft/hard iron effects many times myself on this forum so people could understand what is going on and why it is so important to calibrate compass free of any disturbances and far away from steel, magnets, electromagnetic fields etc. A lot of people believe that avoiding magnets is all you need to be weary about.

Now - as far as agreeing to disagree. We do disagree wrt re-calibrating after travelling. Let's stop at that. You don't calibrate because, based on your knowledge, you believe it unnecessary. I do, because, based on my knowledge and recommendations from DJI, I choose additional insurance. Is it always necessary? It is not. Is it wise? I believe it is.

As far as the case of Nilesh. If he were starting from the point with magnetic interference (let's say reinforced concrete) we would see discrepancy you noted (you said about 25 degrees) but this discrepancy would become smaller and smaller as Spark was raising up away from magnetic disturbance source (compass would slowly "unwind"). Such situation could confuse Spark.

I did observe this discrepancy (not sure about 25 degress, but yes, it was there) you noted but interpreted it differently. During first 12 seconds of AC raising it was difficult for me to interpret data because Nilesh was changing yaw while flying forward. This was changing direction of flight and, within such short time span and with his movements back and for and with Spark inertia you cannot deduct much.However, once he stopped changing yaw and AC kept flying forward and raising, I can see discrepancy quite clearly. But this discrepancy does not go down. It keeps constant until, eventually, Spark gets confused.

Hence I do not agree that the root cause, in that case, was interference at the ground level. I believe the root the cause was un-calibrated compass which did not show magnetic north correctly from the beginning and constantly, throughout the flight, until Spark's firmware could no longer tolerate that. Spark's commanded position consistently differed from physical position during flight forward and Spark needed to compensate all the time. At some point it just gave up. That's all.

Mirek
2018-7-18
Use props
Mirek6
Second Officer
Flight distance : 239319 ft
Canada
Offline

BudWalker Posted at 2018-7-18 14:08
Maybe you'll find this interesting. I normally fly at home where DatCon computes the geoDeclination to be 12.72 degrees. On start up the Yaw is set to a value computed from the magnetometers. Then, when gpsHealth gets to 4 (out of 5) that Yaw value is then adjusted to reflect the geoDeclination
[view_image]
Without re-calibrating I flew at another location 1200 miles distant where the GeoDeclination is 8.61. It can be seen that the Yaw adjustment reflects that of the new location.

BudWalker,

Thanks - this I did not know. Interesting. I do not use .dat files for my analysis.
Perhaps I should start :-)

The page you gave me does not mention Spark. It talks about tool for .dat analysis for higher end drones. It also says that Mavic Air .dat is encoded.

Do you know if the tool supports .dat analysis for Spark?

Mirek
2018-7-18
Use props
Mirek6
Second Officer
Flight distance : 239319 ft
Canada
Offline

BudWalker Posted at 2018-7-18 14:08
Maybe you'll find this interesting. I normally fly at home where DatCon computes the geoDeclination to be 12.72 degrees. On start up the Yaw is set to a value computed from the magnetometers. Then, when gpsHealth gets to 4 (out of 5) that Yaw value is then adjusted to reflect the geoDeclination
[view_image]
Without re-calibrating I flew at another location 1200 miles distant where the GeoDeclination is 8.61. It can be seen that the Yaw adjustment reflects that of the new location.

Bud Walker,

One more thing about your graphs which I find interesting. I now understand that Spark's wait to compensate for magnetic declination until GPS signal gets strong enough is necessary. Correcting before GPS is firm could lead to errors and wrong direction of true north as seen by Spark. This is logical and good design.

When I am talking about re-calibrating compass after travelling long distances I do not mean re-calibrating for magnetic declination. I mean - how does magnetometer ensure that it does point to true magnetic North wherever it is on Earth?

Properties of geomagnetic field are different in all three axis in different places, field is of different strength. This needs to be taken into account so magnetometer can properly figure out where the magnetic north is. Poorly calibrated compass can show 13 degree declination in a point which has 12 degree declination. Can't it?

Mirek
2018-7-18
Use props
BudWalker
First Officer
Flight distance : 150877 ft
  • >>>
United States
Offline

Mirek6 Posted at 2018-7-18 17:17
BudWalker,

I am scratching my head.

You've made several posts. I'll respond to them here in just this post.

I wasn't involved in the Nilesh incident. I didn't see the logs. You probably had me confused with msinger. I'll be glad to review the logs but will probably come to the same conclusion msinger did.

Yes, CsvView/DatCon works for the Spark. One of the hardest things in developing CsvView/DatCon has been staying up to date with the documentation. I'll fix that.

When I am talking about re-calibrating compass after travelling long distances I do not mean re-calibrating for magnetic declination. I mean - how does magnetometer ensure that it does point to true magnetic North wherever it is on Earth?

I don't understand this question. The compass (magnetometers) don't point to true north. The FC uses the declination value it computed to obtain a true heading. That declination was computed when gpsHealth first gets > 4.


Do you understand that in the plots I presented that there was no calibration between the flights that were 1200 miles apart. The only way the FC could correct for declination was for declination to be computed from the GPS coords.


The bottom line is that a calibration does not and can not compensate for magnetic effects that are external to the AC. It's mathematically impossible. One again, if you think you know how this can be done then you should publish your results.
2018-7-19
Use props
12Next >
Advanced
You need to log in before you can reply Login | Register now

Credit Rules