Compass Calibration, A Complete Primer

[solentlife]

You're right, magnetic anomalies external to the AC will have an effect on the compass. But, the problem is that the data due to an external anomaly is ambiguous. This means that any attempt at calibrating in the presence of external anomalies will see those effects but that data can not then be used (since it's ambiguous) to compensate for those effects. It still remains that detecting and then compensating for magnetic anomalies external to the AC is impossible.

Agreed but what happens is because the anomaly adds its effect to the variation / deviation that we would normally be calibrating for - we end up with bad calibration. As you say - it is impossible to separate them. As it is impossible to separate variation and deviation in our scenario.

I've looked at several of these types of incidents - literally dozens. By looked at I mean actual data from the one or both flight logs. They were all caused by a launching from a geomagnetically distorted site. None of these incidents were caused by a flawed compass calibration. If interested you could take a look at these example incidents.

Compass error

Looking for Trouble ??

The distorted site will if the pilot tries to calibrate his AC - will lead to bad calibration or even as I have seen - REFUSED calibration as the calibration falls outside of programmed limits.
I think we are discussing around the same matter - but looking at it from different sides. If you launch from a distorted site - as the AC flies away - the effect of that site lessens and the AC then warns of error or gets into trouble. Yes - we all have read many reports of such. It comes down to launching from a bad site and particularly if the pilot has mistakenly calibrated there.

Nigel

 

[solentlife]

I agree with virtually everything that you wrote in that post, although it appears to me that you and N017RW are saying the same thing. I don't understand your last point though. The statement was that declination (variation) is the difference between true north and magnetic north. Are you objecting to the word "declination" or are you trying to make some other point?

I was just highlighting that the accepted and internationally recognised term for earths magnetic effect is Variation, and giving example. Variation is the difference of Mag to True North ... never argued against that.

If I was to not repeat Var'n vs Dec'n - then any new reader not wanting to go back through loads of posts will be able to understand instead of maybe left in the cold.

Nigel

 

[solentlife]

I'm actually really happy that good and constructive discussion is here.

Magnetism is a factor in so many aspects of our lives and we are not aware or consider. It has fundamental rules and cannot be avoided.
But it also leads to some bizarre ideas for some. I am no expert on it as it is such a vast subject, but if with discussion we can arrive at solutions to questions and why we have incidents - that's great.

I am always open to discussion and explanation. We never progress forward unless we are.

May you enjoy good flights.

Nigel

 

[BudWalker]

Agreed but what happens is because the anomaly adds its effect to the variation / deviation that we would normally be calibrating for - we end up with bad calibration. As you say - it is impossible to separate them. As it is impossible to separate variation and deviation in our scenario.




The distorted site will if the pilot tries to calibrate his AC - will lead to bad calibration or even as I have seen - REFUSED calibration as the calibration falls outside of programmed limits.
I think we are discussing around the same matter - but looking at it from different sides. If you launch from a distorted site - as the AC flies away - the effect of that site lessens and the AC then warns of error or gets into trouble. Yes - we all have read many reports of such. It comes down to launching from a bad site and particularly if the pilot has mistakenly calibrated there.

Nigel

Sorry, I may have contributed to some confusion about labels when I introduced the label magnetic anomaly. So, just to be clear a compass calibration does not, and can not, compensate for any type of magnetic effects that do not rotate with the AC. I think this includes your definition of VARIATION. It also includes declination, cars nearby, buildings nearby, concrete rebar, iron ore mines, watches, bracelets, body piercing jewelry, etc.

I can't tell if you accepted the fact that declination is determined from GPS coords. If not, I can provide data that will show this is the case. As stated above, compensating for declination is not, and can not be done by a compass calibration.

We are not discussing around the same matter. None of the incidents that I referred to were caused by a flawed compass calibration. IMHO an incident caused by calibrating in a bad spot (and then flying from that spot) just doesn't happen. If you think otherwise then please provide an example where there is at least some compelling evidence. Every time I see an incident where this could've happened I ask for the .DAT log file - many times I get the .DAT file. So far, it just hasn't been the case that the incident was caused by a flawed calibration.

I think part of the reason for the bad-calibration-causes-incident misconception is that it's easy to confuse the symptoms with what happens when the AC is launched from a geomagnetically distorted site. There is one crucial difference though. In a launch-from-a-geomagnetically-distorted-site incident the Yaw value is incorrect at launch. That would not be the case with a bad-calibration-causes-incident if you believe the compass is OK at launch because it's been calibrated at the launch site.

 

[solentlife]

OK - we now have to disagree ...

Any compass no matter what its source / construction / type - if it is magnetic - MUST and CANNOT ignore magnetic effects outside of the 'vehicle' it is fixed to.

As you rotate the 'vehicle' - the lines of external magnetic flux and the internal (deviation) flux intersect at changing angles causing the resulting effect on the compass to change. The act of calibration compares this with programmed limits and if within limits accepts it as the new calibration.

Geomag data is averaged ... fact. It is good enough to fly with .. it will initialise the system ... but it is not absolute.

It is simply physically impossible to exclude external influences from a magnetic compass - which is what you imply in your last post. It makes no difference whether its MEMs as these are or a hand held needle job.

I can only assume that your "So, just to be clear a compass calibration does not, and can not, compensate for any type of magnetic effects that do not rotate with the AC." has mistakenly lost the factor that the magnetic source ON the AC is rotating while AC is effected by external magnetism whether it be Variation or metal object near. The vector addition of the two creates a factor that is used as calibration.

As to incidents caused by moving away from areas of local anomaly - yes they do happen. In fact the act of carrying the aircraft away from that anomaly is similar to flying away from it !
Trouble is the reverse is seen in some cases where people calibrate in bad area and then suffer Compass Warning / Error in flight ... I have seen it with my own little eye's and its not fantasy !!

I do not completely disagree with you - but must question parts as they do not agree with magnetic practice in real terms.

And we were doing so well !!

Nigel

 

[BudWalker]

OK - we now have to disagree ...

Any compass no matter what its source / construction / type - if it is magnetic - MUST and CANNOT ignore magnetic effects outside of the 'vehicle' it is fixed to.

As you rotate the 'vehicle' - the lines of external magnetic flux and the internal (deviation) flux intersect at changing angles causing the resulting effect on the compass to change. The act of calibration compares this with programmed limits and if within limits accepts it as the new calibration.

Geomag data is averaged ... fact. It is good enough to fly with .. it will initialise the system ... but it is not absolute.

It is simply physically impossible to exclude external influences from a magnetic compass - which is what you imply in your last post. It makes no difference whether its MEMs as these are or a hand held needle job.

I can only assume that your "So, just to be clear a compass calibration does not, and can not, compensate for any type of magnetic effects that do not rotate with the AC." has mistakenly lost the factor that the magnetic source ON the AC is rotating while AC is effected by external magnetism whether it be Variation or metal object near. The vector addition of the two creates a factor that is used as calibration.

As to incidents caused by moving away from areas of local anomaly - yes they do happen. In fact the act of carrying the aircraft away from that anomaly is similar to flying away from it !
Trouble is the reverse is seen in some cases where people calibrate in bad area and then suffer Compass Warning / Error in flight ... I have seen it with my own little eye's and its not fantasy !!

I do not completely disagree with you - but must question parts as they do not agree with magnetic practice in real terms.

And we were doing so well !!

Nigel

I'm going to try something different here and guess where you don't understand what I'm saying. I never said that magnetometers can ignore or exclude parts of the geomagnetic field. What I did say is the calibration process can't compensate for magnetic effects that don't rotate with the AC. Do you see the difference in these statements? It's one thing to measure the change in field strength caused by those external effects but it's quite another thing to then use that data to compensate for those external effects. The problem is that there isn't enough information in the measurements, i.e. - it's effectively ambiguous. Do you see the difference here?

If you're going to make the claim that
Trouble is the reverse is seen in some cases where people calibrate in bad area and then suffer Compass Warning / Error in flight
then it needs to be backed up with some kind of evidence.

Just to make sure there is no misunderstanding here I'll repeat what I said.
A compass calibration does not, and can not, compensate for any type of magnetic effects that do not rotate with the AC.

 

[N017RW]

This is the entire point of rotating in two planes (H/V), to characterize the Fe distortion(s).
Once established it can be corrected with algorithms to ‘round out’ the sensor’s response. This is better described as [distortion] ‘compensation’.

This understanding is easily available with research.

Not sure why this thread was revived.

Once ‘dynamic’ Deviation correction could be applied you should then be able to travel some distances, with varying deviation, and not need to re-cal.

My personal experience is 600 miles.
No re-cal away, or upon return home.

 

[sar104]

This is the entire point of rotating in two planes (H/V), to characterize the Fe distortion(s).
Once established it can be corrected with algorithms to ‘round out’ the sensor’s response. This is better described as [distortion] ‘compensation’.

This understanding is easily available with research.

Not sure why this thread was revived.

Once ‘dynamic’ Deviation correction could be applied you should then be able to travel some distances, with varying deviation, and not need to re-cal.

My personal experience is 600 miles.
No re-cal away, or upon return home.

I think it was revived because of the question of calibrating after moving geographical location, which previously was recommended, and now is causing some confusion.

 

[N017RW]

I think it was revived because of the question of calibrating after moving geographical location, which previously was recommended, and now is causing some confusion.

Still smile at your Avatar!

 

[sar104]

Still smile at your Avatar!

Right - I never got around to changing it. That was Ian's design, if I recall.

 

[N017RW]

Yea. I recall same. Classic!

 

[solentlife]

This is the entire point of rotating in two planes (H/V), to characterize the Fe distortion(s).
Once established it can be corrected with algorithms to ‘round out’ the sensor’s response. This is better described as [distortion] ‘compensation’.

This understanding is easily available with research.

Not sure why this thread was revived.

Once ‘dynamic’ Deviation correction could be applied you should then be able to travel some distances, with varying deviation, and not need to re-cal.

My personal experience is 600 miles.
No re-cal away, or upon return home.

Thank you .... at last ....

The rotation is to sense the effects of the influences external and internal ....

Nigel

 

[solentlife]

I think it was revived because of the question of calibrating after moving geographical location, which previously was recommended, and now is causing some confusion.

It was revived because of an idiot over on DJI Forum ... who cannot read plain english.

Nigel

 

[solentlife]

I'm going to try something different here and guess where you don't understand what I'm saying. I never said that magnetometers can ignore or exclude parts of the geomagnetic field. What I did say is the calibration process can't compensate for magnetic effects that don't rotate with the AC. Do you see the difference in these statements? It's one thing to measure the change in field strength caused by those external effects but it's quite another thing to then use that data to compensate for those external effects. The problem is that there isn't enough information in the measurements, i.e. - it's effectively ambiguous. Do you see the difference here?

If the effects of the 'dance' where basically two lines of force .... the external magnetic flux 'variation' + local anomaly and the AC's own flux 'deviation' .... interact and produce influence on compass heading where not compensated for - as you imply in this... why do we do the 'dance'.
You are correct that that rotation of the AC is important ... without that rotation - same as swinging a boat / ship round successive headings to obtain errors .... we would not be able to calibrate as we have no means of access to the MEMs or manual adjustment. So it has to be electronically assessed by us doing the 'dance'.

If you're going to make the claim that
Trouble is the reverse is seen in some cases where people calibrate in bad area and then suffer Compass Warning / Error in flight
then it needs to be backed up with some kind of evidence.

No need for me to recite others posts ... as I have seen it myself ... guy took off after mistakenly calibrating at a bad site - despite warnings to not do it ... his excuse was he'd 'read it on DJI Forum that it was necessary' !!
As his flight progressed and his magnetic environment then drastically changed from the bad one he had calibrated in ... he had compass warnings and then a fight to fly it back ... Without others help - I am sure he would have joined the ranks of the Fly Away brigade. Personally I was advising him to land it and go fetch - but he insisted he was going to get it back.

Just to make sure there is no misunderstanding here I'll repeat what I said.
A compass calibration does not, and can not, compensate for any type of magnetic effects that do not rotate with the AC.

It is impossible to compensate only for one set of influences ... it's like saying a boat can compensate for current direction but not needed for wind drift ... you cannot separate them.

The rotation of the magnetics of the AC causes a Vector Addition of forces ... external and internal that offsets the compass. The calibration routine senses and calibrates ... If it falls outside programmed limits - we get the failed calibration as quite often seen if we try it in a really bad location.
The reason it can sense and calibrate is because the effects of the interaction follows a reasonably set pattern that creates a near sinusoidal wave (not exactly but near enough for this topic !). The vertical and horizontal results are set and we fly. Use of the programmed Geo data Variation is purely a stepping stone .. as already explained. A supreme amount of work goes into deriving the Variation curves on the earth and is a result of painstaking manual and electronic techniques ... involves ships, aircraft, satelites - even people on foot. Because the Variation lines of flux do not follow easily found lines or linear ... they bend, distort, even turn back .. look at the diagram earlier - which is actually a simplified representation for illustration.

The only time any of the above would not be correct is within a zero magnetic environment and then the compass would be useless anyway.

Cheers .... now I'm off flying !!

Nigel

 

[BudWalker]

If the effects of the 'dance' where basically two lines of force .... the external magnetic flux 'variation' + local anomaly and the AC's own flux 'deviation' .... interact and produce influence on compass heading where not compensated for - as you imply in this... why do we do the 'dance'.
You are correct that that rotation of the AC is important ... without that rotation - same as swinging a boat / ship round successive headings to obtain errors .... we would not be able to calibrate as we have no means of access to the MEMs or manual adjustment. So it has to be electronically assessed by us doing the 'dance'.



No need for me to recite others posts ... as I have seen it myself ... guy took off after mistakenly calibrating at a bad site - despite warnings to not do it ... his excuse was he'd 'read it on DJI Forum that it was necessary' !!
As his flight progressed and his magnetic environment then drastically changed from the bad one he had calibrated in ... he had compass warnings and then a fight to fly it back ... Without others help - I am sure he would have joined the ranks of the Fly Away brigade. Personally I was advising him to land it and go fetch - but he insisted he was going to get it back.



It is impossible to compensate only for one set of influences ... it's like saying a boat can compensate for current direction but not needed for wind drift ... you cannot separate them.

The rotation of the magnetics of the AC causes a Vector Addition of forces ... external and internal that offsets the compass. The calibration routine senses and calibrates ... If it falls outside programmed limits - we get the failed calibration as quite often seen if we try it in a really bad location.
The reason it can sense and calibrate is because the effects of the interaction follows a reasonably set pattern that creates a near sinusoidal wave (not exactly but near enough for this topic !). The vertical and horizontal results are set and we fly. Use of the programmed Geo data Variation is purely a stepping stone .. as already explained. A supreme amount of work goes into deriving the Variation curves on the earth and is a result of painstaking manual and electronic techniques ... involves ships, aircraft, satelites - even people on foot. Because the Variation lines of flux do not follow easily found lines or linear ... they bend, distort, even turn back .. look at the diagram earlier - which is actually a simplified representation for illustration.

The only time any of the above would not be correct is within a zero magnetic environment and then the compass would be useless anyway.

Cheers .... now I'm off flying !!

Nigel

No one is disputing that VARIATION can be ignored. What you don't understand is that VARIATION isn't determined during the compass calibration. VARIATION determined at the beginning of the flight after the GPS coords are known.

Compensating for magnetic effects that don't rotate with the AC is mathematically impossible. If you know how this could be done then I urge you to publish your idea. You'll be famous.

And, please, it's not necessary to explain that external and on board magnetic effects are combined and the magnetometers can't separate them. This is a well understood concept and we are aware of it's implications.

Finally, your comment about it not being necessary to look at possible bad-calibration-causes-incident scenarios says a lot. Basically, what you're saying is that you don't need to look at evidence because you know it's true. When I see these posts it's always something like we know it was caused by a bad calibration because that's what causes these incidents. There is never any attempt at evidence based analysis. It's a text book example of confirmation bias and your description of the incident you witnessed is a good example. By your description it's also possible that the cause was just that the launch site was geomagnetically distorted and the calibration was not flawed.

 

[N017RW]

As I’ve stated numerous times, I have personally flown at a location 600 miles from the cal. site, and back at home again, with no re-cal.

If the local/ambient mag environment was so crucial or somehow factored-in, how could this be possible?

I hear a lot of jargon being used to dispute this and other points which only obfuscates the discussion.

It’s so easy to research magnetometer compensation theory and the associated mathematical algorithms yet it goes ignored and avoided by some.

 

[solentlife]

As I’ve stated numerous times, I have personally flown at a location 600 miles from the cal. site, and back at home again, with no re-cal.

If the local/ambient mag environment was so crucial or somehow factored-in, how could this be possible?

I hear a lot of jargon being used to dispute this and other points which only obfuscates the discussion.

It’s so easy to research magnetometer compensation theory and the associated mathematical algorithms yet it goes ignored and avoided by some.

I too fly without recalibrating despite significant distance from original location. No problem - I agree.
Because :
Its practical vs theoretical. In practical terms no re-calibration generally needed. In theory it is. Your observation is a perfectly valid practical matter.

But I would choose to recalibrate if I came over to USA from my usual Baltic location - to be sure that my P3 had best setup. Personal choice.

Nigel

 

[sar104]

I too fly without recalibrating despite significant distance from original location. No problem - I agree.
Because :
Its practical vs theoretical. In practical terms no re-calibration generally needed. In theory it is. Your observation is a perfectly valid practical matter.

But I would choose to recalibrate if I came over to USA from my usual Baltic location - to be sure that my P3 had best setup. Personal choice.

Nigel

But since, as I thought had already been agreed, calibration is only zeroing out the magnetic field of the aircraft itself which, presumably, has not changed just because you moved location, what would you be trying to fix with recalibration in that situation?

 

[BudWalker]

As I’ve stated numerous times, I have personally flown at a location 600 miles from the cal. site, and back at home again, with no re-cal.

If the local/ambient mag environment was so crucial or somehow factored-in, how could this be possible?

I hear a lot of jargon being used to dispute this and other points which only obfuscates the discussion.

It’s so easy to research magnetometer compensation theory and the associated mathematical algorithms yet it goes ignored and avoided by some.

You might 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

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.

 

[solentlife]

No one is disputing that VARIATION can be ignored. What you don't understand is that VARIATION isn't determined during the compass calibration. VARIATION determined at the beginning of the flight after the GPS coords are known.

Where do you get this from ... I have NEVER said that Compass Calibration determines Variation.

What I said is that the lines of flux of Variation crossing the lines of flux of the AC's deviation when AC is rotated causes a changing vector result ..... THAT is what the 'dance' is for !! To calibrate to compensate for the VECTOR results !!

Compensating for magnetic effects that don't rotate with the AC is mathematically impossible. If you know how this could be done then I urge you to publish your idea. You'll be famous.

Please keep it sensible ... as I keep saying .... the ROTATION of your AC causes VECTOR changes in the interaction of the external Variation and on-board Deviation ...

Even that bastion (Wikipedia) of the uneducated states clearly :

"Solid state compasses :

Small compasses found in clocks, mobile phones, and other electronic devices are solid-state microelectromechanical systems (MEMS) compasses, usually built out of two or three magnetic field sensors that provide data for a microprocessor. Often, the device is a discrete component which outputs either a digital or analog signal proportional to its orientation. This signal is interpreted by a controller or microprocessor and either used internally, or sent to a display unit. The sensor uses highly calibrated internal electronics to measure the response of the device to the Earth's magnetic field."

And, please, it's not necessary to explain that external and on board magnetic effects are combined and the magnetometers can't separate them. This is a well understood concept and we are aware of it's implications.

So why do you ignore it in your 'AC rotation' claims ?

Finally, your comment about it not being necessary to look at possible bad-calibration-causes-incident scenarios says a lot. Basically, what you're saying is that you don't need to look at evidence because you know it's true. When I see these posts it's always something like we know it was caused by a bad calibration because that's what causes these incidents. There is never any attempt at evidence based analysis. It's a text book example of confirmation bias and your description of the incident you witnessed is a good example. By your description it's also possible that the cause was just that the launch site was geomagnetically distorted and the calibration was not flawed.

Again - you are twisting words ... calibration is not 'flawed' - it is what it is ... If there is unusual magnetic influences when the guy calibrates - as long as the results lie with the programmed limitations - it will accept. It is not 'flawed' - it has an error in it that is allowing for that local anomaly. When the AC leaves that location or the effects of that anomaly - the calibration then is out of synch with its situation.

Let me ask a question then :

If calibration as you suggest is not influenced by external unusual magnetic effects - why is it advised by all and his dog to calibrate in a clean area away from local influences such as iron work, pipes, cables, metal objects... even down to not having your mobile or other gear nearby ?

Sorry but calibration is a serious matter and needs care. To 99.9999% of users - they don't care what or why ... they just want to calibrate and fly ...

Nigel