Fixing DJI's Compass Problem

[ianwood]

So you are totally ignoring toddsmi? You can't back up that it affects 100% of P2 as you have only heard from those with a problem. Have you offered beta testing to any European people with zero declination? Have dji told you how 3.05 works? I.e. Does it have declination table to detect location of individual operators & correct for them? etc etc

Not ignoring anyone. If he posts up video showing the P2 tracking perfectly on a line right from start up, we'll need to look at it. So far, no one has shown any video showing a P2 that is immune to the issue despite many initial claims to contrary. Even if there was a video showing this clearly, it would be an anomaly specific to that P2 as all the other data points to one conclusion.

As for your zero declination, you already asked this and I already answered it. Look up the declination of DJI's headquarters and you will have your answer.

DJI has not explained how they fixed it. The lookup table is one of a couple of ways to do it. Bottom line is declination errors are a fact if you don't correct for it. That is not debatable. It has been a fundamental part of navigation long before there was a Phantom.

 

[Mlaudisa]

So you are totally ignoring toddsmi? You can't back up that it affects 100% of P2 as you have only heard from those with a problem. Have you offered beta testing to any European people with zero declination? Have dji told you how 3.05 works? I.e. Does it have declination table to detect location of individual operators & correct for them? etc etc

Not ignoring anyone. If he posts up video showing the P2 tracking perfectly on a line right from start up, we'll need to look at it. So far, no one has shown any video showing a P2 that is immune to the issue despite many initial claims to contrary. Even if there was a video showing this clearly, it would be an anomaly specific to that P2 as all the other data points to one conclusion.

As for your zero declination, you already asked this and I already answered it. Look up the declination of DJI's headquarters and you will have your answer.

DJI has not explained how they fixed it. The lookup table is one of a couple of ways to do it. Bottom line is declination errors are a fact if you don't correct for it. That is not debatable. It has been a fundamental part of navigation long before there was a Phantom.

Dear Ian,

I for one wish to thank you for your determinate efforts at developing a solution. You put together a strong, factual, data-rich case that illustrates the issue; you solicited the input of this community; and you mounted a vocal campaign to catch the attention of the manufacturer. As a result, we will - hopefully soon - receive new firmware to fix or reduce the problem. It is much appreciated indeed, regardless as to whether there might be any other root causes at play, whether in components, design, or software. It was interesting to hear that in your opinion, all Phantoms are similarly affected, even though in my case I felt my first one never exhibited detectable symptoms.

Again thank you.

 

[Andy T]

I just checked DJI BBB rating and found one in Austin Texas which must be their head office here or something. They have an F rating. I think we should all consider fileing a complaint based on their dragging their butt on this matter

 

[ToddSmi]

So you are totally ignoring toddsmi? You can't back up that it affects 100% of P2 as you have only heard from those with a problem. Have you offered beta testing to any European people with zero declination? Have dji told you how 3.05 works? I.e. Does it have declination table to detect location of individual operators & correct for them? etc etc

Not ignoring anyone. If he posts up video showing the P2 tracking perfectly on a line right from start up, we'll need to look at it. So far, no one has shown any video showing a P2 that is immune to the issue despite many initial claims to contrary. Even if there was a video showing this clearly, it would be an anomaly specific to that P2 as all the other data points to one conclusion.

As for your zero declination, you already asked this and I already answered it. Look up the declination of DJI's headquarters and you will have your answer.

DJI has not explained how they fixed it. The lookup table is one of a couple of ways to do it. Bottom line is declination errors are a fact if you don't correct for it. That is not debatable. It has been a fundamental part of navigation long before there was a Phantom.

Just to be clear, I have read every page in this thread and I am not on the same page as Pazz whatsoever. I am very appreciative of your personal efforts and the efforts of the Drunken Sparrow to bring light to the MD issue. As far as I'm concerned, it's a job very well done in getting the 3.05 fix created and I am very much looking forward to updating my affected model as soon as it's available, so thank you for your tireless efforts on this!

Edit: For what it's worth, I'm a regular in the V+ and P2 threads on RC groups and I can tell you that there are many experienced pilots in high MD areas that claim their particular model is not affected even when being presented with a video and description of the whole MD j-hook and tbe related effects. This doesn't take away from the great work you have done in bringing a fix to the many folks whom are affected! Major Kudos! I have an enquiring mind, so I can't help but wonder why not all units are affected alike, but out of respect for differences of opinion, I will refrain from additional comments on this topic in this thread going forward.

 

[ElGuano]

I just checked DJI BBB rating and found one in Austin Texas which must be their head office here or something. They have an F rating. I think we should all consider fileing a complaint based on their dragging their butt on this matter

That's about 8 months out of date. The TX office was split off from DJI and those folks have an (ongoing?) legal dispute with DJI. Long story short, no longer affiliated with DJI.

PS, how did I miss the July 14 update?

UPDATE: As of July 14, 2014, there are 139,765 P2 users of all sizes and flavors reporting the same issue.[/color] If you are affected, make sure you say so here! Apparently Barack Obama crashed his P2 in the rose garden after it hooked into a trellis on July 4th weekend. He wasn't impressed. Says he wants DJI to fix it ASAP. (Washington DC is -12 magnetic declination.)

What's up with that, Ian?

 

[gr8pics]

Enter your location here http://magnetic-declination.com/

And how do i know what value it is?
It only says POSITIVE...

 

[SeanAxiom]

I missed last few pages here, and really confused about the debate here. The problem is for "real" and "does exist", DJI got to address the issue since they have sold faulty product, it would be much cheaper for them to work on a firmware then settling a multi-million dollar lawsuit filed on "faulty-equipment-causing-harm" basis.

I still cannot understand Pazz's argument about this. I, as one of the affected owner, have been facing this problem since I received my P2 8 months ago. This is a common problem for many users as indicated here, Phantom Owners's Facebook Page, YouTube and other platforms. It represents a very good portion of users with the same problem in specific areas, I have met 4 other Phantom2 users in my area with identical problem. This problem needs to be addressed by DJI or they would face some serious challenges by end-users, and as this topic continued here, DJI took actions to address the problem, so, what's Pazz trying to argue about is really confusing.

 

[sar104]

Enter your location here http://magnetic-declination.com/

And how do i know what value it is?
It only says POSITIVE...

And right above where it says "Declination is POSITIVE", it says "Magnetic declination: xx˚ xx' EAST".

 

[gr8pics]

And how is 2 degr 28 east translated into the scale of +/-8 or higher?

 

[robby666111]

And how is 2 degr 28 east translated into the scale of +/-8 or higher?

It's not. You're very unlikely to have a magnetic declination problem with a Phantom 2 here.

 

[Andy T]

And how is 2 degr 28 east translated into the scale of +/-8 or higher?

Sounds like you are at 2 degrees 26 minutes east, so you are not in an area if high magnetic declination

 

[gr8pics]

I still didnt understand how to read the numbers, but i guess its not important for me anyway then

 

[Andy T]

Edit: For what it's worth, I'm a regular in the V+ and P2 threads on RC groups and I can tell you that there are many experienced pilots in high MD areas that claim their particular model is not affected even when being presented with a video and description of the whole MD j-hook and tbe related effects. This doesn't take away from the great work you have done in bringing a fix to the many folks whom are affected! Major Kudos! I have an enquiring mind, so I can't help but wonder why not all units are affected alike, but out of respect for differences of opinion, I will refrain from additional comments on this topic in this thread going forward.

Well at the 10 minute mark of the video below you can see a prominent DJI employee flying a vision + and experiencing the toilet bowl effect. You would think that he would pick an unaffected model when demonstrating their equipment for Adobe representatives
https://www.youtube.com/watch?v=yZZFnlgCV0I

 

[robby666111]

I still didnt understand how to read the numbers, but i guess its not important for me anyway then

The magnetic declination number is the "difference" between what your compass (any compass analogue or digital) reads and true north. Thus when navigating in an area of high magnetic declination one needs to compensate the reading on the compass by this value to get true north. Map readers and been doing this for decades, it's not a new phenomenon. Also magnetic declination changes over time due to the uneven distribution of iron in the earth's core. For example here in NZ the declination is increasing by a rate of approximately 2° per year. Eventually it will slow and reverse, and one day you might expect Shenzhen (DJI HQ) to have a significant magnetic declination (currently <1). It's at this point we can probably expect a full and final fix for this issue in their firmware .

Cheers
Rob.

 

[gr8pics]

Aha, thanks for clarifying

 

[ToddSmi]

Edit: For what it's worth, I'm a regular in the V+ and P2 threads on RC groups and I can tell you that there are many experienced pilots in high MD areas that claim their particular model is not affected even when being presented with a video and description of the whole MD j-hook and tbe related effects. This doesn't take away from the great work you have done in bringing a fix to the many folks whom are affected! Major Kudos! I have an enquiring mind, so I can't help but wonder why not all units are affected alike, but out of respect for differences of opinion, I will refrain from additional comments on this topic in this thread going forward.

Well at the 10 minute mark of the video below you can see a prominent DJI employee flying a vision + and experiencing the toilet bowl effect. You would think that he would pick an unaffected model when demonstrating their equipment for Adobe representatives
https://www.youtube.com/watch?v=yZZFnlgCV0I

Ok, I was planning to censor myself in this thread but since you posted...

I will say that you make a good point, and I agree that there does appear to be a j hook just after 10 minute mark. But on the other hand there were a lot of things about this presentation that didn't represent DJI in the most professional light so I'm not sure we can infer too much from this. I've read through this entire thread as well as the "beta tester" compass thread to better understand your and Ian's point of view and you make very good arguments for "all P2's being affected equally". And just to reiterate, I am very grateful that so many folks have volunteered their time to try and do something about this problem resulting in a big improvement through new firmware! However, I have a hard time dismissing my own observations and those of other experienced pilots. But don't take my word for it. If you haven't already, spend some time in the P2 thread and V+ thread on rcgroups where they have discussed this issue as well, get to know some of the posters on there, and see if it doesn't start to affect your view a little bit on the topic of whether or not all P2 in high MD areas are equally affected. I'm not talking about posters like Pazz that are continually trying to stir the pot, but rather folks that seem to be reasonably minded and don't have a horse in the race either way.

 

[Erwan]

Last but not least the main point to consider is still any firmware update and any schedule to fix it ?!

 

[ianwood]

What's up with that, Ian?

Consider it a test to see who is looking at what!

 

[Andy T]

*snip*
I will say that you make a good point, and I agree that there does appear to be a j hook just after 10 minute mark. But on the other hand there were a lot of things about this presentation that didn't represent DJI in the most professional light so I'm not sure we can infer too much from this
*snip*

Haha I kind of liked the casual nature of the video and that he was live feeding from the phantom to Google hangouts with voice as well!.
My theory is that many who deny they have an issue may not be seriously affected and also may have a stake in the success of the phantom. My dealer sure didn't like talking about the issue and never really admitted there was an issue (but you could tell he knew about it). There is also no way of knowing if some of the people defending the product are just commenting to make the people that are complaining about the issue look bad and to trivialize the issue. Bottom line, there is an issue that needs to be fixed -period.
And no problem, glad you posted

 

[ToddSmi]

*snip*
I will say that you make a good point, and I agree that there does appear to be a j hook just after 10 minute mark. But on the other hand there were a lot of things about this presentation that didn't represent DJI in the most professional light so I'm not sure we can infer too much from this
*snip*

Haha I kind of liked the casual nature of the video and that he was live feeding from the phantom to Google hangouts with voice as well!.
My theory is that many who deny they have an issue may not be seriously affected and also may have a stake in the success of the phantom. My dealer sure didn't like talking about the issue and never really admitted there was an issue (but you could tell he knew about it). There is also no way of knowing if some of the people defending the product are just commenting to make the people that are complaining about the issue look bad and to trivialize the issue. Bottom line, there is an issue that needs to be fixed -period.
And no problem, glad you posted

Agreed, the comments from many online posters must be taken with a grain of salt. And there are some that will defend DJI to the death in spite of any facts or evidence to the contrary, which does call into question their motives and credibility. In my experience, I've also found quite a few folks that know what they are talking about, and don't appear to come in with an agenda. These are the people whose opinions I take seriously. But the most important point is one you and I can unequivocally agree on...it is undeniable that there is a widespread MD problem and it's awesome that a fix is on the way! Hope it gets released soon!