Hi
It's me again. I tested my spark inside a concrete building with high densed steel re-inforcement (Bunker) and with lots of temp steel props with very limited lighting.

Please see below the link for the flight record for your info and comments
http://www.phantomhelp.com/LogViewer/EOC8BIKT906Y974KVXTC/

Spark displayed warning about compass error and suggested me to calibrate the compass which I didn't do.

The flight time was very short and at lower altitute due to narrow space. But again Spark really impressed me with it's ability to fly without GPS and mulfunctioning compass. It started to drify horizontaly at times but did't go upwards on it's own. One thing noted while flying with compass error/interference is that AC refuse to respond to your controller at times or over reacting.

From my previous tests and this one, I can only assume that if your Spark fly upwards without you pushing the sticks then there is something wrong with the sensors at the bottom that tells the AC you are too close to an obstruction and forcing it to fly upwards to avoid it.

It is not advisable to fly your Spark in high wind, without enough GPS signal or compass error as Spark easily get very upset if you ignore the manufacturers recommendations and run away from you.

Unless you are an experienced pilot and very fluent with controller sticks - Always be on the safe side.

Thank you
Ram
Inexperienced Drone Pilot

Thanks for sharing your observations.

Good test very brave.
A couple of things one yes VPS will cause Aircraft to pop up clear of obstruction and you can clearly see this in your log.
The other thing is your app asked you to calibrate compass was offset many here don’t believe this will happen and choose to tell others to just calibrate even when there are no warnings, which we know can be detrimental advice. I think we now know these compass problems are not down to firmware issues.
Good job.

Thank you for sharing your experiences and for your "explorer" spirit. Doing such things in a controlled manner can teach a lot indeed.

Good input and I believe if anyone have the opportunity to fly spark in a closed spacious  environment (indoor court/hall) they should do it to experience ATTI mode first hand and feel how the AC behave to the RC sticks. Of course as OP mentioned you do not have the wind effect but it is still has it's benefit. BTW use the props guard just in case.

hallmark007 Posted at 2018-1-13 16:02
Good test very brave.
A couple of things one yes VPS will cause Aircraft to pop up clear of obstruction and you can clearly see this in your log.
The other thing is your app asked you to calibrate compass was offset many here don’t believe this will happen and choose to tell others to just calibrate even when there are no warnings, which we know can be detrimental advice. I think we now know these compass problems are not down to firmware issues.

Thank you for your comments and complements.
I didn’t advise calibration of the compass. It’s only my observations shared with others. So that other drone flyers will be careful.
I also mentioned about the consequences of ignoring manufacturers recommendations.

Thank you again for sharing your thoughts with us.

rimza Posted at 2018-1-13 16:29
Good input and I believe if anyone have the opportunity to fly spark in a closed spacious  environment (indoor court/hall) they should do it to experience ATTI mode first hand and feel how the AC behave to the RC sticks. Of course as OP mentioned you do not have the wind effect but it is still has it's benefit. BTW use the props guard just in case.

Thank you for sharing your thoughts.

It’s better to train on a cheap drone in a closed environment. I also have three mini helicopters. One of them is 3D. It’s a nightmare to just lift them off the floor and keep it hovering in one place. If I manage to fly it without hitting anything then flying a drone become so easy.

hallmark007 Posted at 2018-1-13 16:02
Good test very brave.
A couple of things one yes VPS will cause Aircraft to pop up clear of obstruction and you can clearly see this in your log.
The other thing is your app asked you to calibrate compass was offset many here don’t believe this will happen and choose to tell others to just calibrate even when there are no warnings, which we know can be detrimental advice. I think we now know these compass problems are not down to firmware issues.

He was flying in a bunker and got compass errors, does that prove that no compass errors are caused by firmware issues? No. Your logic is flawed. Not to mention not he or anyone else advacated recalibrating the compass in this case.

What we do know - If the AC is calibrated correctly, it won't be worse off.

What we don't know - The firmware does not have unknown bugs affecting the compass or other unknown conditons may affect state of the compass calbration or that the application might miss conditions where calibration may be needed.

What I found useful about the test was that there was an apparent lack of response to pilot input, plenty of people here say to these reports, that yes it was responding, you just didn't know to fly in ATTI.

Great observations - thanks for sharing.

djiuser_pVzOCZc Posted at 2018-1-14 01:07
He was flying in a bunker and got compass errors, does that prove that no compass errors are caused by firmware issues? No. Your logic is flawed. Not to mention not he or anyone else advacated recalibrating the compass in this case.

What we do know - If the AC is calibrated correctly, it won't be worse off.

“What we do know - If the AC is calibrated correctly, it won't be worse off.”

What we do know by calibrating a good compass you can’t make it better, but you can make it worse.

All moves made by OP were recorded there was no lack of response, flying in Atti mode requires much bigger stick movements and as you can see from log this did not happen.

OP also has done test to recreate yaw error which he preformed in high wind conditions at low altitude and in sport mode 100%, and in that test you can clearly see the effects of such flying.

The fact that compass error yaw error has been with the spark since its release shows that new firmware has nothing to do with this problem.

hallmark007 Posted at 2018-1-14 03:01
“What we do know - If the AC is calibrated correctly, it won't be worse off.”

What we do know by calibrating a good compass you can’t make it better, but you can make it worse.

"What we do know by calibrating a good compass you can’t make it better, but you can make it worse."

No one is talking about calibrating a good compass.  If you think that the App not reporting an issue with the compass is evidence of "good compass" - There are ample first hand reports around the forum to contradict this.

"The fact that compass error yaw error has been with the spark since its release shows that new firmware has nothing to do with this problem."

It does nothing of the kind. Making the assumption that because there have always been yaw errors, that yaw errors have always had the same cause, is again a flawed argument.

djiuser_pVzOCZc Posted at 2018-1-14 03:17
"What we do know by calibrating a good compass you can’t make it better, but you can make it worse."

No one is talking about calibrating a good compass.  If you think that the App not reporting an issue with the compass in evidence of "good compass" - Their are ample first hand reports around the forum to contradict this.

Reading anything I have said about compass you will see I always advise people to do necessary checks.

Regarding firmware being responsible for compass problems, as I said before this is a complete urban myth, there is absolute no proof as to how this could be caused by firmware and if it was and has been there from the release of Spark Mavic and Phantom all these aircraft receive same warnings yet after more than 30 FW updates we see no difference in how any of these aircraft react to same warnings.

But one thing we do see in most of these incidents is opening line. “ I calibrated my compass before I started flying”

There are fair amount of evidences available on this forum to suggest that some issues and fly away s caused by the OP error than the AC itself. But that isn't enough to hide the fact that Spark is weak on managing interferences due to its hardware, firmware etc.

hallmark007 Posted at 2018-1-14 03:37
Reading anything I have said about compass you will see I always advise people to do necessary checks.

Regarding firmware being responsible for compass problems, as I said before this is a complete urban myth, there is absolute no proof as to how this could be caused by firmware and if it was and has been there from the release of Spark Mavic and Phantom all these aircraft receive same warnings yet after more than 30 FW updates we see no difference in how any of these aircraft react to same warnings.

"Reading anything I have said about compass you will see I always advise people to do necessary checks."

This is not true. Unless the "necessary checks"  is not calibrating unless instructed by the App.  

You keep huffin puffin about facts and "proof" and you offer absolutely none yourself.  You use the Mavic and the Phantom to support your argument when they use different hardware and firmware, so they  "react to same warnings" ,this has nothing to do with what we've been discussing,  its not surprising AND does absolutely nothing to support your case.  

"But one thing we do see in most of these incidents is opening line. “ I calibrated my compass before I started flying”"

Not true.  There are a few, but not most.

Its easy for anyone see that doing proper calibration won't make things worse and potential to make things better.

You know what bugs the spark firmware does or does not have?  Yea Right.

djiuser_pVzOCZc Posted at 2018-1-14 04:43
"Reading anything I have said about compass you will see I always advise people to do necessary checks."

This is not true. Unless the "necessary checks"  is not calibrating unless instructed by the App.  

Read and you will see what I have said.

https://forum.dji.com/forum.php? ... &fromuid=260008

I would always tell people when getting magnetic interference move or calibrate compass, to shut down aircraft and move to another location, not to calibrate compass yes that’s right.
But unlike you telling people that they should calibrate compass after each firmware simply on the basis of an urban myth, if you have any other information then let’s us see it, I think you will find most people would not calibrate compass more than twice in a year, yet we can have up to 10 FW updates in a year and your recommendation is for all of us to calibrate compass and your clearly told in your manual not to calibrate.
If people have a bad compass I agree they should calibrate, but it’s ridiculous without proof to be telling everyone they should ignore manual and calibrate even if there is not a problem with their compass.

hallmark007 Posted at 2018-1-14 05:04
Read and you will see what I have said.

https://forum.dji.com/forum.php?mod=viewthread&tid=104222&fromuid=260008

Your #1 "Never calibrate Compass unless prompted to"

You state this as an absolute.

Like I've said before, there is ample evidence in forum and elsewhere that this is just bad advice.

"10 FW updates in a year and your recommendation is for all of us to calibrate compass and your clearly told in your manual not to calibrate"

Absolutely, and I clearly stated my reasons to you before.

"but it’s without proof to be telling everyone they should ignore manual and calibrate even if there is not a problem with their compass."

You don't know that there was not a problem with the compass.  Again, the manual is not infallible and there is plenty of evidence of omissions and inaccurate information in the manual.  I would advise people to recalibrate after a firmware update, but unlike you I make arguments based on theories from first hand reports and I don't repeatedly re-phrase and parrot the same flawed arguments over and over.  Nor do I try to put words into your mouth, which you have tried to do repeatly to me to setup your straw men.

"ridiculous without proof"

It's not.  It does not sound like you have much experience with proofs, if you did you would know that providing a valid proofs for a firmware the size of the spark's would fill a library.

People should always think for themselves and not blindly follow.

djiuser_pVzOCZc Posted at 2018-1-14 20:43
Your #1 "Never calibrate Compass unless prompted to"

You state this as an absolute.

The one thing we do know, is in almost all recent cases of yaw error compass problems is that users calibrated their compass, so there’s some proof for you and your recommendations.

hallmark007 Posted at 2018-1-15 02:40
The one thing we do know, is in almost all recent cases of yaw error compass problems is that users calibrated their compass, so there’s some proof for you and your recommendations.

Are you referring to this?

https://forum.dji.com/thread-129453-1-1.html

Oh wait, he didn't.   And even if he did it wouldn't be proof.  It would be another example of a flawed argument to try to protect your precious ego,  007.  

djiuser_pVzOCZc Posted at 2018-1-15 03:26
Are you referring to this?

https://forum.dji.com/thread-129453-1-1.html

I think you will find he did. I can show you two other cases from the last couple of days same thing, using your advice .
It’s nothing to do with ego I’ll leave that up to you. It’s called experience.
https://forum.dji.com/forum.php? ... &fromuid=260008

djiuser_pVzOCZc Posted at 2018-1-15 03:26
Are you referring to this?

https://forum.dji.com/thread-129453-1-1.html

Take a look at this thread some very experienced drone pilots see what they think.

https://forum.dji.com/forum.php? ... &fromuid=260008

hallmark007 Posted at 2018-1-15 03:34
I think you will find he did. I can show you two other cases from the last couple of days same thing, using your advice .
It’s nothing to do with ego I’ll leave that up to you. It’s called experience.
https://forum.dji.com/forum.php?mod=redirect&goto=findpost&ptid=129453&pid=1131495&fromuid=260008

20 flights ago - And he didn't say weather or not prompted by your infallible DJI software.

So are you are implying that his recalibration  20 flights ago caused the issue? Wow, now that's reaching and you have nerve to say this not about your ego, 007.
  

djiuser_pVzOCZc Posted at 2018-1-15 03:44
20 flights ago - And he didn't say weather or not prompted by your infallible DJI software.

So are you are implying that his recalibration  20 flights ago caused the issue? Wow, now that's reaching and you have nerve to say this not about your ego, 007.

No your assuming that calibration will avert the situation , again bad information, read the thread above you will find most experienced drone pilots feel the same way. There not making it up as they go, but through experience.

hallmark007 Posted at 2018-1-15 03:49
No your assuming that calibration will avert the situation , again bad information, read the thread above you will find most experienced drone pilots feel the same way. There not making it up as they go, but through experience.

What does P4P have to do with this?  Speaking from experience myself, I don't mindlessly follow advice from self appointed experts.  You cherry pick information, you ignore whatever you don't agree with, you lie, and you causally and repeatedly employ every fallacy in the book.

If this what your experience has taught you, then I would't put much stock in it.

Why does the aircraft switches to P-gps mode from ATTI, when there is no sats and some errors going on?

Raimenzio Posted at 2018-1-15 04:04
Why does the aircraft switches to P-gps mode from ATTI, when there is no sats and some errors going on?

It no longer has gps to hold Aircraft position so goes to Atti mode , in Atti mode pilot still has full control altitude will still be constant Aircraft will drift on the wind so pilot must try to counteract these movements it can be difficult particularly if flying in strong winds and if there are many obstacles, best course of action is always to try to land safely as quick as you can.

Compass problems going into Atti mode is a different matter but same outcome going into Atti mode.

Why would a compass error disable GPS?
Orientation (compass data) and position (GPS data) are different but related.
When the Spark is moving and getting position information from the GPS that tells the Spark that it is travelling in a different direction from what it's getting from compass data, The Spark programming can't work with the data conflict.
The solution is to drop data from one source and since the Spark can't fly without a compass but can without GPS, it's the GPS data that gets dropped.

djiuser_pVzOCZc Posted at 2018-1-15 04:03
What does P4P have to do with this?  Speaking from experience myself, I don't mindlessly follow advice from self appointed experts.  You cherry pick information, you ignore whatever you don't agree with, you lie, and you causally and repeatedly employ every fallacy in the book.

If this what your experience has taught you, then I would't put much stock in it.

I think what your saying is you have no experience and try to make it up as you go along.

What has the P4Pro to do with anything. It has plenty and compass in P4Pro works the same way as in the spark and you will have noticed from that thread that almost all those guys would not advocate calibrating compass like you are, and why because of experience, and instruction regarding calibration is the very same in the phantom manual.
Again you offer no basis for your theory except to say you don’t except advice from “as you call them” self appointed experts. But you will find most of these guys are here to help and give the best advice to those who are looking for it.

hallmark007 Posted at 2018-1-15 04:18
It no longer has gps to hold Aircraft position so goes to Atti mode , in Atti mode pilot still has full control altitude will still be constant Aircraft will drift on the wind so pilot must try to counteract these movements it can be difficult particularly if flying in strong winds and if there are many obstacles, best course of action is always to try to land safely as quick as you can.

Compass problems going into Atti mode is a different matter but same outcome going into Atti mode.

It's all makes sense to me. All i saw in the log was that aircraft was going from P-GPS to ATTI, then back to P-GPS. Ok, I maybe just took straight forward the called "GPS" mode, it has to use vision system instead of compass and gps in closed enviroments.

hallmark007 Posted at 2018-1-15 04:28
I think what your saying is you have no experience and try to make it up as you go along.

What has the P4Pro to do with anything. It has plenty and compass in P4Pro works the same way as in the spark and you will have noticed from that thread that almost all those guys would not advocate calibrating compass like you are, and why because of experience, and instruction regarding calibration is the very same in the phantom manual.

P4P has different hardware and different firmware and different bugs.

So - it in fact it has nothing to do with bugs that are present in spark firmware.

As to the opinions of others on P4P's compass, it is has no relevance to spark and it's compass.

Anything else, 007?   

It doesn't make sense that it switches to P-GPS with 0 sats. I'd also expect somewhat better error message then "Other(9)".

djiuser_pVzOCZc Posted at 2018-1-15 04:39
P4P has different hardware and different firmware and different bugs.

So - it in fact it has nothing to do with bugs that are present in spark firmware.

Again your making up stuff , you know nothing about hardware in P4Pro or spark if that’s what your saying. Maybe you will tell us why over on the P4 forum and Mavic forum that same advice is given by those with the necessary experience. I don’t think you will be able .
Regarding bug in spark pertaining to compass, well it must have been there from its release because nothing has changed with any FW/SW updates, so maybe not a bug.

munchausen Posted at 2018-1-15 04:42
It doesn't make sense that it switches to P-GPS with 0 sats. I'd also expect somewhat better error message then "Other(9)".

The other message is to tell you that Aircraft won’t let you climb above 16ft because of weak gps, yes it could just say weak gps,

hallmark007 Posted at 2018-1-15 04:54
Again your making up stuff , you know nothing about hardware in P4Pro or spark if that’s what your saying. Maybe you will tell us why over on the P4 forum and Mavic forum that same advice is given by those with the necessary experience. I don’t think you will be able .
Regarding bug in spark pertaining to compass, well it must have been there from its release because nothing has changed with any FW/SW updates, so maybe not a bug.

As much amusement as I get in getting you to make a fool of yourself, I'm also mildly disappointed by your inability to make a well reasoned argument - I had for hoped for better.

BTW - the parts for P4P and Spark are readily  available, and it does not take EE know that they are different.  At this point, I don't see a point in continuing to rub your nose in it.

djiuser_pVzOCZc Posted at 2018-1-15 06:09
As much amusement as I get in getting you to make a fool of yourself, I'm also mildly disappointed by your inability to make a well reasoned argument - I had for hoped for better.

BTW - the parts for P4P and Spark are readily  available, and it does not take EE know that they are different.  At this point, I don't see a point in continuing to rub your nose in it.

What your trying to do is make an argument about something you know nothing about. We are still waiting for you to show any proof for calibration after every firmware update and can show us nothing to back this up, it’s ridiculous and I think you know this now.
As you said you don’t take advice from anyone , your lack of experience shows your just a one trick pony who has tried to troll me for giving information based on experience.
It’s not a coincidence that when phantom Mavic Spark have a problem with compass that we see exactly the same thing happening, but you in limited knowledge decide to tell us all that calibration is the answer to this with no proof and no basis for what your saying, flying in the face of all other experience drone pilots are saying including the manufacturers of the drones.
Yes you know much more than all of above, yet no evidence no proof by advising what is a ridiculous notion . I don’t need to make any argument here , you keep giving out your urban myths and if someone asks why, you can just tell them it’s because you said so who needs to know whether it’s going to help or hinder.
I do not wish to discuss this any further until you can show that what your are putting forward is no more than an urban myth.

djiuser_pVzOCZc Posted at 2018-1-15 06:09
As much amusement as I get in getting you to make a fool of yourself, I'm also mildly disappointed by your inability to make a well reasoned argument - I had for hoped for better.

BTW - the parts for P4P and Spark are readily  available, and it does not take EE know that they are different.  At this point, I don't see a point in continuing to rub your nose in it.

What your trying to do is make an argument about something you know nothing about. We are still waiting for you to show any proof for calibration after every firmware update and can show us nothing to back this up, it’s ridiculous and I think you know this now.
As you said you don’t take advice from anyone , your lack of experience shows your just a one trick pony who has tried to troll me for giving information based on experience.
It’s not a coincidence that when phantom Mavic Spark have a problem with compass that we see exactly the same thing happening, but you in limited knowledge decide to tell us all that calibration is the answer to this with no proof and no basis for what your saying, flying in the face of all other experience drone pilots are saying including the manufacturers of the drones.
Yes you know much more than all of above, yet no evidence no proof by advising what is a ridiculous notion . I don’t need to make any argument here , you keep giving out your urban myths and if someone asks why, you can just tell them it’s because you said so who needs to know whether it’s going to help or hinder.
I do not wish to discuss this any further until you can show that what your are putting forward is no more than an urban myth.

I started this thread in the hope of passing the information from my experiment so the experts can analyse and share their thoughts so everyone can learn something new.

All I see here is couple of people hijacking the thread and filling it with garbage to satisfy their egos.

Very disappointed ☹️

Ram-UK Posted at 2018-1-15 13:08
I started this thread in the hope of passing the information from my experiment so the experts can analyse and share their thoughts so everyone can learn something new.

All I see here is couple of people hijacking the thread and filling it with garbage to satisfy their egos.

Apologies mate got carried away.

hallmark007 Posted at 2018-1-15 06:32
What your trying to do is make an argument about something you know nothing about. We are still waiting for you to show any proof for calibration after every firmware update and can show us nothing to back this up, it’s ridiculous and I think you know this now.
As you said you don’t take advice from anyone , your lack of experience shows your just a one trick pony who has tried to troll me for giving information based on experience.
It’s not a coincidence that when phantom Mavic Spark have a problem with compass that we see exactly the same thing happening, but you in limited knowledge decide to tell us all that calibration is the answer to this with no proof and no basis for what your saying, flying in the face of all other experience drone pilots are saying including the manufacturers of the drones.

"What your trying to do is make an argument about something you know nothing about."
It doesn't seem to stop you from saying what defects do or don't exist in the spark's firmware. The fact that you use the P4P and its firmware is bazzar, but not surpising in that you can't grasp even the basics of software development.  

"it’s ridiculous and I think you know this now."
If you say this enough times it will make it true.

"We are still waiting for you to show any proof."
The feeling is mutal mate.

"As you said you don’t take advice from anyone"
Putting words in my mouth to set up a straw man?  I think that you're just incapable of not lying and resorting to childish tactics.

What I actually said was that I don't mindlessly take advice from self appointed experts.  Stick to your dogma, protect your precious ego 007.

hallmark007 Posted at 2018-1-15 03:42
Take a look at this thread some very experienced drone pilots see what they think.

https://forum.dji.com/forum.php?mod=viewthread&tid=129001&fromuid=260008

That is a funny link.  I should do a crash analysis if I fly my model airplane into a tree.  Drone took off and ran into a tree with the pilot, when it was "too late", finally taking corrective action.  But he still hit the tree.

Photographers discovering their pilot skills when the autopilot disconnects.  They likely wonder why they can't let their attention wander from the flying drone like they do with their camera/tripod combo.  Once the camera is in a position, won't it stay there?

hallmark007 Posted at 2018-1-15 13:12
Apologies mate got carried away.

I have a few questions for you. Appreciate if you could explain to me. Thanks

1. GPS position no match
2. Other(9)
3. Can I manually switch to ATTI if so how ?

Ram-UK Posted at 2018-1-16 05:22
I have a few questions for you. Appreciate if you could explain to me. Thanks

1. GPS position no match

1/ gps no match , weak gps signal usually lasts short time, also have seen this a lot when Aircraft is flying fast in windy conditions.

2/ Other (9) again weak gps usually at take off failsafe won’t let Aircraft exceed altitude of 16ft, Other (8) is usually seen when weak gps is between 5metre and 30 metre failsafe won’t let Aircraft exceed 30 metre.
3/ No you can’t switch to Atti mode, same on Mavic , you need to using phantom for switch to Atti .