Mirek6 Posted at 2018-6-26 15:53
BrilhasMuito,

No connection with Star Wars here :-).

Mirek,

The Excel Program lists as well the Inputs ? I really am impressed how you could also see the RC Configuration.



Really Big Welldone, today its late, if you have time I wish to learn about how Speed Error Works. Even if you direct me to a Youtube Tutorial its enough.

Cheers and Good Night Buddy !

AlexQB Posted at 2018-6-26 14:06
Hello Mirek,
First of all, wooowww !! You left me very impressed! Many thanks for the accurate analysis of the flight, especially for the final moments. What privilege you spent so much time analyzing my fateful flight.

Hey Alex,

I told you I have a friend who will surely come to Help.

Mirek is a Real Genius I can confirm.

Hope that DJI will help you out Brother.

DMX_MT Posted at 2018-6-26 15:59
Mirek,

The Excel Program lists as well the Inputs ? I really am impressed how you could also see the RC Configuration.

DMX_MT

I use Excel to read .csv file. Yes - you do have stick movements recorded there. And plethora of other information.
.csv is a text file but it is darn hard to read without help of Excel.

I appreciate your comments.
Have a good night.

RGMGFitness Posted at 2018-6-26 11:52
I'm not expert on 'Compass Interference' and I didn't see anything in the DJI SPARK manual regarding 'Rocks'.  But in other DJI products like the inspire I've seen specific notes reading; "The ideal place to calibrate is a wide open location that is free of anything metalilic within a 20 ft radius.  Keep away from drainage pipe, irrigation systems, rocks, sidewalks (or other concrete structures), etc.

Obviously, a lot of rock up there on that mountain.  And not being a geologist in all likely hood that mountain could contain all types of metals potentially interfering with signal.  Again, food for thought...I'm not an expert.    Curious, if you were getting any 'Compass Calibration' errors?  Also, did you do a 'Compass Calibration' while on the mountain?

Hello RGMG, infact you are right !

You are always right BOSS !

Mirek said that the rocks may have caused the Magnetic Interference. I never knew that, Thank You Both !

Thanks for your Contribution, it really helped us !

DMX_MT Posted at 2018-6-26 15:59
Mirek,

The Excel Program lists as well the Inputs ? I really am impressed how you could also see the RC Configuration.

DMX_MT,

You say:  "if you have time I wish to learn about how Speed Error Works. Even if you direct me to a Youtube Tutorial its enough".

I do not have pointers to YouTube videos but I was browsing through some other drone forums. Speed error is nothing magic. It simply means that horizontal speed which is measured by variety of Spark sensors is not consistent with the changes of drone's GPS location. For example speed shows 10 km/hour but GPS position is not moving.

If you get Speed errors in any drone, the GPS navigation is compromised and drone must drop to ATTI to keep sane. It is compromised because drone's controller does not know what is faulty, sensors, GPS positioning system? All bets are off. ATTI to the rescue :-).

As you can imagine from the above, there may be multiple sources of speed error. Issue with VPS sensors if flying low, issues with IMU accelerometer, issues with GPS receiver, issues with firmware.
And perhaps many more.

Speed error, yaw error and other errors are just indicators that some mismatch exists. The trick is to figure out what mismatch because it may be different with each case. There is no universal answer.
For example: I am certain you saw compass errors while the root cause was something completely different which just threw on-board controller off. IMU gyroscope was faulty and disagreed with compass and Spark's controller interpreted it as compass error. Happens all the time.

This is bad news. The good news is - because of this and the complexity of Spark design and subsystem interactions we can become one-day detectives and have fun :-).

Cheers!

Mirek   

Mirek6 Posted at 2018-6-26 15:47
Oh well, my previous post was cut off - too long :-).

Continuing:

Wooooww!

You are a real genius! You left me impressed with your thesis. Your explanations have a lot of sense.

As for the IMU calibration, I did it for the second time about 15 days before, since one afternoon the drone made some strange movements and I thought it was convenient to recalibrate it. The following days everything was perfect so I did not give it more importance until now when you commented, I thought about it.

Calibration of the compass also had been done recently.

See what they tell me from DJI Support.

Thank you again for your help!

DMX_MT Posted at 2018-6-26 16:01
Hey Alex,

I told you I have a friend who will surely come to Help.

Also thank you for your help. Thanks to you i will sleep a little better today!

JJBspark Posted at 2018-6-26 09:41
Hi,

Just in addition to this thread ; a speed error says something about high RPM of the motors, ECS will increase rpm if flying at high altitude (thin air) to keep at desired level height.

JJB,

You say:
"Its similar to Aircraft of Helicopter max flying ceiling, at certain height you cannot go any higher. (due to aerodynamics and/or engine power)
Don`t remember where i have read this fact, if i find it again will update this post."

Please do. I am very interested to see this interpretation,

However, in this case, please note how horizontal speed and GPS speed gets totally mismatched at the beginning when Spark start's its mini toilet bowl effect and turns counterclockwise in a circle with no stick input.
GPS positioning barely changes, horizontal speed shows almost nothing and GPS speed jumps like on drugs :-).  Something weird is happening here. Spark cannot figure it out, throws the towel with Speed Error and after 14 seconds of circular dance drops to ATTI.

DMX_MT Posted at 2018-6-26 16:04
Hello RGMG, infact you are right !

You are always right BOSS !

Yeah, it's not always an obvious issue but something to keep in mind.  

AlexQB Posted at 2018-6-26 14:06
Hello Mirek,
 First of all, wooowww !! You left me very impressed! Many thanks for the accurate analysis of the flight, especially for the final moments. What privilege you spent so much time analyzing my fateful flight.

Hi Alex,

Can u tell us your stick setting? I can adjust that in my program, nice to learn from that after reading the good comments of Mirek.

Thanks,
Cheers
JJB

Mirek6 Posted at 2018-6-26 13:08
First – thank you @DMX_MT for directing me to this thread. It is very interesting.

Alex,

Hi Mirek,

interesting analysis, have few questions :

Distance to HP doesnot change much in the period 4m8s and 4m16.s , within 1 meter.
I don`t see high variations in the LatLon positions and drone pitch and roll angles are normal for a GPS stabilized hover.
If Spark cannot handle 1 meter GPS hold position (due to wind fluctuations) and give it up with a speed error, DJI should re-design this great little craft!

Indicating speed in logs are calculated values form the X and Y axis values, not sure how they are measured but guess its kind of inertia related speed.In normal sitaution this calculated value equals the GPS LatLon versus time calculated speed. What i see is that when GPS data is corrupt the speed indication in the log shows low values. In the period 4m8 to 4m16 i see normal values.
At 4m17s log (my ID 2570 and further) speed and calculated values shows different values, not earlier.

PS we use csv files wich gives converted values of the encryped flightlogs, not all values are for sure 100% correct, speed value is a calculated value.
What i see is that bewteen 4m2.1s and 4m6.3 VPS height is indicating 4 - 5 meters ; flying low over a rock? (picking up magn interference?)
See that VPS height line (teal colored)

cheers
JJB

Mirek6 Posted at 2018-6-26 16:15
DMX_MT,

You say:  "if you have time I wish to learn about how Speed Error Works. Even if you direct me to a Youtube Tutorial its enough".

Dear Mirek,

Thank you for explaining this to me. It very clear now what the Speed Error is.

I asked you as you know everything in depth, and I am that kind of person who goes to check the Troubleshooting first when I buy something new.

Some Errors I didn't know what they mean, and Thank to You now I know.

Very much Obligated !

You certainly are the BEST PRO ANALYZER we ever had on DJI FORUM !

RGMGFitness Posted at 2018-6-26 17:42
Yeah, it's not always an obvious issue but something to keep in mind.

You contributions are 'PURE GOLD' !

Really Helpful !

RGMGFitness Posted at 2018-6-26 17:42
Yeah, it's not always an obvious issue but something to keep in mind.

Someone sent me a 'friend request' on the forum.  I accidentally deleted it and doesn't appear you can recover it once a selection is made so have no idea who it was.  If that was you feel free to resend.  

sorry to read that

RGMGFitness Posted at 2018-6-26 11:52
I'm not expert on 'Compass Interference' and I didn't see anything in the DJI SPARK manual regarding 'Rocks'.  But in other DJI products like the inspire I've seen specific notes reading; "The ideal place to calibrate is a wide open location that is free of anything metalilic within a 20 ft radius.  Keep away from drainage pipe, irrigation systems, rocks, sidewalks (or other concrete structures), etc.

Obviously, a lot of rock up there on that mountain.  And not being a geologist in all likely hood that mountain could contain all types of metals potentially interfering with signal.  Again, food for thought...I'm not an expert.    Curious, if you were getting any 'Compass Calibration' errors?  Also, did you do a 'Compass Calibration' while on the mountain?

I am a geologist. Certain rock types contain iron oxides and could certainly interfere with magnetic readings. The summit of that mountain is made of gabbro, which is chemically very similar to basalt. I don't know about the specific composition of that mountain, but Gabbros can contain iron oxides including magnetite, which is (surprise, surprise) magnetic. It would be unusual for a gabbro to contain enough iron oxides to influence a compass, but there were reports of basalt influencing marine navigation in Tasmania, so it's not out of the question that that is the cause of the compass on the spark going wonky.

JJBspark Posted at 2018-6-26 23:00
Hi Mirek,

interesting analysis, have few questions :

JJB,

Good questions and analysis.

You say: “Distance to HP does not change much in the period 4m8s and 4m16.s , within 1 meter.
I don`t see high variations in the LatLon positions and drone pitch and roll angles are normal for a GPS stabilized hover.“

Yes to first sentence. However variations of pitch and roll are there. Consistent with unstable air (wind) and Spark hunting for position, or both.
Pitch variations in 12 seconds switch semi-randomly from -7.9 to plus 11 degrees (quite a wobble), and roll is between -3.8 to 7.7 degrees.

At 4m 1.8s the craft has finished its decent and it should hover in place with strong GPS signal. Even if there is some breeze.
Instead, its starts hunting in a circle. I suggested both wind (its circular path is somewhat consistent with wind direction in which Spark flew away a bit later) and magnetics interference from rock containing hematite or magnetite.
If it was gust of wind only, Spark would overshoot its commanded position and than start returning along ellipse while being pushed by wind and constantly correcting its position.
Here, Spark circled. Either wind was shifting (possible) or we had a considerable mismatch between IMU and compass causing toilet bowl effect.

I concluded that it was speed error since, as you mentioned, GPS position was hardly changing yet calculated GSP speed was very different than horizontal speed. I do not know how both speeds are derived. But they did not match.
We have also seen before that speed/yaw/magnetic errors are linked. Complex circuitry on board of Spark may misinterpret some mismatching signals erroneously. For example, if Spark is confused because of magnetic interference, its controller may throw in yaw errors, weak GPS signal etc. Once problem begins and deepens, controller and firmware may get confused. This is why I usually do not take any particular metrics at face value but compare to other metrics and try to see patterns.

Thank you for explaining how speed in logs is derived – makes sense. If we knew these details, the analysis would be easier. In this case we have issues with Spark combined with possible inaccuracies of how logs are calculated.

Yes at 4m2 to 4m 14s Spark was definitely flying low. I confirmed it with video. Magnetic interference is distinct possibility.

DJI should also look at stability of this Spark’s IMU. The final root cause may be a perfect alignment of several problems – some wind, some magnetic interference from rocks, some issue with IMU and firmware which got all confused.

One last thing – did you notice that Spark switched to ATTI around 4m 15 seconds (starts its drift with no stick input – outside of pilot truing to raise it up) yet logs show P-GPS until 3 seconds before crash?  

Cheers!

Mirek

Mirek6 Posted at 2018-6-27 05:08
JJB,

Good questions and analysis.

Hi Mirek,

Nice to have a decent conversation.

Agree that after descent it should be (more) stable than in this log. Little circling very good seen by you!
Mismatch signals evident, that`s why errors. I would have expected to see an general IMU error instead (wich doesn`t exist i think) or in this case a compass or gps error and not a Speed error. With this minor circling with verly low speed....nah, mayby DJI should look at this and rewrite some code. Having said that, these errors (yaw,compass, speed) are not seen in the GoApp on the screen! Wish DJI could give  a general pre-warning message to inform and make us (the OP) alert of the upcoming action wich is ATTI. (ok. not always but often).

Speed errors are also seen when the motor(s) RPM is too high, that`s why i thought that was mayby the case flying at a relative high altitude. Have to look in other files to see the correct msg is that occurs, "speed" is mentioned and some other word as well.

We both know that DJI will not inform us about al these errors, so not going to ask them for a desciption of all of specific speed errors.

i will have another look at the file for your last remark about raising height.

EDIT:at 4m17.7 msg: aircarft GPS signal to weak
at 4m19.6 switching to OPTI mode (vps height active)
at 4m20.2 back to GPS mode
at 4m23.6 msg: weak GPS signal
at 4m24.6 EXIT GPS into ATTI
at  4m25.4 msg: exit GPS mode.... flight log file is written every 0.1 tick of a second, not always in the right order (if you think in seconds: within 1 seconds all is correct)

Input RC for throttle up/dn is correctly done by Spark (action always fraction later than the inputs, depens of counteracting movement of the drone)
So i do not see strange things at the end of the flight. (see chart, my pitch is up/dn because OP did not use RC mode 2)





cheers
JJB

JJBspark Posted at 2018-6-27 06:33
Hi Mirek,

Nice to have a decent conversation.

JJB,

When you are looking for stick actions, please note that Alex is not using standard Mode 2 configuration. I am not sure if your program accounts for it but this may throw you off :-).

With motors and speed. Perhaps Speed error concerns  both - either horizontal speed mismatch (this case) or vertical speed mismatch (some other case).

I will be tempted to hold Spark in my hand and rev it up still holding. Thrust up would be very strong yet vertical position would be not moving. I wonder if it would throw Speed error in such case. Worth experimenting.

I also read on phantom forums that if horizontal speed does not match changing GPS position, Speed error is displayed  and drone is automatically dropped to ATTI because GPS system may be compromised.

Mirek  

JJBspark Posted at 2018-6-27 06:33
Hi Mirek,

Nice to have a decent conversation.

Continuing my post to you - to help you out a bit.

In Alex case:

Rc elevator = throttle (up/down)
Rc throttle = pitch (back/forth)
Rc aileron = yaw
Rc rudder = roll (sideways)

Alex is left handed so he configured his sticks differently than we are used to.
Perhaps your program already can figure it out. If not - perhaps a nice addition to your program's functionality? :-)

Mirek

Mirek6 Posted at 2018-6-27 06:56
Continuing my post to you - to help you out a bit.

In Alex case:

thanks, i even have this written in my tool. (in the last update a bit is fallen of the line, so have to fix that)Cheers
JJB

JJBspark Posted at 2018-6-27 07:02
thanks, i even have this written in my tool. (in the last update a bit is fallen of the line, so have to fix that)Cheers
JJB

Yep. Alex definitely falls under your configurable Mode N :-)

Hi people!

Very good news received today from DJI Support.

They confirmed that mine is a warranty case. They offer me a 100% off coupon for a new Spark.

Great deal received from DJI Support.

Thank you so much for your help!!!

AlexQB Posted at 2018-7-3 03:05
Hi people!

Very good news received today from DJI Support.

Hi ALex,

Very good news from DJI ! Good that they took their responsibility.

Hope your new Spark will not suffer from this type of errors, so happy and safe flying !

Cheers
JJB

JJBspark Posted at 2018-7-3 03:22
Hi ALex,

Very good news from DJI ! Good that they took their responsibility.

Thanks JJBSpark!

AlexQB Posted at 2018-7-3 03:05
Hi people!

Very good news received today from DJI Support.

Alex,

Thank you for sharing it with us.
Good news for you and another valuable piece of information for the forum.

This pretty much confirms our analysis and allows us to learn so we can help other pilots in the future more effectively.

Mirek

Mirek6 Posted at 2018-7-3 08:35
Alex,

Thank you for sharing it with us.

Hi Mirek
Thank you for your accurate analysis. You really helped me so much!

This accident made me learn a lot more than I had done before about the Spark.

Thanks again!

Mirek6 Posted at 2018-6-27 05:08
JJB,

Good questions and analysis.

Is there any information that this crash occurs only with the new firmware?

Great news! Hope you're up flying again soon!

DMX_MT Posted at 2018-6-26 07:12
Kingram why don't you get a Mavic Air ?

When arms are closed its almost smaller than the Spark, and its far better.

it's loud and the Mavic Pro 2 will be out soon. bad advice.

Mirek6 Posted at 2018-6-26 15:45
Alex,

Hmmm...

you can replace the IMU module yourself. it's a $30 part.

bobsma Posted at 2018-7-4 14:35
it's loud and the Mavic Pro 2 will be out soon. bad advice.

How much will be the Mavic Pro 2 cost Under $1000 ? Will you pay the rest for him, if hes on a Budget and Price Hunting ?


Lol its Loud ! As if not all Drones are mostly loud due to the High Rev Propellers....

I gave Kingram some advise and links to a deal I was offered. Thats helping others, not spreading nonsense, Bobby.

djiuser_bC5M8YoiYr6N Posted at 2018-7-3 23:08
Is there any information that this crash occurs only with the new firmware?

djiuser,

Latest firmware is not new. It is now seven months old. If it was not replaced or fixed yet, it probably is as good as it gets. I would not second guess DJI here. I use latest firmware with no problems.

I cannot say what failed in this case. It may have been h/w it may have been firmware bug.
We will never know. We did recognize that Spark's behaviour seemed to be anomalous and it turned out to be true. This is what really matters - anything else is just guessing.

bobsma Posted at 2018-7-4 14:38
you can replace the IMU module yourself. it's a $30 part.

And why would you do it if DJI replaces your entire drone for free - as is in this case?
With additional warranty which you would lose if you replaced IMU yourself? :-)

AlexQB Posted at 2018-7-3 03:05
Hi people!

Very good news received today from DJI Support.

Hi everyone,

There has been an inconvenience in my case.

DJI says that the warranty covers my case, but since the version of my drone is not the European one, they can not ship to Europe, yes to Asian countries.

I bought the aircraft from a Chinese company but made the shipment from Amsterdam (CE).

Does anyone have any idea how to solve it? I told them that I can take care of the shipping costs. I am now waiting for your answer.

AlexQB Posted at 2018-7-5 04:41
Hi everyone,

There has been an inconvenience in my case.

I had a similar problem but instead of Asia it was the USA. The original AC was bought in The US, they couldn't send the replacement o Europe. So on the order (the way they replace your AC is they give you a coupon for a 100% free new Spark usable in the shop) I had to give a US address. My wife actually told me about a site that provides US addresses for European buyers. I used it, got a US address, "bought" the new Spark, shipped it there. After that I payed somewhat of a huge fee to ship it to Romania (around 155$).
I'm sure you can find something similar for Asia (even though Asia's not a country). Good luck!

Antipaxi Posted at 2018-7-5 05:29
I had a similar problem but instead of Asia it was the USA. The original AC was bought in The US, they couldn't send the replacement o Europe. So on the order (the way they replace your AC is they give you a coupon for a 100% free new Spark usable in the shop) I had to give a US address. My wife actually told me about a site that provides US addresses for European buyers. I used it, got a US address, "bought" the new Spark, shipped it there. After that I payed somewhat of a huge fee to ship it to Romania (around 155$).
I'm sure you can find something similar for Asia (even though Asia's not a country). Good luck!

Hi Antipaxi!



Thanks a lot for your answer. Yours is a nice solution. I will wait the answer of DJI and then i will decide what to do.

AlexQB Posted at 2018-7-3 03:05
Hi people!

Very good news received today from DJI Support.

HI Alex, congrats on getting your spark replaced.  DJI has come through once again.

Sparky_17 Posted at 2018-7-5 05:56
HI Alex, congrats on getting your spark replaced.  DJI has come through once again.

Hi Sparky,

Thanks a lot!

AlexQB Posted at 2018-7-5 04:41
Hi everyone,

There has been an inconvenience in my case.

Hi Alex,

What a weird story by DJI !
Are there EU an ASIA versions of the drone??

Hope DJI will 'repair' this mistake for you.

cheers
JJB