Day 2 on Mavic - failure and crash

Cyberpower678 · Apr 10, 2017

Yep this is definitely turning into an issue with DJI.

Chipfatbob · Apr 10, 2017

Spoken with DJI. I've raised a case and they are sending a replacement drone out, at no charge.

Gary cato · Apr 10, 2017

Chipfatbob said:
Spoken with DJI. I've raised a case and they are sending a replacement drone out, at no charge.

Good response and fast lets hope this is not a QC issue and we do not see a lot more of these in next few weeks

Andrew F · Apr 10, 2017

Chipfatbob said:
Many thanks for all the efforts - hugely appreciated.

I bought the drone with the extra flight pack, which included 2 additional batteries. Luckily, I numbered them as they came out of the box. Battery number 2 was being used during the crash flight. I've tested all three batteries and while batteries 1 and 3 click in and seat immediately, battery 2 has a slightly faulty catch on the right hand side which doesn't click in properly and required an additional push. Due to the faulty catch, it is also under load which means it has a propensity to pop out on the right hand side.

I will take this up with DJI. Frankly, having just spent a huge (to me!) wedge on this drone, I was diligent in checking everything before flight. If there is an issue with batteries not seating properly in the Mavic, when they visually (and audibly) seem to be, this is a design flaw and at the very least needs to be issued as a notice of awareness by DJI.

I'll keep you posted with their response.

Best,

Jon

Jon,

Thanks for sharing. As a community, it's good for all of us to work together and to share our wins and losses. For most of us, the cost of the Mavic is big financially or emotionally. I never feel "carefree" while flying. I'm glad you found the problem and have been able to isolate it so that you can get flying again sooner than later and have at least one less worry. Cheers!

JLane · Apr 10, 2017

Chipfatbob said:
battery 2 has a slightly faulty catch on the right hand side which doesn't click in properly and required an additional push. Due to the faulty catch, it is also under load which means it has a propensity to pop out on the right hand side.

Great work in diagnosis this. It sounds like the first plausible clue as to why there seem to be so many Mavics dropping from the sky recently. Although, I suppose there is also a chance that this faulty latch could be a result of the crash, and not the cause of it...

Cyberpower678 · Apr 10, 2017

JLane said:
Great work in diagnosis this. It sounds like the first plausible clue as to why there seem to be so many Mavics dropping from the sky recently. Although, I suppose there is also a chance that this faulty latch could be a result of the crash, and not the cause of it...

The fact that we have 3 reports in 24 hours on this forum is already an indication of a problem.

Robbyg · Apr 10, 2017

Cyberpower678 said:
The fact that we have 3 reports in 24 hours on this forum is already an indication of a problem.

Now the question becomes is it just a bad batch of batteries or is it random?
I guess we need a thread to track which batches these batteries came from.

Rob

Gary cato · Apr 10, 2017

Robbyg said:
Now the question becomes is it just a bad batch of batteries or is it random?
I guess we need a thread to track which batches these batteries came from.

Rob

Yes the common denominator seems to be so far they were all new with little mileage and not long opened, i think they were pretty much in the first week of use.

gllkoop · Apr 10, 2017

and no one think of collecting more data ?!
what FW were on the mavic thag crashed ?! (it could be some bug !!)
what is th serial numbers ? ( if this is a bad batch ...?)

JLane · Apr 11, 2017

Cyberpower678 said:
The fact that we have 3 reports in 24 hours on this forum is already an indication of a problem.

Oh yeah, I wasn't questioning whether or not there's a problem. I agree with you guys; The number of Mavics falling out of the sky is worrying (just read a new one on the DJI forums).

What I was trying to say is: The faulty latching that @Chipfatbob discovered may not have in fact been the cause of the crash. It's possible the crash impact is what has caused the latch to be faulty.

In other words, the latch may have been working properly, and the battery 100% seated correctly, and yet the Mavic shut down mid-flight. This would then point more towards an internal fault, rather than just a problem with battery installation.

I just don't want us to be jumping to the wrong conclusions as we try to figure out what the heck is going on with these things ;-)

Cyberpower678 · Apr 11, 2017

gllkoop said:
and no one think of collecting more data ?!
what FW were on the mavic thag crashed ?! (it could be some bug !!)
what is th serial numbers ? ( if this is a bad batch ...?)

How is a battery disconnect a FW bug?

Cyberpower678 · Apr 11, 2017

JLane said:
Oh yeah, I wasn't questioning whether or not there's a problem. I agree with you guys; The number of Mavics falling out of the sky is worrying (just read a new one on the DJI forums).

What I was trying to say is: The faulty latching that @Chipfatbob discovered may not have in fact been the cause of the crash. It's possible the crash impact is what has caused the latch to be faulty.

In other words, the latch may have been working properly, and the battery 100% seated correctly, and yet the Mavic shut down mid-flight. This would then point more towards an internal fault, rather than just a problem with battery installation.

I just don't want us to be jumping to the wrong conclusions as we try to figure out what the heck is going on with these things ;-)

But still 3 free falls is worrying. It could mean the latch is too hard to engage. Remember the Karma was recalled for the battery issue because of difficult latches. (The Karma should have been recalled for more than just a battery latch).

gllkoop · Apr 11, 2017

Cyberpower678 said:
How is a battery disconnect a FW bug?

do you know it is a battry problem ?
i mean do you actually KNOW? would you sign on it as a resvoled ticket?
I am QA team leader, I am dealing with FW all day long.
I also flying quads for 6 years.

I would get the
Date, hour, location, FW, calibration(last one), phone, OS, Aircraft serial,
battery Serial, battery voltage at incident, remote battery voltage........

we have at least 3 incident just in past 36 hour, we might all reeady have some more from the passing
week after following this forum,
if we are going to have all the information we might get the resone of why after this product is out
for 6 month with no known issue all of a sudden in the past week I have read about 15 crashes, some was record by video.
we had a release of a FW last week,
the RTH code WAS TOUCHED, they fixed somthing over there, it is highly possible they created a bug.

Cyberpower678 · Apr 11, 2017

gllkoop said:
do you know it is a battry problem ?
i mean do you actually KNOW? would you sign on it as a resvoled ticket?
I am QA team leader, I am dealing with FW all day long.
I also flying quads for 6 years.

I would get the
Date, hour, location, FW, calibration(last one), phone, OS, Aircraft serial,
battery Serial, battery voltage at incident, remote battery voltage........

we have at least 3 incident just in past 36 hour, we might all reeady have some more from the passing
week after following this forum,
if we are going to have all the information we might get the resone of why after this product is out
for 6 month with no known issue all of a sudden in the past week I have read about 15 crashes, some was record by video.
we had a release of a FW last week,
the RTH code WAS TOUCHED, they fixed somthing over there, it is highly possible they created a bug.

I can say for certain the FW is not responsible here. We would have way more incidents than just 3, including me. All these seem to be happening with new Mavics, less than a week old, which points to a batch issue, not a FW. The newest firmware didn't touch anything in the battery. The battery was not updated, so it should continue to just deliver power. To conclude, we already have a DAT file of a fall down where the drone managed to stay on. The battery was also not in place correctly, and we confirmed a sudden momentary loss of power. It was a quick drop, not long enough to kill the computer, but the motors were stalled. @Robbyg discovered this.

gllkoop · Apr 11, 2017

i dont know, your telling me battery fulty after a crash.... a fw can couse a cut in power as well.
all I am saying is get the data in case we need it.
if you sure about battery batch, please provide the serial number of all the involved cashes.

i am willing to start an open source project of we service wich would be a ticket system
for the community, in the end we are the biggest QA team in the world, we need tools.

Cyberpower678 · Apr 11, 2017

gllkoop said:
i dont know, your telling me battery fulty after a crash.... a fw can couse a cut in power as well.
all I am saying is get the data in case we need it.
if you sure about battery batch, please provide the serial number of all the involved cashes.

i am willing to start an open source project of we service wich would be a ticket system
for the community, in the end we are the biggest QA team in the world, we need tools.

I'm an electrical engineer and a software engineer. I know that FW and HW go hand in hand, however when you do a lot of debugging, you get a talent observing patterns that let's you make targeted debugging decisions with reasonable accuracy. If the FW .550 was really the culprit, it would have dropped my Mavic in the sky and we would most certainly see a post about the Mavic falling from the sky after upgrading. But instead we have threads about Mavics falling from the sky and they usually are brand new Mavics. We can thus reasonably deduce either the latch on the Mavic is faulty, or the DJI battery has a faulty clip. Or this could all just be a coincidence and we have 3 users who made the same rookie mistake, but 95% sure it's not the firmware.

@Robbyg, my counterpart, can more than likely confirm this.

Robbyg · Apr 11, 2017

I would not rule anything out but it does seem coincidental that the OP found a defective latch on a drone that lost power. I get the idea that the latch was damaged by the accident but a damaged latch is so rare in most crashes that the odds of it happening in this crash were already slim. Just seems too coincidental.

Now add in the other battery failures from new drones and the odds get slimmer.

As Cyberpower points out if it was a firmware bug we would have seen a lot more crashes last weekend. Luckily it was limited to new drones only and hopefully all of them are covered by a 30 day return policy.

Next weekend should give a more definitive answer to the question.

Rob

taus · Apr 11, 2017

Would you guys mind having a look at my case?

My Mavic dropped out of the sky after 5 minutes of total airtime.

This is the log: https://app.airdata.com/main?flight=5eedb4845cfd43d7a795dd306b937125&page_id=GENERALNotifications

Dat files are here: https://1drv.ms/f/s!AkNV4fqh0kOrspsFRKlh6m6RvRgdLQ

And the video from the craft is here:

I saw it drop like a stone as soon as I lost connection. Badly broken. The battery was still in the craft when I found it.

Thank you in advance...

BudWalker · Apr 11, 2017

taus said:
Would you guys mind having a look at my case?

My Mavic dropped out of the sky after 5 minutes of total airtime.

This is the log: https://app.airdata.com/main?flight=5eedb4845cfd43d7a795dd306b937125&page_id=GENERALNotifications

Dat files are here: https://1drv.ms/f/s!AkNV4fqh0kOrspsFRKlh6m6RvRgdLQ

And the video from the craft is here:

I saw it drop like a stone as soon as I lost connection. Badly broken. The battery was still in the craft when I found it.

Thank you in advance...

FLY003 is the .DAT for this incident. It appears to be different in that there are no records showing a decline in voltage. Also, the end of the recording is much more abrupt than other loss-of-power incidents. Usually, the final 2 or 3 records are corrupted with these loss-of-power incidents. Here, the records were all intact with the last record being just a partial record.

I did look at FLY004 to see if it was a continuation of FLY003. FLY004 is so short that DatCon can't figure out what the aircraft type it is. So I coaxed it some and it appears to be a .DAT created indoors.

Thank you for including FLY000 and FLY001. These appear to have been done at the factory. Might be interesting to see what those contain. Neither had a motorStart.

taus · Apr 11, 2017

Thank you so much BudWalker! I really appreciate you taking the time. It is really tough having waited for 5 months for the drone and then getting 5 minutes of use....

Very strange that there even is a FLY004 if they are done in sequence. I guess FLY004 could have been the result of powering on the drone indoors after the incident? Because no flying is possible after the crash for sure!

So, what do you think it could be?

And can we rule anyting out? I know it was not a collision, but is that obvious from the DAT file?

The fact that it goes into GPS Atti mode right before the crash is strange too I think. Because it had a very strong GPS signal just before. It looses GPS, then connection to the controller and then falls down like a rock...