Good Evening guys, [question also posted on PhantomPilots]


Trying to help out a friend, we both fly phantom 4 drones and just tonight his drone fell out of the sky. The red and green lights appear to have turned off and connection was lost, motors powered off. He cannot recall but the battery lights seem like they were still on when it landed on the ground, but the battery was out of the shell from the impact.


It happened shortly after take off (22 seconds in) and about 20 feet above the ground, some serious damage but we are looking into some possible causes for this as the unit is out of warranty. Battery has 27 charges on a regular capacity unit 5350 mah grey p4.


I uploaded the logs to UAVdata of the crashing flight, I dont see anything out of the ordinary except for the battery temp being near 50C.


First few flights before crash -- where battery temp looks high to me (uploaded logs under my account)
http://app.airdata.com/main?share=iOpIQL
http://app.airdata.com/main?share=bOpFye

Final crash log
http://app.airdata.com/main?share=NAOPLx&page_id=GENERAL

The following errors were on his part and im just trying to make sure it was not a drone issue:
- flew about 2 hours total today
- battery was charged erratically
- flown from 95% charge to 25% charge, landed and put on charger immediately up to 65%
- then thrown back into drone at 65%


I deduced the above and he has a rough re-collection of yes being rushed to get some shots in the nice weather. The battery was not left to cool between charges, but would it cause this ??


Could the random charges and high starting temp (near 50c) cause the battery to just turn off randomly, or was it a battery disconnect ??


We fired up the drone just now, motors turn on, sensors work, but camera is fully smashed off.
This is a P4 regular 12mp model.


Any info would be greatly appreciated before we start working on this bad boy (out of warranty).

The AirData logs are not much good in diagnosing a crash, go to PhantomHelp and upload your flight log there, then return and post a link to the resultant log. Then one of the members will examine it and provide you with an opinion.

Geebax Posted at 2018-2-28 21:10
The AirData logs are not much good in diagnosing a crash, go to PhantomHelp and upload your flight log there, then return and post a link to the resultant log. Then one of the members will examine it and provide you with an opinion.

Thanks for the reply.

Is this what you mean ?
Crash flight
http://www.phantomhelp.com/logviewer/5Q9A2JHV023Z192MMYQE/

Flights leading to crash
http://www.phantomhelp.com/logviewer/DS159L9R84VV8BTHT0I9/
http://www.phantomhelp.com/logviewer/KU01HN8MF5TRFCSD12H5/
http://www.phantomhelp.com/logviewer/VWP1SQXMQ8JQRFIDC565/
http://www.phantomhelp.com/logviewer/BSK17LSM63DQ6BCDR015/
http://www.phantomhelp.com/logviewer/L0LRITTDGCFHHJE3282V/

My guess is the battery crapped out.  It looks like you have cell deviation in the battery.

ibuyufo Posted at 2018-2-28 22:30
My guess is the battery crapped out.  It looks like you have cell deviation in the battery.

That was my thought, popped battery back in after crash. Drone fired up again.
Im thinking in the quick take off maybe it didnt clip in correctly ??

Would the onboard battery have the programming to just turn off midflight if it sensed high battery temp ?

Looking for some DJI input as to what would have the power to crash the drone, then appear to be perfectly normal.

It has the gimbal ripped off now, but all sensors etc are showing normal. The pilot took off battery from charger at 65% rater than letting it fully charge/balance, cool down etc.
I just did not know it would have the ability to kill power to the drone = crash land.

Why takeoff battery is not at 100% ? Just learn to charge this batteries

Vendex Posted at 2018-2-28 22:51
Why takeoff battery is not at 100% ?

I am helping him fix it, I gave him crap for that already.

1. abused the charge process over 1-2 hours, battery on and off the charger, in and out of craft.
2. takes off with battery under 50% at times, flies for 2-3 minutes, lands, back on charger ? then resumes.
3. Battery is showing high temperatures, today was a cold day (3-4C, batt starting flight at 45C).
4. Day ends with drone crash due to something killing power to drone in the air.

Linuxglobal Posted at 2018-2-28 21:16
Thanks for the reply.

Is this what you mean ?

Yes, that's the link. I don't see anything wrong there at all, I suspect the battery was not fully pushed home. There have been some instances lately where the battery connector has got hot and the connection failed. Look for burnt contacts on the battery socket inside the aircraft and also on the battery connector of the battery that was being used at the time.

Don't waste your time waiting for DJI to comment on this, even if you send them your log, they never come back and tell you the results unless you are making a warranty claim, and even then, they mostly don't comment either. Apparently, it is not important to advise customers as to what went wrong.

Geebax Posted at 2018-2-28 22:55
Yes, that's the link. I don't see anything wrong there at all, I suspect the battery was not fully pushed home. There have been some instances lately where the battery connector has got hot and the connection failed. Look for burnt contacts on the battery socket inside the aircraft and also on the battery connector of the battery that was being used at the time.

He will be dropping it off with me soon to scout out what it needs, gimbal motors, landing gear. I will post pics of the process for those who wish to see it unfold. I Wanted to thank you for your input and the others, helps me avoid crashes with my birds as well. Im super meticoulous and take it slow so knock on wood.

Post repair: I will recommend a new battery if the drone connections look good / free of burns.
He mentioned the drone lost connection on app, and all lights turned off and this was at night so it was easy to confirm the lights did in fact turn off.
To me it sounds like the battery circuit board either said "no more juice for you" or the battery was not fully clipped in to the craft. It flew out upon ground impact so its hard to say.

I will recommend velcro straps for post repair.

Its good to get info from other members as I was stumped as to what caused this as well.
Easiest response = pilot error, but it seems to stem from rushing to get that shot, dark outside, and possible weak battery connection.

Linuxglobal Posted at 2018-2-28 22:58
He will be dropping it off with me soon to scout out what it needs, gimbal motors, landing gear. I will post pics of the process for those who wish to see it unfold. I Wanted to thank you for your input and the others, helps me avoid crashes with my birds as well. Im super meticoulous and take it slow so knock on wood.

Post repair: I will recommend a new battery if the drone connections look good / free of burns.

Although I suspect the battery can shut off supply, I doubt it is used by DJI. If it decided the battery level was too low, it would perform an emergency landing and would notify you of that. It would also appear in the logs.

Geebax Posted at 2018-2-28 23:04
Although I suspect the battery can shut off supply, I doubt it is used by DJI. If it decided the battery level was too low, it would perform an emergency landing and would notify you of that. It would also appear in the logs.

I was not there to see it so Im going by the logs and his recollection of events.The events described took place over about 2 hours tonight, with 1 battery being run down, then flown, charged, reflown again right away.

Apparently the battery was lit up green bars on the ground after the crash, supporting a bad connection to drone / not clicked in fully.

That would be the only possibility for this thing to just shut down all systems it seems.

Some pilot errors I would maybe try to avoid
- start with full or 95% battery  (cooled to room temp) not 45C
- verify battery fully clipped into place (pull on tabs)
- give drone hover time upon start up ( dont jump into flight)
- avoid night time flights due to lack of vision sensors

Linuxglobal Posted at 2018-2-28 23:09
I was not there to see it so Im going by the logs and his recollection of events.The events described took place over about 2 hours tonight, with 1 battery being run down, then flown, charged, reflown again right away.

Apparently the battery was lit up green bars on the ground after the crash, supporting a bad connection to drone / not clicked in fully.

Personally, I think the battery connector on the P4 range is not as good as the blade type connector used in the P3 series. And you never heard of a P3 melting it's contacts.

Geebax Posted at 2018-2-28 23:12
Personally, I think the battery connector on the P4 range is not as good as the blade type connector used in the P3 series. And you never heard of a P3 melting it's contacts.

i think it has a different logic too, i dont have a p3 but my inspire1 has very thick blade connectors and 4 little balance ?? ports next to them. Instead of the p4 style where the little pins carry all the power, battery makes decisions.  

It may be a design flaw but the entire DJI lineup basically uses that connector now, mavic, p4, p4p, i2.
Overall I think we have an agreement, battery connection failure, or debris/foreign object affected battery fully seating.

Connection issue not occur overnight, but certainly almost killed the drone.

Linuxglobal Posted at 2018-2-28 23:17
i think it has a different logic too, i dont have a p3 but my inspire1 has very thick blade connectors and 4 little balance ?? ports next to them. Instead of the p4 style where the little pins carry all the power, battery makes decisions.  

It may be a design flaw but the entire DJI lineup basically uses that connector now, mavic, p4, p4p, i2.

The blade type connectors carry current better, and are self-wiping when the connection is being made. The other thing is that they have been in use in Automobiles for decades, and are well proven. The small set of 4 connections next to the blades are a serial communication port to allow the main control board to chat to the battery.

I'm sorry to read the crash. It is a pity that data analysis will not be provided if the warranty expired. If you need any further assistance, please contact the support: https://www.dji.com/support

DJI Susan Posted at 2018-3-1 01:34
I'm sorry to read the crash. It is a pity that data analysis will not be provided if the warranty expired. If you need any further assistance, please contact the support: https://www.dji.com/support

Thank you for the reply, the crash is very minor compared to what could have happened.

Due to the machine being older, I will be ordering the parts online to swap the motor and yaw arm. All else is operating and bird still flies just camera-less at the moment.

Durable P4 afterall, just that battery

You may be able to find some more information in the aircraft logs, the last part of the log may not have been transmitted to the controller to be recorded by the Go App.

I don't see anything wrong in the log above, the battery certainly still has power in it, maybe not the amount shown by the % figure but it was not the cause of the crash.  Partial charging during the day should be OK, but I would advise always charging until the 4th LED starts flashing and always start the day at 100% having charged it to 100% within the last 24 hours.  I wouldn't consider that temperature to be high unless it had been sitting unused and exposed to the weather for several minutes, although it may be high enough that it would have refused to charge until it had cooled down, which may be why it was used at a low charge.

I suggest you check all the main power connections inside the aircraft and make sure that none of the solder joints have cracked or the cables fractured just before the solder joints, also check the battery connectors carefully and the battery latch mechanism.  There was a reason it fell out of the sky, unless it was just that the battery was not latched in properly then you need to find the reason before flying again.

Nigel_ Posted at 2018-3-1 02:31
You may be able to find some more information in the aircraft logs, the last part of the log may not have been transmitted to the controller to be recorded by the Go App.

I don't see anything wrong in the log above, the battery certainly still has power in it, maybe not the amount shown by the % figure but it was not the cause of the crash.  Partial charging during the day should be OK, but I would advise always charging until the 4th LED starts flashing and always start the day at 100% having charged it to 100% within the last 24 hours.  I wouldn't consider that temperature to be high unless it had been sitting unused and exposed to the weather for several minutes, although it may be high enough that it would have refused to charge until it had cooled down, which may be why it was used at a low charge.

Almost forgot about the onboard logs, I will load them in the DJI assistant 2 once the craft is here.

Perhaps the logs can show those last few seconds after signal was lost to the craft.

Cant thank you guys enough for all the great info.

I agree with post #6

Battery charge is critical I agree, however I will pull the DAT files to ensure that after the repair the bird will not fall out of the sky.

A new battery will be recommended as well, maybe keep this one for firmware upgrades while on the ground.

Linuxglobal Posted at 2018-3-1 06:43
Battery charge is critical I agree, however I will pull the DAT files to ensure that after the repair the bird will not fall out of the sky.

A new battery will be recommended as well, maybe keep this one for firmware upgrades while on the ground.

Might be worth opening the battery to look for loose connections/broken wires etc. If that is where the problem was then it will give piece of mind when you find it and if you are not going to use it for flight then you don't need to worry about damaging it.

Nigel_ Posted at 2018-3-1 07:47
Might be worth opening the battery to look for loose connections/broken wires etc. If that is where the problem was then it will give piece of mind when you find it and if you are not going to use it for flight then you don't need to worry about damaging it.

I have to get his input on cutting apart the battery. The crash is probably a bad idea to fly with the shock anyways.

Pins will be the first step and im hoping to see some gold in the DAT onboard files. Maybe some sort of failure, battery cell issue, something. Maybe a shutdown command.

Linuxglobal Posted at 2018-3-1 09:49
I have to get his input on cutting apart the battery. The crash is probably a bad idea to fly with the shock anyways.

Pins will be the first step and im hoping to see some gold in the DAT onboard files. Maybe some sort of failure, battery cell issue, something. Maybe a shutdown command.

Honestly, I would not bother with the flight logs on the aircraft. Over several years of looking at flight logs, I have never seen anything that remotely resembles a 'shutdown' command. However urban myth keeps telling you that the battery 'shut down' in flight, but these ideas mostly come from people who have no electronic experience. If you think about it, there is no valid reason for a battery shutdown command.

The most likely scenario is a broken power connection between the battery and the aircraft. It may have broken for only an instant in time, but that is enough for the processors inside the aircraft to re-boot, shutting down the motors, lights and other features while the processor re-initializes. That takes several seconds to do, and when complete, the aircraft is not left with its motors running, so the aircraft crashes to earth.

That disconnection could be due to the battery not being fully inserted into its socket, or even a bad solder joint inside the aircraft or battery. Members have posted photos on this forum of cold solder joints to the power connections on the batteries, so it is a real possibility.

In any event, I would not trust the battery that was in the aircraft at the time, so it would not be a great loss to dissasemble it and check the connections inside. Batteries are the most critical components of the aircraft, any suspicions about the reliability and it is best to discard them.

Mark The Droner Posted at 2018-3-1 15:15
https://www.youtube.com/watch?v=2ue3tXm-GtQ

That video, which I have seen before, is quite informative, but what is the relevance with regard to this situation?

See post #6

The battery took off already warm so the sag would explain why the voltage drop wasn’t as much as a cool battery was.

From the video it shows that let’s say the same battery was left to cool and then flown at 47% charge, perhaps upon take off it would show less % and possibly drop voltage a lot faster to warn the pilot the battery is low on power.

The warm temp and low charge fooled the drone into thinking all is well.

I still think the battery ejected to kill power fully.

Pics emailed now, updated shots of the drone. I will add them to main page as well.

You took off with 47%.. many people who take off at even 80% get a battery failure.

djiuser_fiQgdFqHaUQg Posted at 2018-3-2 00:37
You took off with 47%.. many people who take off at even 80% get a battery failure.

The person in question ... took off at 47% and lower in some cases.

I know what you mean.
Im usually worried when i take off at 90%.

My buddy was playing with fire, which is odd since he had a P3 before.
Seems like a lot of battery abuse, probably motor heat was involved too in other flights, this was only 22 seconds in.
Im breakding down the parts list since after the pics, there is a lot more damaged.
- gimbal arm + cable
- 1X CCW motor
- Gimbal Roll motor (possible pitch + yaw too)
- Landing gear
- Bottom shell cover
- new battery from DJI store

I will do several hoover tests with it after the repair to see if we can replicate the drop.

djiuser_fiQgdFqHaUQg Posted at 2018-3-2 00:37
You took off with 47%.. many people who take off at even 80% get a battery failure.

The log clearly shows that the battery had enough charge to fly and, was warm enough, it also was able to easily cope with the take off power without significant drop in voltage.  There is no reason to think the fall was due to low battery charge.  My P4 has taken off at 17% without issues, it does not have the same issues as the P3 and P3 batteries.  The only common problem with P4 batteries is that the % figure can be very inaccurate if the battery hasn't been charged to 100% in the 24 hours before flight and that can lead to unexpected critical landings, that wasn't the case here because we can see that it did have charge in it and there wasn't a critical landing.

https://www.google.com/search?rl ... .0....0.PDM4xFyBE80

Mark The Droner Posted at 2018-3-2 02:46
https://www.google.com/search?rlz=1C1AVNE_enUS623US633&ei=UiuZWr3bEoGyggeu14mYDA&q=%22false+reading%22+battery+site%3Aforum.dji.com&oq=%22false+reading%22+battery+site%3Aforum.dji.com&gs_l=psy-ab.12...17048.18161.0.19958.8.8.0.0.0.0.69.488.8.8.0....0...1c.1.64.psy-ab..0.0.0....0.PDM4xFyBE80

I'm not going to check all those, but a quick look shows that most of them landed due to critical voltage, they did not fall out of the sky.  A quick check of the first one shows what was expected, the battery had not been charged to full in the previous 24 hours and did not have the % charge claimed, then reached critical voltage when put under load.

The issue in this thread is that it fell out of the sky without reaching critical voltage - totally different.

Sorry - I thought it was obvious that the phrase "false reading" kind of throws a monkey wrench into that

Nigel_ Posted at 2018-3-2 01:49
The log clearly shows that the battery had enough charge to fly and, was warm enough, it also was able to easily cope with the take off power without significant drop in voltage.  There is no reason to think the fall was due to low battery charge.  My P4 has taken off at 17% without issues, it does not have the same issues as the P3 and P3 batteries.  The only common problem with P4 batteries is that the % figure can be very inaccurate if the battery hasn't been charged to 100% in the 24 hours before flight and that can lead to unexpected critical landings, that wasn't the case here because we can see that it did have charge in it and there wasn't a critical landing.

If you power on the battery from 100% it doesn't matter if you land and take off in the same one flight over and over again.

But if you power on the battery from 50% 70% 80% or even 95%. The discharge while flying can be a problem and a disaster where you can experience the power drop from 70% to 4% all of a sudden or even loss of voltage to 0. It only takes 1 cell drop to drop your copter. Let your battery discharge the best way, discharge has to be done from full power so it recognizes the process well.

djiuser_fiQgdFqHaUQg Posted at 2018-3-2 17:06
If you power on the battery from 100% it doesn't matter if you land and take off in the same one flight over and over again.

But if you power on the battery from 50% 70% 80% or even 95%. The discharge while flying can be a problem and a disaster where you can experience the power drop from 70% to 4% all of a sudden or even loss of voltage to 0. It only takes 1 cell drop to drop your copter. Let your battery discharge the best way, discharge has to be done from full power so it recognizes the process well.

The important thing to note is that the lights went off. There are two obvious ways this can occur, either the power connection was broken, or the onboard processor reset due to low voltage, but as low voltage for the processor means it has to get below 3 volts, that is less likely.

djiuser_fiQgdFqHaUQg Posted at 2018-3-2 17:06
But if you power on the battery from 50% 70% 80% or even 95%. The discharge while flying can be a problem and a disaster where you can experience the power drop from 70% to 4% all of a sudden or even loss of voltage to 0. It only takes 1 cell drop to drop your copter. Let your battery discharge the best way, discharge has to be done from full power so it recognizes the process well.

If you start the day with the battery freshly charged to 100% then you can turn it on and off whenever you like and give it partial charges, and the % will be fine, although after several partial charges it would be wise to land a bit early.

The only time you will see a big sudden drop is if it has self discharged since the last charge, then the % figure can be very inaccurate, however the voltages will still show the actual charge so checking the voltages can tell you if you will have a problem or not.

The only time a P4 is likely to fall out of the sky instead of perform a critical landing is if you use the battery having not charged it for several months.

For the P3 the batteries are different, don't mix up rules for using the P4 and P3.

Nigel_ Posted at 2018-3-3 04:09
If you start the day with the battery freshly charged to 100% then you can turn it on and off whenever you like and give it partial charges, and the % will be fine, although after several partial charges it would be wise to land a bit early.

The only time you will see a big sudden drop is if it has self discharged since the last charge, then the % figure can be very inaccurate, however the voltages will still show the actual charge so checking the voltages can tell you if you will have a problem or not.

If you turn it off even on the same day you turn it on again, battery intelligence won't be calibrated.

Nigel_ Posted at 2018-3-3 04:09
If you start the day with the battery freshly charged to 100% then you can turn it on and off whenever you like and give it partial charges, and the % will be fine, although after several partial charges it would be wise to land a bit early.

The only time you will see a big sudden drop is if it has self discharged since the last charge, then the % figure can be very inaccurate, however the voltages will still show the actual charge so checking the voltages can tell you if you will have a problem or not.

'The only time a P4 is likely to fall out of the sky instead of perform a critical landing is if you use the battery having not charged it for several months.'

I am not sure about that, there is still the issue of the battery or connection failing, and that has been clearly documented here on this forum.

Lots of good info from users on here, as an updated I will be contacting DJI as the drone seems to have some serious issues now.

I ordered all the parts to get the drone fixed up.

Check out this shocking footage in this video as it just cuts out voltage and then restarts the craft, no warning, no errors recorded.
The flight was taken with the original battery that caused it to fall out of the sky.



The next stage is to test it with a new p4/p4p battery.

Another follow up, my p4p is being used in the field at the moment so i had to improvise as I had no P4 batteries around.

I had a spare p4pro drone, plugged in the crash victim battery and guess what ..... it failed in the p4p as well. Just randomly cut out, lights on and everything.

Therefore I can say the mystery is solved, battery failre at 29 charges caused the crash.