Mavic Crash Battery Overcharge

[ksull72487]

I just asked this question of one of the DJI reps who has direct access to the engineers. They said there is no need to drain the batteries to 10%. It will not "teach" anything about the lower end of the curve so to speak. Never taking them below 20-30% will not have any adverse effect nor create any kind of "false" bottom to the capacity. While these kinds of chips may be in smartphones, etc, they are not in the DJI batteries.

So I guess agree to disagree. I'll go by what DJI says since they know what tech is in their batteries.

Not disagreeing with you. It's safer to fly and land with more of a charge. But preventative maintenance to let it drain down a bit every so often on the battery itself (you can do that without ever turning the motors on). A lot of this is subjective as technology isn't perfect but in a perfect world yes there is no need to drain it to 10% etc. In a non perfect world it may not be a bad idea every so often. I'm just putting the info. out there. Take the advice as you see fit. In time the tech will change.

The reason they recommend storing a battery at half is to prevent overcharge, cell inflation etc. When temperature changes so does the battery. So at 50% it has room to go up and down with out straining the battery. A 100% charge puts a strain on it to hold the voltage etc. The battery will discharge itself over the course of 10 days or 3 days. It can be set in the app. Leads me to believe DJI has a chip in the battery etc. But again I'm speculating based on my knowledge in the field. I haven't dissected it. Batteries discharge naturally over time. But seeing they have a setting for it. I have to go back on what I stated it's possible the chip is integrated into the drone. It's definitely in the battery in some way shape or form.

But again. I do not disagree with you. Your advice is correct get her home on the low battery warning. I'm just trying to over analyze and put some additional information out there which may or may not help with battery life and insure proper function. Overtime these things do have an expiration date with use. But again even DJIs information on how many flights you should get from a battery is subjective but accurate based on what they consider.

But this is an extremely important subject. No power means a dead stick. So with what Honda has said. He's absolutely right get it back sooner than later or on your 30% warning. Based on my experience after a ton of flights and usage, I'm stating it may be good to fully cycle that battery and let her get down to around 10% by just leaving it on after a flight in your living room etc. but only after many uses or 6 months etc.

We would both be millionaires if we could answer the question and produce a battery that would last forever and never have what I call a hiccup etc. lol. DJI has done an excellent job in my opinion. But nobody's done it. Apple same thing. A battery is almost like the weather. Things can go wrong or change fast.

Then we have so many things that draw from it. Sensors, Processors, GPS, Lighting which blinks meaning draw increases and decreases, Motors a gimbal and a camera etc. So many things on that board that require power and may fluctuate in what they require coming off that pack. Seriously is a lot going on. So a properly maintained battery is important.


Sent from my iPhone using MavicPilots

 

[jackgarner]

Not disagreeing with you. It's safer to fly and land with more of a charge. But preventative maintenance to let it drain down a bit every so often on the battery itself (you can do that without ever turning the motors on). A lot of this is subjective as technology isn't perfect but in a perfect world yes there is no need to drain it to 10% etc. In a non perfect world it may not be a bad idea every so often. I'm just putting the info. out there. Take the advice as you see fit. In time the tech will change.

The reason they recommend storing a battery at half is to prevent overcharge, cell inflation etc. When temperature changes so does the battery. So at 50% it has room to go up and down with out straining the battery. A 100% charge puts a strain on it to hold the voltage etc. The battery will discharge itself over the course of 10 days or 3 days. It can be set in the app. Leads me to believe DJI has a chip in the battery etc. But again I'm speculating based on my knowledge in the field. I haven't dissected it. Batteries discharge naturally over time. But seeing they have a setting for it. I have to go back on what I stated it's possible the chip is integrated into the drone. It's definitely in the battery in some way shape or form.

But again. I do not disagree with you. Your advice is correct get her home on the low battery warning. I'm just trying to over analyze and put some additional information out there which may or may not help with battery life and insure proper function. Overtime these things do have an expiration date with use. But again even DJIs information on how many flights you should get from a battery is subjective but accurate based on what they consider.


Sent from my iPhone using MavicPilots

Dji's smart batteries definitely have some goodies in them

DJI Mavic Battery Disassemble

 

[ksull72487]

Dji's smart batteries definitely have some goodies in them

DJI Mavic Battery Disassemble

Bingo!!! Good post!!! That's what I thought! Justified $89. I'll be looking closer at those pics. But that's exactly what I was anticipating your 3 cells and then the board.

That definitely confirms it. I can't make out all of it or the chips but the battery is definitely learning or anticipating. I knew it would have to. Too many chips on board requiring power and way to many pins in the battery itself.

Let it drain to around 10% on idle every so often (motors not spinning). Gurantee it calibrates. If it never sees the lower voltages after use it may give u a small variation in reading + or -.

But as Honda has said. Get it back before your well under 30% for safety.

I may dissect one myself or wait till a cleaner pic comes up to make out the board.

Battery has firmware that's guaranteed too. There's a processor and memory I can't make out but the way it sits on the board in those pics it's there. Nothing else would look like that. I almost want to ask the guy to take close ups of the silicone lol. Whether DJI updates it with the drone. I cannot confirm. I would assume they do.

I've spoke a lot on the subject but I want to say that the feedback has been great. I also want to say no criticism intended. That DJI has done a great job with the tech for as little problems as they have had. The batteries obviously will be worth purchasing from them vs an aftermarket one if one pops up.


Sent from my iPhone using MavicPilots

 

[maikm]

The battery will discharge itself over the course of 10 days or 3 days. It can be set in the app. Leads me to believe DJI has a chip in the battery

As in the previous post already found, there are plenty of chips. There's a reason DJI calls this product "Intelligent Flight Battery". I also watched it receiving separate firmware updates on Dec 26th along with the drone and the controller. All three of my batteries received a firmware update each. So it's established that there is running software in the batteries, and so I'm sure that software includes the usual self-calibration and monitoring to keep the displayed percentage as correct as possible. That's industry standard, why wouldn't they.

But the above info about the discharge is not correct. It will discharge itself after a period of 3 to 10 days inactivity. If you haven't touched a button for that long, it will discharge itself down to 65% over the course of 1 to 2 days.

 

[ksull72487]

As in the previous post already found, there are plenty of chips. There's a reason DJI calls this product "Intelligent Flight Battery". I also watched it receiving separate firmware updates on Dec 26th along with the drone and the controller. All three of my batteries received a firmware update each. So it's established that there is running software in the batteries, and so I'm sure that software includes the usual self-calibration and monitoring to keep the displayed percentage as correct as possible. That's industry standard, why wouldn't they.

But the above info about the discharge is not correct. It will discharge itself after a period of 3 to 10 days inactivity. If you haven't touched a button for that long, it will discharge itself down to 65% over the course of 1 to 2 days.

Thanks! So for short if your battery never sees 10% I'd say every once in a great while I'd get it there so it's firmware can recognize the values at those levels. Couldn't hurt, probably helps.


Sent from my iPhone using MavicPilots

 

[DerekG]

Just wanted to add that other than the damage to your Mavic (yuck) this is an incredibly interesting case study for someone with some engineering experience

Basically, as the ice began to accrue, the props became both heavier and less aerodynamic

Sensing the loss of lift the flight controller increases the PWM signal to the affected motors, which "should" increase the motor RPM (as the ESC's will be drawing more energy)

However, because of the ice (increased prop weight / decreased aerodynamic profile) the RPM's remain constant even as the power draw increases - The heavier, less efficient props are now drawing more energy to produce the same amount of lift

Eventually the motor's are spinning at 100% power just to keep the quad hovering... At this point if you had pushed the throttle stick to 100% it couldn't have risen if it wanted to (all due to ice buildup... nothing with the battery yet)

After 25 seconds or so at this power load, the battery detected a overcurrent event which was interpreted by the flight controller as LAND NOW

Once the Mavic began descending the battery voltage recovered as the motors no longer needed to spin at 100% power


Even if you had a battery that contained infinite energy the quad would have begun descending on its own soon after that alert, as even at 100% power it couldn't produce enough lift to remain airborne - Additionally (though it was cold outside) you may have run the risk of overheating a ESC / melting a motor wire / overheating a motor / etc, as that much energy in such a compact package certainly wouldn't have been good for an extended period of time. I'd be curious how flying 100% in sport mode on a nice 60 degree day for example) would compare in terms of power draw.

 

[maikm]

I'm guessing after it hit the ground for the first time, some ice came off of one of the props, causing an inbalance in weight which caused the uncontrolled yawing.

 

[grizzard]

Just wanted to add that other than the damage to your Mavic (yuck) this is an incredibly interesting case study for someone with some engineering experience

Basically, as the ice began to accrue, the props became both heavier and less aerodynamic

Sensing the loss of lift the flight controller increases the PWM signal to the affected motors, which "should" increase the motor RPM (as the ESC's will be drawing more energy)

However, because of the ice (increased prop weight / decreased aerodynamic profile) the RPM's remain constant even as the power draw increases - The heavier, less efficient props are now drawing more energy to produce the same amount of lift

Eventually the motor's are spinning at 100% power just to keep the quad hovering... At this point if you had pushed the throttle stick to 100% it couldn't have risen if it wanted to (all due to ice buildup... nothing with the battery yet)

After 25 seconds or so at this power load, the battery detected a overcurrent event which was interpreted by the flight controller as LAND NOW

Once the Mavic began descending the battery voltage recovered as the motors no longer needed to spin at 100% power


Even if you had a battery that contained infinite energy the quad would have begun descending on its own soon after that alert, as even at 100% power it couldn't produce enough lift to remain airborne - Additionally (though it was cold outside) you may have run the risk of overheating a ESC / melting a motor wire / overheating a motor / etc, as that much energy in such a compact package certainly wouldn't have been good for an extended period of time. I'd be curious how flying 100% in sport mode on a nice 60 degree day for example) would compare in terms of power draw.

Excellent analogy. The Mavic after a reasonable amount of time stopped the RTH Command as it was impossible to ascend to RTH altitude because of lift loss, and began a "no available lift" descent to a controlled crash. When it reached awareness that it was at the landing height it had no way to stop the descent and plunged directly into the ground and bounced back to landing height momentarily. The fact that it was able to rise to landing height might have resulted from ice being thrown off the rear props on the initial impact. With the stick all the way down it did land but obviously not in full control as it did a spin on the ground, which may have been due to a broken prop or some of the ice dislodged. I did notice ice on the forward props but don't remember seeing ice on the rear, or it wasn't as obvious. (I was busy looking for one leg extension as well as the rear height extension.

The lesson for me as a pilot is: Don't fly the drone in conditions that you yourself would not fly in!!! It would have been illegal in my case as vision was limited to about 200'. Guess I forgot what I learned in the distant past, as I remember scraping snow from the wings of a Tri-Pacer at one time to make it flyable to Arizona.

Mentioning mist and icy conditions in DJi's Manual might be a good warning to be included as a flight precaution. Your suggestion as to measuring the temps on a warm day in Sport Mode would be very interesting.

 

[grizzard]

I'm guessing after it hit the ground for the first time, some ice came off of one of the props, causing an inbalance in weight which caused the uncontrolled yawing.

Exactly my thoughts.

 

[ksull72487]

Exactly my thoughts.

Yeah I'll admit I went off on a tangent about battery maintenance. I'd agree. This wouldn't have solved the issue or prevented it. ICE can form easily. Ground temps are always much warmer. A few hundred feet and you could be talking 10 degrees plus humidity and winds. Friction should prevent ice build up although as the blade spins but at the same time there are way too many variables and the temperature from one place to another can change fast. Personally I'm not flying my Mavic in anything lower than 37 degrees Fahrenheit. Reason being the average car considers that the temp in which Ice can form and I think is a reasonable buffer to potentially avoid ice build up. If I did fly in a lower temperature I'd keep the altitude lower as ground temps are always warmer and that's where your average weather forecast is going to be coming from about 20 feet above ground. I think that would be sufficient advice for those who do fly in extremely cold weather. Low Altitude.


Sent from my iPhone using MavicPilots

 

[erkme73]

Would the application of hydrophobic coatings to the props prevent this kind of phenomenon? I wouldn't fly in these conditions (I live in FL), but if it's something I had to do, I'd look into something like this:

To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.

Accept third party cookies

 

[grizzard]

Would the application of hydrophobic coatings to the props prevent this kind of phenomenon? I wouldn't fly in these conditions (I live in FL), but if it's something I had to do, I'd look into something like this:

To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.

Accept third party cookies

Cools stuff .... now if they use that on commercial aircraft props .... I'm good to go! Must be a catch though ..

 

[Capt Drone]

Same exact thing happened to me only I didn't crash. Got the warning like you did and everything. There was plenty of visible ice on the craft when I landed.

 

[grizzard]

Same exact thing happened to me only I didn't crash. Got the warning like you did and everything. There was plenty of visible ice on the craft when I landed.

This may be more of a problem than people realize. It's been wintertime since the Mavic started shipping and people being normal want to get out as soon as they can and fly ... and they do, regardless of the weather. Some crash and we have no idea why and likely sent back as defective or MIA. I see no reference to icing in DJi's Mavic Pro manual and really think there should be some reference to the danger. When the Dewpoint matches the temperature you have fog. If the temperature is below freeaing you have ice crstals forming on surfaces. Add motion and you increase the accumulaton (props/wings/etc). Pilots are accutely aware of the danger and are always cognizant of the dewpoint/temp relationship. Probably we should be too.

 

[Anthony]

While DJI don't seem to list temperatures etc they do say DO NOT USE in Snow, Fog & Rain in the user guide ...

 

[grizzard]

Good catch. I should re-read I think. Now if they would just expand on the reason why, then it might stick in my brain!!. But I'm betting where they developed and tested the Mavic the weather might be different. Just guessing though.

 

[BudWalker]

The battery over discharge error doesn't need icing conditions to happen. Here is a flight with my Mavic where the temp was 52.2 degrees and dew point 43.5 degrees. I had switched to Sport mode, applied 80% elevator, and then added 90% throttle. The result was that cellVolt3 dropped below 3.6 which then triggered the battery error.



Motor data didn't show any evidence of prop icing.

 

[grizzard]

Indeed .... good discussion on this. I think I did see this battery error once before on another day, only this time icing gave me a semi-controlled crash. You didn't have to do a RTH did you? I think icing only becomes a threat when the outside temps are below freezing ... AND the air is saturated.

 

[BudWalker]

No, I just backed off on the elevator and throttle and the battery error went away. The humidity at the time was 63%.

 

[TLT]

Just want to add that I saw the same warning on my Mavic. Was flying it over the ocean off the Pacific Coast of Oregon on a foggy morning about 40 degrees out full speed in sport mode. Saw the warnings, popped it out of sport mode and slowly brought it back to land. Guess I got lucky. Very happy it it didn't just lower itself into the ocean! I think from now on I will only fly it over water in temps above 40 and dry air to ensure it's safety. It was certainly unnerving when it happened.