The reason the V3 is failing & falling out of the sky

[robinb]

With all the reports around us of the V3's falling out of the sky or upgraded P2 with new ESC's and motors having the same problem.

I do not think its a "thin wire" issue, though this is a smaller problem.

Quite simply I think its a problem to with an incompatibility with the ESC and the new 2312 motors.

The motors are driven by the ESC's in a 3 phase way with activation of each coil inside the motor being energised in turn 120 degrees apart to give the direction and drive of the motor.

The control circuits need to know the position of the motor so it can correctly drive the next coil required to turn the motor.
To do this the position is determined by back EMF and sensed and reported to the CPU (Silicon General F330).

However with the higher power motor the detection of the position is getting misinturpreted so the mosfets are being energised incorrectly.

This has the effect of huge amounts of extra current being forced into the motor at the wrong time causing, either
A stutter of flight with minor loss of control for a couple of seconds.
A shut down of the motor and inadvertent reversal of motor or full stop, causing the propeller to unscrew and fly off.
A catastrophic failure of the mosfets due to excess currents.

The last 2 and possibly the first can cause the quad to destroy itself as we have seen.

The reason for the misinterpretation of motor positioning could be software or component tolerances of the ESC.
Either way DJI need to come up with a solution very quickly.

 

[CapnBob]

The jury is still out on this, and your assessment may be spot-on, or not.

However, I agree wholeheartedly that the motor leads are not the issue. They are an extension of the windings, which are of a heavier gauge wire than the original, "V1", or whatever you want to call it. Something else is afoot here, and it cannot be remedied with thicker lead wire. Since we are actually talking about a frequency component in the D.C. to the motors here (pulsation via ESC speed control), there may be a standing wave voltage issue causing current spikes in certain areas. I did the "V3" upgrade, and am keeping a close eye on things.

With all the reports around us of the V3's falling out of the sky or upgraded P2 with new ESC's and motors having the same problem.

I do not think its a "thin wire" issue, though this is a smaller problem.

Quite simply I think its a problem to with an incompatibility with the ESC and the new 2312 motors.

The motors are driven by the ESC's in a 3 phase way with activation of each coil inside the motor being energised in turn 120 degrees apart to give the direction and drive of the motor.

The control circuits need to know the position of the motor so it can correctly drive the next coil required to turn the motor.
To do this the position is determined by back EMF and sensed and reported to the CPU (Silicon General F330).

However with the higher power motor the detection of the position is getting misinturpreted so the mosfets are being energised incorrectly.

This has the effect of huge amounts of extra current being forced into the motor at the wrong time causing, either
A stutter of flight with minor loss of control for a couple of seconds.
A shut down of the motor and inadvertent reversal of motor or full stop, causing the propeller to unscrew and fly off.
A catastrophic failure of the mosfets due to excess currents.

The last 2 and possibly the first can cause the quad to destroy itself as we have seen.

The reason for the misinterpretation of motor positioning could be software or component tolerances of the ESC.
Either way DJI need to come up with a solution very quickly.

 

[captej26]

With all the reports around us of the V3's falling out of the sky or upgraded P2 with new ESC's and motors having the same problem.

I do not think its a "thin wire" issue, though this is a smaller problem.

Quite simply I think its a problem to with an incompatibility with the ESC and the new 2312 motors.

The motors are driven by the ESC's in a 3 phase way with activation of each coil inside the motor being energised in turn 120 degrees apart to give the direction and drive of the motor.

The control circuits need to know the position of the motor so it can correctly drive the next coil required to turn the motor.
To do this the position is determined by back EMF and sensed and reported to the CPU (Silicon General F330).

However with the higher power motor the detection of the position is getting misinturpreted so the mosfets are being energised incorrectly.

This has the effect of huge amounts of extra current being forced into the motor at the wrong time causing, either
A stutter of flight with minor loss of control for a couple of seconds.
A shut down of the motor and inadvertent reversal of motor or full stop, causing the propeller to unscrew and fly off.
A catastrophic failure of the mosfets due to excess currents.

The last 2 and possibly the first can cause the quad to destroy itself as we have seen.

The reason for the misinterpretation of motor positioning could be software or component tolerances of the ESC.
Either way DJI need to come up with a solution very quickly.

Yep.....What he said...HEHEHE :shock:

 

[flyNfrank]

With all the reports around us of the V3's falling out of the sky or upgraded P2 with new ESC's and motors having the same problem.

I do not think its a "thin wire" issue, though this is a smaller problem.

Quite simply I think its a problem to with an incompatibility with the ESC and the new 2312 motors.

The motors are driven by the ESC's in a 3 phase way with activation of each coil inside the motor being energised in turn 120 degrees apart to give the direction and drive of the motor.

The control circuits need to know the position of the motor so it can correctly drive the next coil required to turn the motor.
To do this the position is determined by back EMF and sensed and reported to the CPU (Silicon General F330).

However with the higher power motor the detection of the position is getting misinturpreted so the mosfets are being energised incorrectly.

This has the effect of huge amounts of extra current being forced into the motor at the wrong time causing, either
A stutter of flight with minor loss of control for a couple of seconds.
A shut down of the motor and inadvertent reversal of motor or full stop, causing the propeller to unscrew and fly off.
A catastrophic failure of the mosfets due to excess currents.

The last 2 and possibly the first can cause the quad to destroy itself as we have seen.

The reason for the misinterpretation of motor positioning could be software or component tolerances of the ESC.
Either way DJI need to come up with a solution very quickly.

Is this an echo I hear from what I have tried explaining from the get go?

So many got on the kick of saying the new motor has a thin wire. Well they need to place a new motor next to the older one, and then they'll see what thin wires are.

robinb, as I say you pretty much have repeated what I have said for some time, except a few things. It's more then likely the root to the issue is in the propulsion system that drives the ESC's. And the worse part to this is, we can not do anything on our side to remedie this where it needs to be "if" we choose to use the new motor combo. At some ppoint someone might be able to come up with a band-ade on some other area, but doing that will most likely only cause something else to go wrong.

I have data showing how the voltage from the battery was effected while going through a current draw. But there needs to be much more data then I have.

 

[mede8er]

Interesting hypothesis Robin...I'm awaiting the outcomes of hunch and burl's attempts to relieve wire tension with their current fix, if it works consistently in the field that would be great, if it too fails, however, it will give your explanation more validation...

One way or another the lack of response on DJI's end smacks of denial....

 

[robinb]

Interesting hypothesis Robin...I'm awaiting the outcomes of hunch and burl's attempts to relieve wire tension with their current fix, if it works consistently in the field that would be great, if it too fails, however, it will give your explanation more validation...

One way or another the lack of response on DJI's end smacks of denial....

I don't think tension has anything to do with it, though sharp turns or kinks in cable can cause problems especially when large amounts of RF energy (EMF) is being generated by the ESC's and motors.

The timing problem does account for the type of problems being seen out there.

 

[robinb]

So many got on the kick of saying the new motor has a thin wire. Well they need to place a new motor next to the older one, and then they'll see what thin wires are.

robinb, as I say you pretty much have repeated what I have said for some time, except a few things. It's more then likely the root to the issue is in the propulsion system that drives the ESC's. And the worse part to this is, we can not do anything on our side to remedie this where it needs to be "if" we choose to use the new motor combo. At some ppoint someone might be able to come up with a band-ade on some other area, but doing that will most likely only cause something else to go wrong.

I have data showing how the voltage from the battery was effected while going through a current draw. But there needs to be much more data then I have.

I think if DJI were to do a redesign of the ESC the replacement of the mosfets and wire counting points would be much better at the end nearest the motor. This would allow for shorter motor wires to be used. Also this would help reduce the timing delay in the back EMF measurement as ever uSecond counts when the motors are going full tilt.

The propulsion system as you call it are all on the ESC, the main board tells the ESC to go and the speed it needs to go at for each motor, the ESC's then do their thing, which have there onboard CPU the positional timing are critical for the motors. Any failure in this will result in what you have personaly seen along with other failures we have seen on line.

 

[Mori55]

Since I did the upgrade awhile ago on my plus that came with esc labeled v2 and had the single beep and the correct red light pattern , I thiught I was ok to go ! I flew it a couple of times and it was fine. Now I'm concerned. It also seems that there's no rhyme or reason to what v2+ had the updated esc. I purchased mine when it first came out. Others have posted that they had bought theirs in later and they had the older esc with the 4 beeps.
Have no idea on what to do now. Remove the new motors and put the old ones back ? That's alot of money down the drain.

 

[flyNfrank]

Since I did the upgrade awhile ago on my plus that came with esc labeled v2 and had the single beep and the correct red light pattern , I thiught I was ok to go ! I flew it a couple of times and it was fine. Now I'm concerned. It also seems that there's no rhyme or reason to what v2+ had the updated esc. I purchased mine when it first came out. Others have posted that they had bought theirs in later and they had the older esc with the 4 beeps.
Have no idea on what to do now. Remove the new motors and put the old ones back ? That's alot of money down the drain.

You have the same layout I had. until my crash. I sure wish I could put my old motor back on while things get ironed out!

 

[flyNfrank]

So many got on the kick of saying the new motor has a thin wire. Well they need to place a new motor next to the older one, and then they'll see what thin wires are.

robinb, as I say you pretty much have repeated what I have said for some time, except a few things. It's more then likely the root to the issue is in the propulsion system that drives the ESC's. And the worse part to this is, we can not do anything on our side to remedie this where it needs to be "if" we choose to use the new motor combo. At some ppoint someone might be able to come up with a band-ade on some other area, but doing that will most likely only cause something else to go wrong.

I have data showing how the voltage from the battery was effected while going through a current draw. But there needs to be much more data then I have.

I think if DJI were to do a redesign of the ESC the replacement of the mosfets and wire counting points would be much better at the end nearest the motor. This would allow for shorter motor wires to be used. Also this would help reduce the timing delay in the back EMF measurement as ever uSecond counts when the motors are going full tilt.

The propulsion system as you call it are all on the ESC, the main board tells the ESC to go and the speed it needs to go at for each motor, the ESC's then do their thing, which have there onboard CPU the positional timing are critical for the motors. Any failure in this will result in what you have personaly seen along with other failures we have seen on line.

My opinion is, dji should return to the old motor combo, and let the aftermarket mess with aftermarket parts. DJ should simply turn the quad's performance back up some closer to what we use to have before the vPlus came along. They restricted almost every area. The straight on speed was quicker. the ascent was quicker, the descent was faster, the tilt angle was greater. Not sure if I forgot about anything? But that's all they need to do rather then this mess they have now.

 

[mij119]

Does anyone have some kind of data on HOW MANY of these instances have occurred? I'm not talking about upgrades, but new out of the box V3s. Quite frankly, aside from a couple upgrades on this forum, and one or two similar instances on the DJI forum, I don't see this as a widespread problem. Maybe I just don't frequent enough forums but if I remember correctly, there were V2s that fell out of the sky occasionally too, apparently for no reason.

If this were a hardware problem, I would think it would be far more common than what we are seeing. The undersized wire theory is just silly IMHO.

 

[Mori55]

Well I'd like know about all v3 motor/esc failure. Whether it was a upgrade or new V3.

 

[Luap]

Well there is a simple way to find out whether this issue is real or just magnified.
Tip over you V3 and see whether your ESC fails.
I have a P2V and tipped it over at least 7 times and this was never an issue for the ESC - only the props got damaged.

 

[Mori55]

But is that the "test" or will it just fail later ? I know I've tipped my phantoms numerous times over the years and it was never a big deal. Now they're talking about not doing any abrupt maneuvers ?? That's just crazy having to worry about that while flying.
Just wondering how much of this real or isolated cases ?

 

[robinb]

Well there is a simple way to find out whether this issue is real or just magnified.
Tip over you V3 and see whether your ESC fails.
I have a P2V and tipped it over at least 7 times and this was never an issue for the ESC - only the props got damaged.

The original P2V+ can tip over and NOT burn out the ESC's.

Happened a couple of time to me and its NOT an issue.

 

[N017RW]

Assuming the construction is sound and components are of acceptable quality, ESC failure is 99% attributed to overcurrent.

There is not enough design margin between it's current spec and demand.

This would seem to be further suggested by the sensitivity to a stalled prop/motor causing failure.

I too have stalled my rev/gen 1.0 P2 motors and not had any problems.

The jury's still out but there seems to be a lot of overcomplicated speculation at work here with no data to back it up.

Occam's Razor (in the house).

 

[robinb]

when you have two competing theories that make exactly the same predictions, the simpler one is the better