Here's a decent reference - it helped me understand what's happening in these little guys - a lot different in some respects than their industrial big brothers.
Inside the Electronic Speed Control | Model Aviation
Inside the Electronic Speed Control | Model Aviation
- Joined
- May 27, 2017
- Messages
- 14,458
- Reactions
- 19,469
- Location
- Los Alamos, NM
- Site
- www.mavicpilots.com
I'm so misunderstood.I don't have a database and I'm certainly not organized enough to have notes. I usually end up searching the forum for my own posts until I find what I'm looking for. Gimme a break you guys.
I thought that you had mentioned that you had organized records. I was impressed, anyway. And you always seem to have examples on hand far quicker than I can find them. Now I don't have an excuse.
Interesting - my only niggle would be that this entire branch of discussion has been altered from my original observation of a terminal breaking loose.
A battery pack sliding out of the connector could possibly show anomalous readings from God only knows what - my assertion was that it was not from any inductive/capacitive effects.
It has gotten subverted into a discussion about the battery pack coming loose which is not the same thing. The connector pins sliding out is a function of many physical vectors while the wire breaking loose has only one.
So as interesting as the discussion is as far as what the graph looks like when a battery disconnects mechanically, I'm not convinced it has much to do with a battery that has become disconnected by a wire breaking internally.
- Joined
- May 27, 2017
- Messages
- 14,458
- Reactions
- 19,469
- Location
- Los Alamos, NM
- Site
- www.mavicpilots.com
Are you thinking that a wire breaking might be a more discontinuous effect than a battery disconnect?
I believe it would be a quicker event given the photo I posted showed it sheared away - that's why I mentioned the wire is under tension.
Do you think physically sliding out of a tight connector would be faster and/or a "cleaner break"?
There have also been incidents where the battery didn't become dislodged. My favorite theory is that there is short circuit protection circuitry gets confused and disconnects the cells from the terminal. I really don't know much about batteries - I'm just spit balling here.
I have seen that theory before and I think it has some merit. A short is simply a big current rush, if teh motor draws a lot of juice it could potentially look that way.
I just don't know if that circuitry is in there - it would make sense to be and I wouldn't be surprised if it were.
- Joined
- May 27, 2017
- Messages
- 14,458
- Reactions
- 19,469
- Location
- Los Alamos, NM
- Site
- www.mavicpilots.com
No - I think you are likely correct that a terminal break would generally be cleaner. However, there are only three kinds of variability in a physical disconnect that I can think of; gradually increasing resistance as the contact area or contact pressure decreases, arcing in an inductive system, or a period of decreasing make/break duty cycles, and none of those seems a likely candidate for the apparent reproducibility of the observed data.
Callum
Well-Known Member
- Joined
- Oct 10, 2017
- Messages
- 536
- Reactions
- 265
- Age
- 39
Thanks.
That link is a few years old and still refers to a scalar system. The system is detecting rotor position by monitoring emf in the third winding. There is no compensation for flux current.
Vercor control has been used in industrial drives for a while. When I read that FOC had made it into RC that's what caught my attention.
If that's what's in the mavic its not field orientated control as I know it. DJI should really update the specs if that's the case.
www.roboteq.com/index.php/applications/100-how-to/359-field-oriented-control-foc-made-ultra-simple
Field-Oriented Control of Small DC Motors put Drones on a Rising Flight Path | DigiKey
I think you're mistaking some very broad terminology on my part as technical descriptions. I normally don't throw textbooks at folks I don't know in a 30,000 ft discussion. When I say it acts like a stepper I intend that to impart a general sense of what's happening to make the motor go round and round. Comparing the ESC to a VFD is also functionally correct in that it controls speed by varying frequency.
This wasn't intended to be any kind of dissertation or thesis - it was a discussion on a terminal breaking loose.
Callum
Well-Known Member
- Joined
- Oct 10, 2017
- Messages
- 536
- Reactions
- 265
- Age
- 39
Is that what this thread was?I think you're mistaking some very broad terminology on my part as technical descriptions. I normally don't throw textbooks at folks I don't know in a 30,000 ft discussion. When I say it acts like a stepper I intend that to impart a general sense of what's happening to make the motor go round and round. Comparing the ESC to a VFD is also functionally correct in that it controls speed by varying frequency.
This wasn't intended to be any kind of dissertation or thesis - it was a discussion on a terminal breaking loose.
It has became a bit side tracked.
I did eventually understand what you meant by stepper. It's just not a term I've heard in the UK unless referring to a stepper motor.
Perhaps I'm getting confused by DJIs use of "FOC" but thought you were confusing vector with a scalar system.
If it was a true vector control it would have current sensing on the motor and could explain the voltage decay we see on power failure.
Anyway back to the original topic.
Have you head back from DJI?
No word from DJI yet, maybe running it through the lawyers first
Sensing the field currents is where it would get far too complicated in reporting the average current and voltage being drawn. You have a single source for your power and Occam's Razor would suggest that is what is used to report power consumption - not some mishmash of three phase pseudo sine wave applied to inductors with magnets involved
So here's a thing: Rumor has it that DJI uses the ESC to utilize the otherwise unoccupied fields that are actually acting like generators at that moment to feed back into the battery to help battery life. I didn't mention it because it is a function of the ESC and if you lose power the ESC can't actually set the configuration to allow that use.
But - maybe I'm wrong. Maybe this is what is being seen as a voltage step in the battery failures mentioned. If so then I guess I would concede that in a broad sense the back EMF is the reason for the step. I still however am hesitant to think this phenomenon could last the relatively enormous time span of 20ms or more given the current consumption of the motors - IOW the fields that are being powered would consume the energy being generated and returned to the battery (which in theory is unplugged).
So - still a can of worms.
Sensing the field currents is where it would get far too complicated in reporting the average current and voltage being drawn. You have a single source for your power and Occam's Razor would suggest that is what is used to report power consumption - not some mishmash of three phase pseudo sine wave applied to inductors with magnets involved
So here's a thing: Rumor has it that DJI uses the ESC to utilize the otherwise unoccupied fields that are actually acting like generators at that moment to feed back into the battery to help battery life. I didn't mention it because it is a function of the ESC and if you lose power the ESC can't actually set the configuration to allow that use.
But - maybe I'm wrong. Maybe this is what is being seen as a voltage step in the battery failures mentioned. If so then I guess I would concede that in a broad sense the back EMF is the reason for the step. I still however am hesitant to think this phenomenon could last the relatively enormous time span of 20ms or more given the current consumption of the motors - IOW the fields that are being powered would consume the energy being generated and returned to the battery (which in theory is unplugged).
So - still a can of worms.
Callum
Well-Known Member
- Joined
- Oct 10, 2017
- Messages
- 536
- Reactions
- 265
- Age
- 39
Was thinking some more about it at work. Here's my theory. I'm not saying it's correct, just how I imagine it.No word from DJI yet, maybe running it through the lawyers first
Sensing the field currents is where it would get far too complicated in reporting the average current and voltage being drawn. You have a single source for your power and Occam's Razor would suggest that is what is used to report power consumption - not some mishmash of three phase pseudo sine wave applied to inductors with magnets involved
So here's a thing: Rumor has it that DJI uses the ESC to utilize the otherwise unoccupied fields that are actually acting like generators at that moment to feed back into the battery to help battery life. I didn't mention it because it is a function of the ESC and if you lose power the ESC can't actually set the configuration to allow that use.
But - maybe I'm wrong. Maybe this is what is being seen as a voltage step in the battery failures mentioned. If so then I guess I would concede that in a broad sense the back EMF is the reason for the step. I still however am hesitant to think this phenomenon could last the relatively enormous time span of 20ms or more given the current consumption of the motors - IOW the fields that are being powered would consume the energy being generated and returned to the battery (which in theory is unplugged).
So - still a can of worms.
Let's take DJI at their word and assume they have a FOC (vector) control system.
According to them they describe the ESC as "FOC sinusoidal driver".
Here's a block diagram for a FOC driver, not DJIs but pretty standard.
In a FOC system you monitor current and possibly voltage on the phases to adjust phase angle for lower flux current and higher torque.
The micro processor which calculates the torque and flux vectors samples at a rate of around 100-200 micro seconds and adjusts phase angle continually to keep vectors at reference.
In the event of a power failure, after a few hundred micro seconds the ESCs could detect an anomaly and shut output.
This is less than a millisecond and no power is being supplied to motors.
We are talking 20 milliseconds on the flight logs but there is a possibility ESC can react much quicker.
Without the motors draining any residual energy it's possible that we are viewing the decay in voltage of the coasting motors.
Another test would be to pull battery at a low RPM and then high RPM. If back emf in the motor is causing the slow decay you should see a difference depending on inertia of motor.
Anyway that's me done. Just my theory, thanks for humouring me, and if nothing else I've learned a bit. Cheers.
Callum
Well-Known Member
- Joined
- Oct 10, 2017
- Messages
- 536
- Reactions
- 265
- Age
- 39
It might not detect a power failure as such. Its monitoring rotor position, flux and torque current in real time. All three phases are permanently energised 120 Deg apart and PI regulators adjust phase in response to the previous rotations feedback.
In the event of a power failure the feedback may not be what it is expecting with the given output.
I'm not sure what it's response to this condition would be but it may be to stop all activity if it does not know where in the cycle it is or has no torque current feedback to create the correct phase angle.
I'm sure it would also do the same if single phasing.
Last edited:
[QUOTE=".
The MP is deader than a doornail but I noticed an oddity in that the battery was still showing on - two solids and 1 blinky LED. So I popped it into the charger and got the expected 1 LED blink but then it went out and the charger showed red.
[/QUOTE]
Interesting as this is exactly what my battery did after a crash and a fall on to soft mud, Mavic said cell error and could not fly with that battery, did the reduce to 5% power, recharge (tech advice from DJI) and then insert in Mavic again but still same error, eventually returned battery under warranty
The MP is deader than a doornail but I noticed an oddity in that the battery was still showing on - two solids and 1 blinky LED. So I popped it into the charger and got the expected 1 LED blink but then it went out and the charger showed red.
[/QUOTE]
Interesting as this is exactly what my battery did after a crash and a fall on to soft mud, Mavic said cell error and could not fly with that battery, did the reduce to 5% power, recharge (tech advice from DJI) and then insert in Mavic again but still same error, eventually returned battery under warranty
jkirk
Member
- Joined
- Jun 7, 2017
- Messages
- 22
- Reactions
- 12
- Age
- 69
Thanks for sharing your experience Brojon. I always feel we the consumers are part of an on-going field trial that can only stand to gain, sometimes painful & costly knowledge from events, but also very important to communicate to the QC group at DJI...."Hey DJI, we love our drones but get on the ball with learning from your customers"!!! Upgrade your solder process..... You can bet if mine ever takes a nose dive or self-wills into deep 6 I'll be squalling like a cut cat to DJI.
Interesting as this is exactly what my battery did after a crash and a fall on to soft mud, Mavic said cell error and could not fly with that battery, did the reduce to 5% power, recharge (tech advice from DJI) and then insert in Mavic again but still same error, eventually returned battery under warranty[/QUOTE]
You misunderstand - my battery would not charge and showed and error when attempting it because it was "disconnected" internally.
A cell error is only possible if it can power up the aircraft since each cell is monitored internally in the battery.
BJR981S
Well-Known Member
haha, why not run it like a serious airline!
I figure, damaged blades aren't worth using if they put unnecessary stress on the motors and burn through battery faster. And you cannot remove the weird reflections in your footage if your gimbal cover is scratched. So I account for and track each crash as a way to measure my own improvement and the durability of the MP
I love data points.
I agree with you Damaged Blades are one of the major causes of fall from the sky issues. This is consistent across all quad vendors.
Scratched or damaged blades cause vibrations that will upset the IMU to the point of an internal reset in flight.
- Joined
- May 27, 2017
- Messages
- 14,458
- Reactions
- 19,469
- Location
- Los Alamos, NM
- Site
- www.mavicpilots.com
Similar threads
DJI Drone Deals
1. Mini 2
2. Mini 3 Pro
3. Mini 4 Pro
4. Air 2s
5. Air 3
6. Avata 2
7. Mavic 3 Pro
8. Mavic 3 Classic
2. Mini 3 Pro
3. Mini 4 Pro
4. Air 2s
5. Air 3
6. Avata 2
7. Mavic 3 Pro
8. Mavic 3 Classic
Forum statistics
