Solution for Inductive spike from BLDC motor

Thread Starter

Samsul Arifin

Joined Jun 14, 2021
1
Hi, i want to build regenerative system in my electric scooter, i try to rectifying back emf of three phase wire and convert it with dc/dc converter then save the power in supercapacitor. But when i try this, my mosfet was broken, because the inductive spike from bldc motor, for several time it can take over 2kV for several ms in period. I try to fix this using TVS diode and varistor, but it seems useless. Anyone know how to fix this? Sorry for my english and Thankyou for your expert.

This is what i get in my scope
IMG_20210615_110918.jpg
 

LowQCab

Joined Nov 6, 2012
1,104
You can not make a BLDC Motor generate Current like a Generator.
They do not work that way.
It is NOT possible.
Hopefully you didn't damage your ESC.
.
.
.
 

shortbus

Joined Sep 30, 2009
8,814
You can not make a BLDC Motor generate Current like a Generator.
They do not work that way.
It is NOT possible.
Hopefully you didn't damage your ESC.
.
.
.

You should tell the people making them that. "Both brushed and brushless DC motors can be operated as generators. However, there are some important points to consider when designing the drive." from - https://drive.tech/en/stream-content/dc-motors-as-generators And there are many other Google hits saying the same basic thing.
 

Ian0

Joined Aug 7, 2020
3,252
Firstly, there's no such thing as a "brushless DC motor". DC motors have brushes. Motors without brushes require AC to make them rotate. The phrase "Brushless DC motor" was invented by a marketing department somewhere to refer to the combination of an AC motor and a DC-powered inverter.
The type of motor used in this combination, is usually (though not always) a three-phase permanent magnet synchronous motor. It makes an excellent generator, just connect 6 diodes in a three-phase bridge.

But this post isn't about making a generator, it's about regeneration - returning some of the kinetic energy to the battery.
The inverter in the system, using either Hall-effect sensors or current sensors, switches the MOSFET bridge to keep the torque positive. All that is required for regeneration to take place is that the MOSFETs are switched at different times in order to make the torque negative.
Each motor phase will then act as a voltage source, and its inductance and the two MOSFETs will resemble a synchronous boost converter with the battery as a load.
 

MaxHeadRoom

Joined Jul 18, 2013
23,632
All of the BLDC and PM AC motors I have used will generate when back fed.
In fact I rely on that for `scope aligning the commutation tracks (previously hall effect).
The term BLDC is because the motor represents a DC brushed motor, turned inside-out.
Hence only two windings are energized at any one time.
The 3ph version is practically identical and either motor can be used in either way as long as it has correct commutation.
The BLDC version has slightly higher torque capability.
Incidentally, when both motor types are back-fed (rotated) they both produce 3ph AC sine wave.
 
Last edited:

LowQCab

Joined Nov 6, 2012
1,104
I'm not pointing fingers, it just seems that everyone is missing the point ........

The Thread-Starter is trying to "harvest" the "noise" present in his Motor's-Windings,
while the Motor is being powered by the factory configured ESC,
and store it in a Super Capacitor, not in the Battery.

He didn't state what he intended to do with the Power stored in the Super-Capacitor.

He is not trying to figure out how to use his ESC in Reverse, ( Braking-Mode ),
which would have somewhat of a regenerative effect.

He is adding components to the existing ESC / Motor Circuitry,
it ain't gonna work.
He may as well simply charge the Super-Capacitor from the Battery.

Although, now that I think about it, 3-Zener-Diodes with a
Voltage rating in excess of the Battery-Voltage, might be able to collect something.
But they will probably just interfere with the feed-back to the ESC.
.
.
.
 

shortbus

Joined Sep 30, 2009
8,814
What has all this got to do with what the Thread-Starter is trying to accomplish ??
It has to do with your statement-
You can not make a BLDC Motor generate Current like a Generator.
But they will generate when back driven, like going down a hill or just coasting.


The Thread-Starter is trying to "harvest" the "noise" present in his Motor's-Windings,
while the Motor is being powered by the factory configured ESC,
and store it in a Super Capacitor,
You do have a higher reading ability than me, I guess. I never saw or still don't see where he said any of that in the original post. I take that "noise" as being what happens when he lets off of the throttle and is coasting.
 
Last edited:

Ian0

Joined Aug 7, 2020
3,252
When it's coasting it should generate a set of three 120°-offset sinewaves, with a peak-to-peak voltage rather less than the battery voltage. If the motor is not being driven, there is nothing that would generate narrow, high voltage pulses. That's why I think @LowQCab 's interpretation is probably correct.
There's little point in harvesting energy from coasting - from an energy point of view, keeping the kinetic energy as kinetic energy is about 20% more efficient than using it charge the battery, and then using the motor to regain the speed.
Regeneration is only a good idea if you actually want to slow down!
 

MaxHeadRoom

Joined Jul 18, 2013
23,632
To get an idea of the output, here is the Renco sheet on aligning encoder commutation tracks when installing a new encoder.
The 3ph sine waves can be seen from the stator of a BLDC motor when back-fed rotated.
 

Attachments

MaxHeadRoom

Joined Jul 18, 2013
23,632
Also that article tends to dwell on the act of using 3ph for BLDC, in place of the normal BLDC practice of energizing two winding's.at a time. As without feedback, they tend to cog in low rpm's.
I have used the BLDC motors extensively in CNC applications with torque mode drives and under PID control, they are capable of extremely smooth response down to zero RPM and with slightly more torque than the 3ph versions.
As smooth as any high quality DC brushed versions that they replaced. ;)
.
 

shortbus

Joined Sep 30, 2009
8,814
My comment was meant to show the fact that a BLDC motor uses a trapezoidal wave not the sine wave that a couple of the others claim.

And that the ringing shown in the original post is a normal part of driving a BLDC.
 

Ian0

Joined Aug 7, 2020
3,252
My comment was meant to show the fact that a BLDC motor uses a trapezoidal wave not the sine wave that a couple of the others claim.
But the article from Digikey that you referenced says how much better it is to drive a “BLDC“ motor with sine waves.
BLDC motors are proving a popular alternative to conventional brushed types because of their performance and reliability advantages. For many applications, trapezoidal control is satisfactory, but if the designer is tasked with enhancing efficiency, reducing electrical and acoustic noise and improving torque delivery, sinusoidal control should be considered.

While sinusoidal control adds complexity and cost, development tools, capable MCUs and integrated driver ICs have considerably eased the design process, making sinusoidal control more practical and less daunting. Better yet, the development tools’ flexibility and driver ICs’ adaptability allow designers to fine tune the motor for their application, and focus more on product differentiation.
 

Ian0

Joined Aug 7, 2020
3,252
From my experience, it is only better (3ph) when used open-loop.
I can believe that - I imagine it is a little bit tricky synchronising a sinewave to the Hall sensors. My main experience is with steppers (another permanent magnet synchronous motor), which are mainly run open-loop, and they do run rather nicely from sinewaves.
 

MaxHeadRoom

Joined Jul 18, 2013
23,632
The 3phase version does not usually use Hall (or equivalent) sensors.
A common method used in ECM motors, HVAC fans etc, is to initially exercise the motor to find the location of the rotor and from then on use electronic feedback,
.
 

Ian0

Joined Aug 7, 2020
3,252
Looking back at the first post, the pictures shows spikes of 220V coming from a battery-powered scooter.
As these motors are all driven by triple half-bridges, which will have the outputs clamped to the power supply rails by the body diodes, where are the spikes coming from?
 
Top