RC ESC for EV motor

[Psi]

Oh man that YEP unit is comically small, 164 grams.

Noname ebike units are cheaper, although they all have hall connections:
https://www.aliexpress.us/item/3256804130980146.html
https://www.aliexpress.us/item/2255800947428853.html
https://www.aliexpress.us/item/3256804704639103.html

Yes, it is small, but remember, it can pump out 51V at 100A or 5kW pretty much continues if you keep it cool.
And this application only requires spinning an unloaded motor for educational purposes.
And being set in Heli mode means it will slow start up, like the main rotor blades of a heli.

Many small RC brushless motors will draw 2kW and are around the size of a golf ball.
You cant really judge things by physical size when it comes to ESCs and motors.

But i do agree that something for a Ebike/Escooter is a better fit especially if it has lots of motor settings for PID timing etc..
And if it's also cheaper its a no-brainer.

[Robotec]

I would go for VESC, it is completely open source and can be configured with its own program through UART:

https://vesc-project.com/

example of device (84V 200 A MAx for 160 euros): https://es.aliexpress.com/item/1005005197381369.html?spm=a2g0o.productlist.main.47.139a7e34YfxN9u&algo_pvid=7811e574-9de7-48c1-aa1b-28b031c3ff5c&algo_exp_id=7811e574-9de7-48c1-aa1b-28b031c3ff5c-23&pdp_npi=3%40dis%21EUR%21215.07%21159.15%21%21%21234.00%21%21%402100b88516899196291956748d0775%2112000032097774746%21sea%21ES%21175983338&curPageLogUid=aAgdpRUvPdmb

there are hundreds of devices compatible with the project from chinese or you can go with the originals :https://trampaboards.com/1-x-vesc-6-mkvi---225-each--p-27529.html

[Psi]

open source VESC sounds like a good option, you can mess with the code/timing parameters.

You don't have the potential problem where a closed source ESC might have all the hardware needed but the firmware enters an error state because off odd timing.

I would start at a lower voltage than 84V though. There is a trade off were if you have too much voltage the inrush current will be too high for the ESC and the back EMF stronger than the ESC may be expecting.

[Siwastaja]

Using very low voltages does not hurt if you just want to test/play around. I have rotated 400V rated motors with a 20V power supply just fine. Especially if the motor is efficient, like large motors usually are, it doesn't require much extra voltage beyond the BEMF (which is zero at 0RPM) to generate torque. RPM is of course very limited.

Fundamentally, motor RPM linearly corresponds to voltage at the motor, e.g. a motor rated at 50V 5000 RPM requires 10V to run at 1000 RPM and 1V to run at 100RPM, assuming zero loss. Current that goes into the motor then defines the torque it generates, pretty much linearly.

But because of losses, a little bit extra voltage is needed; motor can be mentally modeled as a voltage source with series resistance, so to actually drive non-zero current into it, you have to input more voltage than the motor's internal voltage source, as there is drop over the R. On highly efficient motors, this R is small so even small changes in input voltage lead to large changes in motor current, which is also why motor controllers should do current sense / active current limitation.

[Berni]

Yep there is no point in going to high voltages if you don't need to spin the motor fast.

Brushless motors this huge have very low resistance windings so applying only a few volts can already send many 10s of amps of current trough them. It is only when the motor starts spinning at a decent RPM and starts generating its own back EMF, that's when you need the higher voltage to overcome the EMF and continue pushing current into the motor.

The voltage required at a given RPM is the Kv rating for the motor, this is the voltage constant, it tells you how many RPM per Volt of input you can get. Hence why running at near 0 RPM needs near 0 Volts. (However you can squeeze even more RPM than Kv would suggest, using field weakening methods)

Similarly BLDC motors also have a Kt, this is the torque constant. It tells you how many Nm of torque you get for every Ampere of input current. Motors can be designed to optimize for one or the other of these. If you use weak magnets and low number of turns you get a motor with a high Kv that can spin really fast but not produce lots of torque. Conversely you can use stronger magnets, more turns on coils, more poles...etc to get a high Kt motor that will produce a lot of toque, but will also have a low Kv so it won't spin very fast.

In general an EV car motor will be designed for more of a high Kt because it needs the torque, while the low Kv can be compensated for by using a very high battery pack voltage. Such low Kv motors are easier to drive since the current control loop does not need to be as fast (lots of turns means lots of inductance, hence slow current change) and since they tend to have high Kt, they will pull the rotor alignment using less current during sensorless startup.

Electric bike ESCs are indeed a good fit since bikes also use high Kt motors, but since they don't need to spin as fast (heck a bike wheel would probably explode at 5000 rpm) means they can get away with much lower battery pack voltages in the 24V to 90V range (since some people want to go unreasonably fast on a bike).

[langwadt]

Yep there is no point in going to high voltages if you don't need to spin the motor fast.

Brushless motors this huge have very low resistance windings so applying only a few volts can already send many 10s of amps of current trough them. It is only when the motor starts spinning at a decent RPM and starts generating its own back EMF, that's when you need the higher voltage to overcome the EMF and continue pushing current into the motor.

The voltage required at a given RPM is the Kv rating for the motor, this is the voltage constant, it tells you how many RPM per Volt of input you can get. Hence why running at near 0 RPM needs near 0 Volts. (However you can squeeze even more RPM than Kv would suggest, using field weakening methods)

Similarly BLDC motors also have a Kt, this is the torque constant. It tells you how many Nm of torque you get for every Ampere of input current. Motors can be designed to optimize for one or the other of these. If you use weak magnets and low number of turns you get a motor with a high Kv that can spin really fast but not produce lots of torque. Conversely you can use stronger magnets, more turns on coils, more poles...etc to get a high Kt motor that will produce a lot of toque, but will also have a low Kv so it won't spin very fast.

In general an EV car motor will be designed for more of a high Kt because it needs the torque, while the low Kv can be compensated for by using a very high battery pack voltage. Such low Kv motors are easier to drive since the current control loop does not need to be as fast (lots of turns means lots of inductance, hence slow current change) and since they tend to have high Kt, they will pull the rotor alignment using less current during sensorless startup.

Electric bike ESCs are indeed a good fit since bikes also use high Kt motors, but since they don't need to spin as fast (heck a bike wheel would probably explode at 5000 rpm) means they can get away with much lower battery pack voltages in the 24V to 90V range (since some people want to go unreasonably fast on a bike).

if EVs didn't use low Kv motors and high voltage batteries the currents needed would be sky high, the powerstage would be quite the challenge and wires would need to be the size of fireshoses

[Siwastaja]

In general an EV car motor will be designed for more of a high Kt because it needs the torque

I think EV motors are balanced designs between speed and torque, where maximum efficiency and minimum motor weight are the targets, and speed/torque will be what it is. This is because torque is not exactly important as there will always be a gearbox. High rpm (6000-8000) + gearing it down for wheels (<1000rpm or so) gives overall minimum weight and maximum efficiency even if you count in the weight and efficiency of the gearbox (which in an EV is a simple fixed reduction gear, much simpler than ICE gearbox, automatic or manual.)

Direct wheel drive motors are different, but they are not usual in normal EVs. They are just very heavy as a lot of iron is required for the torque, and due to that weight you can't really mount them to the wheels directly but need CV joints to decouple the unsprung mass, and once you have that adding a tiny gearbox at the end of the motor is not a big deal anymore and makes the motor so much smaller.

Choosing the voltage/current relationship in a motor is trivial and does not affect the torque/speed characteristics except that multiplier value. It's exactly like when winding a transformer, you just choose number of turns according to voltage you want and the wire gauge (or total section when using parallel wires) according to current, while trying to maximize the fill rate within the available window. You can rewind a motor to a different voltage by just changing this relationship and a 400V 10A motor becomes a 100V 40A motor and nothing else much changes.

[Berni]

The motors in EVs can get most of the gear reduction already in the differential. Since usually they have one motor they need a diff to power 2 driven wheels, you can then choose the gears in the diff to get a desired reduction ratio for free. This is why internal combustion cars tend to have the gearbox be a direct drive 1:1 ratio in 4th or 5th gear and yet the tires don't hit 5000 rpm.

But yeah EV motors tend to spin at about 2x the typical RPM of internal combustion engines since higher RPM means more total power out of a smaller and lighter motor (So they do sometimes have gearboxes for when the ratio is too big for the diff alone to provide). But they are still fairly low Kv since running at ever higher RPM can present other challenges like how to make reliable long lasting bearings and gearboxes to handle that RPM while being low cost and efficient. The motor construction might become more complicated at some point to avoid the rotor tearing itself apart, the core losses etc... So in the end you end up with a Kv that is close to the slow spinning ebike motors.

The high Kv motors are the tiny motors in model aircraft and drones. They don't have a lot of Volts in the battery pack to work with since having 100 cell batteries is not practical, but they do want a lot of power from a light weight motor. So they often go for insane RPM, they don't need a gearbox since they can directly spin a propeller, as a bonus a high RPM propeller is also smaller. As a result they pump insane amounts of current into the motors to get any meaningful torque out of them, the ESCs even have to be specially designed for high Kv motors as they need to have very fast regulation loops to keep up. That is why those ESCs are not that well suited.

[macboy]

RC Drone or plane ESCs never need to spin the motor especially slowly but an RC car ESC does. Not all do it equally well, but some can drive the car very slowly. It works out to much less than 1000 RPM, and that's on a 2 pole motor no less.

[Psi]

A lot of RC ESCs can handle anything from like a 160KV pancake outrunner motor designed to spin a 22 inch prop on a large quad, all the way up to like a 6000KV inrunner designed to spin a 2.5inch prop on a small foamy and scream across the sky at 200kph.   And i've seen up to 11000KV and as low as 80KV in the RC industry

There are ESCs intended for a specific purpose, but in general they are quite forgiving with KV ranges. You just have to set them up right.

[Siwastaja]

A lot of RC ESCs can handle anything from like a 160kV pancake outrunner motor designed to spin a 22 inch prop on a large quad, all the way up to like a 6000kV inrunner designed to spin a 2.5inch prop on a small foamy and scream across the sky at 200kph.   And i've seen up to 11000KV and as low as 80KV in the RC industry

This causes so much physical pain to my eyes, there are these normal SI units for things you know, 160kV WTF? That would be definitely outside of radio control toys by four order of magnitudes. Or maybe you mean 160V/rpm or what? 200kph is what? Knots per hour? Or maybe km/h, if that, then say so? (Sorry for a blunt rant.) And when did that kilovolt become a kelvin-volt?

[Psi]

Sorry, i'm much more used to writing kV than KV.
Fixed.

But there's nothing wrong with kph

[langwadt]

A lot of RC ESCs can handle anything from like a 160kV pancake outrunner motor designed to spin a 22 inch prop on a large quad, all the way up to like a 6000kV inrunner designed to spin a 2.5inch prop on a small foamy and scream across the sky at 200kph.   And i've seen up to 11000KV and as low as 80KV in the RC industry

This causes so much physical pain to my eyes, there are these normal SI units for things you know, 160kV WTF? That would be definitely outside of radio control toys by four order of magnitudes. Or maybe you mean 160V/rpm or what? 200kph is what? Knots per hour? Or maybe km/h, if that, then say so? (Sorry for a blunt rant.) And when did that kilovolt become a kelvin-volt?

short hand for K_v
K for constant and v for velocity, messy but that's how it is

[Siwastaja]

Note the difference between symbol and unit. One does not write current is 5I, just because I is symbol for current; instead, I = 5A. Similarly, K_v = 100 rpm/V, where rpm/V is the unit. KV = kelvin*volt, which makes absolutely no sense.

There is nothing messy in the symbol K_v and it follows the practice of using K for constants like this, where units are then application-specific. But it's a symbol, not an unit. If you want to be concise, just don't use the symbol at all, the unit obviously tells the reader what it's all about, just like we say "1000W heater", not "P=1000W heater" or, incorrectly, "1000P heater".

Note that SI system reuses same letters in both symbols and units. V as symbol is volume, V as unit is volt for example.

[langwadt]

There is nothing messy in the symbol K_v and it follows the practice of using K for constants like this, where units are then application-specific.

sure, but it is when written as Kv when you don't have subscript

[Siwastaja]

There is nothing messy in the symbol K_v and it follows the practice of using K for constants like this, where units are then application-specific.

sure, but it is when written as Kv when you don't have subscript

Yeah, but it's never written after a number, because it's not an unit.

Using standard 7-bit ASCII is not a big sin IMHO. Programmers are for example limited to such symbols in their code, so Kv is fine for a symbol (but still not unit).