Odd question in learning VESC and FOC-based controllers

Joined
Jun 15, 2019
Messages
809
I'm still learning the major differences between the controller types- FOC versus VESC, modified sine, ect.- but in seeing the tiny size of things like the Grin Baserunner versus what high levels of power they can handle, I started to wonder- what is their practical limit? How Big can a VESC get? Common FET-based controllers seem to progressively increase in size, but even the biggest VESC-based controller I've found is maybe the size of a notebook. And why is that?
 
Not sure i understand the question? VESC is the name of the open source software originally developed by benjamin vedder (Vedder ESC), it is a FOC controller
It can and has been scaled to whatever size :D

Axiom VESC with its’ thread here on ES can be looked at as an example of a 100kW high power vesc.

As for the mosfet vs IGBT hardware architecture i’m not the best to answer.. i think each electronic component has its uses, and mosfets mainly for lower power, lower voltage than IGBTs. Maybe arlo1 or some other power electronics guy can chime in here..
 
IGBTs generally have the advantage when you want to run over about 120v. Below that, MOSFETs will have lower losses.
 
CONSIDERABLE SHOUTING said:
I'm still learning the major differences between the controller types- FOC versus VESC,
Those are two completely different things.

FOC is a motor control method.

VESC is a specific controller design chain of hardware / software. Call it a "brand" of controller, if it helps. (all the versions I've read of here on ES happen to use an FOC controller method, as well)



modified sine, ect.- but in seeing the tiny size of things like the Grin Baserunner versus what high levels of power they can handle, I started to wonder- what is their practical limit? How Big can a VESC get?
How well can you design a powerstage? ;)

There's a number of powerstage-design threads (and PS design sections of motor controller design threads) that discuss the complex problems involved; the bigger you go the harder it is to keep things in sync, keep electrical noise down, keep everything at the same voltage under high currents, etc.

Common FET-based controllers seem to progressively increase in size, but even the biggest VESC-based controller I've found is maybe the size of a notebook. And why is that?


All of the controllers you're referring to are FET-based. There are high voltage/etc controllers that use IGBTs, but the stuff you're looking at, in all classes, almost certainly just use FETs. THere's one VESC project I know of, Axiom
https://vesc-project.com/node/144
https://endless-sphere.com/forums/viewtopic.php?f=30&t=89056
that uses IGBTs.

Why are the VESC-based projects physically small?

Dont' know specifically; for that you'd have to ask the developers of the specific VESC projects. But most likely because they're using more advanced drive and control and sensing methods than the brute-force common ebike controllers.

Some of them like Axiom would make fairly large notebooks, if you're referring to the old paper kind
https://endless-sphere.com/forums/download/file.php?id=313995
though perhaps about the size of an old thick laptop-computer notebook. ;)
277250242_2768765926750898_6201572535698116187_n[1].jpg

A question for you: Is there a specific reason you require the controller to be physically large?

(A fairly small controller, given enough heatsinking and good enough design work inside, good enough software, and good enough parts in the current paths, can handle quite a lot of power.)



Typical cheap ebike controllers get big because they're very basic designs, little-changed in the last couple of decades, and cheap to manufacture (which is all that matters for them)...without the optimizations done in smaller / denser controllers, they just add more parallel FETs, and make them bigger to hold them and to handle the heatsinking with just the surface area of the simple cheap extruded casing, without adding external heatsinks bolted to them, or other cooling methods.

Bigger ones that follow the Kelly/Sevcon/etc model of a big thick flat plate of aluminum and a plastic casing are generally intended to be bolted to external heatsinks, as they dont' have enough surface area to get rid of all their waste heat without that (though many people that use them don't have any idea about this, and sites that sell them don't always mention this requirement). The big thick plate has a lot of thermal mass to absorb short heat spikes (like from acceleration from startup from a stop) but for continuous higher power operation there has to be a way to get rid of the heat that builds up in it.... If it doesnt' have a built-in thermal shutdown or rollback....poof.
 
amberwolf said:

Holy shit Amber, that literally answers just about everything I wondered. Looking at the Axiom and companies making MASSIVE DC
car hub motors like Proterra, it seems like possibly Axiom is the largest we will have for some time because everything else needs such specific control software and methods to "know" where the motor "is" in relation to the stator. Or something like that :lol: I'm still figuring this out. Thank you so much!

larsb said:
Cheers Amberwolf! :D

I always find that you add valuable information each time you post - you’re truly a cornerstone of the endless sphere.
Seconded!
 
Back
Top