Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[teddillard]

I hope this is helpful.  I put together two posts trying to distill down the basics on motor terminology here: Basic Motor Types: PMDC, BLDC, AC Induction, Synchronous and Series DC
https://evmc2.wordpress.com/2014/12/04/basic-motor-types-pmdc-bldc-ac-induction-and-synchronous-and-series-dc/

I basically did it for myself, but I know the terminology gets confusing and gets tossed around incorrectly a lot, both here and in general on the web.  I hope it helps, and if there are any errors, please feel free to correct me here or on the post's comment. 

Also, what started this is another thread where we were talking about motor performance and how the controller affects that.  I realized I really knew almost nothing about how "AC" controllers work, or much beyond the basics of PMDC motor controllers even.  So here is what I found and compiled: Motor Speed Controllers – How They Work (PMDC BLDC and AC, Torque and Speed Control)

https://evmc2.wordpress.com/2014/12/04/motor-speed-controllers-how-they-work-pmdc-bldc-and-ac-torque-and-speed-control/

 Hope this is a help to people...  please let me know your thoughts.
 

[CrashCash]

That's absolutely amazing. It even distracted me from the Orion launch. I think my work productivity is going to be zero today. Thanks.

[Doug S]

I'm an EE with some experience in driving electric motors, and it looks like you got pretty much all the basics correct. The only thing I'd add, just to clarify a point of interest to us as EV enthusiasts, is that a PMAC (Permanent Magnet AC) motor, like in many of our rides including my Zero SR, is topologically the same as a BLDC (BrushLess DC) motor. In theory, they're driven differently, but the controllers have gotten very smart and are way beyond any trapezoidal/square wave explanations anyhow. Both the BLDC and PMAC motors have permanent magnets on the rotor, with the coils placed on the stator, which are driven with AC waveforms to produce torque. The frequency of the AC waveform depends on motor RPM (from 0 to ~5700 RPM in the case of the Zero SR), and the magnitude (voltage level) of the AC waveform depends on throttle setting -- the more throttle, the more voltage is applied to produce more torque.

I don't much like the "PMAC" label, because in my mind it implies that the motor is more closely related to the cheap, fixed-frequency, plain-ole "AC" induction motors that are in things like your refrigerator, your washing machine, etc. In particular, these motors have very little starting torque compared to a PMAC motor, which has caused a lot of confusion among EV people that know just enough to be dangerous. Our motors have a LOT of torque at 0 rpm.

[teddillard]

PLEASE feel free to add that as a comment to the post.  I think the biggest source of confusion is exactly that - the BLDC and PMAC distinction, and you just made that pretty clear, pretty simply. 

edit: I have a friend that put it even more simply, so I've added this:

BLDC motors vs PMAC

This is probably the most confusing part for anybody but an electrical engineer.  So a BLDC motor is driven by DC, where a PMAC is driven by AC?  It’s really simpler than that.  The two types are virtually identical in design and structure.  They’re simply driven by a different type of waveform from the controller.  BLDC is trapezoidal.  PMAC is sinusoidal.  What that means, I guess, will have to be the topic of another post (when I actually have it figured out, maybe).  PMAC motors are also those motors that you find everywhere throughout your house, that plug into the wall, and have no speed controls, but that’s not something that is a concern to EV builders.

[frodus]

There's tons of ways to break them up, but here's one I've found easy to remember, self-commutated and externally-commutated.

Self-Commutated (via brushes):
- Series Wound DC (field and rotor windings connected in series to the controller)
- Sepex DC (Separate field and rotor winding connection to controller)
- Permanent magnet DC (Rotor windings connected to controller, Magnets provide field)

Externally-Commutated (Via 3-phase controller):
- AC Synchronous (Includes PMAC, BLDC and have 3 phases and a position encoder)
- AC Asynchronous (Includes AC Induction motors and have 3 phases and speed encoder)

[teddillard]

That just goes to show you EE types think different from ever'buddy else.   

Thanks though - I updated the post to add this: "I’m going to indicate right in the title whether there are brushes or not – or, as some would say “self commutated” (brushes) or “externally commutated” (without brushes via a 3-phase controller)."

...and it's a good thing, because I never stopped to think if Series DC had brushes or not.  As far as SepEx, is anybody using those for EVs anymore? 

[CrashCash]

Is there anything like this level of technical detail concerning charging?

When I got into this, I knew there was CHAdeMO and J1772. Then I talk to Abe Askenazi at AIMExpo and he does a facepalm and grumps about how there's at least half-a-dozen variations on J1772 and introduced me to Level 1/2/3. Kind of like having Newton teaching me F=ma. Especially since electrical physics caused me to flunk out of college.

Is J1772 just a dumb connector that outputs "lotsa DC" into anything it's plugged into? With presumably a circuit breaker somewhere to avoid being a frequent stop for the fire truck?

CHAdeMO appears to want to negotiate the volts/amps with the vehicle's CANBUS, which makes it complex and expensive to support. Is there any public info on how that actually works, or is it proprietary?

I assume Tesla Supercharging is a whole 'nother kettle 'o fish... highly proprietary, and not much relevance here, but I'd love to hear details.

[Cortezdtv]

Awesome write up

I'm no EE so I got a lot to learn

[Doug S]

Is J1772 just a dumb connector that outputs "lotsa DC" into anything it's plugged into? With presumably a circuit breaker somewhere to avoid being a frequent stop for the fire truck?

There's a lot of (mis)information out there to sort through, but as I understand it, plain-jane J1772 doesn't supply DC at all. It's just a source of AC line voltage; you'll need to supply your own charger to convert the available AC to DC voltage/current levels your battery will appreciate. In 2009 they did propose adding DC pins to the connector (the famous "combo plug") but I don't think anybody's using it yet...and in any event, the lowest voltage it's supposed to support is 200, meaning it couldn't charge many if any of our bikes -- the highest battery voltage Zero uses is the 103 volts on the DS, S and SR. DC charging is "better" in that then the conversion from bulk AC line voltage to precise DC voltage/current levels appropriate for your battery, which is the part of the process requiring big, heavy, bulky, heat-creating equipment, is then located off the vehicle.

CHAdeMO appears to want to negotiate the volts/amps with the vehicle's CANBUS, which makes it complex and expensive to support. Is there any public info on how that actually works, or is it proprietary?

Complex and expensive are relative terms. There is a minimum amount of information the charger (which is located off the vehicle for DC charging schemes like CHAdeMO) requires to safely recharge your battery quickly, without damaging or reducing the lifespan of the battery. Be thankful for that complex and expensive comm port! There is information available regarding the CHAdeMO communication protocol, but I haven't seen a very clear and comprehensive description yet. Then again I haven't really looked. :-)

I assume Tesla Supercharging is a whole 'nother kettle 'o fish... highly proprietary, and not much relevance here, but I'd love to hear details.

That's a really interesting point. Their Supercharging hardware and protocol is indeed their own invention, but Elon Musk did say a few months back that they would be releasing all their patent and proprietary information so people could use their IP -- he's convinced that any help they can give to push the EV market forward will be good for everybody, including them. So it sounds like they'd be willing to work with people that wanted to engage the Supercharger network, though I'm sure there would be hoops to jump through to guarantee safety of their equipment as well as the end users' equipment. No hackers allowed, I'm sure.

I'd be very surprised if it was possible to adapt to our machines, though...the Tesla battery pack runs somewhere near 400V, IIRC, and I'd be surprised it their system would accommodate our lower battery voltages. Even a lot of the CHAdeMO stations, which are supposed to work down to 50V, don't obey the spec and aren't compatible with our machines.

[CliC]

The Carotron excerpt implies that closed-loop vector control of externally-commutated motors can't be done without an encoder. But I've worked with several industrial drives that do exactly that. I'm not sure why sensorless vector is not used more often in EVs; maybe the demands of that application are too great. But they do just fine in a variety of industrial applications (though all the ones I've seen can accept encoder inputs as well).

Another thing that the industrial drives often do, that might be useful to hobbyist EVers at least, is self-tune. They will run the motor up and down for a few minutes and build an internal mathematical model of your motor. Not sure whether they use it for some sort of model-predictive control, or just for accurate sensorless position inference, but it's cool nonetheless.

[teddillard]

Last installment: Alltrax SPM controller programming features.  Speed and Torque controls. 
https://evmc2.wordpress.com/2014/12/17/the-controller-thread-last-installment-alltrax-spm-controllers/

Mostly for my own interests, I was wondering if one of the improvements of the SPM line was programming that allowed torque control, since the AXE doesn't.  Short answer: yes.