USA: ACIM, PMAC, PMDC motor efficiency

and electric cars

There is a good article in the wonderful Appliance Design

magazine about driving motors. It was written by Eric Persson,

the Field Applications Director at International Rectifier. The

article is great because it was written from the application side

and will help you understand the differences between an ac

induction motor (ACIM) a permanent magnet dc motor (PMDC)

and a permanent magnet ac motor (PMAC). I had always heard

that PMAC and PMDC motors are the same, you just drive the

DC one by banging the windings on and off while the PMAC you

feed a three sine waves that you have to fabricate. Eric points out

that the motors are indeed different and notes that if you spin a

PMDC you see a trapezoid generated on the windings just like

the trapezoids you drive the thing with. If you use a PMAC motor

like a generator you would see sine waves coming off the

windings. Makes sense to me.

Eric seemed so knowledgeable that I wrote him with a few

questions. One was whether an ACIM or a PMAC motor would

be better for an electric car. Mary Tolikas from Ansoft said that

in theory a PM motor can be more efficient since you don’t have

to use power to create the magnetic field. But when I read Eric’s

article about field weakening, maybe that is why there are few

PMAC motors in the electric vehicle world. Tesla uses an

induction motor. Is it fair to say that a PMAC motor with a

transmission would be better than an ACIM for efficiency? Then

again there are the whole commutation losses– making a sine

wave is way more lossy than trapezoidal commutation. So I could

see PMDC being simple but noisy and cogging. PMAC being

great for some applications but when the drive is large enough

the switching losses kill you. What are the trade-offs for electric

vehicles vis-à-vis motor type? I also noted that Mary said that

when she worked at Delphi she was using Ansoft (and that

doctorate from MIT) to reduce the noise in a reluctance motor

meant for the EV-1. But the EV-1 came out with an induction

motor (ACIM). Eric responded:

Thanks for the feedback – I’m glad you enjoyed the article! It is

always challenging to take a complex subject and summarize it

into a few thousand words with some graphics. I think that is

especially true in the world of motors, drives, and control where

there are SO many factors and variables to consider – which

leads to your questions.

I agree that PM motors are typically more efficient than Induction

Motors (IM). This is due to a combination of rotor losses and the

“excitation penalty” (the overhead current to produce flux) in an

IM. But this generalization is strongly dependent on the size of

the motor. As the size increases, the current required to produce a

given flux density increases linearly – but the cross section

available for copper increases by the square. This means that

larger ACIM (such as those used in EV) can add a lot more

copper cross-section, and more effectively remove any heat from

the stator. Of course this increases the cost of the stator compared

to running skinny wire. But then there is the magnet cost of a

large PM machine.

So as a general rule, I think that somewhere below about 5kW

that it makes sense to use permanent magnets for excitation.

Permanent magnet motors will have higher efficiency and more

torque per amp. Above approximately 5 kW, magnet costs begin

to dominate, and the ACIM becomes much more attractive from

a price/performance standpoint. ACIM is also seen as more

rugged by the automotive industry – no worries about accidental

degradation or demagnetization due to over-temperature or

accidental over-current. And, large PM machines are riskier to

assemble due to the enormous axial forces of the permanent

magnets. To answer your question directly, I am sure a PMAC

would be more energy efficient than an ACIM for an EV

application – but it may be economically unfeasible. By the way,

DOE did a teardown and complete efficiency analysis on the

Toyota hybrid electric motor and transmission. It is publicly

available at http://www.osti.gov/bridge/- just do a search for

Toyota (lots of cool info here!)

Your friend Mary is also correct about the SR motor. It looked

promising on paper as a low-cost, high efficiency synchronous

machine (some appliance-makers tried it as well). But the noise,

due to magnetostriction and structural stress as the magnetization

rapidly commutates, was unacceptable. In addition, SR motors

require a very small air gap, and the manufacturing tolerances

drove the manufacturing costs up to uncompetitive levels.

Great stuff, and also a great tip about the government website.

My pal Otmar Ebenhoech actually has a GM EV-1 motor,

apparently some clueless GM dealers would sell them from the

parts counter, not realizing GM never wanted anyone to get their

hands on one. Otmar tore down his EV-1 motor here. Otmar also

had a comment about Eric’s answer:

Shucks, someone who actually knows the subject. I must say that

seems rather rare these days. Good on him. I appreciate that.

The one thing he did not mention about PMAC motors and on

road EV’s is that the efficiency depends on the peak to average

power ratio. A “normal” car these days runs about a 10 X peak to

cruise power ratio. The eddy current losses of a BLDC/PMAC

(whatever the wankers want to call it today) motor are relative to

the peak torque. Stronger magnets and more iron = more losses.

A single ratio 100kw UNIQ joke (sorry, “motor”) makes over

1500 W of waste heat at highway cruising speed, and that’s with

the power turned off. Not so good for a car that takes about 15kw

to cruise. So if you made a 15 kW PMDC and attached it to a 100

KW Induction for acceleration plus peak you would be golden,

but that’s complicated and a bit expensive. Years ago some

people suggested and even mocked up a light PM machine with

extra windings to beef up the field for higher current. It was a

green box visiting all the shows in the mid 90’s. Can’t remember

the name now but I’m sure they had a patent or some such. A bit

large it was, and surely a bit expensive, but efficiency was in the

cards.

Then again, who cares about a few percent in efficiency when

you can gain a few percent in battery capacity by waiting 3

months? Such is the reality of EV’s these days.

And my pal Dave Ruigh sent a link from good ol’ Circuit Cellar

on some folks trying to adapt the Microchip DSP motor control

system to an electric vehicle. Like Otmar, I think they would

have better luck if they put some bus capacitors on those copper

bars

Source: edn.com

http://www.4evriders.org/2010/01/usa-acim-pmac-pmdc-motor-efficiency-...

-------------- ---

Trexa
Four-Wheel Drive EV Platform

This all-wheel drive platform includes everything you need to

make a vehicle go – a lithium iron phosphate battery pack,

driveline, and power electronics. From a full charge, the battery

stack can provide a top speed of 100mph and a range of 105

miles with 21 kWh of lithium iron phosphate battery power

installed. Recharge time is said to be four hours.

Modular and scalable, the standard Trexa platform will feature an

aluminum, carbon steel tubing and thermoplastic shell containing

open source and user programmable electronics and advanced

battery technology.

The Trexa will have an 80 inch wheelbase with a curb weight of

1,250 lb. The all-wheel drive platform features a programmable

gearless 2-speed transmission that uses regenerative braking.

Prototype is supposed to be available in the 4th Quarter of 2010.

Trexa-ev3.jpg 89 kt Lataa