Limitations of the Conventional Phase Advance Method for

Constant Power Operation of the Brushless DC Motor

J. S. Lawler; University of Tennessee; Knoxville, Tennessee
J. M. Bailey and J. W. McKeever; Oak Ridge National Laboratory; Oak Ridge, Tennessee
Joiio Pinto; Universidade Federal De Matto Grosso do SUI;Brazil

Keywords. Phase advance, Brushless dc motor, Constant power operation

ABSTRACT motor, has some failure modes that may not be acceptable in
electric vehicle applications. If a short circuit occurs in the dc
The brushless dc motor (BDCM) has high-power density supply, then the motor will supply current to the fault so long
and efficiency relative to other motor types. These properties as the permanent magnet rotor continues to rotate. Fuses or
make the BDCM well suited for applications in electric other protection would need to be included to guard against
vehicles provided a method can be developed for driving the such faults. In addition, when transistor-firing signals are lost,
motor over the 4 to 6:1 constant power speed range (CPSR) because of a controller board failure for example, the motor
required by such applications. The present state of the art for can enter deep regenerative braking when operating at high
constant power operation of the BDCM is conventional phase speed. Not only would this be confusing to the vehicle
advance (CPA) [I]. In this paper, we identify key limitations operator but it could also create a traffic hazard. Unless the
of CPA. It is shown that the CPA has effective control over controller board failure automatically activates the brake
the developed power but that the current magnitude is lights, trailing traffic would not he alerted to the rapid
relatively insensitive to power output and is inversely deceleration of the vehicle. To preclude the undesired
proportional to motor inductance. If the motor inductance is regeneration, one might open the dc bus. If this were the case,
low, then the rms current at rated power and high speed may the inverter transistors would experience voltage levels
be several times larger than the current rating. The inductance determined by the back emf of the motor. A high speed, the
required to maintain rms current within rating is derived back emf might have a magnitude several times larger than the
analytically and is found to be large relative to that of BDCM dc supply voltage and therefore the transistors would need to
designs using high-strength rare earth magnets. Thus, the be rated accordingly.
CPA requires a BDCM with a large equivalent inductance. In Section 2, we present the inverter topology and transistor-
firing scheme for high-speed operation of the BDCM by CPA.
1. INTRODUCTION The parameters of an example motor used throughout the
paper are also given. Section 3 provides simulation results for
The size, weight, and efficiency properties of the BDCM are the example motor operating at rated power at three and six
highly desirable in electric vehicle applications. However, times base speed. The example motor i s a ‘‘low’’ inductance
such applications also require a broad CPSR such as 4 to 6 1 . motor, and the current at high speed is twice the rated value.
The present state of the art for driving a BDCM beyond base It is shown that the motor operation in each phase is a mixhue
speed is CPA [I]. In this paper, several important limitations of motoring and regenerative braking. The brakiig action is
of the CPA method are identified. caused by the conduction of the bypass diodes. For low-
Specifically, CPA requires that the equivalent motor inductance motors, the mixture of motoring and regeneration
inductance per phase be sufficiently large. If the motor is extreme. A large motoring component is substantially
inductance is too low, the motor current will exceed its rated cancelled by. a braking component of nearly the same
value when operating at rated power and high speed. magnitude, leaving a modest net motoring component.
Additional cooling would be required for the motor and Section 4 presents a simple fundamental frequency model that
inverter, and the semiconductor ratings would have to be accurately predicts the rms motor current and average motor
increased accordingly. This is unfortunate since high-power- power developed under high-speed conditions. This model is
density BDCMs built with rare earth magnets generally have used to derive a formula for the minimum inductance required
low inductance. A fundamental frequency model of the by CPA to keep the motor within the rms rating over the
BDCM driven by CPA is used to quantify the minimum desired CPSR. Finally, Section 5 contains our conclusions.
required inductance. Detailed inverterhotor simulation
confirms the validity of the simplified fundamental frequency 2. INVERTER TOPOLOGY AND FIRING SCHEME
model. It is also shown that the motor current under coasting
conditions is not significantly smaller than the motor current at The CPA method uses the common three-phase, voltage-fed
rated power. Consequently, the copper losses in the motor are inverter (VFI) topology shown in Fig. 1. Figure 1 also shows
almost independent of the developed power. the motor model used for simulation.
In addition, the CPA uses the conventional voltage fed The bypass diodes of the common VFI make this
inverter, which when combined with a permanent magnet configuration inherently capable of regeneration. This

0-7803-7252-2/02/$10.00 0 2002 IEEE Proceedings IEEE Southeastcon 2002

174
capability is desirable in the case of controlled regenerative the available dc supply voltage can drive current into the
braking, but it also has two undesirable consequences. If a motor). In the vicinity of base speed, operation is a mixhue of
fault develops in the dc supply, the motor will feed current phase advance and current regulation. At a speed only slightly
into the fault so long as the permanent magnets continue to greater than base speed, the current regulation becomes
rotate. In addition, if the motor is operating at high speed, a ineffective and all the control is accomplished by phase
loss of transistor firing signals will result in uncontrolled advance. In this work we consider only speeds at which all
regenerative braking until the motor slows to the speed where control is achieved through phase advance.
the back emf magnitude drops below the level of the dc supply The firing scheme for the CPA method at high speed is
voltage. Guarding against the consequences of such failures shown in Fig. 2, which indicates the timing of the phase A
would require additional components. transistor (QI and Q4) gate signals relative to the trapezoidal
line-to-neutral back emf, e,. The phase B and C back emfs
have the same shape but are delayed from phase A by 120’
and 240”,respectively. The firing of phase B and C transistors
is analogous to that in Fig. 2 but with the appropriate delays
applied. Note that the switching frequency during pure phase
advance is at the fundamental electrical frequency consistent
with motor speed. Pulse width modulation is not necessary.
Transistor QI is fired S, degrees ahead of the instant that
the phase A back emf, e,, reaches its positive maximum.
Fig. 1. Common voltage-fed inverter topology and
motor model used with conventional phase advance. 0, is called the “advance angle.” Transistor Q4 is fired
8,degrees ahead of the instant that e, reaches its most
The phase-to-neutral back emf waveforms of the BDCM are
tmpezoidal in shape with 1200of 6oo of transition negative value. The magnitude of the emf trapezoid increases
top linearly with speed, but the shape is the same in each cycle.
in each half cycle, as in Fig, 2 , The magnitude of the
trapezoidal emf increases linearly with speed, Below base Figure 2 suggests that the gate pulse width of each m i s t o r is
120”, consistent with [I], but we have found that high-speed
speed the BDCM operates in a current regulation mode. The
conduction of the transistors is modulated so that while the performance can be improved by increasing the width to
pbase-to-neutral emf is in the “flat-top” portion of each haif nearly 180”. Although [I] indicates that 8. can be varied from
cycle, the phase current is maintained within a hysteresis band 0 to 60°, we have found that the limiting range is from 6 0 to
about a current set point. The current set point is selected to +120°. An advance angle near 30°, the exact value being
produce the torque necessary to maintain speed. This type of parameter and speed dependent, results in zero average power.
operation is well understood and is therefore not discussed An advance less than this value results in regenerative braking
further here. By definition, base speed is the highest speed at and a greater value results in motoring operation.
which rated torque can be developed without using phase A motor, designed by Oak Ridge National Laboratory
advance. This speed is slightly less than the speed at which (OWL), is used for illustration. This motor is an axial gap
the peak magnitude of the line-to-line emf equals the dc BDCM with samarium-cobalt magnets. The parameters of the
supply voltage because at base speed the dc supply voltage motor are as follows:
equals the line-to-line emf magnitude plus a small additional
amount of voltage necessary to overcome the winding p = number of poles = 12
resistance and inductance. Nb = base speed in rpm = 2600 rpm
N = rotor speed in rpm
V
N
n = relative speed = -
Nb
Qb = base speed in electrical rad/sec
ZaNb = 1633.6 elec radsec
- -~
2 60
” Lo+ L‘ = leakage plus self inductance
= 61.8pH per phase
M = mutual inductance
= 11.8pH (1)
L = equivalent inductance
= Lo + L, + M = 73.6 pH per phase
Fig. 2. Transistor firing scheme in phase A. R = 0.0118 ohms
Eb = peak phase-to-neutral back emf at base speed
Above base speed, the back emf exceeds the dc supply = 74. I6 volts
voltage and the firing must be advanced (i.e., a phase is E. = peak phase-to-neutral back emf at relative speed n
energized during the transition portion of the back emf where = n Eh

Proceedings IEEE SoutheastCoo 2002

175
 P, = rated power = 36,927 watts (49.5 horsepower) Insight into the large current when operating above base
T, = rated toque = 135.6 Nm speed can be seen in Figs. 4 and 5. These figures provide
Vdf = dc supply voltage = 188.7 volts additional detail of the operating condition at six times base
speed while motoring at rated power. Figure 4 shows the
The 188.7-V supply is the voltage required by the motor. phase A motor current and phase-to-neutral back emf. In
Any voltage drop in the inverter would have to he added to addition, the portions of phase A current flowing through the
this value. In this paper, the inverter voltage drops are
phase A bansistors (QI and Q4) and phase A bypass diodes
neglected and the above ideal value is used. (Dl and D4) are shown. Observe that when the bypass diodes
Assuming the classical idealized rectangular phase current conduct, the motor phase current and hack emf are of opposite
waveshape of the BDCM that typifies operation below base
sign. Consequently, bypass diode conduction introduces a
speed (rectangular shape of 120' duration each half cycle), the
braking component. Other than a short period of time,
theoretical peak and rms currents of this motor are introduced by the phase advance, the motor phase current and
back emf are of the same sign during transistor conduction.
Ipk = peak current = = 249 amps This indicates that transistor conduction contributes motoring
Eb power. These conclusions are iiuther supported in Fig. 5.
,I = \I-
r
2
3
Ipk = 203.3 amps.
Figure 5 shows the instantaneous values of total power, the
power in phase A, the power flowing through the phase A
transistors, and the power through the phase A bypass diodes.
This motor was used in the laboratory demonstration testing Observe that although the total power has an average value of
reported in [2,3]. The motor was not designed to be operated 36,927 W and modest ripple, the instantaneous power in phase
beyond 3000 'pm. Here, however, we address high-speed A has substantial ripple, more than 400 kW peak-to-peak and
performance and will simulate the performance at speeds up to an average value of 12,310 W. The average power flowing
six times the base speed of 2600 'pm. through the phase A transistors is 60,775 W of motoring
Models in MATLAB and PSPICE were developed to power, while the average power flowing through the phase A
simulate the performance of the BDCM being driven by the bypass diodes is 48,467 W of braking power. The braking
CPA control method. The simulators include detailed power associated with the bypass diode conduction cancels a
representation of the motor and switching logic, and the like amount of motoring power associated with transistor
actions of inverter transistors and bypass diodes. To conduction, leaving the net contribution of phase A as 12.3
concentrate on performance limitations imposed solely by the kW of motoring power. Phases B and C would exhibit similar
CPA method, all loss mechanisms except winding resistance behavior, and each phase would contribute, on the average,
are neglected. Speed-dependent losses such as the core losses 12.3 kW, which is one third ofthe average total power.
(hysteresis and eddy currents), and rotational losses (friction
and windage) are neglected. Inverter bansistors and bypass
diodes are modeled as ideal devices. In the next section,
simulations at relative speeds of three and six times base speed
are presented for operation at rated power.

3. PERFORMANCE AT THREE AND

SIX TIMES BASE SPEED
For the example motor with a base speed of 2600 Ipm,
relative speeds of n = 3 and 6 correspond to 7800 and 15,600
rpm, respectively. Simulation of the motor phase currents and
the instantaneous total three-phase power over one
fundamental electrical cycle are shown in Fig. 3 for operating
at rated power of 36,927 W at relative speeds of 3 and 6. The
advance angle, 8,, required to achieve rated power is 48.2" at
n = 3 and 49.1" at n = 6.
Note that the average power is the rated value of 36,921 W
for both operating conditions. The instantaneous power is not
smooth but includes ripple that increases in magnitude with The simulation results show that the operation in each phase
speed. In an electric vehicle, this ripple would not be is a mixture of motoring and braking operation at high speed.
objectionable since it would be filtered by the substantial mass For a low-inductance motor, such as the example used here,
of the vehicle being propelled. However, note that the rms this mixture is extreme and there is substantial cancellation of
values of the motor phase currents are 315.5 A and 420.9 A, motoring power by a large braking component. The braking
respectively, for n = 3 and n = 6. The rated rms current is only component is caused by bypass diode conduction. This
203.3 A for this motor. Thus, the rms current increases with observation suggests that if bypass diode conduction can be
speed, and in this case the rms current at full power and at six inhibited, eliminating the braking component, the remaining
times base speed is more than twice as large as the rated value.

Proceedings IEEE Southeastcon 2002

176
current through transistors will contribute predominantly 4. FUNDAMENTAL FREQUENCY MODEL
motoring power. It is also likely that the magnibde of the
transistor current required to produce rated average power will Comparing the relative speed cases of n = 3 and 6 in Fig. 3,
be substantially reduced. This is the basic concept of the Dual shows that the motor current of the BDCM, when driven by
Mode Inverter Control method that is described in [3,4,5]. CPA at high speed, is nearly sinusoidal. The higher the speed,
the more nearly the phase currents approach a pure sinusoid at
n = 6, adv a 49.1 deg the fundamental electrical frequency. Although not displayed,
I
the voltage applied to the motor by the inverter is a "six-step"
trpe waveform, while the motor back emf is trapezoidal as
shown in Fig. 2. Using the fundamental frequency
components of the applied inverter voltage and the back emf
8.1@JOA 1, Pw 4L 5m i o &
results in the simplified fundamental frequency phasor model
a M o r current and back emf shown in Fig. 6.

................... ...... .........

-1mo0 iw 200 MO 4w Mo 600
b. Current flowing through lransktors

....... ... ........ .........

Fig. 6. Per phase of a fundamental frequency
-1wo CPA phasor model at high speed.
0 100 200 900 4Ml M o m
hale $Si
c. Current llowing thmugh bypass diodes Using this simplified model to analyze the condition n = 3,
8, = 48.21 results in an rms motor current of 315.3 A and an
Fig. 4. Phase A current through transistors and bypass average motor power of 36,378 W. For comparison, the
dioding for n = 6 and 8. = 49.1". detailed simulator yielded 315.5 A rms and 36,927 W as
shown in Fig. 3. For n = 6, e,= 49.1", the simplified model
predicts an rms current of 420.9 A and an average power of
36,374 W, while the simulator yielded 420.9 A and 36,927 W.
........... ..... ............................... The conclusion is that the simplified model accurately predicts

wi---J
Ii ~

0 1 " W " m
a. Total paw
rms current and average motor power.
~ .._.__ If the winding resistance is neglected (R = 0), the rms
€
Lm7 : .?Yf
' .... . . '2?'!. current of the simplified model is given by
.......................
O , .
M. .
? .M
. .3. .M. .n . .
m.M
. . 8. .m
b. mas A P D W ~ ~

0 l M m x m " m
E. phnse A porrer flowing lhmylh Vadsmm Observe that for a fixed advance angle,8,, the m current
. . . . . . . . . . ' . . . ,.:. . . . . . I. . . . . . :.., varies with the relative speed, n. Equation (3) provides some
insight into what is required for CPA to provide an infinite
CPSR Note that as the speed becomes infinite

Fig. 5. Instantaneous power waveforms. For the example motor this limiting value is 530.3 A, which
is significantly larger than the 203.3 A rating of the motor.
In the next section, a fundamental frequency model is Consequently, for the example motor, which has an
presented that accurately predicts the motor current and power inductance of 73.6 P H, the CPSR will befin ite. In fact,
as a function of advance angle when operating at high speed simulation shows that the highest speed at which rated power
using CPA. can be produced without exceeding rated current is n = 1.87.
If the inductance in eqn. (4) were sufficiently large to cause
the limiting current value to be 203.3 A, then the CPA would

Proceedings IEEE Southeastcon 2002

177
yield an infinite CPSR. This limiting value of inductance is machine rating, the peak power that can he developed is
192 pH. A finite CPSR, such as 6 1 , will require less than reduced. In the low-inductance, 73.6- p H case, the peak
192 p H . power is approximately 120 kW in the motoring mode and
A desired, finite CPSR requires an inductance that can be about 150 kW in ther egenerativeb raking mode, compared
found by setting n of eqn. (3) equal to the CPSR value and with the 36.9-kW rating of the motor. This provides
recognizing that to have the rms current remain less than or substantial margin for short-term overloads. In the higher
equal to the rated value, 5. the inductance must satisfy inductance, 1 4 9 - p H case, the peak power capability is
reduced to approximately 60 kW in the motoring mode and 70
kW in the braking mode. Thus, the added inductance reduces
the peak power-producing capability of the machine. The
(4) reduction in peak power-producing capability can be
important in electric vehicle applications where short-term
overload is desired for passing or rapid regenerative braking.
For the parameters of the example motor, and a CPSR of 6,
this expression yields 149 p H per phase as the minimum
inductance necessary to keep the rms current from exceeding
the rated value of 203.3 A when operating at six timesb ase
speed. This value is more than twice the inductance of the
example motor.
Figure 7 shows the variation of average motor power and
rms current over the full range of advance angle
(- 60" 5 ea 5 120"), with the nominal motor inductance of
73.6 p H at relative speeds of n = 1.87 and n = 6. Note that
the control is effective in swinging the developed power over
the full rated range of 36.9 kW in both the motoring and
regenerative braking modes. The plots of power vs advance
angle are virtually indistinguishable for the two speeds. Also
note that for a given power level substantially more current is
required at n =.6 than for n = 1.87. A relative speed of 1.87 is
the CPSR of this low-inductance motor since it is the highest
speed at which a rated power of 36.9 kW can be produced
while maintaining motor current within the rating of 203.3
A-,.
~~~
At n = 6, the motor develom the rated Dower with an Adv Anple. ldepl..
rms current of 421 A as shown k Fig. 7, &d in the time
domain simulation of Fig. 3. The figure also shows that the Fig. 7. Average power and rms current vs
rms current is relatively insensitive to advance anale and that advance angle with L = 73.6 pH.
the rms current plots &e fairly "flat," especially achigh speed.
Consequently, at zero power the rms motor current is not
much smaller than at full power in either the braking or
regenerative braking mode. Thus, the copper losses in the
motor are decoupled from the power output and the efficiency
will be low when coasting. Figure 7 shows that the CPSR of a
low-inductancemotor is small when driven by CPA.
Figure 8 shows that the CPSR can be extended if the motor
inductance is sufficiently large. In this figure, the inductance
of the example motor is increased from 73.6 to 149 p H per
phase with all other parameters held constant. The figure
shows rms current and average power vs advance angle for a
relative speed of n = 6. The inductance value of 149 p H was
derived previously as the value required for a CPSR of 6:l.
However, the formula did not include the effect of winding
resistance, which is included in Fig. 8. Note in Fig. 8 that the
36.9 kW of rated power can be developed with a current that is
I I _ - -
onlv sliehtlv lareer than the 203.3 A ratinn of the motor.
Thus, if the inductance is sufficiently large, the CPA can drive
&iv b * r la01

Fig. 8. Average power and rms vs advance

the BDCM over an extended CPSR Observe in the angle for the example motor at 15,600 rpm
comparison of Figs. 7 and 8 that although the higher for L = 149 pH.
inductance is effective in bringing the rms current within the

Proceedings IEEE Southeastcon 2002

I78
 5. CONCLUSIONS 6. I. S. Lawler and J. M. Bailey, Conrlon: Power SpeedRmge Exienrion of
Swfoce Mounred P M Molors. U.S. Patent Number 6,236,179 BI. May
In this paper we have identified several limitations of the 22,2001
CPA method for constant power operation of the BDCM. The BIOGRAHPIES
CPA is especially sensitive to the motor inductance that must
be larger than a threshold value to maintain motor current Jack S. Lawlrr
within rating when operating at rated power at high speed. If Deparlment of Eleehical and Computcr Engineering
The University ofTennessee
the motor inductance is low, additional cooling will be Knoxville, Tennessee 37996
necessary for the motor and inverter components and the
current rating of the inverter will have to be increased. High e-mail: [email protected]
inductance is found to require a higher dc supply voltage and
J. S. Lawlrr received the BSEE degree (1971). M.S. in Systems Science
results in a reduction in the peak power production capability (19721 and Ph!3 in Systems S c i e w (1979) d l from Michigan State
of the motor. A fundamental frequency model was developed University. He has industrial experience in control systems and slecuooics.
that allows easy determination of the reauired inductance. He was a guidance and conuol engineer with the Missile Division of Martin-
The c p ,,Ies ~ the VFI, &ich has some failure Marietta in Orlando, Florida. In rhat position, he had responsibility for the
stability of the initial Nm autopilots of the SAM-D missile that ultimsfely
modes that may not be acceptable in electric k a m e the Patnot. ,Healso has experience as a new product development
applications or that will require the addition of supplementary engineer with the ~ e n e r a ime
l ~ Corporation of Davidson, NO& c d i a n . ~e
components to guard against these failures. Specifically, the joined the Electrical and Computer E?gineering faculty of the University of
Tennessee as an Assistant Professor In 1979 and has the held the rank of
,,,ill feed faults that occur in the inverter or dc supply Professor Since 1988. At the University he has aught comes in control
system so long as the permanent rotor continues to systems, power system and in power elecuonics. Dr. Lawlsr's research
revohe. LOSS of transistor firing signals when motoring at interest includes the control of de-regulated utiliry . syst~ms
. and power
high-speed results in deep regineitive braking that may electronic controls especially permanent magnet motor drives. For the past
create a danger to the vehicle and trailing traffic. Opening the four yean he has been a con~ultantto the Power Electronics and Electric
Machines Reseanh Group of the Oak Ridge National Labamtory working on
dc bus following such a failure results in the inverter brurhless dc motw drives for electric vehicles. Ix. Lawler is a Senior
transistors being exposed to the voltage level of the motor Member ofthe IEEE, Phi Kapp Phi and Sigma Xi.
back emf, which may be much larger than the dc supply
voltage. The transistors may be damaged unless their voltage 1.MUton Bailey
Oak Ridge National Laboratory
rating is based on the maximum motor emf rather than the dc National Transportation Research Center
supply voltage. 2360 Cherahala Boulevard
The highest power density and eifciency BDCMs have Knoxville, Tennessee 37932
inductance that is too low to be compatible with CPA. To e-mail: JMBailG3ulk.edu
realize tbe full benefit of such motors, an alternative control
scheme is required. Keying on the detrimental impacts of Dr. J. M U m Bailey received his B.S. in Physics from Davidson College in
bypass diode conduction under CPA, the authors have 1949, and his Masten in Electrical Engineering from the University of
developed an altemative method of driving the BDCM above Tennessee at Knoxville in 1952. He spent three years as an Instrument
Engineer with E.I. Dupnt in Orange, Texas. He then entered the Georgia
base speed. This alternative is called dual-mode inverter InstiNte of Technology where he graduated with a Ph.D. in Elecmcal
control (DMIC) 161. The DMIC uses thyristors to block the Engineering in 1960. While at Georgia Tech, he received the M.A. Fen1
undesired conduction of the bypass diodes at high speed. It Sigma XI Award for his Doctoral Dksertatian. Today, he is Professor
EmeriNs in the Electrical Engineering Depamnent of the University of
has been shown that when all the loss mechanisms are Tennessee and an adjunct participant at the Oak Ridge Nslional Laboratory.
neglected, the BDCM has an infinite CPSR when driven by He has five patents in the field of motor design and eonfml.
the DMIC. The thyristors also isolate the motor from faults
and avoid undesired regeneration following loss of firing At Martin Mariem in Orlando, Florida, he received a Publication's Award for
his bwk, Syslems Engineers Molhemalicr. In 1994, he received &e most
signals. Preliminary studies of the DMIC, including a prestigious award given by the College of Engineering at the Universirj of
laboratory demonstration,can be found in [2,4]. Temessee;The Nathan W. Dougherty Award for Distinguished Service in the
Engineering Profession. He is at present a senior member ofthe InStiNte of
6. REFERENCES Electrical and Electronics Engineers.

1. Cambier et al.. Bmhlesr DC Molar Udng Phase Timing Ahoncemen:, John W. MeKeever
U.S. Patent Number 5,677,605, October 14, 1997. Oak Ridge National Laboratory
2. I. M. Bailey et SI., D u d Mode Inverier Conlml T a l Verifiolion, National Transportation Research Center
OR"-2000/172, Oak Ridge National Laboratory, 2001. 2360 Cherahala Boulevard
3. I. S. Lawler, J. M. Bailey, I. W. McKeever, and J. Pinto, "Extending the Knoxville, Tennessee 37932
Constant Power Speed Range ofthe Blushless DC Motor through Dual
Mode Inverter Conml - Part II: Laboratory Pmof-of-Principle." e-mail: [email protected]
submined to the IEEE Transactions on Power Elecuonics.
4. 1. S. Lawler, J. M. Bailey, and J. W. McKeever, fitended Conrlanl John W. McKeover received the Bachelor's in Physics in 1960 €" Case
Power Speed Ronge ofthe Bmhless DC M o m lhmugh Duo1 Mode Institute of Technology in Cleveland, Ohio, a Master's in Physics from the
Inverler ConlmLO RNLTTM-2000/130, Osk Ridge National Laboratory. University ofTennessee, Knoxville in 1968, and a Ph.D. in Physics f"the
2001. Uniwnity ofTennessee in 1973. He has 41 yea" work exprience serving
5. J. S. Lawler. J. M. Bailcv. I. W. MeKeever. and J. Pinto. "Extmdine the in technical and umieCt manamnent cauasities at the ~oovemmmtowned
Constant Poker Speed R&ge of the Brushless DC Motor through bul conhctcn operaaah ficilities in b a k Rid& Tennessee and'bas 40 publications
Mode lnverter Conuol - Pan I: Theory and Simulation:' submitted N and 3 patents. From 1960 to 1970 he worked on barrin developmeot for the
the IEEE T m a c t i o n s on Power Electronics. gaseous diffusion process. Between 1970 and 1982 he designed and tested

Proceedings IEEE SontheastCon 2002

179
mtauDg machine componmrr for l e gao centrifuge developed at Oak Ridge JoIo 0. P. Pinto
for he Depamnent of Energy (DOE). Fmm 1982 until 1985 he managed thc Univcnidade Federal do Malo Grossa do SUI
Machine and Halance of Plant Materials Gmup duing which lime he chaired Campo Grande, MS. Brazil
an advanced component derrlopmcnt !cam
JoPo Onofrc Perelr~Pinto was bom in Valpamiso. B w i l io 1966. He
From 1985 Io 1995 he a a n s f e d tcrhnolng) dcvclopcd lor !he WE lo U.S. attended Univenidade Estadual Paulista a1 llha Solteira, SP, Brad in 1986,
industry h n n g this time he managed a supersondueling motor projcel. B where he received his Bachelor's degree in Electrical Engineering in 1990. In
new son-switched invenei dcvclapmca prqrc~, and a high vollagc I)(: 1991, he attended Univenidade Federal de Ubnlindia, MG, Brazil where he
m m i s s i o n EOSI rrducrion prqcct. From 1995 to lhc prcwnl hc has managed received his Master's of Science also in Electrical Engineering in 1993.
Hybnd Electns \'thick prqecrr to ruppon DOES Oflicc of AJvancrJ
Automotive lechnalogicr and callabarativc m a r c h piojrcrr with mdusy 81 He has k e n an assismt professor at Univenidade Federal do Malo Gross0 do
OKNL'r Power Llecmonicr and Elcct"c Ma-hinsg Research Center, which ii SUI at Campo Grande, MS, Brazil since 1994, and is currently on leave of
now locared 31 OHNLr n w Nauonil Transponaiiun Rerwrch Center in absence to pursue lhe Ph.D. degree in The University of Tennessee
Knoxville,Tennessee Knoxville Depamnent of Electrical Engineering since 1997.

Proceedings IEEE Southeastcon 2002

180