You are on page 1of 4

Transverse flux motor drive dynamics

loan-Adrian VIOREL, Loran d SZABO, Mircea GUTMAN Zoltan PUKLUS


Department of Electrical Machines. I ntitute of lntormatics and Electrical
Marketing and l'v1anagement Engine..:ring. Dep<rtment of Automation
Technical University of Cluj Szechenyi Istvan University
R0-400750 CLUJ, P.O. Box 358 H-9026 GYOR, Egyetem H!r I
Romania Hungary
Lorand.Szabo@mae.ulcluj.ro pukltts({]:s=e.lm

Abstract - The transverse nux motor (TFM) can be an If the stator poles are aligned for all phases then the
important competitor on the variable-speed-direct-drive rotor poles must be shifted adequately to obtain the
domain due to its vety high torque density. The transverse necessary spatial displacement. TFRM must have three or
nux reluctance motor (TFRM) is the simplest structure of the
more then three phases in order to avoid start-up
TFM class. The TFRM's mathematical model is developed in
difficulties and to obtain continuous rotation. The TFRM
the paper and the motor's dynamic regime is studied via
has to be supplied by a power electronic converter, which
MATLAB/SIMULINK software environment.
gives unipolar pulses precisely synchronized with the rotor
position. The converter, usually of asymmetric half bridge
l. INTRODUCTION
type, is quite the same as for the SRM [2]. Such a
converter assures an independent current variation on each
The TFR machine is a variant of the TF machine with
phase.
passive rotor. The TFRM has a ring winding with salient
poles on the stator and only salient poles on the rotor. The TFRM behaves as the SRM and its control is quite
similar. The TFRM requires rotor position sensing to
J\ part of a TFRM phase module, given in linear layout
control the current commutation from phase to phase. An
for better understanding. is given in Fig. I.
encoder attached to the shaft supplies usually the rotor
position, but a specific technique for indirect sensing can
be applied.
A performant drive system cannot be obtained only with
sta.tor very well designed and constructed motors. ft requires a
suitable power electronic converter and adequate torque,
speed and current controllers. Computer simulation of the
entire drive system enables verification of the motor design
and of the drive system ability to match the load torque
distinctive over the entire speed range in steady-state and transient
tooth
regime. The steady-state regime can be covered adequately
with the simplified mathematical model used for
design-estimation, or by finite element (FEM) analysis (2.
3 ]. Simulation of the entire drive system requires the
mathematical model of the TFRM and of the other
Fig. 1. A part of a TFRM phase model [2]
components as well as their interconnections. Some
The TFRM has quite the same features as the switched attempts to present the TFRM mathematical model were
reluctance motor (SRM), the main differences consisting in previously made [4, 5, 6], but they did not cover entirely
the homopolar stator ring winding and the equal number of the subject. In this paper the TFRM mathematical model is
poles on the stator and rotor [ 1, 2]. As i n the case of all studied on analytical base, as well as by simulation via the
TFM's each TFRM phase is an independent machine, the MATLAB/SlMUUNK programming environment.
phase modules being placed one after another on the axial
direction. In Fig. 2 the TFRM's phases have a common l l . THE TFRM MATHEMATICAL MODEL
yoke but they can be totally independent.

Stator yoke Stator pole The TFRM has independent phases, then its
mathematical model is given by the equations:
Stator
phase coil
v= R . dA.
t+ (I)
dt
A.=L(O,i)i (2)

i
T ()Wm = .!!__ J-! .:t- (O,i) di
(){)
=

dO 0 2
Rotor salient pole

Flg. 2. A TFR:Vl with common yok for the phases 121 T J dO + T, + 7i


dt
=

393
where v, i, A, and R arc the phase voltage, current, flux ?: R
linkages and resistance and T, 7/, Tr are the kc R = --
---' =--- (12)
t:R -rg
electromagnetic, load and friction torque respectively.
The rotor position electric angle () is given by the where tR is the rotor pole pitch and bRs the rotor slot width
equation: The saturation factor ks depends on the phase current
d() and can be computed in aligned position, as
-=(t)=pQ (5)
dt ks = f(i) (13)
where the number of pole pairs p is equal to the number of
The variable air-gap flux density is:
stator, or rotor, pole pieces, Q?'QR.
The friction torque depends on the machine speed, but B g(B,i)=Ff.Lo P(B)
usually it has smaller value and is neglected against the
load torque at low speed. B g(B,i)= F f.lo (1+PR sin B) (I4)

The TFRM design procedure implies, quite usually, a g


FEM analysis and consequently the variation of the phase
where F is the phase MMF.
flux linkages and torque function of the rotor angular
The maximum value of the air-gap flux density occurs at
position and phase current are known. These can be used
aligned position,
in computing the steady-state or dynamic characteristics of
the TFRM. As was proved in [3, 5] there is an 3D to 2D
equivalence for TFRM and the FEM analysis can be
performed on the 2D equivalent structure.
Based on the given mathematical model (1)(5) and on and consequently:
the flux linkages and torque characteristics computed, via a
3D or 2D FEM analysis, function of phase current and
( . 1+PR sin 8
B g((),l.)=B gmax 1) (16)
rotor angular position in [5] and [6] the TFRM dynamic I+PR
regime was simulated via SIMULINK. and SIMPLORER
The phase inductance is:
respectively. In the SlMULINK program the
characteristics obtained by FEM computation were (B)
L(8,i)= A. +Lsa (17)
introduced as look-up tables. The SIMPLORER program I
was linked to the FLUX 3D FEM software and the
where Lsu is the phase leakage inductance, which is
computation was done in the same time in both
considered constant.
programming environments.
Finally comes . ::
In the following a simplified mathematical model, which
does not require FEM analysis results to operate with, is I+PR sinB +
L(B,i)=M A (i) Ls-
V (18)
proposed. In this model, in order to obtain the inductance . ,. l+PR
and torque variation function of current and rotor angular
position, an air-gap variable equivalent permeance is with the aligned inductance Md,
defined as: B gmax (i) N AP Qs
M
d
()

I = --"-
' -- ---

(19)

P(B,i)=(l+PR sinB ) (6)
g N, Ap and /being the phase number of turns, the stator pole

where is the permeance coefficient area and the phase rms current.
PR with slots
considered only on the rotor, [2]: By now the flux linkages derivative can be computed as:

dA.(B,i) = di + CJL(B,i). dB
L(B,i) i (ZO)
PR:::::!:... fJ kcR. sin L.L ;r
( ) (7) dt dt ()() dt
;r fJ '>R 2
which leads to:

dA.(B,i) 1+PR sinB di )


(8) = Md (i)
( +L Sa +
dt 1+PR dt
(21)
,,," PR cos() .
+ (t)lVl d (l) I
l+PR
(9)

The electromagnetic torque is then:

r= ;(utg-1u-1n) (10) dWm .


T = -- = kr 1 cos B (22)
()()
The equivalent air gap g is:

with
g * = kcR ks g (11)
N PR Bg max
k =-Q2. A . (23)
with the Carter's factor is given by: T 2 S P I+ PR

391
The proposed simplified mathematical model contains dQ
0.3 (K r i cos() - T1 ) (25)
the equations ( 1), (4), (21) and
(22) with the necessary
-=

dt

computations done via (7)-(12), (19) and (23).


ks = f(l),
IlL TFRM DYNAMICS SIMULATION,
Md 2.08 I o-3 f(i), (26)
CONCLUSIONS.
=

K r = 2.4If2 (!)
The entire TFRM drive system is simulated based on
Some results obtained for a run up when the rotor initial
MATLAB/SIMULINK software environment. It means
position is unaligned, fJ=O, and the final one is aligned,
that, for only one phase current duty cycle, the hysteresis
fJ=td2, arc given in the next figures.
current controller and the asymmetric half bridge
The phase voltage variation function of time is shown in
electronic converter were simulated together with TFRM.
Fig. 4a. In Fig. 4b for better understanding purpose a zoom
The considered dynamic regime is the run on regime for
of the conduction period's end is given.
one stroke. The main data of the TFRM are given i n
Table I.

TABLE I.

Rated power [kW] 24


Rated phase voltage [V] 260 100

Rated phase current [A] 170


0
lilY Ill
Maximum phase current [A] 200 "

Efficiency 0.95 100

Power factor 0.65


.21lJ
Moment of inertia (kgm2] 3.29
Number of pole pieces 40
Air-gap diameter [m] 0.33
a)
Length per phase [mm] 60
Rotor slot width [mm] 13 300r---------
.-
Number ofturns per phase 12 200

Phase leakage inductance [mH] 0.314


100

Air-gap length [mm] 0.6


0

The SIMULINK model of the TFRM is given in Fig. 3. 100


200

, '<s-------;------::'5-=----
.3 --
--'-----
___J
4. 5
I)O) 10 X

b)
Fig. 4. Phase voltage variation
The current, Fig. 5, is kept almost constant at its
maximum value.

nor-------
200
-------+CJ 1111

IlD
Fig. 3. The SIMULINK model of the TFRM
J.IO
The model is based on the following equations: _120
:!.
- HJJ
PR cos() )
v-(R+QR O. Md i Ill
di I+PR
(24)
dt
(l+ PR sin B)+ Lsa
l+PR
/:;----;;;';;
0 5:-7-
- --;-
';:-
1 5---
2 ::':
2 5:---7-----: :" 5--.:-----:5
3'::-
t!J x1o
'

Fig. 5. Phase current variation

395
The plot of the rotor speed versus time is given in Fig. 6. Since the current is kept almost constant at the
maximum value, ks is constant too, and consequently Md
and Kr have a fixed value computed with the maximwn
04 stator MMF.
035 The proposed mathematical model is simple and quite
03 accurate. It is well fitted for dynamic regime study, as it
comes out from the given simulation results.

IV. ACKNOWLEDGEMENTS
0.15

0.1 The work was possible due to the support given by the
Joint Hungarian - Romanian Intergovernmental S & T
005
Co-operation Program under grant 17/2002.
0 2--2 . 5--35. --7-5
1 E --
o The researches were supported also by the National
[sJ 10"
Council of Scientific Research in Higher Education
Fig. 6. Rotor speed variation (Romanian Ministry of Education and Research) through a

The angular displacement of the TFRM is given m grant offered to the authors.

Fig. 7.
V. REFERENCES

0.035 . [1] Henneberger, G., Viorel, I. A., "Variable reluctance


o.w electrical machines," Shaker Verlag, Aachen,
Germany, 2001.
o025.
[2] Viorel, LA., Henneberger, G., Blissenbach, R.,
Lowenstein, L., "Transverse flux machine, its

0.015 behaviour, design, control and applications,"


Mediamira, Cluj, Romania, 2003.
0.01
[3] Crivii, M., Viorel, I.A., Jufer, M., Husain, I., "3D to
0005 2D equivalence for a transverse flux reluctance
motor," in: Proceedings of the International
2
1 [sJ
2.5 3 3.5 4' '4 5
Conference on Electrical Machines, JCEM '02,
x 104
Brugge, Belgium, on CD-ROM.
Fig. 7. Angular displacement variation
[4] Viorel, LA., Blissenbach, R., Henneberger, G.,
The torque increases quite fast and varies almost Popan, A., "The transverse flux machine
cosinusoidally since the current is kept constant, Fig 7. mathematical model" (unpublished).
[5] Kruse, R., Pfaff, G., Pfeiffer, C., Wehner, H.l.,
Hopper, E., "New concept of a low-speed direct servo
drive," in: Proceedings of the Power Conversion and
Intelligent Motion, PCIM '98, vol. Intelligent Motion,
pp. 1-10.
[6] Lowenstein, L., Henneberger, G., "Dynamic
simulation of a crankshaft starter alternator," in:
GPC- Advanced Propulsion and Emission
Technology Proceedings, Detroit, USA, 2001,
pp 141-146.
[7] Kruse, R., "Calculation methods for a transverse flux
reluctance motor", in: Proceedings of the
'100o!---:0:'::-- :':15---=2- .2.5--::'3'::-5 _
---:
5 -'- c---
....4 .....,.. 5
l[sJ x 104
OPTIM '2000 International Conference, Braov,

Fig. 8. Torque variation


Romania, 2000, pp. 387-392.

396

You might also like