The present invention relates to a magnetic bearing control device
and, in particular, to a magnetic bearing control device capable of suppressing
a ripple current flowing through an electrolytic capacitor to a low level.
A magnetic bearing is used in a rotary apparatus such as a turbo-molecular
pump for use in a semiconductor manufacturing process. A conventional magnetic
bearing exciting circuit will be described on the basis of a construction example
of the magnetic bearing of a turbo-molecular pump.
Fig. 6 is a sectional view showing a turbo-molecular pump as a construction
example of a magnetic bearing.
In Fig. 6, the turbo-molecular pump is equipped with a rotating body
103 provided with a plurality of rotary vanes 101a, 101b, 101c, ... arranged in
a number of stages and serving as turbine blades for discharging gas.
To rotatably support the rotating body 103, there are arranged an
upper radial electromagnet 105a, a lower radial electromagnet 107a, and an axial
electromagnet 109a to thereby form a magnetic bearing. Further, there are provided
an upper radial sensor 105b, a lower radial sensor 107b, and an axial sensor 109b.
In each of the upper radial electromagnet 105a and the lower electromagnet
107a, four electromagnets are formed by electromagnet windings constructed as shown
in the cross-sectional view of Fig. 7. These four electromagnets are opposed in
twos to form a magnetic bearing of two axes: the X-axis and Y-axis.
More specifically, there is realized an opposite-polarity arrangement
with electromagnet windings 111 wound around two adjacent core protrusions constituting
one pair to thereby form one electromagnet. This electromagnet constitutes one
pair together with the electromagnet formed by the electromagnet windings 113
wound around the core protrusions opposed thereto with the rotating body 103 therebetween,
each attracting the rotating body 103 in the positive direction or the negative
direction with respect to the X-axis.
Similarly, regarding the Y-axis direction orthogonal to the X-axis,
one pair is formed as electromagnets opposed to each other with respect to the
Y-axis direction using two electromagnet windings 115 and two electromagnet windings
117 opposed thereto.
As shown in the longitudinal sectional view in Fig. 8, the axial
electromagnets 109a are formed as one pair by two electromagnet windings 121 and
123 with the armature 103a of the rotating body 103 therebetween. The two electromagnets
109a respectively formed by the electromagnet windings 121 and 123 apply an attracting
force to the armature 103a in the positive direction or the negative direction
with respect to the rotation axis.
The upper radial sensor 105b and the lower radial sensor 107b consist
of four sensing coils corresponding to the electromagnets 105a and 107a and arranged
in the X- and Y-axis directions, and are adapted to detect a radial displacement
of the rotating body 103. These sensors are constructed so as to transmit their
respective detection signals to a magnetic bearing control device (not shown) .
On the basis of these sensor detection signals, the magnetic bearing
control device individually adjusts through PID control or the like the attracting
forces of the ten electromagnets constituting the upper radial electromagnet 105a,
the lower radial electromagnet 107a, and the axial electromagnets 109a, whereby
the rotating body 103 is supported in a magnetically levitated state.
Next, a magnetic bearing exciting circuit for exciting the electromagnets
of the magnetic bearing constructed as described above will be described. Fig.
9 shows an example of a magnetic bearing exciting circuit controlling an electric
current flowing through electromagnet windings by a pulse width modulation system.
In Fig. 9, one end of the electromagnet winding 111 forming one electromagnet
is connected to the positive electrode of a power source 133 through a transistor
131, and the other end thereof is connected to the negative electrode of the power
source 133 through a transistor 132.
Then, the cathode of a current regeneration diode 135 is connected
to one end of the electromagnet winding 111, and the anode thereof is connected
to the negative electrode of the power source 133. Similarly, the cathode of a
diode 136 is connected to the positive electrode of the power source 133, and the
anode thereof is connected to the other end of the electromagnet winding 111. Between
the positive electrode and the negative electrode of the power source 133, there
is connected an electrolytic capacitor 141 for stabilization.
Further, an electric current detecting circuit 139 is provided on
the source side of the transistor 132, and the electric current detected by this
electric current detecting circuit 139 is input to a control circuit 137.
The exciting circuit 110 constructed as described above is provided
in correspondence with the electromagnet winding 111, and similar exciting circuits
110 are also formed in correspondence with the other electromagnet windings 113,
115, 117, 121, and 123. Thus, in the case of a five-axis control type magnetic
bearing, ten exciting circuits 110 are connected in parallel to the electrolytic
capacitor 141.
In this construction, when both the transistors 131 and 132 are turned
on, the electric current increases, and when both of them are turned off, the electric
current decreases. When one of them is turned on, a flywheel current is maintained.
By causing the flywheel current to flow, the hysteresis loss is reduced, making
it possible to keep the power consumption at a low level.
Further, it is possible to reduce high frequency noise such as higher
harmonic. Then, by measuring this flywheel current with the electric current detecting
circuit 139, it is possible to detect electromagnet current iL flowing
through the electromagnet winding 111 (See Japanese Patent No. 3176584).
The control circuit 137 compares a current command value with the
detection value obtained by the electric current detecting circuit 139 to determine
the pulse width within one switching period through pulse width modulation, and
transmits signals to the gates of the transistors 131 and 132.
When the current command value is larger than the detection value,
both the transistors 131 and 132 are turned on only once for a period of time corresponding
to a pulse width time TP1 within one switching period Ts
(for example, Ts = 100 µs), as shown in Fig. 10. At this time, the electromagnet
current iL increases, and at the same time, a pulse-shaped positive
electric current ic1 flows through the electrolytic capacitor 141. That
is, the electric current ic1 flows out of the electrolytic capacitor
141.
On the other hand, when the current command value is smaller than
the detection value, both the transistors 131 and 132 are turned off only once
for a period of time corresponding to a pulse width time TP2 within
one switching period Ts, as shown in Fig. 11. At this time, the electromagnet
current iL decreases, and at the same time, a pulse-shaped negative
electric current ic2 flows through the electrolytic capacitor 141. That
is, the electric current ic2
flows into the electrolytic capacitor 141.
As shown in Fig. 5A, in the conventional magnetic bearing exciting
circuit, there is, depending upon the levitation control of the rotating body 103,
the danger of the current ic1 (or the current ic2) for ten
axes being instantaneously superimposed and flowing as a ripple current. Thus,
it is necessary to use a large electrolytic capacitor as the electrolytic capacitor
141, resulting in an increase in circuit size and high cost.
The present invention has been made in view of the above problem
in the prior art. It is an object of the present invention to provide a magnetic
bearing control device capable of suppressing a ripple current flowing through
the electrolytic capacitor to a low level.
According to the present invention, there is provided a magnetic
bearing control device comprising: a rotating body, magnetic bearing means for
controlling the radial position and/or axial position of the rotating body by means
of an electromagnet; exciting circuits including switching elements for effecting
connection and disconnection between the electromagnet and a power source, arranged
in a plurality of rows and connected to the power source in parallel; a capacitor
arranged in the power source; and control means for pulse-controlling the switching
elements, characterized in that the exciting circuits are divided into a plurality
of groups so that the magnitude of the electric current flowing to the capacitor
through the exciting circuits may be less than a predetermined value, and that
a region in one switching period where pulse generation is effected differs from
group to group or there is a partial overlapping of the regions between groups.
By dividing the exciting circuits into a plurality of groups and
making the region where pulse generation is effected during one switching period
different from group to group, it is possible to prevent an electric current from
flowing simultaneously to the capacitor from all the exciting circuits. Due to
the division into groups, the magnitude of the ripple current is smaller, so that
the effective current also becomes smaller.
Note that as long as the current flowing to the capacitor through
the exciting circuit is less than a predetermined value, there may be a partial
overlapping of the regions where pulse generation is effected during one switching
period between groups. Due to this arrangement, it is possible to achieve a reduction
in capacitor size. The control unit can also be reduced in size, which leads to
a reduction in cost.
Further, the magnetic bearing control device of the present invention
is characterized in that the grouping of the exciting circuits is effected such
that the combination of electric currents flowing to the capacitor is substantially
uniform for each region.
With this structure, the effective current resulting from the ripple
current is minimized.
Furthermore, the magnetic bearing control device of the present invention
is characterized in that exciting circuits exciting electromagnets arranged opposite
to each other so as to vary the radial position or the axial position in the same
direction are in the same group.
Generally speaking, in control by a magnetic bearing, when the current
flowing through one electromagnet increases, the current flowing through the other,
opposite electromagnet is likely to decrease. Thus, when opposing electromagnets
are in the same group, it is possible to reduce the ripple current flowing through
the capacitor.
Embodiments of the present invention will now be described by way
of further example only and with reference to the accompanying drawings, in which:-
- Fig. 1 is a circuit diagram showing an embodiment of the present invention;
- Fig. 2 is a time chart showing the control when the current command value is
larger than the detection value;
- Fig. 3 is a time chart showing the control when the current command value is
smaller than the detection value;
- Fig. 4A is a time chart showing how control is effected by opposing electromagnets
(X-axis iL+, X-axis iL-);
- Fig. 4B is a time chart showing how control is effected by opposing electromagnets
((ic+) = ic due to iL+, (ic-) = ic
due to iL-) ;
- Fig. 4C is a time chart showing how control is effected by opposing electromagnets
((ic+) + (ic-));
- Fig. 5A is a diagram showing the magnitude of a ripple current flowing through
an electrolytic capacitor (when there is no division into first and second halves
as in the prior art);
- Fig. 5B is a diagram showing the magnitude of a ripple current flowing through
an electrolytic capacitor (when there is division into first and second halves);
- Fig. 6 is a sectional view of a turbo-molecular pump;
- Fig. 7 is a cross-sectional view of a radial electromagnet;
- Fig. 8 is a longitudinal sectional view of an axial electromagnet;
- Fig. 9 is a diagram showing an example of a magnetic bearing exciting circuit;
- Fig. 10 is a time chart showing the control when the current command value
is larger than the detection value; and
- Fig. 11 is a time chart showing the control when the current command value
is smaller than the detection value.
An embodiment of the present invention will now be described. Fig.
1 is a circuit diagram showing an embodiment of the present invention. The components
which are the same as those of Fig. 9 are indicated by the same reference numerals,
and a description of such components will be omitted.
In Fig. 1, the exciting circuits 110 for ten axes of a magnetic bearing
are divided into two groups: a group in which the transistors 131 and 132 are controlled
to generate pulses on the first half side of one switching period Ts
(ranging from 0 to Ts/2), and a group in which the transistors are controlled
to generate pulses on the second half side thereof (ranging from Ts/2
to Ts).
In this way, the exciting circuits 110 are divided into two groups.
The time chart of Fig. 2 shows the control when the current command value is larger
than the detection value, and the time chart of Fig. 3 shows the control when the
current command value is smaller than the detection value.
Normally, the on/off duty of the transistors 131 and 132 is approximately
0 to 50%, so that through the division into the first-half group and the second-half
group, it is possible to reduce the peak current flowing through the electrolytic
capacitor 141.
In a specific example, the group in which the transistors 131 and
132 are controlled in the first half consists of the exciting circuits 110 corresponding
to the electromagnet windings 111, 113, 115, and 117 of the lower radial electromagnet
107a.
On the other hand, the group in which control is effected in the
second half consists of the exciting circuits 110 corresponding to the electromagnet
windings 111, 113, 115, and 117 of the upper radial electromagnet 105a and the
electromagnet windings 121 and 123 of the axial electromagnet 109a.
Further, in Fig. 7, when the rotating body 103 is to be controlled
in the positive direction with respect to the X-axis, iL+ is increased,
and iL- is decreased. Conversely, when it is to be controlled in the
negative direction with respect to the X-axis, iL+ is decreased, and
iL- is increased. Fig. 4A shows how this control is effected.
In this way, in the case in which the rotating body 103 is controlled
by opposing electromagnets, when the current flowing through one electromagnet
increases, the current flowing through the electromagnet on the opposite side tends
to decrease. Thus, by putting these opposing electromagnets in the same group,
it is possible to reduce the ripple current flowing through the electrolytic capacitor
141.
At this time, the current ic flowing from each exciting
circuit 110 to the electrolytic capacitor 141 is as shown in Fig. 4B, and accordingly
the combined current ic flowing to the electrolytic capacitor 141 is
as shown in Fig. 4C. Thus, the effective value of the combined current ic
can be made small.
Thus, it is desirable for the electromagnet windings 111 and 113
of the upper radial electromagnet 105a, the electromagnet windings 115 and 117
of the upper radial electromagnet 105a, the electromagnet windings 111 and 113
of the lower radial electromagnet 107a, the electromagnet windings 115 and 117
of the lower radial electromagnet 107a, and the electromagnet windings 121 and
123 of the axial electromagnet 109a to be grouped as pairs.
By thus grouping the windings, it is possible, as shown in Figs.
5A and 5B, to suppress the ripple current flowing through the electrolytic capacitor
141 to approximately 1/2 as compared with that in the conventional construction.
Calculation of the effective current based on Figs. 5A and 5B, using
as a reference the magnitude of the ripple current flowing in the case in which
division into two groups is effected as in the present invention, gives the following
results: the effective current when grouping is effected as in the present invention
is to be expressed by Equation 1, whereas the effective current in the case of
the conventional arrangement is to be expressed by Equation 2. Here, the pulse
width is T0, and the duty D = T0 / Ts.
[Equation 1]
Effective current Σ Ic(rms) = sqrt(io2 × D + io2
× D) = sqrt(2) sqrt(D) × io
[Equation 2]
Effective current Σ Ic(rms) = sqrt( square(2 × io) × D) = 2 sqrt(D)
× io
Thus, the effective current is reduced, so that it is possible for
the electrolytic capacitor 141 used to be a small one. The control unit can also
be reduced in size, which leads to a reduction in cost.
It is to be noted here that the grouping is not restricted to the
above-mentioned combination. Any combination will do as long as the magnitude of
the ripple current flowing through the electrolytic capacitor 141 is substantially
uniform in both the first-half group and the second-half group. Thus, there is
a certain degree of freedom regarding the combination.
While in the above embodiment of the present invention the pulse
generation for the exciting circuits 110 is effected in two groups: the group in
which the pulse generation is controlled in the first half of one switching period
Ts, and the group in which it is controlled in the second half thereof,
it is also possible to effect the pulse generation in three groups: first, middle,
and last portions. In this case also, the exciting circuits 110 are grouped such
that the magnitude of the ripple current flowing through the electrolytic capacitor
141 is substantially uniform in each of the three groups.
Further, as long as the magnitude of the ripple current flowing to
the electrolytic capacitor 141 through each exciting circuit 110 is less than a
predetermined value, it is possible to realize a partial overlapping, between different
groups, of the regions where pulse generation is effected during one switching
period.
As described above, in accordance with the present invention, the
exciting circuits are divided into a plurality of groups, and the region where
pulse generation is effected during one switching period differs from group to
group, whereby there is no danger of an electric current flowing simultaneously
to the capacitor from all the exciting circuits.
Due to the grouping, the magnitude of the ripple current is reduced,
so that the effective current is also reduced. Thus, the capacitor size can be
reduced. The control unit can also be reduced in size, which leads to a reduction
in cost.