 |
Description  |
|
|
BACKGROUND OF THE INVENTION
a. Field of the Invention
The present invention relates to a control system for a magnetic type
bearing for floating a high-speed rotating member such as a spindle for
use in a turbo-pump, a compressor, a turbine or a machine tool and further
a traveling member such as a tenter.
b. Description of the Related Art
Means for floatingly holding a rotating member and a traveling member have
utilized a magnetic type bearing employing an electromagnet. The magnetic
type bearing has less loss than that of a conventional lubricated
hydraulic bearing, maintains a dry, clean atmosphere and in particular is
useful under vacuum.
In the magnetic type bearing, in order to establish a float position of the
rotating member and the traveling member, there is provided a system in
which a position of a floating member is measured to determine a current
value flowing in the electromagnet on the basis of the measured signal so
that a magnitude of a magnetic force produced from the electromagnet is
determined.
Referring to FIG. 7 showing a block diagram illustrating the above manner,
a position sensor 11 measures a position (displacement) of the floating
member and may be, for example, an eddy current type displacement meter. A
position feed-back gain circuit 12 proportionally multiplies a magnitude
of a signal obtained by the position sensor 11 to a required magnitude. A
control circuit 13 is a processing circuit for converting a signal
obtained by the position feed-back gain circuit 12 to a proper signal to
supply the signal to an electromagnet 14 and may be, for example, a PID
(Proportion, Integration and Differentiation) circuit, a phase
compensation circuit or a combination thereof. The electromagnet 14
includes a coil wound on an iron core and produces a magnetic force for
floating the member in response to a current supplied from the control
circuit 13.
A simple position feed-back system has the control circuit 13 formed of
only a proportional element (P element). The transfer function of an input
I and a magnetic force F of an output of the electromagnet 14 is given by
the following first-order lag system which depends on resistors and
inductances of a coil and an iron core.
F/I=K.sub.M /(1+T.sub.M .multidot.S) (1)
where K.sub.M is a gain of the electromagnet 14, T.sub.M is a time constant
of the electromagnet 14, and S is a Laplacian operator. Accordingly, the
transfer function of the force F exerted on the floating member with
respect to the displacement D measured by the position feed-back system is
as follow:
F/D=K.sub.F .multidot.K.sub.P .multidot.K.sub.m /(1+T.sub.M .multidot.S)
(2)
where K.sub.F is a proportional gain of the position feed-back gain circuit
12 and K.sub.P is a proportional gain of the control circuit 13. In order
to observe a frequency characteristic of the force/displacement (F/D) of
the position feed-back system, the Laplacian operator is set to S=j2.pi.f
in which f is a frequency (Hz) and j=.sqroot.-1 and is substituted in the
equation (2). The force/displacement (F/D) is a complex number as follow:
F/D=K.sub.R (f)+j.multidot.K.sub.I (f) (3)
The real part K.sub.R of the force/displacement (F/D) in the above equation
(3) means stiffness dependent on the frequency f and an imaginary part
K.sub.I thereof means attenuation dependent on the frequency f. The
first-order lag as described in the equation (2) has always a negative
imaginary part and the attenuation forms an unstabilizing force for the
floating member.
FIG. 8 is a graph showing a relation of the force/displacement (F/D), that
is, a relation of a value of the imaginary part of the equation (3) and
the frequency f. A dashed line A shown in FIG. 8 corresponds to the
equation (2) and shows the above-described state. A characteristic
frequency fc determined by the floating member and the position feed-back
system increases divergently and the system can not operate due to the
attenuation of the characteristic frequency fc, particularly the
attenuation of the floating member if a value of the frequency f=fc shown
in FIG. 8 is large.
Thus, in order to give the attenuation effect to the force/displacement
(F/D) of the position feed-back system, the control circuit 13 comprises a
differential element (D element) or a position compensation element
disposed in parallel with the proportional element (P element). In this
description, the differential element is taken up by way of example. When
the differential element (D element) is added to the control circuit 13,
the following first-order lag is added to the circuit.
Differential Element=K.sub.D .multidot.S/(1+T.sub.D .multidot.S) (4)
where K.sub.D is a gain of the differential element and T.sub.D is a time
constant. The force/displacement (F/D) of the position feed-back system
including only the differential element is as follow:
F/D=K.sub.F .multidot.K.sub.D .multidot.K.sub.M .multidot.S/{(1+T.sub.D
.multidot.S)(1+T.sub.M .multidot.S)} (5)
Since the numerator of the equation (5) is an equation of a first degree of
S and the denominator thereof is an equation of a second degree of S, the
imaginary part of the equation (5) is given by a one-dot chain line B
shown in FIG. 8. That is, the attenuation effect is given to the floating
member in a low frequency range and the unstabilizing operation is given
to the floating member in a high frequency range. In order to hold the
position of the floating member, the control circuit 13 requires both of
the proportional element and the differential element. The
force/displacement (F/D) of the position feed-back system of the control
circuit 13 is given by
F/D=K.sub.F .multidot.{K.sub.P +K.sub.D .multidot.S/(1+T.sub.D
.multidot.S)}.multidot.K.sub.M /(1+T.sub.M .multidot.S) (6)
The force/displacement (F/D) is also shown by a solid line C of FIG. 8 and
has the same characteristic as described above. When the characteristic
frequency fc determined by the floating member and the position feed-back
system is placed in a low frequency range having the attenuation effect,
stabilization can be obtained and operation can be made without occurrence
of vibration.
When it is considered that the magnetic type bearing having the above
characteristic is employed as a bearing 16 of a rotating member 15 shown
in FIG. 9(a) to float the rotating member 15, the following phenomenon
occurs. The rotating member 15 has unlimited number of characteristic
frequencies the first five of which are shown in FIGS. 9(b), (c), (d), (e)
and (f). The attenuation of material of the rotating member 15 itself acts
on unstabilization with respect to the characteristic frequency less than
a rotational number of the member and acts as the attenuation operation
with respect to the characteristic frequency larger than the rotational
number.
Accordingly, it is necessary to set the characteristic frequency less than
the rotational number within the frequency range in which the attenuation
effect of the force/displacement (F/D) of the position feed-back system of
the magnetic type bearing is brought. However, since the number of the
characteristic frequencies of the rotating member 15 is unlimited as shown
in FIGS. 9(b) , (c), (d), (e) and (f), the characteristic frequency
certainly exists in the frequency range in which the unstabilizing
operation of the force/displacement (F/D) is effected. Accordingly, when
the unstabilizing operation of the position feed-back system of the
magnetic type bearing is larger than the attenuation of the characteristic
frequency by the rotating member 15 itself, operation is destabilized and
vibration of the rotating member increases divergently, so that the
rotating member can not be rotated.
As described above, heretofore, in order to hold the position of the
floating member, the position of the floating member is measured and the
measured signal is fed back to produce force from the electromagnet.
However, the force is destabilizing force which vibrates the floating
member. Thus, even if processing such as the PID and phase compensation is
provided in the control circuit 13, the force is the stabilizing force
(attenuation) in the low frequency range, while the force still contains a
large destabilizing force in a middle and high frequency range.
Accordingly, the floating member such as the rotating member having the
unlimited number of characteristic frequencies certainly includes the
characteristic frequency existing in the frequency range in which the
destabilizing force is produced and divergent vibration occurs by means of
the magnetic type bearing.
SUMMARY OF THE INVENTION
It is an object of the present invention to provide a control system for a
magnetic type bearing in which destabilizing force produced by the
magnetic type bearing in a specified frequency range is changed to
stabilizing force (attenuation force) to prevent occurrence of divergent
vibration so that a floating member is floated stably.
BRIEF DESCRIPTION OF THE DRAWINGS
FIGS. 1 to 5 show a first embodiment of the present invention, in which:
FIG. 1 is a block diagram showing a configuration of a control system for a
magnetic type bearing;
FIG. 2 is a plan view showing a mounting state of a first and second
position sensors;
FIGS. 3(a) and (b) are waveform diagrams showing output signals of the
first and second sensors, respectively;
FIGS. 4(a) and (b) are characteristic diagrams showing gain-frequency
characteristics of a first and second filters, respectively; and
FIG. 5 is a characteristic diagram showing an attenuation characteristic of
the magnetic type bearing;
FIG. 6 is a block diagram of a control system for a magnetic type bearing
showing a second embodiment of the present invention;
FIG. 7 is a block diagram showing a configuration of a conventional control
system for a magnetic type bearing;
FIG. 8 is a characteristic diagram showing an attenuation characteristic of
the magnetic type bearing controlled by the conventional control system;
and
FIGS. 9(a)-(f) show a rotating member and characteristic frequencies.
DETAILED DESCRIPTION OF THE INVENTION
Embodiments of the present invention are now described.
FIG. 1 is a block diagram showing a configuration of a first embodiment of
the present invention. In FIG. 1, numerals 1 and 2 denote first and second
position sensors, respectively, for measuring a position (displacement) of
a floating member and which may be, for example, eddy current displacement
type sensors. The first and second position sensors 1 and 2 are disposed
at opposite positions around a rotating member 3 as shown in FIG. 2. In
FIG. 2, the rotating member 3 is shown as one example of the floating
member, while it is not limited to the rotating member.
In FIG. 1, numeral 4 denotes a first filter to which a signal of the first
position sensor 1 is supplied, numeral 5 denotes a second filter to which
a signal of the second position sensor 2 is supplied, numeral 6 denotes an
adder which adds signals passed through the first and second filters,
numeral 7 denotes a position feedback gain circuit for proportionally
multiplying the addition output of the adder 6 to a required magnitude,
and numeral 8 denotes a control circuit for converting a signal from the
position feedback gain circuit 7 to a proper signal to supply it to an
electromagnet 9. The control circuit 8 may be, for example, a PID
(Proportion-Integration-Differentiation) circuit, a phase compensation
circuit or a combination thereof. The electromagnet 9 also includes a coil
wound on an iron core and produces a magnetic force for floating the
rotating member 3 in response to a current supplied by the control circuit
8.
When a signal supplied to the first filter 4 from the first position sensor
1 is a and a signal supplied to the second filter 5 from the second
position sensor 2 is b, the signal a has an inverted amplitude of that of
the signal b with respect to operation of the floating member, that is, is
shifted 180.degree. in phase from the signal b as shown in FIGS. 3(a) and
(b). Accordingly, when the signal a has, for example, a value of +5, the
signal b has a value of -5.
FIGS. 4(a) and (b) show gain characteristics of the first and second
filters 4 and 5, respectively. As shown in FIG. 4(a), the first filter 4
has a cut-off characteristic (the gain thereof is zero) in a predetermined
frequency range in which the floating member is to be stabilized, that is,
in a range from a frequency f.sub.c1 to a frequency f.sub.c2, while the
second filter 5 has a passing characteristic (the gain thereof is 1) in
the range from the frequency f.sub.c1 to the frequency f.sub.c2 reversely
as shown in FIG. 4(b).
The signal a from the first position sensor 1 and the signal b from the
second position sensor 2 are supplied to the adder 6 through the first and
second filters 4 and 5, respectively, and are added to each other in the
adder 6. More particularly, the signal a from the first position sensor 1
is supplied as a first signal c to the adder 6 through the first filter 4
and the signal b from the second position sensor 2 is supplied as a second
signal d to the adder 6 through the second filter 5 so that the two
signals c and d are added to each other. The addition signal is supplied
through the position feedback gain circuit 7 to the control circuit 8.
When the force/displacement (F/D) of the magnetic type bearing is expressed
by a complex function such as the equation (3), the force/displacement
(F/D) of the magnetic type bearing in the path provided with the first
position sensor 1 in the frequency range from f.sub.c1 to f.sub.c2 is as
follows:
F/D=0 (7-1)
The force/displacement (F/D) in the other frequency range is as follow:
F/D=K.sub.R (f)+j.multidot.K.sub.I (f) (7-2)
Since the polarity of the signal b from the second position sensor 2 is
quite opposite to that of the signal a, the force/displacement (F/D) in
the frequency range from f.sub.c1 to f.sub.c2 is as follows:
F/D=-K.sub.R (f)-j.multidot.K.sub.I (f) (8-1)
The force/displacement (F/D) in the other frequency range is as follows:
F/D=0 (8-2)
Finally, both are added and the added value in the frequency range from
f.sub.c1 to f.sub.c2 is as follows:
F/D=-K.sub.R (f)-j.multidot.K.sub.I (f) (9-1)
The value in the other frequency range is as follows:
F/D=K.sub.R (f)+j.multidot.K.sub.I (f) (9-2)
Accordingly, the attenuation characteristic of the magnetic type bearing is
as shown by a solid line D of FIG. 5, that is, the destabilizing force
shown by a dashed line E of FIG. 5 in the frequency range of f.sub.c1
-f.sub.c2 is changed to stabilizing force. Accordingly, the characteristic
frequency in the frequency range is stabilized and occurrence of the
divergent vibration is prevented.
Further, the cut-off frequency range f.sub.c1 -f.sub.c2 of the first filter
4 and the passing frequency range f.sub.c1 -f.sub.c2 of the second filter
5 are not required to be identical with each other in the strict sense and
the frequency ranges may be shifted by deviation at the boundaries of the
band width due to adjustment. The gains of the first and second filters 4
and 5 may be different from each other.
In the first embodiment, a signal portion of the signal a from the first
position sensor 1 in the frequency range which produces the unstabilizing
force is cut off in the first filter 4 and a signal portion of the signal
b from the second position sensor 2 in the frequency range which is
inverted 180.degree. with respect to the signal a and produces the
stabilizing force passes through the second filter 5. Both the signals
from the first and second filters 4 and 5 are added and fed back, and
accordingly force produced by the magnetic type bearing is changed to the
stabilizing force.
According to the embodiment, the position of the floating member is
measured by the first and second position sensors 1 and 2 from two
opposite directions and the measured signals are passed through the first
and second filters 4 and 5 having opposite passing and cut-off
characteristics, respectively, the signals passed through the filters
being added and fed back. Accordingly, there can be provided the control
system for the magnetic type bearing which converts the unstabilizing
force produced by the magnetic type bearing to the stabilizing force to
prevent the divergent vibration of the floating member so that the
floating member can be floated stably.
FIG. 6 is a block diagram showing a second embodiment of the present
invention, in which the same elements as those of FIG. 1 are given like
numerals.
In FIG. 6, numeral 10 denotes a polarity inversion circuit which inverts
the polarity of a signal.
The signal from the first position sensor 1 is divided into two signals.
One signal a passes through the first filter 4 as it is, while the other
signal b is polarity-inverted by the polarity inversion circuit 10 and
subsequently passes through the second filter 5.
In this case, the signals a and b supplied to the first and second filters
4 and 5, respectively, have quite opposite amplitude with respect to
movement of the floating member in the same manner as shown in FIGS. 3(a)
and (b), that is, are shifted 180.degree. in phase each other.
In the second embodiment, since the configuration of the latter part
subsequent to the first and second filters 4 and 5 are identical with the
configuration of the first embodiment shown in FIG. 1, it is understood
that the second embodiment attains the same effect as that of the first
embodiment.
More particularly, since a signal portion of one signal in the frequency
range which produces the unstabilizing force is cut off by the first
filter 4 while a signal portion of the inverted signal of the other signal
in the frequency range which produces the stabilizing force is passed
through the second filter 5 and the addition of both the signals are fed
back, force produced by the magnetic type bearing is all changed to the
stabilizing force.
According to the second embodiment, the signal from the position sensor 1
is divided into two signals. One signal a thereof passes through the first
filter 4 having a cut-off band corresponding to the frequency band in
which stabilization is desired. The other signal b is polarity-inverted
and passes through the second filter 5 having a passing band corresponding
to the cut-off band. Both the signals from the filters 4 and 5 (that is,
first and second signals c and d) are added and fed back to the magnetic
type bearing to convert the signals to force by means of the
electromagnet. Accordingly, there can be provided the control system for
the magnetic type bearing which changes the unstabilizing force produced
by the magnetic type bearing in the specified frequency range to the
stabilizing force (attenuation force) and can prevent the occurrence of
the divergent vibration so that the floating member can be floated stably.
Further, the present invention is not limited to the first and second
embodiments. For example, in the second embodiment, while the other signal
is inverted just after the signal from the position sensor 1 is divided
into two signals, the other signal may be inverted after the second filter
5, that is, before adder 6. Furthermore, while the stabilization is
attained in one frequency range from f.sub.c1 to f.sub.c2 in the first and
second embodiments, the stabilization may be attained in a plurality of
frequency ranges or all frequency ranges above f.sub.c1 in accordance with
characteristics of the floating member and the magnetic type bearing. It
is a matter of course that various modifications can be made without
departing from the gist of the present invention.
* * * * *
|
|
 |
|
 |
Description |
|
