JP4875335B2 - High frequency power supply device and control method of high frequency power supply - Google Patents

Description

  The present invention relates to a high-frequency power supply apparatus that supplies power to a load such as a plasma processing apparatus that performs plasma etching and plasma CVD, for example.

  Conventionally, as a high-frequency power supply device that supplies high-frequency power to a load, a method of controlling traveling wave power output from the high-frequency power supply device to be constant (hereinafter referred to as PF control) or load-side power obtained by subtracting reflected wave power from traveling wave power A method of controlling the frequency constant (hereinafter referred to as PL control) has been used.

FIG. 7 is a block diagram showing a configuration of a conventional high-frequency power supply device 50 using PF control and a connection relationship among the high-frequency power supply device 50, the matching unit 3, and the load 5.
The conventional high frequency power supply device 50 is a power supply device for supplying high frequency power to the load 5 via the transmission line 2, the matching unit 3, and the load connection portion 4, and the output power is controlled as follows.

  In the voltage controlled variable attenuator 13, an output signal of an error amplifier 26, which will be described later, is used as a command value, and the level of a high-frequency signal having a predetermined frequency output from the crystal oscillator 12 is varied and output. The amplifying unit 15 receives the direct current power supplied from the AC / DC converting unit 14 and the high frequency signal output from the voltage controlled variable attenuator 13 and amplifies the high frequency signal output from the voltage controlled variable attenuator 13. Traveling wave power PF is output. The power detector 16 attenuates and extracts the traveling wave power PF output from the amplifier 15 and outputs an attenuated traveling wave signal RFpf. Thereafter, the power level converter 17 converts the output of the power detection unit 16 into a DC signal, and then normalizes and outputs it. The error amplifier 26 outputs a signal serving as a command value to the voltage controlled variable attenuator 13 so that the magnitude of the output signal of the power level converter 17 is equal to the magnitude of the output power setting signal. In this way, traveling wave power can be controlled to be constant.

  The matching unit 3 includes a power source side impedance Zo (usually 50Ω) viewed from the input end 301 of the matching unit 3 via the transmission line 2 and a load side viewed from the input end of the matching unit 3 to the load side. This is a device used for the purpose of impedance matching between the high-frequency power supply device and the load 5 by matching the side impedance ZL (impedance of the matching device 3, load connecting portion 4 and load 5).

  The matching unit 3 includes a variable impedance element (not shown) (for example, a variable capacitor, a variable inductor, etc.) inside the matching unit 3. The matching unit 3 includes a variable impedance element such as a variable capacitor and a variable inductor. It has a function to change impedance. More specifically, for example, the impedance (output impedance) of the high frequency power supply device viewed from the output end 101 of the high frequency power supply device is designed to be 50Ω, for example, and the high frequency power supply device is impedance matched with the transmission line 2 having a characteristic impedance of 50Ω. If the impedance matching unit 3 is connected to the input end of the impedance matching unit 3, the impedance matching unit 3 changes the impedance of the variable impedance element so that the load side impedance ZL viewed from the input end of the impedance matching unit 3 is converted to 50Ω. Change.

  The load 5 is an apparatus for processing (etching, CVD, etc.) a workpiece such as a wafer and a liquid crystal substrate that is provided with a processing unit and is carried into the processing unit. In order to process the workpiece, the load 5 introduces a plasma discharge gas into the processing portion, and discharges the plasma discharge gas using high-frequency power (voltage) supplied from a high-frequency power supply (hereinafter referred to as plasma). This is referred to as “discharge” to change from a non-plasma state to a plasma state. Then, the workpiece is processed using plasma. Such a load accompanied by discharge is called a discharge load.

FIG. 8 is a block diagram showing a configuration of a conventional high-frequency power supply device 51 using PL control and a connection relationship between the high-frequency power supply device 51, the matching unit 3, and the load 5.
The conventional high frequency power supply device 51 is a power supply device for supplying high frequency power to the load 5 via the transmission line 2, the matching unit 3, and the load connection unit 4, and the output power is controlled as follows. The matching unit 3 and the load 5 are the same as those in FIG.

  In the voltage controlled variable attenuator 13, an output signal of an error amplifier 26, which will be described later, is used as a command value, and the level of a high-frequency signal having a predetermined frequency output from the crystal oscillator 12 is varied and output. The amplifying unit 15 receives the direct current power supplied from the AC / DC converting unit 14 and the high frequency signal output from the voltage controlled variable attenuator 13 and amplifies the high frequency signal output from the voltage controlled variable attenuator 13. Traveling wave power PF is output. The power detection unit 16 attenuates and extracts the traveling wave power PF and the reflected wave power PR output from the amplification unit 15 and outputs an attenuated traveling wave signal RFpf and an attenuated reflected wave signal RFpr. Thereafter, the power level converter 17 converts the output of the power detection unit 16 into a DC signal, and then normalizes and outputs it. Thereafter, the load-side power calculation unit 18 obtains and outputs a signal corresponding to the load-side power (traveling wave power-reflected wave power). The error amplifier 26 outputs a signal serving as a command value to the voltage controlled variable attenuator 13 so that the magnitude of the output signal of the load-side power calculator 18 is equal to the magnitude of the output power setting signal. By doing so, the load-side power can be controlled to be constant.

By the way, when the impedance matching is performed between the high frequency power supply devices 50 and 51 and the load 5 shown in FIGS. 7 and 8, the high frequency power output from the high frequency power supply devices 50 and 51 is efficiently supplied to the load 5. Is done. However, since the internal impedance of the load 5 varies depending on the state of the plasma discharge, the impedance of the load varies to a high impedance or a low impedance as compared with the matching.
Then, the impedance is not matched between the high-frequency power supply devices 50 and 51 and the load 5, and the reflection coefficient at the input terminal 301 of the matching device increases, so that the output is made from the high-frequency power supply devices 50 and 51. Thus, part or all of the traveling wave power directed to the load 5 is reflected to generate reflected wave power PR from the matching unit 3 to the high frequency power supply device, so that the high frequency power is not efficiently supplied to the load 5.

  Normally, the matching unit 3 returns to the matching state in order to perform impedance matching, but reflected wave power is generated during the period from the non-matching state to the matching state. The reflected wave power PR generated for the above reason returns into the high frequency power supply device.

  Next, an operation after high frequency power is output from the high frequency power supply device in the system including the high frequency power supply device described above will be described.

(1) In a system including the above-described high-frequency power supply device, discharge is started when high-frequency power is supplied from the high-frequency power supply device to the load 5 and discharge is possible. At this stage, since there is no matching, the reflected wave power is large and almost totally reflected. That is, the absolute value | Γ | ≈1 of the reflection coefficient Γ. The reflection coefficient Γ refers to the ratio (ratio) of the reflected wave to the traveling wave, and is represented by Expression (1). Further, when the absolute value of the reflection coefficient Γ is | Γ | and the phase angle is θ, the reflection coefficient Γ is expressed by Expression (2). Further, when the formula (2) is changed, the formula (3) is obtained. When the formula (3) is collected, the reflection coefficient can be expressed in the form of “real part + imaginary part”.
Reflection coefficient Γ = RFpr / RFpf (1)
Reflection coefficient Γ = | Γ | · exp (jθ) (2)
Reflection coefficient Γ = | Γ | ((cos θ) + j (sin θ)) (3)

(2) The matching unit performs a matching operation so as to reduce the reflected wave power. As a result, the alignment target point is finally reached. The alignment target point is a position on the polar chart where the absolute value of the reflection coefficient | Γ | = 0 or within an allowable range set by the absolute value of the reflection coefficient | Γ | (for example, | Γ | ≦ 0. The position of 1) is said.

(3) When the load conditions (power supply amount, etc.) change, the impedance of the load changes and reflection may increase. In that case, the matching operation is performed again by the matching unit to reach the matching target point. This is repeated thereafter.
Japanese Patent Laying-Open No. 2005-077248

  As described above, regardless of the PF control or the PL control, the output power is controlled based on the detection signal of the power detection unit 16. Therefore, in order to perform control with high accuracy, the detection accuracy of the power detection unit 16 needs to be good. However, since the detection accuracy of the power detection unit 16 has characteristics as shown in (1) below, (2 ) And (3) cause problems peculiar to each control method.

(1) Detection accuracy of the power detection unit 16:
The detection sensitivity of the power detection unit 16 of the high-frequency power supply device is determined by two specifications of the degree of coupling and the directionality. The degree of coupling represents the amount of attenuation of high-frequency power. The directivity is the magnitude of leakage of traveling wave power PF and reflected wave power PR when the input / output characteristic impedance (mainly determined by the structure) of the power detection unit 16 and a load 50Ω are connected to the power detection unit 16 alone. The detection accuracy of the power detection unit 16 is determined by this directionality.

FIG. 9 shows an ideal reflection coefficient calculation value (becomes a perfect circle) obtained when an “ideal power detection unit” (a power detection unit with no error in the detection value) is used when performing PF control. FIG. 4 is a diagram showing a comparison between an example of a reflection coefficient calculation value calculated from a reflected wave power detection value of an “actual power detection unit” on a polar chart. The center of the circle in FIG. 9 is | Γ | = 0 (matched and no reflection), and the outermost circle represents | Γ | = 1 (total reflection).
The trajectory indicated by (I) is a reflection obtained by using the “ideal power detector” in the case of reflected wave power with an absolute value of reflection coefficient | Γ | = a (0 <a <b <1). Indicates the coefficient calculation value.
The trajectory indicated by (II) is a reflection obtained by using the “ideal power detection unit” in the case of reflected wave power with an absolute value of reflection coefficient | Γ | = b (0 <a <b <1). Indicates the coefficient calculation value.
The trajectory indicated by (III) is a reflection obtained by using the “actual power detection unit” in the case of reflected wave power with an absolute value of reflection coefficient | Γ | = a (0 <a <b <1). Indicates the coefficient calculation value.
The trajectory indicated by (IV) is a reflection obtained by using the “actual power detection unit” in the case of reflected wave power with an absolute value of reflection coefficient | Γ | = b (0 <a <b <1). Indicates the coefficient calculation value.

  As can be seen from FIG. 9, there is a difference between the reflection coefficient calculation value obtained using the “ideal power detection unit” and the reflection coefficient calculation value obtained using the “real power detection unit”. That is, the “real power detection unit” indicates that there is an error in the detected value of the reflected wave power. Further, as the absolute value | Γ | of the reflection coefficient increases from 0, the error between the reflected wave power detected by the “ideal power detector” and the reflected wave power detected by the “real power detector” Becomes larger. The amount of error generated increases or decreases depending on the phase angle θ, and the increase / decrease tendency shows the same tendency even if the value of | Γ |

  If the directivity of the power detector 16 is -40 dB or less, the error in detection accuracy is as small as ± 2% when | Γ | = 0 and ± 7.8% when | Γ | = 0.98. However, when the frequency increases or the amount of power passing through the power detection unit 16 increases, it is the limit to ensure the directivity of about −20 dB. When the directivity is −20 dB, the error becomes an order of magnitude as large as ± 20% when | Γ | = 0 and ± 78.1% when | Γ | = 0.98.

(2) Problems in the region where the absolute value | Γ | of the reflection coefficient is large:
In the case of the above-described high-frequency power supply device of the PF control system, the traveling wave power PF is controlled to be constant even when the error in the detection signal of the reflected wave power increases as | Γ | increases from 0. For this reason, the effect of causing an error in the reflected wave power display value is sufficient. However, in the above-described high frequency power supply apparatus of the PL control system, the error of the detection signal of the reflected wave power increases as | Γ | increases from 0, so that PL = PF-PR near the outer periphery of the polar chart. The amount of power supplied to the load fluctuates because the amount of power supplied to the load becomes unstable even though the amount of PR in the equation expressed by . Therefore, the plasma state changes and the load side impedance ZL fluctuates. As a result, the disappearance of the plasma may occur due to the fact that the plasma cannot be maintained.

  In addition, if the amount of power supplied to the load becomes unstable and the load side impedance ZL changes, the matching unit tries to match. However, since the load side impedance ZL changes, the matching target point is unsatisfactory. May become stable. In this case, the target position of the variable impedance element provided in the matching unit becomes unstable and cannot be accurately determined. Therefore, the variable impedance element provided in the matching unit may cause a hunting phenomenon.

Therefore, PL control is not suitable in a region where the absolute value | Γ | of the reflection coefficient is large, and PF control is more desirable.

(3) Problems in the region where the absolute value | Γ | of the reflection coefficient is small:
FIG. 10 is a diagram illustrating an example of reflected wave power detection characteristics in the vicinity of the matching target point of the power detection unit 16.
The locus indicated by (V) uses the “ideal power detector” when the reflected wave power is the absolute value of the reflection coefficient | Γ | = m (m is the reflection coefficient indicating the allowable range of the matching target point). The reflection coefficient calculation value obtained in this way is shown.
The trajectory indicated by (VI) indicates a reflection coefficient calculation value obtained using the “actual power detection unit” in the case of reflected wave power at which the absolute value of the reflection coefficient | Γ | = m.
As shown in FIG. 10, since the absolute amount of reflected wave power is smaller than the traveling wave power near the matching target point, the error in the reflected wave power detection value due to directionality is small. Therefore, in the high frequency power supply device of the PL control system, the error amount for PR in the expression represented by PL = PF-PR is small, so that the influence on the amount of power supplied to the load is small and a stable amount of power can be obtained. Can supply load.

  Since the impedance of the load changes depending on the electric power supplied from the high frequency power supply device, matching can be achieved at high speed in combination with the matching operation of the matching unit if the supplied electric power is stable.

  On the other hand, since the traveling wave power PF is controlled to be constant as described above in the case of the traveling wave power constant control type high frequency power supply device, if the reflected wave power increases, the amount of power supplied to the load decreases. The plasma state changes and the load side impedance ZL changes. As a result, the extinction phenomenon of the plasma may occur due to the absence of the alignment state or the state where the plasma cannot be maintained.

  In addition, if the amount of power supplied to the load becomes unstable and the load side impedance ZL changes, the matching unit tries to match. However, since the load side impedance ZL changes, the matching target point is unsatisfactory. May become stable. In this case, the target position of the variable impedance element provided in the matching unit becomes unstable and cannot be accurately determined. Therefore, the variable impedance element provided in the matching unit may cause a hunting phenomenon. When such a hunting phenomenon occurs in the matching process, for example, only a part of the variable impedance element provided in the matching unit performs an operation for changing the impedance (for example, a rotation operation), and thus wear progresses partially. Problems such as shortening the life of the variable impedance element may occur.

  Therefore, as the output power control method of the high frequency power supply device near the matching target point, the PL control is more preferable than the PF control.

  Therefore, as proposed by the present applicant in Japanese Patent Application No. 2004-082028, the absolute value of the reference reflection coefficient (hereinafter referred to as the reference reflection coefficient absolute value) is determined, and the reflection coefficient absolute value is larger than the reference reflection coefficient absolute value. It can be considered that PF control is performed in a region where the reflection coefficient is large, and PL control is performed in a region where the reflection coefficient absolute value is smaller than the reference reflection coefficient absolute value. For example, PL control may be performed when the reflection coefficient absolute value is 0.1 or less, and PF control may be performed when the reflection coefficient absolute value exceeds 0.1.

  However, the absolute value of the reference reflection coefficient is not clearly defined. In general, as shown in FIG. 11, a region suitable for PF control (hereinafter referred to as PF control compatible region), a region suitable for PL control (hereinafter referred to as “PF control suitable region”). , Referred to as a PL control adaptation region), and an intermediate region that is intermediate between the PF control adaptation region and the PL control adaptation region.

  FIG. 11 is a diagram showing the intermediate region on the polar chart. In FIG. 11, the shaded area is the intermediate area, the outside of the intermediate area is the PF control compatible area, and the inside of the intermediate area is the PL control compatible area. In this intermediate area, the degree suitable for PF control increases toward the outside, and the degree suitable for PL control increases toward the inside, but the reference reflection coefficient absolute value cannot be clearly determined. There is no clear reference reflection coefficient absolute value reference, but PL control consumes more power than PF control, so the reference reflection coefficient absolute value should be as small as possible to reduce power consumption. Is desirable.

  By the way, in the region where the absolute value | Γ | of the reflection coefficient is small, there is a region where the impedance of the load fluctuates rapidly due to slight condition fluctuations, for example, slight fluctuations in high-frequency power and fine movement of the matching device. There is a case where this region exists (hereinafter, this region is referred to as a variation region). In such a fluctuation region, the amount of reflected wave power detected by the high-frequency power supply device increases simultaneously with a sudden change in the impedance of the load. As a result, since the amount of power supplied to the load fluctuates, a hunting phenomenon may occur in the plasma extinction phenomenon or the matching process, as in the region where the absolute value | Γ | of the reflection coefficient is small. Therefore, as the output power control method of the high-frequency power supply device in the fluctuation region, the PL control is more preferable than the PF control.

  If there is a variable region in a region smaller than the reference reflection coefficient absolute value, PL control suitable for the variable region can be performed. However, as described above, it is desirable to make the absolute value of the reference reflection coefficient as small as possible. Therefore, if the absolute value of the reference reflection coefficient is set without considering the fluctuation region, all or part of the fluctuation region is changed to the reference reflection coefficient. It may enter an area larger than the absolute value. If so, the control in the fluctuation region cannot be performed well, and it is necessary to set the reference reflection coefficient absolute value in consideration of the fluctuation region. That is, on the polar chart, it is necessary to set the reference reflection coefficient absolute value so that the area inside the reference reflection coefficient absolute value includes the PL control compatible area and the fluctuation area.

  FIG. 12 is a diagram showing an example of the fluctuation region on the polar chart. The fluctuation area 90 shows an example of the fluctuation area where the whole area is included in the PL control compatible area. The fluctuation area 91 is an example of a fluctuation area in which a part of the area is included in the PL control compatible area and the other area is included in the intermediate area. The fluctuation area 92 shows an example of the fluctuation area where the whole area is included in the intermediate area.

  Note that FIG. 12 shows the PL control compatible region, the intermediate region, and the PF control compatible region as in FIG. 11, but the oblique lines of the intermediate region are not drawn. Further, FIG. 12 shows a case where the fluctuation region exists inside the intermediate region on the polar chart. In FIG. 12, the variable region is shown as a circle for the sake of explanation, but the size and shape of the variable region are not uniform and vary depending on the situation.

  Using FIG. 12 as an example, how to set the reference reflection coefficient absolute value will be described.

  When only the fluctuation region 90 exists, since the fluctuation region 90 exists in the PL control compatible region, the minimum value of the reference reflection coefficient absolute value is a boundary value between the PL control compatible region and the intermediate region. The optimum value of the absolute value of the reference reflection coefficient may be larger than the boundary value, but the extent to which it is best to enter the intermediate region depends on the situation, so it can be determined at the time of actual operation or experiment Good.

  When the fluctuation region 91 exists, the reference reflection coefficient absolute value needs to be set so that at least the fluctuation region 91 is included because the fluctuation region 91 exists not only in the PL control compatible region but also in the intermediate region. In the case where the variable region 92 exists, as in the case of the variable region 91, it is necessary to set the reference reflection coefficient absolute value so as to include at least the variable region 92. Further, there are cases where the reference reflection coefficient absolute value is further increased in the intermediate region to be the optimum value. However, since it varies depending on the situation, it may be determined during actual operation or experiment.

  Thus, on the polar chart, it is necessary to set the reference reflection coefficient absolute value so that the area inside the reference reflection coefficient absolute value includes the PL control compatible area and includes the fluctuation area. In some cases, the absolute value of the reference reflection coefficient needs to be set larger than in the case where no exists.

  FIG. 13 is a diagram illustrating the absolute value of the reference reflection coefficient when the variation region 92 illustrated in FIG. 12 exists.

  When there is no fluctuation region 92, even if the reference reflection coefficient absolute value = S1 can be obtained, as shown in this figure, when the fluctuation region 92 exists, it is necessary to set the reference reflection coefficient absolute value = S2 at a minimum. Therefore, as compared with the case where the variable region 92 does not exist, the region where the PL control is performed is enlarged only by the region filled with diagonal lines.

  Since it is better to perform the PL control on the fluctuation area 92 in the enlarged area, it is considered that the extra power consumption increases in the area excluding the fluctuation area 92 from the enlarged area compared to the case where the PF control is performed.

  As described above, in the conventional high-frequency power supply device, if PL control is used, a hunting phenomenon may occur in the plasma extinction phenomenon or the matching process in the region where the absolute value | Γ | of the reflection coefficient is large. there were. In addition, when PF control is used, there is a problem that a plasma annihilation phenomenon or a hunting phenomenon may occur in the matching process in a region where the absolute value | Γ | of the reflection coefficient is small.

  The present invention has been conceived under the above circumstances, and does not cause a plasma extinction phenomenon or a hunting phenomenon in a matching process over a wide region of the absolute value | Γ | of the reflection coefficient. Furthermore, an object of the present invention is to provide a high-frequency power supply device that does not increase excessive power consumption even when there is a fluctuation region.

The high frequency power supply device provided by the first invention is
In the high frequency power supply device that controls the power value of the high frequency power supplied to the load to be constant,
High-frequency output means as a supply source of high-frequency power;
Detecting means for detecting a traveling wave from the high-frequency output means toward the load and a reflected wave from the load side toward the high-frequency output means;
A power signal output means for obtaining a signal corresponding to the traveling wave power and a signal corresponding to the reflected wave power based on the traveling wave and the reflected wave detected by the detection means;
Load-side power calculating means for calculating a signal corresponding to load-side power by subtracting a signal corresponding to the reflected-wave power from a signal corresponding to the traveling-wave power;
A reflection coefficient calculating means for calculating a reflection coefficient based on the high frequency information detected at the output terminal of the high frequency power supply device;
A region setting means for setting one inner region defined by a relational expression on the coordinates relating to the reflection coefficient and one outer region paired with the inner region;
When the reflection coefficient calculated by the reflection coefficient calculation means enters the inner region, a signal corresponding to the load side power is output as a power signal, and the reflection coefficient calculated by the reflection coefficient calculation means is when exiting the further outside the outer region enters the outer region exits said side region, the control target switching means for outputting a signal corresponding to the traveling wave power as a power signal,
Control means for controlling the output of the high-frequency output means so that the power signal output from the controlled object switching means is equal to the signal corresponding to the output power setting value;
The inner region set in the region setting means has a large detection error amount of the reflected wave so as to include the coordinate origin of the coordinates when the reflection coefficient is indicated on the coordinates relating to the reflection coefficient and detected by the detection means. The area is set so that the boundary of the area is closer to the coordinate origin as the part is,
The outer region set in the region setting means is set so as to surround the inner region when the reflection coefficient is indicated on the coordinates related to the reflection coefficient,
The control object switching means is characterized in that, when the outer region and the inner region overlap, the overlapping region is handled as an inner region.

  Note that “high-frequency information detected at the output end of the high-frequency power supply device” means information related to the high frequency detected at the output end of the high-frequency power supply device, such as traveling wave, reflected wave, current, voltage, load impedance, or the detected information. This is information obtained based on the information, and the information from which the reflection coefficient can be obtained based on the information.

The high frequency power supply device provided by the second invention is
The inner region is an elliptical region .

The high frequency power supply device provided by the third invention is
In the high frequency power supply device that controls the power value of the high frequency power supplied to the load to be constant,
High-frequency output means as a supply source of high-frequency power;
Detecting means for detecting a traveling wave from the high-frequency output means toward the load and a reflected wave from the load side toward the high-frequency output means;
A power signal output means for obtaining a signal corresponding to the traveling wave power and a signal corresponding to the reflected wave power based on the traveling wave and the reflected wave detected by the detection means;
Load-side power calculating means for calculating a signal corresponding to load-side power by subtracting a signal corresponding to the reflected-wave power from a signal corresponding to the traveling-wave power;
A reflection coefficient calculating means for calculating a reflection coefficient based on the high frequency information detected at the output terminal of the high frequency power supply device;
A region setting means for setting a plurality of inner regions defined by a relational expression on a coordinate relating to a reflection coefficient and a plurality of outer regions paired with each of the inner regions;
When the reflection coefficient calculated by the reflection coefficient calculation means falls within any one of the plurality of inner areas, a signal corresponding to the load-side power is output as a power signal, and the reflection coefficient calculation If the computed reflection coefficient comes into further outside the outer region enters the outer region forming the inner region and the pair out one of the inner side region of said plurality of inner side area means, said traveling wave Control target switching means for outputting a signal corresponding to power as a power signal;
Control means for controlling the output of the high-frequency output means so that the power signal output from the controlled object switching means is equal to the signal corresponding to the output power setting value;
The area setting means includes at least
(1) When the reflection coefficient is indicated on the coordinates related to the reflection coefficient, the boundary of the region is such that the portion where the detection error amount of the reflected wave detected by the detection means is larger, including the coordinate origin of the coordinates. A first inner region whose region is set to be close to
(2) a first outer region in which a region is set so as to surround the first inner region when the reflection coefficient is indicated on coordinates relating to the reflection coefficient;
(3) When the reflection coefficient is indicated on the coordinates related to the reflection coefficient, the region is set such that the outermost boundary of the region is outside the boundary of the first inner region when viewed from the coordinate origin of the coordinates. A second inner region, and
(4) a second outer region in which an area is set so as to surround the second inner region when the reflection coefficient is indicated on coordinates related to the reflection coefficient;
Is set,
The control object switching means is characterized in that, when any one of the outer regions overlaps with any one of the inner regions, the overlapping region is handled as an inner region.

The high frequency power supply device provided by the fourth invention is
The first inner region is an elliptical region .

The high frequency power supply device provided by the fifth invention provides:
The coordinates for the reflection coefficient, is characterized in that the real number portion is u-axis, uv coordinates the imaginary part was v axis.

The control method of the high frequency power source provided by the sixth invention is:
In the control method of the high frequency power source that controls the power value of the high frequency power supplied to the load to be constant,
One inner region defined by a relational expression and one outer region paired with the inner region are set on the coordinates relating to the reflection coefficient,
The traveling wave power output based on the traveling wave and the reflected wave at the output end of the high frequency power supply and the reflected wave power reflected from the load side are detected, and the reflection coefficient is calculated based on the high frequency information at the output end of the high frequency power supply. ,
The computed reflection coefficient, when entering into the inner area performs control for the load-side power minus the reflected power from the traveling-wave power constant, the outer region exits said side region When entering and further outside the outer region, control to make the traveling wave power constant,
When the reflection coefficient is shown on the coordinates related to the reflection coefficient, the inner area includes the coordinate origin of the coordinate, and the boundary of the area is larger in the portion where the reflected wave detection error amount at the output end is larger. The area is set to be close to
The outer region is set so as to surround the inner region when the reflection coefficient is indicated on the coordinates related to the reflection coefficient,
When the outer region and the inner region overlap, the overlapping region is treated as an inner region.
The high frequency power supply control method provided by the seventh invention is:
The inner region is an elliptical region.

The control method of the high frequency power source provided by the eighth invention is:
In the control method of the high frequency power source that controls the power value of the high frequency power supplied to the load to be constant,
A plurality of inner regions defined by a relational expression on the coordinates relating to the reflection coefficient and a plurality of outer regions paired with each of the inner regions are set,
The traveling wave power output based on the traveling wave and the reflected wave at the output end of the high frequency power supply and the reflected wave power reflected from the load side are detected, and the reflection coefficient is calculated based on the high frequency information at the output end of the high frequency power supply. ,
When the calculated reflection coefficient falls within any one of the plurality of inner regions, the load side power obtained by subtracting the reflected wave power from the traveling wave power is controlled to be constant, if the further-out outer region enters the outer region forming the inner region and the pair out one of the inner side region of the inner side region performs control of the traveling wave power constant,
As the plurality of inner regions and the plurality of outer regions, at least,
(1) When the reflection coefficient is shown on the coordinates related to the reflection coefficient, the region boundary is closer to the coordinate origin as the reflected wave detection error amount at the output end is larger. A first inner region with the region set to be
(2) a first outer region in which a region is set so as to surround the first inner region when the reflection coefficient is indicated on coordinates relating to the reflection coefficient;
(3) When the reflection coefficient is indicated on the coordinates related to the reflection coefficient, the region is set such that the outermost boundary of the region is outside the boundary of the first inner region when viewed from the coordinate origin of the coordinates. A second inner region, and
(4) a second outer region in which an area is set so as to surround the second inner region when the reflection coefficient is indicated on coordinates related to the reflection coefficient;
Is set,
When any one of the outer regions overlaps with any one of the inner regions, the overlapping region is treated as an inner region.
The high frequency power supply control method provided by the ninth invention is
The first inner region is an elliptical region.

The control method of the high frequency power source provided by the tenth invention is
The coordinates for the reflection coefficient, is characterized in that the real number portion is u-axis, uv coordinates the imaginary part was v axis.

According to the first and third aspects, PF control can be performed in a region where the absolute value of the reflection coefficient is large, and PL control can be performed in a region where the absolute value of the reflection coefficient is small. Therefore, the absolute value of the reflection coefficient is large. An appropriate control method can be selected for each case. Therefore, the problem that occurs in the region where the absolute value | Γ | of the reflection coefficient is large and the problem that occurs in the region where the absolute value | Γ |

  That is, in the region where the absolute value of the reflection coefficient is large, PF control can be performed to avoid the hunting phenomenon in the plasma extinction phenomenon or the matching process. Also, in the region where the absolute value of the reflection coefficient is small, such as near the alignment target point, by performing PL control, it is possible to avoid the plasma extinction phenomenon and the hunting phenomenon in the alignment process. In other words, appropriate power control can be performed over a wide region of the absolute value | Γ | of the reflection coefficient.

  Furthermore, avoiding hunting and plasma extinction in the alignment process leads to a reduction in time required for alignment. When the time required for matching is reduced, the amount of operation of the driving part of the matching unit in the matching process is reduced as compared with the conventional case. Therefore, the consumption of the driving part of the matching unit is reduced and the life can be extended. For example, the wear of the sliding portion between the shaft and the bearing of the variable capacitor or variable inductor provided in the matching unit is reduced as compared with the conventional case, so that the life of the variable capacitor or variable inductor can be extended.

Furthermore, since the time required for alignment is reduced, the processing time of a workpiece such as a wafer or a liquid crystal substrate can be shortened. Therefore, not only productivity can be improved, but also power consumption can be reduced.
Furthermore, since the time required for matching is reduced, abnormal discharge that tends to occur in the non-matching state is also reduced. Therefore, damage to the wafer caused by abnormal discharge can be reduced.

Further, according to the third invention, even when a variable region exists, a region dedicated to the variable region can be provided to perform PL control suitable for the variable region. Therefore, the area provided so as to include the center (origin) of the polar chart can be determined without considering the existence of the fluctuation area. That is, even when the fluctuation area exists outside the area provided so as to include the center (origin) of the polar chart, it is not necessary to enlarge the area so that the fluctuation area is included. For this reason, it is possible to make the region to which PL control, which consumes more power than PF control, is applied as small as possible. Therefore, it is possible to reduce power consumption while selecting an appropriate control method.

In addition, since the setting data of the inner area and the outer area set by the area setting means is defined by a relational expression, an optimum area can be set according to the amount of error generated in the detection means . For example, where the amount of error generated in the detection means is large, a relational expression that brings the boundary value of the region closer to the center of the polar chart may be defined.

The effect of the sixth invention is the same as the effect of the first invention. The effect of the eighth invention is the same as the effect of the third invention.

According to the first and third inventions, the outer region is made larger than the inner region. Therefore, since the reference value for switching from PL control to PF control is different from the reference value for switching from PF control to PL control, even if the reflection coefficient fluctuates near the boundary of the inner region, PL control and PF control Thus, the so-called hunting phenomenon, in which the power supply is alternately and rapidly switched, can be prevented, and more stable power control can be performed.

In the first and third inventions, in order to determine how to control when the areas overlap, any one of the outer areas and any of the inner areas are defined. ing.

Further, the third invention shows that a plurality of areas need only be provided when the fluctuation area exists outside the area provided so as to include the center (origin) of the polar chart. That is, the area provided so as to include the center (origin) of the polar chart is always provided, and other areas may be provided as necessary. According to the third invention, as described above, the area provided so as to include the center (origin) of the polar chart can be determined without considering the existence of the fluctuation area.

In addition, since the amount of error generated by the detection means varies depending on the phase angle, it is possible to improve the accuracy by determining the region in consideration of the amount of error rather than the circular region having the center of the polar chart as the center of the circle. . Therefore, according to the first and third invention, as the partial detection error amount is large, by setting to be close to the boundary of the region in the coordinate origin, it is to improve the accuracy.

According to the second and fourth invention, since the elliptical region, taking into account the amount of error, Ru can be simplified simplification and control settings.

  Hereinafter, details of the present invention will be described with reference to the drawings.

  FIG. 1 is a block diagram showing a configuration of a high frequency power supply device 1 according to the present invention and a connection relationship among the high frequency power supply device 1, a matching unit 3 and a load 5.

The high-frequency power supply device 1 according to the present invention uses a transmission line 2 (for example, a coaxial cable), a matching unit 3 and a load connection unit 4 (for example, electromagnetic waves leak) as a high-frequency power (traveling wave power) output from an internal amplification unit 15. The power supply device supplies power to the load 5 via a copper plate shielded so as not to feed back the traveling wave power to be output or the load-side power to be controlled to be equal to the output power set value.
The high frequency power output from the amplifying unit 15 of the high frequency power supply device 1 and directed to the load 5 is referred to as traveling wave power PF. On the contrary, the high frequency power directed from the load side to the high frequency power supply device is referred to as reflected wave power PR. In general, this type of high frequency power supply device outputs high frequency power having a frequency of several hundred kHz or more.

  The ON / OFF control unit 11 receives an ON signal or an OFF signal, and controls so that a high-frequency signal is output from a crystal oscillator 12 described later when the ON signal is input.

  The crystal oscillator 12 outputs a high-frequency signal having a predetermined frequency. The high-frequency signal referred to here is not limited to a sine wave signal, and includes, for example, a rectangular wave signal, a triangular wave signal, and the like.

  The voltage-controlled variable attenuator 13 receives a high-frequency signal output from the crystal oscillator 12 and an output signal from an error amplifier 26 described later, and uses the output signal from the error amplifier 26 as a command value to change the level of the input high-frequency signal. To output. Thereby, the output level of the traveling wave power output from the amplifying unit 15 is varied.

  The AC / DC conversion unit 14 converts AC power supplied from a commercial power source into DC power and supplies the DC power to the amplification unit 15 described later.

  The amplifier 15 receives the DC power supplied from the AC / DC converter 14 and the high frequency signal output from the voltage controlled variable attenuator 13 and amplifies the high frequency signal output from the voltage controlled variable attenuator 13. Traveling wave power PF is output. Therefore, the output level of the amplifying unit 15 is controlled by the output of the error amplifier 26 described later. The traveling wave power output from the amplifying unit 15 is output to the outside of the high frequency power supply device via the transmission line 2 connected to the high frequency power output connector 101 as an output end of the high frequency power supply device.

  Note that the portion constituted by the amplifier 15, the ON / OFF controller 11, the crystal oscillator 12, the voltage controlled variable attenuator 13, and the AC / DC converter 14 is an example of the high-frequency output means of the present invention.

  The power detection unit 16 is provided at the output end of the high-frequency power supply device 1 (on the transmission line between the amplification unit 15 and a current / voltage detection unit 20 described later in the drawing) and travels from the amplification unit 15 toward the load. In addition, the reflected wave returning from the outside of the high frequency power supply device is attenuated and extracted, and the attenuated traveling wave signal RFpf and the attenuated reflected wave signal RFpr are output. For example, a directional coupler is used for the power detection unit 16. The power detection unit 16 is an example of the detection means of the present invention.

  The output terminal of the high frequency power supply device 1 is substantially equivalent not only to the output terminal 101 in the drawing, but also between the amplification unit 15 and the output terminal 101 of the high frequency power supply device, and also on the transmission line 2. In this specification, these positions are collectively referred to as an output end of the high-frequency power supply device.

  The power level converter 17 converts the attenuated traveling wave signal RFpf and the attenuated reflected wave signal RFpr output from the power detection unit 16 into a DC signal, and then normalizes the converted signal to the traveling wave power signal Vpf corresponding to the traveling wave power and the reflection. The reflected wave power signal Vpr corresponding to the wave power is output. The power level converter 17 is an example of the power signal output means of the present invention.

The load side power calculation unit 18 obtains and outputs a load side power signal Vpl corresponding to the load side power by subtracting the reflected wave power signal Vpr from the traveling wave power signal Vpf. When the above relationship is expressed by an arithmetic expression, Expression (4) is obtained. The load side power calculation unit 18 is an example of the load side power calculation means of the present invention.
Vpl = Vpf−Vpr (4)

  The current voltage detection unit 20 is provided at the output end of the high frequency power supply device 1 (on the transmission line between the power detection unit 16 and the output end 101 of the high frequency power supply device in the drawing), and the high frequency current I and the high frequency voltage E. Are detected and output. Since the load side impedance ZL ′ viewed from the load side from the current / voltage detection unit 20 and the load side impedance ZL viewed from the input end of the matching unit 3 are the same or substantially the same, practically ZL = ZL ' In addition, the impedance when the load side is viewed from the power detection unit 16 is also referred to as a load side impedance ZL ′.

FIG. 2 shows a circuit configuration example of the current / voltage detector 20.
In FIG. 2, the current / voltage detection unit 20 includes a current detection unit 202 that detects current and a voltage detection unit 203 that detects voltage.

  The current detection unit 202 includes a current transformer CT provided between the power detection unit 16 and the output terminal 101 of the high frequency power supply device, and a resistor R1 provided between both ends of the secondary coil of the current transformer CT. ing. Therefore, the voltage corresponding to the current on the transmission line 201 between the power detection unit 16 and the output terminal 101 of the high-frequency power supply device is applied between one end (ground side) and the other end (opposite side of the ground) of the resistor R1. Occur between. Then, the voltage at the other end of the resistor R1 is detected, and the detected voltage is output as a signal corresponding to the high-frequency current I. For convenience, this signal is a high-frequency current I.

  The voltage detection unit 203 is provided between the transmission line between the power detection unit 16 and the output terminal 101 of the high-frequency power supply device and the ground (GND), and includes a capacitor C1 and a capacitor C2. Therefore, the voltage on the transmission line 201 between the power detection unit 16 and the output terminal 101 of the high frequency power supply device is divided by the capacitor. Then, the divided voltage is output as a signal corresponding to the high-frequency voltage E from the connection point between the capacitor C1 and the capacitor C2. For convenience, this signal is a high-frequency voltage E.

  In the example shown in FIG. 2, the voltage detection unit 203 is configured using two capacitors. However, the configuration is not limited to this configuration, and a configuration using resistors, inductors, and transformers (transformers). It may be.

Returning again to FIG.
The load impedance calculation unit 21 calculates the load side impedance ZL ′ when the load side is viewed from the current voltage detection unit 20 using the high frequency current I and the high frequency voltage E output from the current voltage detection unit 20, and outputs the calculation result. To do. As a specific calculation method, first, an absolute value | I | of the high-frequency current I, an absolute value | E | of the high-frequency voltage E, and a phase difference θie between the high-frequency current I and the high-frequency voltage E are obtained. ) To calculate the absolute value | ZL ′ | of the load side impedance ZL ′. Note that the high-frequency current I and the high-frequency voltage E are divided in the current-voltage detection unit 20, and are calculated in consideration thereof.
| ZL '| = | ZL | = | V | / | I | (5)

Next, assuming that the resistance component of the load side impedance ZL ′ is Ri, the reactance component is Xi, and Z1 ′ = Ri + jXi, Ri and Xi are given by equations (6) and (7).
Ri = | ZL ′ | · cos (θie) = | ZL | · cos (θie) (6)
Xi = | ZL ′ | · sin (θie) = | ZL | · sin (θie) (7)

The reflection coefficient calculation unit 22 represents the reflection coefficient Γ in the form of “real part + imaginary part” from the resistance component Ri and reactance component Xi of the load side impedance ZL ′ output from the load impedance calculation unit 21. The real part u1 and the imaginary part v1 are calculated. Specifically, the reflection coefficient Γ is expressed by Expression (8), where the characteristic impedance is Z0. Therefore, by solving the equation (8), it is possible to calculate the real part u1 and the imaginary part v1 when the reflection coefficient Γ is expressed in the form of the expression (9).
Γ = (ZL′−Z0) / (ZL ′ + Z0)
= (ZL-Z0) / (ZL + Z0) (8)
Γ = u1 + jv1 (9)
Thus, when the reflection coefficient Γ is expressed in the form of “real part + imaginary part”, the reflection coefficient Γ is expressed on the uv coordinates with the real part being the u axis (horizontal axis) and the imaginary part being the v axis (vertical axis). Can be represented. That is, it is possible to display the uv coordinates on the polar chart with the origin of the uv coordinates (u1 = 0, v1 = 0) as the center of the polar chart. In addition, when Expression (8) is modified, an expression for obtaining the load side impedance ZL ′ from the reflection coefficient Γ and the characteristic impedance Z0 is obtained as shown in Expression (10). The reflection coefficient calculation unit 22 is an example of the reflection coefficient calculation means of the present invention.
ZL ′ = ZL = Z0 ((1 + Γ) / (1-Γ)) (10)

The region setting unit 23 sets a relational expression that defines a region serving as a reference for determining whether to perform PF control or PL control using the reflection coefficient Γ. This region is defined by using a relational expression on the uv coordinates, representing the reflection coefficient Γ in the form of “real part + imaginary part”. The region setting unit 23 is an example of a region setting unit according to the present invention.

  FIG. 3 shows an example of area setting data set by the area setting unit 23. FIG. 4 shows the relational expression uv coordinates set in FIG.

  In FIG. 3, the relational expression of the inner area is a relational expression that defines a boundary of the inner area described later, and the inner area indicates an area inside the boundary defined by this relational expression. The relational expression of the outer area is a relational expression that defines the boundary of the outer area, which will be described later, and the outer area excludes the inner inner area from the area inside the boundary defined by this relational expression. In other words, the outer region is provided so as to surround the inner region. (The idea is that the outer region includes the inner region. However, as described later, when the outer region and the inner region overlap, the overlapping region is the inner region. (Because it is an area, the outer area substantially excludes the inner area inside as described above).

  3 and 4 show three regions, region 1 to region 3, the number of regions is not limited. In addition, since the area 1 is a basic area for determining which control method, PL control or PF control, is always provided. However, the fluctuation region may not exist. In this case, regions such as region 2 to region 3 corresponding to the variable region are not set by the region setting unit 23. Further, the size of the area is an example, and in practice, an appropriate value is set according to the situation.

  Incidentally, as described above, the vicinity of the center of the polar chart (near the origin of the uv coordinates) and the fluctuation region are regions suitable for PL control. Therefore, when the reflection coefficient Γ (specifically, the reflection coefficient Γ output from the reflection coefficient calculation unit 22) enters these regions, the region near the center (including the center) of the polar chart is used to perform PL control. A special area called “inner area” is provided in the area corresponding to the fluctuation area. Basically, PL control is performed when the reflection coefficient Γ enters the inner region, and PF control is performed when the reflection coefficient Γ exits the inner region. However, when the reflection coefficient Γ fluctuates in the vicinity of the boundary of the inner region, a so-called hunting phenomenon occurs in which the PL control and the PF control are alternately switched rapidly.

  Therefore, in order to prevent this hunting phenomenon, a special region called “outer region” is provided outside the inner region, and the reflection coefficient Γ is from outside the inner region of any region set by the region setting unit 23. PL control is performed when entering the inside, and PF control is performed when the reflection coefficient Γ exits from the outside of one of the regions set by the region setting unit 23. When the control is performed in this manner, the switching reference value from the PL control to the PF control is different from the switching reference value from the PF control to the PL control, so that the hunting phenomenon can be prevented.

Next, a method for setting the inner region and the outer region will be described.
First, the region 1 near the center of the polar chart (near the origin of the uv coordinates) will be described.
As described with reference to FIGS. 9 and 10, PF control is preferable when the absolute value of the reflection coefficient is large, and PL control is preferable when the absolute value of the reflection coefficient is small, from the relationship of the amount of error generated in the detection unit. For this purpose, the reference reflection coefficient absolute value is generally determined, an area having a smaller reflection coefficient absolute value than the reference reflection coefficient absolute value is set as an inner area, and an outer area is set outside the area. That is, the inner region and the outer region may be circular regions with the center of the polar chart as the center of the circle. In this case, the region 1 may be set by a relational expression representing a circle.

However, as described with reference to FIG. 9, since the amount of error generated in the detection unit varies depending on the phase angle, it is better to determine the region in consideration of the amount of error than the circular region having the center of the polar chart as the center of the circle. preferable. For this reason, when the amount of error generated in the detection unit is as shown in FIG. 9, it is preferable to use a substantially elliptical region such as region 1 in FIG. 4 rather than a circular region.
That is, in the case of FIG. 9, the error amount is maximum when the phase angle is about 135 degrees. Therefore, as shown in FIG. In order to make it less susceptible to the influence of the amount, the absolute value of the reflection coefficient is deformed in a direction smaller than that of other portions. In other portions, the boundary of the region may be deformed according to the amount of error that occurs, but the amount of error that occurs varies depending on the phase angle and is not uniform. Therefore, the region 1 has a distorted circular shape. However, in order to simplify the control, the region 1 is an elliptical region having a similar shape.
Further, when the region 1 is elliptical as described above, the center of the ellipse deviates from the center of the polar chart due to the magnitude relationship of the error amount. Therefore, as shown in FIGS. 3 and 4, the center of the ellipse of the region 1 is deviated from the origin of the uv coordinates. Of course, a circular region may be adopted rather than an elliptical region for the purpose of simplifying the setting and simplifying the control.

Next, a method for setting the inner region and the outer region relating to the variable region will be described using the region 2 shown in FIG. 4 as an example.
As described above, the inner region of the region 2 is a region corresponding to the variable region. By setting in this way, as described above, when the reflection coefficient Γ output from the reflection coefficient calculation unit 22 enters the inner region, it can be regarded as having entered the fluctuation region. It is possible to determine whether or not. However, the variable region is not clearly defined, and varies depending on the actual operation and the situation during the experiment. For this reason, the setting of the fluctuation region must be performed by roughly estimating the region. In other words, the variable region cannot be set with high accuracy.

  Therefore, it is more effective to improve the convenience of setting the inner region corresponding to the variable region by simplifying the setting and making the control easier, instead of increasing the accuracy. Therefore, as shown in FIG. 4, when the area set on the uv coordinates is a rectangle, it is easy to determine whether or not the reflection coefficient Γ output from the reflection coefficient calculation unit 22 has entered the inner area and is set. Is preferred because of its simplicity. Of course, like the region 1, the variable regions such as the region 2 and the region 3 may be regions having other shapes such as a circular region and an elliptical region.

  Note that when there are variable regions such as the region 2 and the region 3, the region 1 may be set so as to include these variable regions. That is, the area 1 may be larger than those shown in FIGS. This simplifies setting and simplifies control. However, since the PL control consumes more power than the PF control, from the viewpoint of power saving, when the fluctuation region exists, as shown in FIG. 3 and FIG. It is preferable to provide an area and make the area to which PL control, which consumes more power than PF control, is applied as small as possible.

  Next, a case where a plurality of regions overlap each other as in region 1 and region 3 shown in FIGS. 3 and 4 will be described. As described above, the vicinity of the center of the polar chart and the fluctuation region are regions suitable for PL control. In the example shown in FIGS. 3 and 4, the inner region of region 1 is suitable for PL control because it is near the center of the polar chart, and the inner region of region 2 and region 3 corresponds to the variable region. Suitable for PL control. When each region exists independently without overlapping, it may be controlled by the method described above, but when a plurality of regions overlap, such as region 1 and region 3 shown in FIGS. The outer region of another region enters into the inner region of a certain region.

  In such a case, it is not possible to determine which one of PL control and PF control should be selected by the above-described method. Therefore, when the inner region and the outer region overlap in this way, they may be regarded as the inner region.

  FIG. 5 is a diagram illustrating an inner region and an outer region when a plurality of regions overlap. In FIG. 5, the above concept is applied to FIG. 4, but scales and the like are omitted. In this way, the two regions of region 1 and region 3 shown in FIG. 4 may be merged into a single region.

Returning again to FIG.
The region determination unit 24 selects a suitable control method according to the real part u1 and the imaginary part v1 of the reflection coefficient Γ output from the reflection coefficient calculation unit 22. Specifically, when the reflection coefficient Γ output from the reflection coefficient calculation unit 22 enters from the inside of one of the regions set by the region setting unit 23 to the switch circuit 25 described later, Command to output the load side power signal Vpl used for the PL control. In addition, when the reflection coefficient Γ goes out from the outside of one of the regions set by the region setting unit 23, the traveling wave power signal Vpf used for PF control is output to the switch circuit 25. To command.

In addition, what is necessary is just to do as follows at the time of the output start of the high frequency electric power which is an initial state, for example.
(A) When the reflection coefficient Γ is in the inner region of any region set by the region setting unit 23, the switch circuit 25 is instructed to output the load-side power signal Vpl.
(B) The traveling wave power signal Vpf is output to the switch circuit 25 when the reflection coefficient Γ is outside the inner region of any region set by the region setting unit 23. To

  The switch circuit 25 receives the traveling wave power signal Vpf output from the power level converter 17 and the load side power signal Vpl output from the load side power calculation unit 18 and proceeds based on a command from the region determination unit 24. The internal switch is switched to output either the wave power signal Vpf or the load side power signal Vpl. Note that the portion configured by the region determination unit 24 and the switch circuit 25 is an example of the control target switching means of the present invention.

The error amplifier 26 includes a traveling wave power signal Vpf output from the power level converter 17 or a load power signal Vpl output from the load power calculator 18 and an output power setting signal Vset corresponding to the output power setting value. And the power control signal Vpc serving as a command value is output to the voltage controlled variable attenuator 13 so that the magnitude of the traveling wave power signal Vpf or the load side power signal Vpl is equal to the magnitude of the output power setting signal Vset. . When the above relationship is expressed by an arithmetic expression, Expression (11) or Expression (12) is obtained. The error amplifier 26 is an example of the control means of the present invention.
Vpc∝ (Vset−Vpf) (11)
Vpc∝ (Vset−Vpl) (12)

  As described above, the voltage controlled variable attenuator 13 receives a high frequency signal output from the crystal oscillator 12 and an output signal of an error amplifier 26 described later as an input, and an output signal of the error amplifier 26 as a command value. Since the signal level is varied and output, the output of the amplifying unit 15 can be controlled.

  Next, in the system including the high frequency power supply device according to the present invention, the operation after high frequency power is output from the high frequency power supply device will be described with reference to FIG.

  FIG. 6 is a diagram showing an example of a change in reflection coefficient on the polar chart in the matching operation process of the system including the high-frequency power supply device according to the present invention. FIG. 6 shows the change of the reflection coefficient on the polar chart described in FIG.

(1) P1: in FIG.
In the system including the above-described high-frequency power supply device, discharge is started when high-frequency power is supplied from the high-frequency power supply device to the load 5 to enable discharge. At this stage, since there is no matching, the reflected wave power is large and almost totally reflected. That is, the absolute value | Γ | ≈1 of the reflection coefficient Γ. Since the reflection coefficient at this time is outside the entire region, it is controlled by PF control. In FIG. 6, since only the vicinity of the center of the polar chart is enlarged, the original position of P1 is at another position. In that sense, a line from P1 to P2 is represented by a dotted line.

(2) P2 in FIG. 6:
The matching unit 3 performs a matching operation so as to reduce the reflected wave power. In the process, the position indicated by P2 is reached, but the reflection coefficient at this point is outside the entire region, and is controlled by PF control.

(3) P3 in FIG. 6:
The matching unit performs a matching operation to further reduce the reflected wave power. In the process, it reaches the position indicated by P3 and enters the outer region of the region 2, but at this point, it remains controlled by the PF control.

(4) P4 in FIG. 6:
The matching unit performs a matching operation to further reduce the reflected wave power. In the process, when the position indicated by P4 is reached and the inside of the area 2 is entered, the control is switched to the PL control.

(5) P5 in FIG. 6:
The matching unit performs a matching operation to further reduce the reflected wave power. In the process, it reaches the position indicated by P5 and goes out of the inner area of the area 2, but at this time, it remains controlled by the PL control.

(6) P6 in FIG. 6:
The matching unit performs a matching operation to further reduce the reflected wave power. In the process, when it reaches the position indicated by P6 and goes outside the area 2 outside the area 2, it switches to PF control.

(7) P7 in FIG. 6:
The matching unit performs a matching operation to further reduce the reflected wave power. In the process, it reaches the position indicated by P7 and enters the outer region of the region 1, but at this time, it remains controlled by the PF control.

(8) P8 in FIG. 6:
The matching unit performs a matching operation to further reduce the reflected wave power. In the process, when it reaches the position indicated by P8 and enters the inner region of region 1, it switches to PL control. Thereafter, the matching unit performs a matching operation so as to further reduce the reflected wave power. As a result, the alignment target point is finally reached and the alignment state is reached.

(9) P9 in FIG. 6:
While the PL control is performed during the matching state, if the load condition (power supply amount or the like) changes, the load impedance may change and reflection may increase. In this case, it reaches the position indicated by P9 and goes out of the inner area of the area 1, but at this time, it remains controlled by the PL control. Note that at the position indicated by P9, as described in FIG. 5, it is considered that the two areas of the area 1 and the area 3 are combined into one area. The position shown is a position that has moved from the inner area of the area 3 to the outer area.

(10) P10 in FIG. 6:
When the reflected wave power further increases and reaches the position indicated by P10 and goes out of the area outside the area 3, the control is switched to PF control.

(11) P11 in FIG. 6:
Further, the reflected wave power increases and reaches the position indicated by P11. Since the reflection coefficient at this point is outside the entire region, it remains controlled by the PF control.

(12) The subsequent operations are the same as the control described so far, and the matching unit performs the matching operation so as to reduce the reflected wave power. In the process, when the reflection coefficient enters any region, either PL control or PF control is selected according to the state.

  By switching the control method by the method described so far, it is possible to prevent the plasma extinction phenomenon or the hunting phenomenon in the matching process from occurring over a wide area of the absolute value | Γ | of the reflection coefficient. Further, even when there is a fluctuation region, it is possible to prevent an increase in excessive power consumption.

  In addition, this invention is not limited to said embodiment. For example, in the above embodiment, as a specific example of the reflection coefficient, the reflection coefficient Γ output from the reflection coefficient calculation unit 22 is used. However, as shown in Expression (1), the attenuated traveling wave signal RFpf and the attenuated reflected wave signal RFpr are used. You may ask for it.

  In addition, when there are a plurality of variable regions, a plurality of regions corresponding to the plurality of variable regions are necessary as the regions set in the region setting unit 23. However, if there are multiple fluctuating areas adjacent to each other or if there are multiple fluctuating areas close to each other, for the purpose of simplifying the settings and simplifying the control, A certain variable region may be regarded as one, and the regions set in the region setting unit 23 may be combined into one.

  Further, the reflection coefficient Γ and the load side impedance ZL may be input from an external device such as the matching unit 3.

FIG. 1 is a block diagram showing a configuration of a high frequency power supply device 1 according to the present invention and a connection relationship among the high frequency power supply device 1, a matching unit 3 and a load 5. FIG. 2 shows a circuit configuration example of the current / voltage detector 20. FIG. 3 shows an example of setting data for the area set by the area setting unit. FIG. 4 shows the area set in FIG. 3 on the polar chart. FIG. 5 is a diagram illustrating an inner region and an outer region when a plurality of regions overlap. FIG. 6 is a diagram showing an example of a change in the reflection coefficient on the polar chart in the matching operation process of the system including the high-frequency power supply device according to the present invention. FIG. 7 is a block diagram showing a configuration of a conventional high-frequency power supply device 50 using PF control and a connection relationship among the high-frequency power supply device 50, the matching unit 3, and the load 5. FIG. 8 is a block diagram showing a configuration of a conventional high-frequency power supply device 51 using PL control and a connection relationship between the high-frequency power supply device 51, the matching unit 3, and the load 5. FIG. 9 shows an ideal reflection coefficient calculation value (becomes a perfect circle) obtained when an “ideal power detection unit” (a power detection unit with no error in the detection value) is used when performing PF control. FIG. 8 is a diagram showing a comparison of reflection coefficient calculation values calculated from reflected wave power detection values of the “real power detection unit” on a polar chart. FIG. 10 is a diagram illustrating reflected wave power detection characteristics in the vicinity of the matching target point of the power detection unit 16. FIG. 11 is a diagram showing the intermediate region on the polar chart. FIG. 12 is a diagram showing an example of the fluctuation region on the polar chart. FIG. 13 is a diagram illustrating the absolute value of the reference reflection coefficient when the variation region 92 illustrated in FIG. 12 exists.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 High frequency power supply device 2 which concerns on this invention Transmission line 3 Matching device 4 Load connection part 5 Load 11 ON / OFF control part 12 Crystal oscillator 13 Voltage control variable attenuator 14 AC / DC conversion part 15 Amplification part 16 Power detection part 17 Electric power Level converter 18 Load side power calculation unit 20 Current voltage detection unit 202 Current detection unit 203 Voltage detection unit 21 Load impedance calculation unit 22 Reflection coefficient calculation unit 23 Region setting unit 24 Region determination unit 25 Switch circuit 26 Error amplifier

Claims (10)

In the high frequency power supply device that controls the power value of the high frequency power supplied to the load to be constant,
High-frequency output means as a supply source of high-frequency power;
Detecting means for detecting a traveling wave from the high-frequency output means toward the load and a reflected wave from the load side toward the high-frequency output means;
A power signal output means for obtaining a signal corresponding to the traveling wave power and a signal corresponding to the reflected wave power based on the traveling wave and the reflected wave detected by the detection means;
Load-side power calculating means for calculating a signal corresponding to load-side power by subtracting a signal corresponding to the reflected-wave power from a signal corresponding to the traveling-wave power;
A reflection coefficient calculating means for calculating a reflection coefficient based on the high frequency information detected at the output terminal of the high frequency power supply device;
A region setting means for setting one inner region defined by a relational expression on the coordinates relating to the reflection coefficient and one outer region paired with the inner region;
When the reflection coefficient calculated by the reflection coefficient calculation means enters the inner region, a signal corresponding to the load side power is output as a power signal, and the reflection coefficient calculated by the reflection coefficient calculation means is when exiting the further outside the outer region enters the outer region exits said side region, the control target switching means for outputting a signal corresponding to the traveling wave power as a power signal,
Control means for controlling the output of the high-frequency output means so that the power signal output from the controlled object switching means is equal to the signal corresponding to the output power setting value;
The inner region set in the region setting means has a large detection error amount of the reflected wave so as to include the coordinate origin of the coordinates when the reflection coefficient is indicated on the coordinates relating to the reflection coefficient and detected by the detection means. The area is set so that the boundary of the area is closer to the coordinate origin as the part is,
The outer region set in the region setting means is set so as to surround the inner region when the reflection coefficient is indicated on the coordinates related to the reflection coefficient,
The control object switching means, when the outer region and the inner region overlap, treats the overlapping region as an inner region.   The high frequency power supply device according to claim 1, wherein the inner region is an elliptical region. In the high frequency power supply device that controls the power value of the high frequency power supplied to the load to be constant,
High-frequency output means as a supply source of high-frequency power;
Detecting means for detecting a traveling wave from the high-frequency output means toward the load and a reflected wave from the load side toward the high-frequency output means;
A power signal output means for obtaining a signal corresponding to the traveling wave power and a signal corresponding to the reflected wave power based on the traveling wave and the reflected wave detected by the detection means;
Load-side power calculating means for calculating a signal corresponding to load-side power by subtracting a signal corresponding to the reflected-wave power from a signal corresponding to the traveling-wave power;
A reflection coefficient calculating means for calculating a reflection coefficient based on the high frequency information detected at the output terminal of the high frequency power supply device;
A region setting means for setting a plurality of inner regions defined by a relational expression on a coordinate relating to a reflection coefficient and a plurality of outer regions paired with each of the inner regions;
When the reflection coefficient calculated by the reflection coefficient calculation means falls within any one of the plurality of inner areas, a signal corresponding to the load-side power is output as a power signal, and the reflection coefficient calculation If the computed reflection coefficient comes into further outside the outer region enters the outer region forming the inner region and the pair out one of the inner side region of said plurality of inner side area means, said traveling wave Control target switching means for outputting a signal corresponding to power as a power signal;
Control means for controlling the output of the high-frequency output means so that the power signal output from the controlled object switching means is equal to the signal corresponding to the output power setting value;
The area setting means includes at least
(1) When the reflection coefficient is indicated on the coordinates related to the reflection coefficient, the boundary of the region is such that the portion where the detection error amount of the reflected wave detected by the detection means is larger, including the coordinate origin of the coordinates. A first inner region whose region is set to be close to
(2) a first outer region in which a region is set so as to surround the first inner region when the reflection coefficient is indicated on coordinates relating to the reflection coefficient;
(3) When the reflection coefficient is indicated on the coordinates related to the reflection coefficient, the region is set such that the outermost boundary of the region is outside the boundary of the first inner region when viewed from the coordinate origin of the coordinates. A second inner region, and
(4) a second outer region in which an area is set so as to surround the second inner region when the reflection coefficient is indicated on coordinates related to the reflection coefficient;
Is set,
The control object switching means, when any one of the outer regions and any one of the inner regions overlaps, treats the overlapping region as an inner region.   The high frequency power supply device according to claim 3, wherein the first inner region is an elliptical region. The coordinates for the reflection coefficient is high-frequency power supply device according to any one of claims 1 to 4, characterized in that the real number portion is u-axis, uv coordinates the imaginary part was v axis. In the control method of the high frequency power source that controls the power value of the high frequency power supplied to the load to be constant,
One inner region defined by a relational expression and one outer region paired with the inner region are set on the coordinates relating to the reflection coefficient,
The traveling wave power output based on the traveling wave and the reflected wave at the output end of the high frequency power supply and the reflected wave power reflected from the load side are detected, and the reflection coefficient is calculated based on the high frequency information at the output end of the high frequency power supply. ,
The computed reflection coefficient, when entering into the inner area performs control for the load-side power minus the reflected power from the traveling-wave power constant, the outer region exits said side region When entering and further outside the outer region, control to make the traveling wave power constant,
When the reflection coefficient is shown on the coordinates related to the reflection coefficient, the inner area includes the coordinate origin of the coordinate, and the boundary of the area is larger in the portion where the reflected wave detection error amount at the output end is larger. The area is set to be close to
The outer region is set so as to surround the inner region when the reflection coefficient is indicated on the coordinates related to the reflection coefficient,
When the outer region and the inner region overlap each other, the overlapping region is handled as an inner region.   The method for controlling a high frequency power supply according to claim 6, wherein the inner region is an elliptical region. In the control method of the high frequency power source that controls the power value of the high frequency power supplied to the load to be constant,
A plurality of inner regions defined by a relational expression on the coordinates relating to the reflection coefficient and a plurality of outer regions paired with each of the inner regions are set,
The traveling wave power output based on the traveling wave and the reflected wave at the output end of the high frequency power supply and the reflected wave power reflected from the load side are detected, and the reflection coefficient is calculated based on the high frequency information at the output end of the high frequency power supply. ,
When the calculated reflection coefficient falls within any one of the plurality of inner regions, the load side power obtained by subtracting the reflected wave power from the traveling wave power is controlled to be constant, if the further-out outer region enters the outer region forming the inner region and the pair out one of the inner side region of the inner side region performs control of the traveling wave power constant,
As the plurality of inner regions and the plurality of outer regions, at least,
(1) When the reflection coefficient is shown on the coordinates related to the reflection coefficient, the region boundary is closer to the coordinate origin as the reflected wave detection error amount at the output end is larger. A first inner region with the region set to be
(2) a first outer region in which a region is set so as to surround the first inner region when the reflection coefficient is indicated on coordinates relating to the reflection coefficient;
(3) When the reflection coefficient is indicated on the coordinates related to the reflection coefficient, the region is set such that the outermost boundary of the region is outside the boundary of the first inner region when viewed from the coordinate origin of the coordinates. A second inner region, and
(4) a second outer region in which an area is set so as to surround the second inner region when the reflection coefficient is indicated on coordinates related to the reflection coefficient;
Is set,
A method of controlling a high-frequency power source, wherein when any one of the outer regions overlaps with any one of the inner regions, the overlapping region is treated as an inner region.   The method of controlling a high frequency power supply according to claim 8, wherein the first inner region is an elliptical region. The coordinates are related to the reflection coefficient, the control method of the high-frequency power source according to any one of claims 6-9, characterized in that the real number portion is u-axis, uv coordinates the imaginary part was v axis.

Images