JP3904464B2 - High voltage pulse generator for pulse laser - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention provides a high voltage generator for a pulse laser having a regenerative circuit for using reflected energy generated by impedance mismatch as energy for generating a next pulse in a high voltage generator used in an exposure gas laser apparatus or the like. The present invention relates to a voltage pulse generator.
[0002]
[Prior art]
With the miniaturization and high integration of semiconductor integrated circuits, improvement in resolving power is demanded in the projection exposure apparatus for production. For this reason, the wavelength of the exposure light emitted from the exposure light source is being shortened, and a KrF excimer laser device having a wavelength of 248 nm from a conventional mercury lamp is used as a light source for semiconductor exposure. Furthermore, as a next-generation light source for semiconductor exposure, an ArF excimer laser device having a wavelength of 193 nm and fluorine (F) having a wavelength of 157 nm are used. ₂ ) Gas laser devices that emit ultraviolet rays, such as laser devices, are promising.
In the KrF excimer laser device, fluorine (F ₂ ) Gas, krypton (Kr) gas, and mixed gas composed of noble gas such as neon (Ne) as buffer gas, and in an ArF excimer laser device, fluorine (F ₂ ) Gas, argon (Ar) gas, and mixed gas composed of noble gas such as neon (Ne) as buffer gas, fluorine (F ₂ ) In laser equipment, fluorine (F ₂ ) A laser gas, which is a mixed gas composed of a rare gas such as helium (He) as a gas and a buffer gas, generates a discharge inside a laser chamber sealed at several hundred kPa, thereby exciting the laser gas as a laser medium. .
[0003]
Inside the laser chamber, a pair of main discharge electrodes for exciting the laser gas are disposed facing each other at a predetermined distance in a direction perpendicular to the laser oscillation direction.
A high voltage pulse is applied to the pair of main discharge electrodes, and when the voltage applied between the main discharge electrodes reaches a certain value (breakdown voltage), the laser gas between the main discharge electrodes breaks down and main discharge starts. The laser medium is excited by this main discharge.
Therefore, such an exposure gas laser apparatus performs pulse oscillation by repeating main discharge, and the emitted laser light becomes pulse light. At present, the repetition frequency of the laser pulse of the laser apparatus used for exposure is about 2 kHz. However, in recent years, a repetition frequency of 4 kHz or more has been demanded in order to increase throughput and decrease variations in exposure amount.
[0004]
FIG. 6 shows an example of a high voltage pulse generator (hereinafter sometimes referred to as a discharge circuit) for generating a discharge in the laser chamber and exciting the laser gas in the gas laser apparatus for exposure as described above.
The discharge circuit of FIG. 6 is composed of a two-stage magnetic pulse compression circuit using three magnetic switches SR1, SR2 and SR3 composed of saturable reactors. The magnetic switch SR1 is a magnetic switch for reducing a switching loss in the solid-state switch SW which is a semiconductor switching element such as an IGBT, and is usually called magnetic assist (hereinafter referred to as magnetic assist).
The first magnetic switch SR2 and the second magnetic switch SR3 constitute a two-stage magnetic pulse compression circuit.
The circuit configuration and operation will be described below with reference to FIG.
First, the voltage of the high voltage power supply HV is adjusted to a predetermined value Vin, and the main capacitor C0 is charged. At this time, the solid switch SW is turned off. When the charging of the main capacitor C0 is completed and the solid switch SW is turned on, the voltage applied to both ends of the solid switch SW is mainly applied to both ends of the magnetic assist SR1.
When the time integral value of the charging voltage V0 of the main capacitor C0 applied to both ends of the magnetic assist SR1 reaches a limit value determined by the characteristics of the magnetic assist SR1, the magnetic assist SR1 is saturated and the magnetic switch is turned on, the main capacitor C0, the magnetic assist A current flows through the loop of SR1, the inductance LL, the primary side of the step-up transformer Tr1, and the solid switch SW. At the same time, a current flows through the secondary side of the step-up transformer Tr1 and the loop of the capacitor C1, and the charge stored in the main capacitor C0 is transferred to be charged in the capacitor C1.
Here, a combination of the inductance of the circuit loop and the parasitic inductance of the main capacitor C0 is represented as an inductance LL.
[0005]
Thereafter, when the time integral value of the voltage V1 in the capacitor C1 reaches a limit value determined by the characteristics of the magnetic switch SR2, the magnetic switch SR2 is saturated and the magnetic switch enters, and the capacitor C1, the capacitor C2, and the magnetic switch SR3 enter the loop. A current flows, and the charge stored in the capacitor C1 is transferred to charge the capacitor C2.
Thereafter, when the time integral value of the voltage V2 in the capacitor C2 reaches a limit value determined by the characteristics of the magnetic switch SR3, the magnetic switch SR3 is saturated and the magnetic switch is turned on, and the capacitors C2, the peaking capacitor Cp, and the magnetic switch SR3 A current flows through the loop, and the charge stored in the capacitor C2 is transferred to charge the peaking capacitor Cp.
[0006]
Corona discharge for preionization occurs on the outer peripheral surface of the dielectric tube 12 starting from the point where the dielectric tube 12 in which the first electrode 11 is inserted and the second electrode 13 are in contact with each other. As the charging of the capacitor Cp proceeds, the voltage Vp increases. When Vp reaches a predetermined voltage, corona discharge is generated on the surface of the dielectric tube 12 in the corona preionization part. By this corona discharge, ultraviolet rays 6 are generated on the surface of the dielectric tube 12, and the laser gas 2 as the laser medium between the main discharge electrodes E and E is preionized.
As the charging of the peaking capacitor Cp further proceeds, the voltage Vp of the peaking capacitor Cp increases. When this voltage Vp reaches a certain value (breakdown voltage) Vb, the laser gas between the main discharge electrodes E and E is broken down. The main discharge starts, the laser medium is excited by this main discharge, and laser light is generated.
Such a discharge operation is repeatedly performed by the switching operation of the solid switch SW and the high-voltage power supply operation, whereby pulse laser oscillation at a predetermined repetition frequency is performed.
Here, the pulse width of the current pulse flowing through each stage is set by setting the inductance of the capacity transfer type circuit of each stage composed of the magnetic switches SR2 and SR3 and the capacitors C1 and C2 to be smaller as it goes to the subsequent stage. The pulse compression operation is performed so as to be narrowed sequentially, and a strong discharge with a short pulse is realized between the main discharge electrodes E and E.
[0007]
The discharge circuit has the following problems.
FIG. 7 shows voltage waveforms of the voltage V2 applied to the capacitor C2 and the voltage Vp applied to the peaking capacitor Cp in FIG.
In general, in the case of a discharge excitation type gas laser device, since the discharge impedance is small between the discharge impedance between the main discharge electrodes E and E and the peaking capacitor Cp, impedance mismatch occurs, and the reflected energy is reflected by the magnetic pulse compression circuit. Oscillates in the reverse direction and in the forward direction.
That is, after discharging in the discharge circuit of FIG. 6, voltage oscillation occurs as shown in FIG. 7 due to impedance mismatch, the voltage Vp of the peaking capacitor Cp is inverted and a reverse voltage is generated, and Cp → SR3 → C2 A current flows through the loop and a reverse voltage is generated in the capacitor C2.
Similarly, a current flows through a loop of C2 → SR2 → C1, and a reverse voltage is generated in the capacitor C1. The reverse voltage generated in the capacitor C1 passes through the step-up transformer Tr1 and charges the main capacitor C0 in the reverse direction. The main capacitor C0 charged in the reverse direction charges the capacitor C1 in the reverse direction via the step-up transformer Tr1. Thereafter, a current flows in the forward direction of the capacitor C1 → the capacitor C2 and the capacitor C2 → the peaking capacitor Cp. As described above, the current oscillates in the magnetic pulse compression circuit in the power supply device.
[0008]
Further, as shown in FIG. 7, since the discharge between the main discharge electrodes E and E often starts during the charging from the capacitor in the previous stage of the peaking capacitor Cp (C2 in the case of FIG. 6) to the peaking capacitor Cp. The charge remaining in the capacitor in front of the peaking capacitor Cp (C2 in the case of FIG. 6) flows between the main discharge electrodes E and E, and a reverse voltage is generated in the capacitor in front of the peaking capacitor Cp (in the case of FIG. 6, C2). To do.
The charges generated at this time are reflected as reflected energy, and as in the case of the impedance mismatch described above, the magnetic pulse compression circuit travels in the reverse direction and vibrations travel in the forward direction are generated.
Due to the vibration, a voltage is applied between the main discharge electrodes E and E within a short time after the occurrence of the main discharge, which has a problem of adversely affecting the main discharge.
[0009]
In order to avoid the problem of vibration through the magnetic pulse compression circuit having the reflected energy as described above, various circuit configurations have been proposed.
FIG. 8 is a configuration example of a discharge circuit of a gas laser device having a circuit for processing charges generated by a reverse voltage with a resistor.
In FIG. 8, a series circuit composed of a diode D1 and a resistor R is connected in parallel with the capacitor C1 having the circuit configuration of FIG. The direction of the arrow shown in FIG. 6 indicates the direction of current flow when a reverse voltage is generated in the capacitor C1. In this way, in the circuit shown in FIG. 8, the current flowing when the reverse voltage is generated is consumed by the resistor R, so that a voltage is applied between the main discharge electrodes E and E within a short time after the main discharge. To prevent it.
On the other hand, when the solid switch SW is closed to transfer electric charge from the main capacitor C0 to the capacitor C1 in order to generate a discharge between the main discharge electrodes E and E, the current is opposite to the direction of the arrow, that is, the reverse direction of the diode D1. Flowing into. Therefore, the current does not flow to the series circuit composed of the diode D1 and the resistor R, and is not consumed by the resistor R.
However, since the current generated by the reverse voltage is processed by the resistor, heat is generated in the resistor, and the amount of heat generated from the laser device increases. Therefore, there is a problem that a large cooling structure is required and the apparatus becomes large. In addition, since the current generated by the reverse voltage is consumed as heat, there is a problem that the current consumption of the entire laser increases.
[0010]
In order to avoid the above problems, the current that flows when the reverse voltage is generated is not consumed by the resistor as in the circuit configuration of FIG. Or has been proposed.
FIG. 9 and FIG. 10 show configuration examples of the discharge circuit of the gas laser device having a regeneration circuit configuration for energy regeneration.
In the circuit configuration of FIG. 9, an intermediate tap is provided on the primary side of the step-up transformer TR1, a series circuit of an inductance LL, a magnetic assist SR1, and a main capacitor C0 is disposed on the intermediate tap, and a solid switch SW is provided at one end of the primary winding. The diode D2 is provided at the other end.
In the circuit configuration of FIG. 9, the solid state switch SW is opened when, for example, the electric charge charged in the main capacitor C0 has been transferred to the capacitor C1. After discharging, the reflected energy transferred while extending the pulse width is charged in the capacitor C1 so that a reverse voltage is generated in the capacitor C1. Since the solid switch SW is in the off state, the current when the reverse voltage is generated in the capacitor C1 flows in the direction of the arrow in FIG. 9 via the step-up transformer TR1, and reversely shifts so as to charge the main capacitor C0 in the forward direction. . In this way, the reflected energy is regenerated in the main capacitor C0.
When the solid switch SW is turned on to transfer electric charge from the main capacitor C0 to the capacitor C1 in order to generate a discharge between the main discharge electrodes E and E, a voltage is applied to the reverse side of the diode D2, so that the current is The flow proceeds to the capacitor C1 through the step-up transformer TR1 without flowing through the diode D2.
[0011]
In the circuit configuration of FIG. 10, a series circuit including an inductance L1 and a diode D3 is connected in parallel to the main capacitor C0.
In the circuit configuration of FIG. 10, for example, the solid switch SW is turned OFF when the reflected energy ends the reverse charging of the main capacitor C0.
Note that the OFF timing of the solid switch SW is a period from the time point until the main capacitor C0 is recharged in the forward direction. During that time, the magnetic assist SR1 does not saturate and does not flow in the direction of the solid switch SW.
The reflected energy transferred while extending the pulse width after the discharge is charged to the capacitor C0 so that a reverse voltage is generated in the capacitor C0. Since the solid switch SW is in the off state or the magnetic assist SR1 is not saturated, the current when the reverse voltage is generated in the capacitor C0 flows through the series circuit including the inductance L1 and the diode D3 in the direction of the arrow in FIG. The main capacitor C0 is charged in the forward direction. In this way, the reflected energy is regenerated in the main capacitor C0.
[0012]
The circuit configuration including the regenerative circuit illustrated in FIGS. 9 and 10 does not consume current as shown in FIG. 8, but regenerates reflected energy. There is no increase in size. Further, since the reflected energy is not consumed as heat but is effectively used again, the current consumption of the entire laser is reduced.
However, the circuit configuration including the regenerative circuit illustrated in FIGS. 9 and 10 has the following problems.
As described above, the charge of the main capacitor C0 is transferred to the capacitor C1 via the step-up transformer TR1, and the turn ratio of the step-up transformer TR1 is set so that the charge of the main capacitor C0 is completely transferred to the capacitor C1. Has been.
On the other hand, in order to shorten the charging time to the peaking capacitor Cp as much as possible, to accelerate the rise of the applied voltage to the main discharge electrodes E, E and to increase the discharge start voltage, the initial pulse input to the magnetic pulse compression circuit Is preferably as short as possible. That is, the charge transfer rate from the main capacitor C0 to the capacitor C1 is preferably as fast as possible. Here, as the number of turns on the primary side and the secondary side of the step-up transformer TR1 increases, the transition speed becomes slower. Therefore, both the numbers of turns are set to be minimum.
[0013]
In the case of the turn ratio set in this way, when the charge is transferred as reflected energy from the capacitor C1 to the main capacitor C0, all the charge of the capacitor C1 does not transfer to the main capacitor C0, and some residual charge exists in the capacitor C1. To do.
That is, in the case of FIG. 10, it is difficult to define the turn ratio of the step-up transformer TR1 so that both forward and reverse transitions are performed 100%, and there is inevitably residual charge in the capacitor C1.
[0014]
In FIG. 9, the number of turns of the primary side winding of the step-up transformer TR1 used differs between the time of pulse generation and the time of regeneration, but the secondary side is common. In the optimal turns ratio between the primary side winding and the secondary side winding used at the time of regeneration, the number of turns on the primary side corresponding to the secondary side, which is the minimum required number, is often not a natural number. Therefore, in order to make the primary side and the secondary side winding number at the time of regeneration into a natural number, the secondary side winding number increases. If it does so, the transfer speed of the electric charge from the main capacitor | condenser C0 to the capacitor | condenser C1 at the time of a pulse generation will become slow.
[0015]
In order to solve the above problem, as shown in FIG. 11, it is conceivable to newly provide a regenerative transformer and regenerate the inversion energy through the regenerative transformer.
The high voltage pulse generator shown in FIG. 11 is obtained by adding a regenerative circuit including a magnetic switch SR4 including a regenerative transformer Tr2, a diode D4, and a saturable reactor SR4 to the circuit shown in FIG.
The primary side of the regenerative transformer Tr2 is connected in parallel with the capacitor C1, and the secondary side of the regenerative transformer TR2 is connected in series with a diode D4 and a magnetic switch SR4. The secondary side, the diode D4 and the magnetic switch SR4, Is connected in series to the main capacitor C0.
[0016]
The circuit shown in FIG. 11 operates as follows.
As shown in FIG. 12, when the time from when the solid switch SW is turned on until the magnetic assist SR1 is saturated (magnetic assist time T in FIG. 2) elapses, the charge is transferred from the main capacitor C0 to the capacitor C1. Starts and current flows (refer to the current waveform in the dashed line in the figure).
As described above, the voltage pulse shifts while being compressed by the magnetic pulse compression circuit, and the peaking capacitor Cp is charged. The voltage waveforms of the capacitors C1 and C2 and the peaking capacitor Cp at this time are shown as V1, V2, and Vp in FIG.
Discharging is started during the charging of the peaking capacitor Cp (“discharge start” in the figure), and the current then shifts in the opposite direction as reflected energy as described above.
That is, as shown in FIG. 12, the voltage pulse reverses through the magnetic pulse compression circuit and shifts while being expanded. The voltage waveforms of the capacitors C2, C1 and capacitor C0 at this time are indicated by V2kb, V1kb, V0kb in FIG. 12 (broken lines in FIG. 12).
Hereinafter, the voltage charged in the capacitors C1 and C1 by the reflected energy is referred to as a kickback voltage, and the phenomenon in which the reflected energy shifts in the opposite direction is referred to as kickback.
In the operation of FIG. 12, the solid switch SW needs to be in the on state at least during the period A in order to transfer the charge from the main capacitor C0 to the capacitor C1. Further, during the regenerative operation, it is necessary to charge the main capacitor C0 through the regenerative circuit, and during that time, the solid switch SW must be off. Therefore, it is necessary to turn off the solid switch SW at least during the period B. .
[0017]
If there is no regenerative circuit, the main capacitor C0 is charged so as to have a polarity opposite to that at the time of initial charging (when the solid state switch SW is turned on) [V0kb in FIG. Refer to “No regenerative circuit”].
On the other hand, when a regenerative circuit is provided as shown in FIG. 11, it operates as follows.
When the capacitor C1 is charged in the reverse direction due to the reflected energy, the solid switch SW is in the off state, and the voltage charged to the capacitor C1 (kickback voltage) is generated on the secondary side of the regenerative transformer Tr2, and this voltage Is applied to a series circuit of a diode D4, a magnetic switch SR4, and a main capacitor C0 provided on the secondary side of the regenerative transformer Tr2.
For this reason, as shown in FIG. 12, the main capacitor C0 is charged so as to have the same polarity as at the time of initial charging (when the solid switch SW is on). That is, the reflected energy is regenerated in the main capacitor C0 [see V0 kb (with regenerative circuit) in FIG. 12].
In the circuit of FIG. 11, as described above, the regenerative transformer Tr2 for regenerating reflected energy is provided separately from the step-up transformer Tr1, so that the step-up transformer Tr1 does not intervene in the circuit operation during regeneration. Therefore, the number of turns of the regenerative transformer can be made large on both the primary side and the secondary side, and the turn ratio can be set freely. Therefore, the turn ratio can be set so that some residual charges do not exist in the capacitor C1. Can be precisely defined.
Of course, since it is not necessary to change the turns ratio and the number of turns of the step-up transformer TR1, the turn ratio and the number of turns of the step-up transformer TR1 are set so that the charge transfer speed from the main capacitor C0 to the capacitor C1 is as fast as possible. It becomes possible. For this reason, the residual voltage of the capacitor C1 can be reduced.
[0018]
[Problems to be solved by the invention]
Incidentally, in order for the regenerative circuit shown in FIGS. 9, 10, and 11 to perform the regenerative operation, it is necessary to appropriately set the on / off timing of the solid switch SW.
For example, in the circuit of FIG. 11, the solid state switch SW must be in the off state by the time (C) of FIG. 12, and the solid state switch while the charge is transferred from the main capacitor C0 to the capacitor C1. SW must be in the on state. That is, in FIG. 12, the solid switch SW must be in the on state from the time (time (A)) to the time (B) when the solid switch SW is turned on.
In summary, during the period A in FIG. 12, the solid state switch SW is in the on state and must be in the off state at any point in the period B.
[0019]
Here, in a gas laser device used as an exposure light source, it is required that the laser output energy be constant, and output energy stabilization control is performed.
The exposure gas laser apparatus 1 including the discharge circuit 1a that performs the control as described above includes a controller 2, a monitor module 3, and a beam splitter 4 as shown in FIG.
Part of the laser light emitted from the exposure gas laser device 1 is extracted by the beam splitter 4 and guided to the monitor module 3 for measuring the laser output energy. The monitor module 3 measures the laser output energy and transmits the measurement result to the controller 2. The controller 2 adjusts the voltage of the high voltage power supply HV of the discharge circuit (high voltage pulse generator) shown in FIG. 13 so that the laser output energy is constant. That is, for each laser pulse, the voltage of the high voltage power supply HV, that is, the voltage V0 of the main capacitor C0 varies, and the laser output energy becomes constant.
[0020]
The magnetic assist time of the magnetic assist SR1 is constant in the VT product of the magnetic switch that is a saturable reactor. Therefore, when V0 varies as described above, the magnetic assist time T also varies so that the VT product becomes constant. . For this reason, the length of the period A in FIG. 12 also varies.
As described above, the on-off control of the solid-state switch SW needs to be performed corresponding to the periods A and B. At this time, it is necessary to consider the fluctuation of the period A caused by the fluctuation of the magnetic assist time T described above. is there. Further, since the period B is a function of voltage as in the case of A, it is necessary to consider it.
[0021]
Next, the configuration of the control device that can control the period A and the period B will be described.
FIG. 14 shows a configuration example of a control device for performing on-off control of the solid switch SW. The controller in FIG. 14 is equivalent to the controller 2 in FIG. 13, and receives data on the measurement result of the laser output energy transmitted from the monitor module 3.
Based on the received laser output energy data, the controller 2 calculates and determines the charging voltage to the main capacitor C0 for the next discharge by the charging voltage value calculation unit 2a.
Then, a charging voltage value adjustment command is transmitted to the high voltage power supply HV so that the charging voltage to the main capacitor C0 becomes the determined voltage value, thereby controlling the high voltage power supply HV.
At the same time, the controller 2 sends the charging voltage value data of the main capacitor C0 determined by the charging voltage value calculation unit 2a to the switch on-off timing control unit 2b.
The switch on-off timing control unit 2b calculates the magnetic assist time T based on the received charging voltage value data and the VT product of the magnetic assist SR1 stored in the switch on-off timing control unit 2b in advance. Find the length of.
That is, the current pulse width Ti in FIG. 11 is determined almost constant from the value of the circuit constant, and is stored in advance in the switch on-off timing control unit 2b. The switch on-off timing control unit 2b obtains the length of the period A from the sum of the calculated magnetic assist time T and the current pulse width Ti stored in advance.
The period B is determined by the charging voltage value data and the circuit constants of the magnetic pulse compression circuit. Therefore, for example, the relationship between the voltage value and the period B is obtained in advance and stored in a storage circuit or the like. Then, the period B corresponding to the charging voltage value data is read from the storage device.
[0022]
The on-off timing control unit 2b controls the on-off frequency of the solid-state switch SW based on the set repetition frequency. The on-off timing control unit 2b turns on the solid-state switch SW based on the periods A and B obtained as described above. Starting from the time when the signal is transmitted (time (A) in FIG. 12), after the lapse of period A and before the end of period B (between time (B) and time (C) in FIG. 12), an off signal is sent to the solid state switch SW. Send.
Note that the off signal is transmitted with a delay from the time (B) in FIG. 12 in consideration of the difference between the time when the on signal is transmitted to the solid switch SW and the time when the solid switch SW is actually turned on and the fluctuation (jitter) of the shift. It is desirable to do. Further, the off signal may be transmitted prior to the time point (C) in consideration of the deviation between the time point when the off signal is transmitted to the solid state switch SW and the time point when the solid state switch SW is actually turned off. desirable.
[0023]
FIG. 15 shows a second configuration example of a control device for performing on-off control of the solid-state switch SW.
The configuration example of the control device shown in FIG. 15 is a modification of the first configuration example shown in FIG. 14, and the difference is that in the first configuration example, the main capacitor C0 from the charging voltage value calculation unit 1a of the controller 2 is different. In the second configuration example, the voltage value of the main capacitor C0 is directly detected by the voltage detector Vs, and this detection data is calculated using the charging voltage value data. To calculate the magnetic assist time T.
That is, the voltage detector Vs transmits the measured charging voltage value data of the main capacitor C0 to the switch on-off timing control unit 2b. As described above, the switch on-off timing control unit 2b determines the length of the period A based on the received charging voltage value data and the VT product of the magnetic assist SR1 stored in the switch on-off timing control unit 2b in advance. In addition, the period B is obtained as described above.
[0024]
The on-off timing control unit 2b controls the on-off frequency of the solid-state switch SW based on the repetition frequency. The on-signal is sent to the solid-state switch SW based on the period A and the period B calculated as described above. Is sent to the solid state switch SW before the end of the period B (between the time point (B) and the time point (C) in FIG. 12), starting from the time point (time point (A) in FIG. 12). Send.
As in the case of the first configuration example, the off signal is calculated in consideration of a shift between the time when the on signal is transmitted to the solid switch SW and the time when the solid switch SW is actually turned on and a variation (jitter) of the shift. It is desirable to transmit after the time (B) in FIG. Further, the off signal may be transmitted prior to the time point (C) in consideration of the deviation between the time point when the off signal is transmitted to the solid state switch SW and the time point when the solid state switch SW is actually turned off. desirable.
While the regenerative operation is performed by the configuration and control as described above, the solid switch SW can be reliably turned off.
[0025]
In the configuration example shown in FIG. 14 described above, the magnetic assist time T is calculated based on the charging voltage value data and the VT product of the magnetic assist SR1 stored in the switch on-off timing control unit 2b in advance. The period A is obtained by calculation using T and the stored current pulse width Ti, and the transmission timing of the off signal is obtained.
For this reason, the switch on-off timing control unit 2b requires complicated calculation means and storage means, and as a result, the controller has become larger and more complicated.
On the other hand, in recent years, the pulse interval has become very short as the repetition frequency is increased (4 kHz or more). For this reason, in the configuration example shown in FIG. 14, the switch on-off timing control unit 2b must perform calculations, commands, and the like for controlling the on-off timing of the solid-state switch SW at very short pulse intervals. .
That is, high-speed calculation is required, and the switch on-off timing control unit 2a is complicated and expensive.
[0026]
Also in the configuration example shown in FIG. 15, the switch on-off timing control unit 2b is complicated and expensive for the same reason as in FIG.
In particular, the timing detection in the configuration example shown in FIG. 15 is performed mainly when the charging voltage calculation unit 2a sends a charging command based on the charging voltage value data to the high voltage power supply HV as shown in FIG. After detecting the voltage value of the capacitor C0, the switch on-off timing control unit 2b performs calculation and the like. For this reason, the time required for the calculation by the switch on-off timing control unit 2b is further shorter than that in the case of FIG. 14, so the switch on-off timing control unit 2b in FIG. Compared to that, it becomes more complex and expensive.
As the voltage detector, a voltage probe, a voltage dividing resistor, or the like is used. However, there are individual differences in these measurement characteristics. Therefore, initial adjustment for adjusting such individual differences is required.
Further, in the configuration examples of FIGS. 14 and 15, since it is necessary to allow for drift and jitter of the turn-on time of the solid switch SW switch (time point A ′ in FIG. 12) as described above, the period B is shortened. It becomes difficult to control the off signal of the switch SW.
The present invention has been made to solve the above-mentioned problems of the prior art, and the object of the present invention is to simplify the controller as compared with the prior art, and sufficiently cope with the increase in speed. It is another object of the present invention to provide a high voltage pulse generator for a pulse laser that can reduce the cost.
[0027]
[Means for Solving the Problems]
In the present invention, the above-described problems are solved as follows.
(1) a main capacitor charged to a high voltage by a charger; With this main capacitor, Magnetic assist consisting of saturable reactor and switch means Are connected in series to the primary side, Magnetic pulse compression circuit was connected to the secondary side With a step-up transformer, In the high voltage pulse generator having a regenerative circuit for regenerating the reflected energy from the load side to the main capacitor, a current flow timing detection means for detecting a current flow start timing of the current flowing through the switch means is provided. The timing for turning off the switch means is determined based on the output of the flow timing detection means.
(2) In the above (1), the timer means is driven by the signal from the current flow timing detecting means, and the switch means is turned off by the timing obtained when the timer counts up.
In the present invention, the current flow timing detection means is provided as described above, and the timing for turning off the switch means is determined based on this output. Therefore, it is not necessary to obtain the magnetic assist time T, and therefore An arithmetic device that performs high-speed arithmetic processing is not required. For this reason, the controller can be simplified as compared with the conventional one, and the speed can be sufficiently coped with, and the cost can be further reduced.
[0028]
DETAILED DESCRIPTION OF THE INVENTION
FIG. 1 is a diagram showing a configuration of a high voltage pulse generator according to a first embodiment of the present invention. This embodiment shows an embodiment in which the present invention is applied to the high voltage pulse generator having the regenerative circuit shown in FIG. 11, and the controller in FIG. 1 is equivalent to the controller 2 in FIG. The laser output energy measurement result data transmitted from the module 3 is received.
In the present embodiment, as in the configuration examples shown in FIGS. 14 and 15, the calculation of the magnetic assist time T is performed to find the timing for transmitting the off signal of the solid state switch SW, and the off signal is not transmitted. In addition, the current flow timing detection means Is is provided in a closed circuit composed of the main capacitor C0, the solid switch SW and the magnetic assist SR1, and the transition current (current flowing through the solid switch SW) flows from the main capacitor C0 to the capacitor C1. The start timing is detected, and the detection result is used to find the timing for transmitting the off signal of the solid state switch SW, and the off signal is transmitted.
[0029]
That is, after the solid switch SW is turned on, the magnetic assist time T elapses, the charge transfer from the main capacitor C0 to the capacitor C1 starts, and the closed circuit including the main capacitor C0, the solid switch SW, and the magnetic assist SR1. When a current flows therethrough, the current flow timing detection means Is detects this current and generates an output. The output of the current flow timing detection means Is is transmitted to the timer 2 c in the switch on-off timing control unit provided in the controller 2.
After receiving this signal, the timer 2c starts counting a clock signal having a predetermined frequency (not shown). When this count value reaches a predetermined value, an output is generated.
The predetermined value set in the timer 2c is set so as to generate an output after a time corresponding to the current pulse width Ti has elapsed after the start of counting. Here, as described above, since the current pulse width Ti is substantially constant from the value of the circuit constant, the predetermined value may be a constant value.
When the timer 2c counts up (after time (B) in FIG. 12), the switch on-off timing control unit 2b transmits an off signal to the solid state switch SW.
[0030]
As described above, in this embodiment, the current flow timing detection means Is is provided and the time (A ′) at which the current pulse rises is measured. Therefore, as in the configuration examples shown in FIGS. Therefore, it is not necessary to obtain the magnetic assist time T, and an arithmetic device for performing high-speed arithmetic processing for that purpose is not necessary. In addition, since the transmission timing of the off signal can be obtained simply by using a timer, there is an advantage that the controller is also small and inexpensive.
As in the case of the first and second configuration examples, the off signal is considered in consideration of a shift between the time when the on signal is transmitted to the solid switch SW and the time when the solid switch SW is actually turned on and a variation (jitter) of the shift. It is desirable to set the predetermined value of the timer 2c so that is transmitted later than the time (B) in FIG.
In addition, the off signal is transmitted prior to the time point (C) in consideration of the difference between the time point when the off signal is transmitted to the solid state switch SW and the time point when the solid state switch SW is actually turned off. It is desirable to set a predetermined value for the timer 2c.
[0031]
FIG. 2 is a diagram showing a second embodiment of the present invention. This embodiment shows a case where the present invention is applied to the regenerative circuit configuration shown in FIG. 9, and the main embodiment is the same as the first embodiment. The current flow timing detection means Is is provided in a closed circuit including the capacitor C0, the solid switch SW, and the magnetic assist SR1, and the flow start timing of the transition current (current flowing through the solid switch SW) from the main capacitor C0 to the capacitor C1 is set. It is intended to be detected. The controller in FIG. 2 is equivalent to the controller 2 in FIG. 13, and receives the data of the measurement result of the laser output energy transmitted from the monitor module 3.
[0032]
In the circuit configuration of FIG. 2, when the solid switch SW is turned on and a time corresponding to the Vt product of the magnetic assist SR1 has elapsed, the charge charged in the main capacitor C0 is transferred to the capacitor C1, as shown in the timing chart of FIG. (Time (A ′) in FIG. 3). For example, the solid switch SW is turned off when the electric charge charged in the main capacitor C0 has been transferred to the capacitor C1 (time (B) in FIG. 3). As described above, the voltage pulse shifts while being compressed by the magnetic pulse compression circuit, and the peaking capacitor Cp is charged. The voltage waveforms of the capacitors C1, C2 and peaking capacitor Cp at this time are shown as V1, V2, Vp in FIG.
Discharging is started during the charging of the peaking capacitor Cp (“discharge start” in the figure), and the current then shifts in the opposite direction as reflected energy as described above. That is, as shown in FIG. 3, the voltage pulse reverses the magnetic pulse compression circuit and shifts while being expanded. The voltage waveforms of the capacitors C2 and C1 and the capacitor C0 at this time are shown as V2kb, V1kb, and V0kb in FIG.
In the operation of FIG. 3, the solid switch SW needs to be in the on state at least during the period A in order to transfer the charge from the main capacitor C0 to the capacitor C1. Further, during the regenerative operation, it is necessary to charge the main capacitor C0 via the regenerative circuit, and during that time, the solid switch SW must be off, so that the solid state is at least between time (B) and time (C) in FIG. It is necessary to turn off the switch SW.
[0033]
As in the first embodiment, the current flow timing detection means Is shown in FIG. 2 starts the transfer of charge from the main capacitor C0 to the capacitor C1, and is a closed circuit including the main capacitor C0, the solid switch SW, and the magnetic assist SR1. When an electric current flows through it, an output is generated. The output of the current flow timing detection means Is is transmitted to the timer 2 c in the switch on-off timing control unit provided in the controller 2.
After receiving this signal, the timer 2c starts counting a clock signal having a predetermined frequency (not shown), and generates an output when the count value reaches a predetermined value.
The predetermined value set in the timer 2c is set so that an output is generated after the time point (B) in FIG.
When the timer 2c counts up, the switch on-off timing control unit 2b transmits an off signal to the solid state switch SW.
[0034]
As in the first embodiment, the off signal in FIG. 3 is taken into account in consideration of the deviation between the time point when the on signal is transmitted to the solid state switch SW and the time point when the solid state switch SW is actually on. It is desirable to set the predetermined value of the timer 2c so that it is transmitted later than the time (B).
In addition, the off signal is transmitted prior to the time point (C) in consideration of the difference between the time point when the off signal is transmitted to the solid state switch SW and the time point when the solid state switch SW is actually turned off. It is desirable to set a predetermined value for the timer 2c.
As described above, in this embodiment, the current conduction timing detection means Is is provided and the time point (A ′) at which the current pulse rises is measured, so that the magnetic assist time T is obtained as in the first embodiment. There is no need, and an arithmetic device for performing high-speed arithmetic processing is not necessary. In addition, since the transmission timing of the off signal can be obtained simply by using a timer, there is an advantage that the controller is also small and inexpensive.
[0035]
FIG. 4 is a diagram showing a third embodiment of the present invention. This embodiment shows a case where the present invention is applied to the regenerative circuit configuration of FIG. 10, and the main embodiment is the same as the first embodiment. The current flow timing detection means Is is provided in a closed circuit including the capacitor C0, the solid switch SW, and the magnetic assist SR1, and the flow start timing of the transition current (current flowing through the solid switch SW) from the main capacitor C0 to the capacitor C1 is set. It is intended to be detected. The controller in FIG. 4 is equivalent to the controller 2 in FIG. 13, and receives data on the measurement result of the laser output energy transmitted from the monitor module 3.
In the circuit configuration of FIG. 4, when the solid switch SW is turned on and the time corresponding to the Vt product of the magnetic assist SR1 has elapsed, the charge charged in the main capacitor C0 is transferred to the capacitor C1, as shown in the timing chart of FIG. (Time (A ′) in FIG. 5).
As described above, the voltage pulse shifts while being compressed by the magnetic pulse compression circuit, and the peaking capacitor Cp is charged. The voltage waveforms of the capacitors C1, C2 and peaking capacitor Cp at this time are shown in FIG. V1, V2, Vp Shown in
[0036]
Discharging is started during the charging of the peaking capacitor Cp (“discharge start” in the figure), and the current then shifts in the opposite direction as reflected energy as described above.
That is, as shown in FIG. 5, the voltage pulse reverses the magnetic pulse compression circuit and shifts while being expanded. The reflected energy charges the capacitor C0 so that a reverse voltage is generated in the capacitor C0. During this time, the solid switch SW is held in the on state, and the solid switch SW is turned off when the reverse charging of the main capacitor C0 is completed.
When the reverse charging of the main capacitor C0 is completed, the voltage reversely charged to the main capacitor C0 flows in the direction of the arrow in FIG. 4 through a series circuit including an inductance L1 and a diode D3 in parallel with the main capacitor C0. Charge C0 in the forward direction. The voltage waveforms of the capacitors C2, C1 and capacitor C0 at this time are shown as V2kb, V1kb, V0kb in FIG.
In the circuit configuration of FIG. 4, the main capacitor C0 is charged in the reverse direction by the reflected energy, and then the main capacitor C0 is charged in the forward direction via the series circuit including the inductance L1 and the diode D3. The switch SW must be on until the reverse charging of the main capacitor C0 is finished, and the timing of turning off the solid switch SW is finished as shown in the timing chart of FIG. This is a period (period D) from when the main capacitor C0 is recharged in the forward direction. During this period, the magnetic assist SR1 does not saturate, and therefore does not flow in the direction of the solid switch SW.
[0037]
As in the first and second embodiments, the current flow timing detection means Is shown in FIG. 4 starts the transfer of charge from the main capacitor C0 to the capacitor C1, and includes the main capacitor C0, the solid switch SW, and the magnetic assist SR1. When current flows in the closed circuit, an output is generated. The output of the current flow timing detection means Is is transmitted to the timer 2 c in the switch on-off timing control unit provided in the controller 2.
After receiving this signal, the timer 2c starts counting a clock signal having a predetermined frequency (not shown), and generates an output when the count value reaches a predetermined value.
Since the saturation time of the magnetic switches SR2 and SR3 is sufficiently shorter than the assist time of the magnetic assist SR1, the predetermined value set in the timer 2c is shown in FIG. In period D Set to generate output.
When the timer 2c counts up, the switch on-off timing control unit 2b transmits an off signal to the solid state switch SW.
As described above, in this embodiment, the current conduction timing detection means Is is provided, and the time (A ′) at which the current pulse rises is measured. Therefore, as in the first and second embodiments, the magnetic assist time T There is no need to obtain an arithmetic unit for performing high-speed arithmetic processing. In addition, since the transmission timing of the off signal can be obtained simply by using a timer, there is an advantage that the controller is also small and inexpensive.
[0038]
【The invention's effect】
As described above, in the present invention, the following effects can be obtained.
Since the current flow timing detection means Is is provided and the flow start timing of the current flowing through the switch means is measured, there is no need to obtain the magnetic assist time T, and an arithmetic device for performing high-speed calculation processing is not required. . In particular, the transmission timing of the off signal can be obtained only by using a timer.
Therefore, the controller can be simplified and it is possible to easily cope with the increase in speed. Further, it can be configured at low cost.
[Brief description of the drawings]
FIG. 1 is a diagram showing a configuration of a high voltage pulse generator according to a first embodiment of the present invention.
FIG. 2 is a diagram showing a configuration of a high-voltage pulse generator according to a second embodiment of the present invention.
FIG. 3 is a time chart for explaining the operation of the regenerative circuit shown in FIG. 2;
FIG. 4 is a diagram showing a configuration of a high voltage pulse generator according to a third embodiment of the present invention.
FIG. 5 is a time chart for explaining the operation of the regenerative circuit shown in FIG. 4;
FIG. 6 is a diagram showing a configuration example of a high voltage pulse generator.
FIG. 7 is a diagram illustrating waveforms of a voltage applied to a capacitor C2 and a voltage applied to a peaking capacitor Cp in FIG.
FIG. 8 is a diagram showing a configuration example of a discharge circuit of a gas laser device having a circuit for processing charges generated by a reverse voltage with a resistor.
FIG. 9 is a diagram showing a configuration example (1) of a high voltage generator of a gas laser device having a regeneration circuit configuration for energy regeneration.
FIG. 10 is a diagram showing a configuration example (2) of the high voltage generator of the gas laser device having a regeneration circuit configuration for energy regeneration.
FIG. 11 is a diagram showing a configuration example (3) of the high voltage generator of the gas laser device having a regeneration circuit configuration for energy regeneration;
12 is a time chart for explaining the operation of the regenerative circuit shown in FIG.
FIG. 13 is a diagram showing a configuration example of a control system of the gas laser apparatus for exposure.
FIG. 14 is a diagram showing a configuration example (1) of a control device for controlling a solid-state switch of a high-voltage pulse generator.
FIG. 15 is a diagram showing a configuration example (2) of the control device for controlling the solid state switch of the high voltage pulse generating device;
[Explanation of symbols]
SR1 Magnetic assist
SR2, SR3 magnetic switch
SR4 magnetic switch
SW Solid switch
C0 main capacitor
Tr1 step-up transformer
Tr2 regeneration transformer
C1, C2 capacitors
Cp peaking capacitor
E Main discharge electrode
D1-D4 diode
Vs voltage detector
Is current flow timing detection means
1 Gas laser equipment for exposure
1a Discharge circuit
2 Controller
2a Charge voltage value calculator
2b switch on-off timing controller
2c timer
3 Monitor module
4 Beam splitter

Claims (2)

A charger and a main capacitor charged to a high voltage by the charger;
A high voltage pulse generator comprising a step-up transformer in which a main capacitor, a magnetic assist composed of a saturable reactor, and a switching means are connected in series on the primary side and a magnetic pulse compression circuit is connected on the secondary side. There,
The high-voltage pulse generator includes a regenerative circuit that regenerates reflected energy from the load side to the main capacitor,
A controller for controlling the switch means;
A current flow timing detecting means for detecting a current flow start timing of the current flowing through the switch means,
The high voltage pulse generator according to claim 1, wherein the controller determines a timing for turning off the switch means based on an output of the current flow timing detecting means. The controller has timer means,
2. The high voltage according to claim 1, wherein the timer means is driven by receiving a signal from the current flow timing detecting means, and the switch means is turned off by the timing obtained when the timer counts up. Pulse generator.

Images