JPH0125657B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention includes a device that causes one of the electrode and the workpiece to be moved in a direction in which the other is pushed, and a device that causes movement in a direction perpendicular to the above direction. The present invention relates to an improvement of an electrical discharge machining device equipped with the same.

Conventionally, for example, in electrical discharge machining equipment, relative movement is applied to the electrode and the workpiece in the direction in which the electrode is pushed into the workpiece, and the distance between the electrode and the workpiece is usually constant in that direction. Machining is carried out while using the servo. Here, in normal electric discharge machining, after rough machining, finishing machining is performed using a plurality of electrodes having the same shape and slightly different dimensions. This is because in rough machining, the machining speed is high but the machined surface is rough, while in finishing machining the machined surface is fine but the machining speed is low, and the side gap between the electrode and the workpiece is narrower in finishing machining. This is due to Therefore, the following devices have been proposed for the purpose of processing from rough machining to finishing machining using a single electrode. In other words, after rough machining is completed, the electrode or the workpiece is given a movement with an element perpendicular to the normal feed direction, for example, a circular orbital motion, and the rough machining is performed in the same way as if an electrode with a large apparent size was used. The finishing process is performed using the same electrode used for the process. This is, for example, as shown in FIG. 1, in which an electrode 10 and a workpiece 12 are placed opposite each other in an insulating liquid, and a pulsed current supplied from a pulsed current supply device 14 is applied to the machining gap. The workpiece 12 is processed. At that time, the electrode 10 is connected to the workpiece 12 by a servo circuit consisting of a voltage differential circuit 16 and an amplifier 18, and a servo mechanism consisting of a hydraulic servo valve 20 and a hydraulic cylinder 22, which are driven by the output signal of the circuit. The material is fed in the direction in which it is pushed in (Z-axis direction), for example, so that the voltage Vd in the processing gap matches the reference value Vs on average, and the processing progresses. After the rough machining is completed to a depth set slightly before the final desired depth, the energy of one pulse of the pulse current supply device 14 is changed to be smaller, and the electrode motion control device 24 is further controlled by a known method. The servo motors 26, 28 are operated, thereby causing the X-Y cross tables 30, 32 to perform continuous circular motion. In this case, it is sufficient to apply a sine wave to the servo motors 26 and 28 having a voltage having a phase difference of π/2 and an amplitude corresponding to the difference in side gap between rough machining and finishing machining. Then, while the electrode 10 and the workpiece 12 are relatively orbiting in a circular motion in this manner, processing is performed again to the desired depth. In this case, the effect is equivalent to expanding the diameter of the electrode 10 by a dimension corresponding to the diameter of the relative orbital circular motion, so the rough machined surface caused by the previous rough machining is removed. be. Here, when using this method to machine a corresponding hole on the workpiece 12 using the electrode 10 having an elliptical cross section as shown in FIG. The amount of workpiece 12 to be removed is
is much smaller in parts with a large radius of curvature than in parts with a small radius of curvature,
Therefore, as the machining progresses, the machined depth of the part with the large radius of curvature and the machined depth of the part with the small radius of curvature become significantly different, as shown in Figure 3. Therefore, when machining deep holes with such equipment, it is not possible to sufficiently remove the rough machined surface from the previous rough machining, and the depth reached by the electrode during finishing machining is It has a major drawback that there are differences in each part depending on the shape of the part.

Therefore, in order to solve the above-mentioned drawbacks, an apparatus that employs the following processing method has been proposed.
FIG. 4 is a diagram illustrating this device. 2nd above
In the machining shown in the figure, it is clear that the cause of the defect pointed out is due to the difference in machining allowance on the X-Y plane seen from the Z-axis. , 32, if the radius of the circular motion applied to As a result, the situation shown in Figure 3 cannot occur. That is, the fourth
The spiral circle shown in the figure shows the locus of relative movement between the electrode 10 and the workpiece 12 according to the above method, and shows the amount of increase in machining allowance for each round as an enlarged value ΔR of the orbital circular motion. If ΔR is made extremely small, the machining stock per hit will be extremely small, and if machining energy is available, machining will proceed at less than the machining capacity, which naturally results in longer machining times and lower machining efficiency. descend. If ΔR is increased to prevent this, the machining allowance will increase, and if the machining allowance exceeds the machining capacity, a situation like the one shown in FIG. 3 described above will occur, which is a problem. Therefore, it is necessary to adjust the machining allowance increase amount ΔR according to the machining state as needed. In addition, in such a machining method, it is expected that the machining powder and sludge will be discharged by utilizing the opening and shortening of the machining gap due to the revolving circular motion, and when the machining condition deteriorates due to short circuits, the machining gap will be opened. It is necessary to increase the clearance and increase the revolution speed to speed up the discharge of processed powder, sludge, etc. However, if the revolution speed is increased too much, it becomes difficult to follow the control system of the device and machining accuracy deteriorates, so it is not preferable to simply increase the speed.

The present invention has been made in view of the above-mentioned conventional problems, and its purpose is to adjust the separation width of the machining gap according to the machining state when one of the electrode and the workpiece is moved in a revolving manner relative to the other. By appropriately controlling the machining allowance increase amount ΔR for each round and optimally controlling the revolution speed, it is possible to perform efficient and good machining, and when the machining condition deteriorates, the machining Processing powder, without affecting accuracy
An object of the present invention is to provide an electrical discharge machining device that can effectively discharge sludge and the like.

In order to achieve the above object, the present invention causes one of the electrode and the workpiece to move approximately in revolution relative to the other, and gradually expands or contracts the movement path by a predetermined amount to control the separation width of the machining gap. In an electrical discharge machining device that performs electrical discharge machining, the separation width is detected from the detected voltage between the electrode and the workpiece, and it is determined whether the electrode is in a machinable state or not machinable on the workpiece during the revolution movement. a discriminating device for detecting whether the machining gap is in a condition or in a machining deterioration condition, and at least two speeds of reduction for reducing the gap width of the machining gap are set, and an expansion speed for increasing the gap width. At least one type of reduction speed is set, and when the discrimination device detects a machinable state, it selects the one with the smaller reduction speed, and when it detects a machinable state, it selects the larger one to determine the machining deterioration state. a motion path determination device that selects the speed of expansion when detecting the rotational motion and controls the expansion or contraction of the path of the orbital motion along a plane substantially perpendicular to the main machining feed; When at least two speeds are set and the discrimination device detects a deteriorating state of machining, it selects the faster revolution speed, and when it detects any other state, it selects the slower revolution speed and performs the revolution motion. The invention is characterized by comprising a revolution speed determining device for controlling the revolution speed of the vehicle.

Next, preferred embodiments of the present invention will be described based on the drawings. Note that the same reference numerals are given to the parts corresponding to those of the conventional device described above, and the explanation thereof will be omitted.

FIG. 5 shows an embodiment of the electrical discharge machining apparatus of the present invention.

Here, the X-Y cross tables 30 and 32 are operated by driving the servo motors 26 and 28 in response to a signal from the electrode motion control device 24, and a relative position is created between the electrode 10 and the workpiece 12. Providing the orbital circular motion is similar to the conventional device shown in FIG. 34 is a differential transformer.
This differential transformer 34 is configured such that its coil portion is fixed to the fixed side of the machine, and its movable iron core moves in the same manner as the electrode 10, and the height of the electrode 10, that is, the position in the Z-axis direction, is detected. be done. In addition, 36 is, for example, No. 1 of Japanese Patent Publication No. 53-32112.
In a circuit similar to Fig. 12, the difference voltage Vd−Vs between the machining gap voltage Vd and the reference voltage Vs, and the differential transformer 3
This is a comparison selection circuit for preferentially selecting the lower one of the four outputs. Here, the operation of the comparison selection circuit 36 is performed by setting in advance the position of the coil portion of the differential transformer 34 fixed to the fixed side of the machine, as detailed in the above-mentioned Japanese Patent Publication No. 53-32112. If the electrode 10 is above the set position, the voltage Vd at the machining gap and the reference voltage Vs
The position of the electrode 10 is controlled according to the differential voltage Vd-Vs between the two, and when the electrode 10 is lowered to the above-mentioned set position, the output of the differential transformer 34 is selected preferentially, and the servo valve 20 and the hydraulic cylinder 2
The electrode 10 is controlled by a servo mechanism consisting of two parts so that it does not fall below that position.

FIG. 6 is an explanatory diagram of the control device 24 of the X-Y cross tables 30, 32. Here, 4
0 is a two-phase oscillator, which generates sine waves with a 90° phase difference.
Outputs e _X and e _Y. 42 is the electrode voltage detection signal
This is a control circuit for controlling the outputs of e _X and e _Y based on Vd and extracting voltages E _X and E _Y corresponding to a desired eccentric radius. The voltages _{of E}
Move the shaft motors 26 and 28. The X-Y cross tables 30 and 32 are equipped with linear potentiometers _R
RY is attached and used to detect the position of the electrode 10 on the X-Y plane. The output voltages _of the linear potentiometers _R
6 and 28 move, and the table position is determined by the control circuit 4.
The outputs _EX and E _Y of the two outputs are controlled to be equal to each other.

FIG. 7 is a detailed explanatory diagram of the two-phase oscillator 40. This two-phase oscillator 40 includes an operational amplifier Q ₁ ,
An integrating circuit consisting of a resistor R and a capacitor C,
It is formed by cascading an operational amplifier Q ₂ , a resistor R, a capacitor C, and a limiting inverting integrator consisting of voltage control Zener diodes ZD1 and ZD2 in a feedback loop that gives the following differential equation. .

RCd/dte _X = e _Y ... ₍ 1) RCd/ _dte Y _{= -e}
It is set larger than RC, making the circuit slightly unstable. Also, Zener diode ZD for voltage limiting
1. ZD2 prevents the waveforms of e _X and e _Y from being distorted and stabilizes the amplitude. The two outputs e _X and e _Y thus obtained have a phase difference of 90° and are expressed by the following equation.

e _X = Esin t/RC (3) e _Y = Ecos t/RC (4) Here, E is the voltage of the voltage limiting Zener diodes ZD1 and ZD2. Further, in order to set the frequency, external terminals are provided at both ends of the resistor R, and these external terminals are connected to external resistors 50 and 52.
is connected. And this external resistance 50,
By setting the resistance value of 52, frequency control can be performed externally. In this embodiment, photoconductive cells are used for the external resistors 50 and 52, and the frequency is controlled by external light.

FIG. 8 is a detailed explanatory diagram of the control circuit 42.
Generally, while electrical discharge machining is being performed smoothly between the electrode 10 and the workpiece 12, the inter-electrode voltage detection signal V'd is within a certain range. As the width of the discharge gap increases, the signal V'd increases, and when the width of the discharge gap exceeds a certain level, a good discharge becomes impossible. The detection signal at this time is defined as the processable voltage _V1 . Furthermore, if a short circuit occurs during electrical discharge machining, the detection signal V'd decreases to below a certain value. The detection signal at this time is defined as short circuit detection voltage _V2 . Then, by setting the machining voltage V ₁ and the short circuit detection voltage V ₂ in advance, and comparing these with the detection signal V'd, it is determined whether the electric discharge machining is being performed properly. Can be done. First, the detection signal V'd is compared with the machining voltage _V1 in the comparator 60, and if the detection signal V'd is larger than the voltage _V1 , the gap length between the electrode 10 and the workpiece 12 is set to such an extent that no discharge occurs. It is determined that the condition has spread to the point where it cannot be processed. Then, the comparator 60 outputs a logic level "0", the gate of the AND gate 62 is closed, and the AND gate 66 connected through the inverter 64 is opened. These AND gates 62,
A clock pulse CP generated by the oscillator 68 at fixed time intervals or a clock pulse CP synchronized with the oscillation cycle of the output of the two-phase oscillator 40 is applied to the clock pulse 66. Therefore, when processing is not possible, the first clock pulse CP is applied.
This clock pulse CP is input to the addition input terminal of the counter 70 via the OR gate 72. Also, when the detection signal V′d is lower than the machinable voltage V ₁ ,
The gap length is a distance that allows sufficient discharge, and is determined to be in a processable state. At this time, the comparator 60
outputs logic level “1” and AND gate 62
is opened, and the clock pulse CP is input to the addition input terminal of the second counter 74. These counters 70 and 74 are connected in cascade, and the output counted by the counter 74 is further counted by the counter 70. Further, the detection signal V'd is compared with the short circuit detection voltage _V2 in the comparator 76. Then, when the detection signal V'd decreases by the short circuit detection voltage _V2 , it is determined that a machining deterioration condition such as a short circuit phenomenon has occurred, and the comparator 76 outputs logic level 0, closes the AND gates 62 and 66, and inverts the AND gate 76 via device 76a
A clock pulse CP similar to that described above is applied to this AND gate 76b that opens b.
Therefore, when the machining is in a deteriorating state, this clock pulse is inputted to the subtraction input terminal U of the second counter 70 via this AND gate. In this way, a determination device is constructed that determines whether or not the electrode 10 is in a machinable state for the workpiece 12, and also determines whether or not the machining state has deteriorated. Furthermore, as is clear from the above circuit configuration, when the process is possible, the AND gate 6
2 is opened and the first and second counters 7
The outputs of 0 and 74 are counted up by one unit for each clock pulse CP. Also, when processing is not possible, the AND gate 66
The outputs of 70 and 74 are counted up by 4 units for each clock pulse CP. In addition, when processing is in a deteriorating state,
Only the AND gate 76b is opened and the counter 70,
The output of 74 is counted down by 4 units in response to one clock pulse CP.

That is, the clock pulse is counted up or down by a predetermined unit in accordance with the clock pulse CP from the oscillator 68, and the rate of expansion or contraction of the orbital radius of the electrode 10 is determined. In the case of CP,
The expansion value or contraction value of the orbital radius of the electrode 10 is determined. Further, the outputs of the counters 70 and 74 are input to multiplication type DA converters _78X and _78Y , and the values of the outputs e , the analog quantities of the multiplied values are output as _EX and _EY . As such a multiplication type DA converter, AD7520 manufactured by Analog Devices, Inc. of the United States is known. With the above configuration, the _peak values of the outputs _E
It increases slightly by 1 unit in response to the input of clock pulse CP, and increases by 4 units in response to the input of clock pulse CP when the machining gap is wide and machining is impossible. Furthermore, when a deteriorating state of machining such as a short circuit is detected, the peak values of the outputs _EX and _EY are decreased by 4 units in response to the input of the clock pulse CP. In this way, the multiplication type AD converters 78X, 78Y function together with the counters 70, 74 as a motion path determination device.

Furthermore, if the low state of the detection signal V'd continues longer than the first-order lag time constant composed of the resistor 80 and the capacitor 82, the comparator 76 continues to output logic level "0", so the capacitor 80
The charge of V2 is discharged through the resistor 80, and the charge voltage of the capacitor 82 becomes lower than the reference voltage _V3 . As a result, the comparator 84 outputs a logic level "1", turning on the transistor 88 and lighting the lamp 86. This lamp 86 uniformly illuminates the photoconductive cells 50, 52 within the closed case 90 to lower their resistance values, thereby increasing the oscillation frequency of the outputs _eX , _eY of the two-phase oscillator 40.
In this way, the resistor 80, capacitor 82, comparator 84, lamp 86, transistor 8
8. The photoconductive cells 50, 52 function as revolution speed determining devices.

The embodiment of the present invention has the above-described configuration, and its operation will be explained below by taking as an example a case where a hole is machined in a workpiece 12 using the electrode 10.

First, at a predetermined machining position, the servo mechanism, that is, the hydraulic servo valve 20 and the hydraulic cylinder 22, are driven so that the electrode 10 vertically descends toward the workpiece 12, and a rough machining process is started. Here, the servo mechanism is controlled as follows so that the machining voltage Vd between the electrode 10 and the workpiece 12 is always within a range that allows good electrical discharge machining. First, when the machining voltage Vd decreases and there is a risk of short circuit, it is natural that the comparison selection circuit 36
The voltage Vd input to the hydraulic servo valve 20 via this circuit 36 and the amplifier 18 also decreases. As a result, the descending speed of the electrode 10 is reduced, the machining gap between the electrode 10 and the workpiece 12 is widened, and the machining gap is controlled to such an extent that good electrical discharge machining can be performed. Then, when the risk of short circuiting or the like is eliminated, the voltage Vd also rises and returns to a value that allows proper electrical discharge machining, so that the descending speed of the electrode 10 increases. Further, if the machining gap becomes too wide for some reason and discharge becomes impossible, the machining voltage Vd increases and the voltage Vd-Vs input to the hydraulic servo valve 20 increases. As a result, the descending speed of the electrode 10 is increased, and the machining gap is controlled to be narrowed to the extent that good electrical discharge machining can be started. Then, when good electrical discharge machining is started, the machining voltage Vd decreases and the descending speed of the electrode 10 decreases. In this way, the electrical discharge machining progresses while the machining gap between the electrode 10 and the workpiece 12 is always maintained in an optimal state. Then, when the electrode 10 reaches a certain depth,
Differential transformer 34 detects this. Then, the output of the differential transformer 34 falls below the value of Vd-Vs, and the comparison selection circuit 36 selects the output of the differential transformer 34.
Select output 4. As a result, the input applied to the servo valve 20 via the amplifier 18 becomes 0, and the lowering of the electrode 10 is stopped. With the above steps, the rough machining process is completed.

Simultaneously with the completion of this rough machining step, the energy per pulse of the pulse current supply device 14 is controlled to a small value, and the electrode motion control device 24 controls the X-
A finishing process is started in which the Y cross tables 30 and 32 are given continuous orbital circular motion. Here, the radius of the orbital circular motion given to the X-Y cross tables 30, 32 is controlled so as to gradually increase by ΔR from 0 as shown in FIG.
Then, this increase amount ΔR is controlled so that it is within the range of the processing capacity of the apparatus and the processing efficiency is maximized. First, the two-phase oscillator 40 outputs sine waves e _X and e _Y whose phases are different by 90 degrees. This output e _X , e _Y
is the multiplication type DA converter 78X of the control circuit 42,
At 78Y, the outputs of the counters 70 and 74 are multiplied, and the drive voltage _EX of the servo motors 26 and 28 is
Output as E _Y. And this output E _X , E _Y
X-Y cross table 3 according to the voltage peak value of
The radius of the orbital circular motion of 0,32, i.e. the electrode 10
The radius of the orbital circular motion with respect to the workpiece 12 is given. Here, the radius of the above-mentioned orbital circular motion gradually increases as follows. First, at the start of finishing processing, the outputs of the counters 70 and 74 are 0, so the multiplication type DA converter 78X,
The outputs _EX and E _Y of 78Y are also 0. Therefore, the orbital circular motion starts from a point where the radius is zero. When the machining voltage detection signal V'd is in the range of V ₁ >V'd> V ₂ , comparators 60 and 7 indicate that the discharge machining is being performed satisfactorily.
It is judged by 6. And comparator 60,7
6 both output logic level "1", AND gate 62 is opened, and counters 74 and 70 start counting clock pulses CP via this AND gate 62. Then, the multiplier type AD converters 78X, 78Y convert _E
By increasing the peak value of E _Y , the radius of the orbital circular motion also increases gradually by ΔR ₁ . Further, when the machining voltage detection signal V'd is in the range of V'd> _V1 , the comparator 60 determines that the machining gap is too wide and the discharge is not being performed satisfactorily, making machining impossible. Then, the comparator 60 outputs a logic level "0", closes the AND gate 62, and
Gate 66 is opened to cause counter 70 to directly count clock pulses CP. Then, as is clear from the drawing, the counters 70 and 74 count one clock pulse CP at a rate four times higher than when they are in the machining ready state.
Therefore, multiplication type DA converters 78X, 78Y
increases the _peak values of _E Increase amount ΔR ₂
(ΔR ₂ =4ΔR ₁ ). As a result, the machining gap between the electrode 10 and the workpiece 12 is rapidly narrowed to a range where electrical discharge machining is possible. Then, when it enters the machining ready state, the machining voltage detection signal V′d
becomes V ₁ >V'd, and the operation switches to the machinable state described above.

Further, if a short circuit occurs between the electrode 10 and the workpiece 12 due to the machining gap being too narrow, for example, the detection signal V'd becomes V ₂ >V'd and is detected by the comparator 76. Outputs logic level “0”. Then, AND gate 62,
66 is closed and AND gate 76b is opened.
As a result, the outputs of the counters 70 and 74 decrease the peak values of EX _{and EY} by 4 units in response to the input of the clock pulse CP, and the radius of the orbital circular motion of the electrode 10 is gradually decreased by ΔR ₂ . In this way, when the machining gap is sufficiently opened and the short circuit is naturally removed, the machining voltage detection signal V′d returns to V′d>V ₂ , and the above-mentioned machining possible state or machining impossible state operation occurs. Switch to . Further, if the short circuit continues even though the machining gap has been sufficiently opened, the charge in the capacitor 82 will be discharged through the resistor 80, and the charge voltage of the capacitor 82 will fall below the set voltage _V3 . Then, the comparator 84 outputs a logic level "1", turning on the transistor 88 and lighting the lamp 86, so that the photoconductive cells 50, 52
The resistance value of decreases, and the outputs e _X , e _Y of the two-phase oscillator 40
oscillation frequency increases. As a result, the circumferential speed increases with the revolution period of the electrode 10, and the effect of discharging processed powder and sludge increases significantly. Furthermore, since the robot revolves with a reduced radius of revolution, there is no overcut phenomenon that degrades machining accuracy, and the machining accuracy can maintain the same value as when it revolves at a low speed. In this way, processed powder and sludge are efficiently discharged, the short-circuit condition is quickly eliminated, and the state is again switched to the processable state or the non-processable state.

As described above, finishing processing is performed while the processing gap between the electrode 10 and the workpiece 12 is always maintained at an appropriate value.

In the above embodiments, processing in which the radius of the revolution of the tool electrode gradually increases is explained, but the present invention is not limited to this, and can also be used in cases where the tool electrode processes the outer periphery of a workpiece. In this case, the radius of the revolution of the tool electrode will gradually decrease, but it is sufficient if the counter used is of a subtraction type instead of an addition type.

In addition, in the embodiment, the workpiece is fixed,
Although the tool electrode is shown in revolution, the same effect can be obtained by fixing the tool electrode and causing the workpiece to revolve.

As described above, according to the present invention, when one of the electrode and the workpiece is orbitally moved relative to the other, the separation width of the machining gap is appropriately controlled according to the machining state, so that an efficient and good process can be achieved. In addition to being able to perform machining, at least two types of revolution speeds are set for the revolution movement, so it is possible to control the machining by selecting one of the set revolution speeds. Moreover, when a machining deterioration condition is detected, the faster revolution speed is selected and the revolution speed is controlled. This allows processing powder, sludge, etc. to be effectively discharged without affecting processing accuracy.
It is possible to return to a good machining state in an extremely short time, and has the effect of improving machining accuracy and machining efficiency.

[Brief explanation of drawings]

Fig. 1 is an explanatory diagram of a conventional electric discharge machining device, Fig. 2 is an explanatory diagram showing an example of its electrode structure, Fig. 3 is a machining explanatory diagram showing a problem when using the electrode shown in Fig. 2, Fig. 4 is an explanatory diagram showing the locus when the electrode is given a revolving circular motion whose radius automatically expands, Fig. 5 is an explanatory diagram showing one embodiment of the electrical discharge machining apparatus of the present invention, and Fig. 6 is An explanatory diagram of the control device for the X-Y crosstable, FIG. 7 is a circuit diagram of its two-phase oscillator, and FIG. 8 is a circuit diagram of its control circuit. The same members in each figure are given the same symbols, 10 is a tool electrode, 12 is a workpiece, 40 is a two-phase oscillator, 5
0, 52 are photoconductive cells, 60, 76, 84 are comparators, 62, 66, 76b are AND gates,
64, 76a are inverters, 68 are oscillators, 70, 7
4 is a counter, 78X and 78Y are multiplication type DA converters, 80 is a resistor, 82 is a capacitor, 86 is a lamp, and 88 is a transistor.

Claims (1)

[Scope of Claims] 1 Electric discharge machining is performed by causing one of the electrode and the workpiece to substantially revolve relative to the other, and by gradually expanding or contracting the movement path by a predetermined amount and controlling the separation width of the machining gap. In the electric discharge machining device, the separation width is detected from the detected voltage between the electrode and the workpiece, and it is possible to determine whether the electrode is in a machinable state or in a machinable state with respect to the workpiece during the orbital movement. a discriminating device for detecting whether machining is in a deterioration state; and at least two types of reduction speeds for reducing the separation width of the machining gap and at least one type of expansion speed for increasing the separation width. When the discrimination device detects a machinable state, it selects the one with the smaller reduction speed; when it detects an unworkable state, it selects the larger one; and when it detects a machining deterioration state, it selects the one with the smaller reduction speed. a motion path determining device that selects the speed of expansion to control expansion or contraction of the path of the orbital motion along a plane substantially perpendicular to the main processing feed; When the discrimination device detects a deteriorating machining state, it selects the faster revolution speed, and when it detects any other state, it selects the slower revolution speed to determine the revolution speed of the revolution motion. An electrical discharge machining device characterized by comprising a revolution speed determining device for controlling.