Method for electro-discharge machining

Abstract

A method for electro-discharge machining is provided. The invention relates to a method and an apparatus for electro-discharge machining (EDM) of a workpiece by means of a series of machining pulses. During the machining time, machining pulses are applied to a working gap between the workpiece and an electrode. First an open circuit voltage is applied and if no discharge occurs within a waiting time do, the open circuit voltage is increased to a second open circuit voltage level and if no discharge occurs within a second waiting time di, the open circuit voltage is further increased to a third open circuit voltage level. An average product of the actual open circuit voltage multiplied by a partial ignition delay is calculated and subtracted from a product of a reference open circuit voltage multiplied by a total ignition delay and a servo setpoint value is adjusted accordingly. The actual axis position is sampled in real time and discharge electrical parameters are adjusted accordingly. The discharge electrical parameters can be adjusted depending on the duration of do and di. Furthermore, a micro-oscillation can be applied to the axis.

Description

Technical Field

[0001] This invention relates to a method for electrical discharge machining (EDM), and more particularly to a method and apparatus for generating machining pulses for EDM to enhance the productivity of an EDM system. Background Technology

[0002] In EDM, a series of machining pulses are applied to the working gap between the workpiece and the electrode, thereby machining the workpiece. To monitor this process, voltage and current curves across the working gap over time are sensed. Based on the pulse voltage curve of the machining pulses applied to the working gap at that time, characteristic values ​​such as ignition voltage, ignition delay time, and average pulse voltage are derived. In particular, the measured ignition delay time can be used to control the distance between the electrodes or the gap width (hereinafter referred to as: gap), as disclosed in DE2250872. The measured ignition delay time is compared with a corresponding reference value, and the difference is used to control the axis position, and thus control the gap width between the electrodes. It is well known that other servo control methods exist, such as gap width control based on average pulse voltage.

[0003] The problem with the process described above is the relatively slow response time of the machine relative to the discharge frequency. In fact, the eigenvalue of the die sinking machine head is less than 70 Hz, causing quill movement to occur in the range of one-tenth to several hundred seconds.

[0004] Since the discharge frequency of the EDM process can typically be 50Hz, in other words, once every 20µs, this means that if the sleeve shaft response time is 10ms, there could be as many as 500 sparks discharging in the wrong place.

[0005] The machine shafts control the gap width by performing small movements in an attempt to reach the setpoint position. However, because they cannot follow the discharge frequency, the machine's performance is affected in terms of material removal rate, electrode wear, and surface quality. It is known from EP0333170 that adjusting the filter frequency of the acquired ignition delay can reduce machine head instability. This alleviates, but does not solve, the problems described above.

[0006] Further known from state-of-the-art technology is the ability to increase the open-circuit voltage, thereby reducing the average ignition delay. However, these known methods lack accuracy because a higher open-circuit voltage alone results in a larger gap.

[0007] For example, JP61090822A specifies the time range of the ignition delay for the machining pulse, and if the ignition delay is long, the open-circuit voltage is increased to keep the delay within two limits. Simply working on the voltage without contributing to the gap width control servo will not solve the problem, as the process will remain at an unfavorable operating point and performance will not be improved.

[0008] - As known from JP02095514A: the gap voltage is gradually increased without discharging at a predetermined voltage. A specific application of this solution is wire discharge machining (WEDM) of materials containing insulating particles (such as diamond), specifically WEDM performed using a diamond-coated grinding wheel. The aim is to prevent wire breakage due to this.

[0009] Furthermore, US2018221977A1 ​​applies a voltage spike or increases the pulsed current to the open-circuit voltage after the jump movement of the machine head to stabilize the process. In reality, after the jump movement, the gap is very clean, and the process oscillates between open-circuit voltage pulses and erosion or short-circuit pulses. Increasing the voltage per pulse, without considering ignition delay, will only increase the gap width and reduce machining accuracy without stabilizing the erosion process. Increasing the current has the disadvantage of increasing surface roughness and the heat-affected zone.

[0010] - Yu, P.-H. et al. (2011), “Improvement of wire electrical discharge machining efficiency in machining polycrystalline silicon with auxiliary-pulse voltage supply,” in The International Journal of Advanced Manufacturing Technology, 57(9–12), 991-1001, https: / / doi.org / 10.1007 / s00170-011-3350-2. This paper describes the destruction of polycrystalline silicon insulation by applying an initial voltage spike to each pulse during machining via WEDM. Applying such a voltage spike to each pulse increases the risk of wire breakage during machining in standard mode, thus requiring reduced machining energy and performance for most applications. Summary of the Invention

[0011] The primary objective of this invention is to improve the performance and stability of the EDM process through electronic means, since adapting discharge parameters can typically be 1000 times faster than moving the machine axis. According to this invention, this is achieved without any loss of reproduction accuracy.

[0012] This invention achieves this objective through the subject matter of the first claim, which discloses a method for electrical discharge machining (EDM) of a workpiece, wherein a plurality of discrete machining pulses are applied to a gap between the workpiece and an electrode, and wherein an open-circuit voltage is applied between the electrode and the workpiece to induce discharge, and wherein an initial open-circuit voltage level U is initially applied. o0 The open-circuit voltage of the current pulse is applied, and if no discharge occurs within the first waiting time d0, the open-circuit voltage is then increased to the second open-circuit voltage level U. o1 And if no discharge occurs within the second waiting time d1, the open-circuit voltage is then further increased to a third open-circuit voltage level U. o2 Furthermore, it calculates the partial open-circuit voltage U based on multiple pulses. o0 U o1 U o2 Multiplied by partial ignition delay t d0 t d1 t d2 The average sum of products W step And where the sum of the average products W step With the average product W tec The average product W is compared. tec It is the previously set reference open-circuit voltage U 0_tec Multiply by total ignition delay t d The product of, and wherein the servo setpoint value for gap width control is adjusted such that if the sum of the average products W step Greater than the average product W tec If the gap width is reduced, and the servo setpoint value is adjusted such that if the average product sum W step Less than the average product W tec If so, the gap width will be increased.

[0013] Another object of the present invention is to adapt one or more electrical parameters as a function of changes in the gap width. For this purpose, and according to the present invention, the shaft position is continuously acquired, and the change in the gap width is derived from the change in said shaft position. Therefore, the electrical parameters can be adapted in real time as a function of changes in the shaft position. These electrical parameters include: initial open-circuit voltage U. o0 Pulse pause t o Pulse current amplitude I and pulse current shape I shape .

[0014] Further advantages are defined in the dependent claims. Attached Figure Description

[0015] Preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings, in which:

[0016] Figure 1 It is a plot of a typical machining pulse;

[0017] Figure 2 It is a plot of typical movement of the machine head with associated EDM pulses;

[0018] Figure 3 An oscilloscope plot of the stepping pulse according to the present invention generated from the pulse generator;

[0019] Figure 4 It is a diagram illustrating electrode wear, material removal rate, and inter-electrode gap width as functions of processing voltage, as known in the art.

[0020] Figure 5 It is a plot of typical movement of a machine head with associated EDM pulses of the present invention;

[0021] Figure 6 It is a plot of voltage acquisitions from step pulse (a) and standard pulse (b);

[0022] Figure 7 It is a block diagram of the gap control loop based on the most advanced technology;

[0023] Figure 8 and 9 This is a block diagram of the gap control loop according to the present invention;

[0024] Figure 10 This is a flowchart of the optimal control algorithm;

[0025] Figure 11 It is a diagram illustrating the Z-axis position of a die-cutting machine as a function of time when machining the cavity. Detailed Implementation

[0026] First, refer to Figure 1 The figure illustrates the voltage and current curves as a function of time for an ideal EDM pulse (or so-called normal pulse) in a known die discharge machine with an anodicly polarized processing electrode. These voltage and current curves are characterized by:

[0027] - At the beginning of the pulse, during the ignition delay time t d During this period, the open-circuit voltage U is obtained and maintained. o ,

[0028] - After ignition, the open-circuit voltage U o The voltage drops sharply to the discharge voltage U. e ,

[0029] - Processing current I e At discharge time t e During the period of flow,

[0030] - During the pause time t o In the middle and during the ignition delay time t d During this period, no processing current flows and no material is removed.

[0031] Material removal occurs at discharge time t e During this period, the ignition delay time t d and pause time t o This results in a loss of machine productivity.

[0032] As stated, the main objective of this invention is to improve the performance and stability of the EDM process without sacrificing any reproduction accuracy.

[0033] Therefore, in the method for electrical discharge machining according to the first embodiment of the present invention, a plurality of discrete electrical discharge machining pulses are applied to the gap between the workpiece and the electrode, wherein an open-circuit voltage is applied between the electrode and the workpiece to induce discharge, wherein initially an initial open-circuit voltage level U o0 The open-circuit voltage of the current pulse is applied, and if no discharge occurs within the first waiting time d0, the open-circuit voltage is increased to the second open-circuit voltage level U. o1 Furthermore, if no discharge occurs within the second waiting time d1, the open-circuit voltage is further increased to a third open-circuit voltage level U. o2 And calculate the partial open-circuit voltage U based on multiple pulses. o0 U o1 U o2 Multiplied by partial ignition delay t d0 t d1 t d2 The average sum of products W step And the sum of the average products W step With the average product W tec The average product W is compared. tec It is the previously set reference open-circuit voltage U 0_tec Multiply by total ignition delay t dT The product of the average products, where the sum of the average products is W step Greater than the average product W tec If the servo setpoint value used for gap width control is increased, and wherein the sum of the average products W step Less than the average product Wtec If so, then reduce the servo setpoint value.

[0034] The steps of the present invention for electrical discharge machining will now be explained in detail.

[0035] The method involves applying and gradually increasing the open-circuit voltage to induce discharge, as follows:

[0036] - First, set the initial open-circuit voltage level U o0 Apply to the gap;

[0037] - If no breakdown occurs within the first waiting time d0, then the initial open-circuit voltage level U is adjusted. o0 Increase to the second open-circuit voltage level U o1 ;

[0038] - If no breakdown occurs within the second waiting time d1, then the second open-circuit voltage level U is adjusted. o1 Increase to the third open-circuit voltage level U o2 .

[0039] In this way, the method facilitates breakdown after the application of open-circuit voltage, thereby minimizing unproductive time. Figure 6 a shows the voltage of a typical ideal pulse according to the present invention. In contrast, Figure 6 b shows the shape of a typical conventional ideal pulse. Furthermore, Figure 3 An exemplary voltage signal of a pulse according to the invention, measured by means of an oscilloscope, is shown.

[0040] Due to ignition delay time t d The duration of the discharge is roughly proportional to the gap between the electrodes, so increasing the open-circuit voltage allows for forced discharge. Therefore, the relatively slow response of the servo system (which cannot adjust the gap width in real time) is mitigated by gradually increasing the open-circuit voltage within the same pulse.

[0041] After an open-circuit voltage is applied, breakdown can occur at any time. Breakdown may occur during the first waiting time d0, or during or after the second waiting time d1. In some cases, breakdown may not occur at all (open-circuit pulse), in which case the open-circuit voltage pulse is stopped.

[0042] Preferably, the waiting times d0, d1, etc., are the maximum open-circuit voltage application time t. dmax The predetermined value or predetermined percentage. For example, the waiting time d0 is t. dmax 10%, and the waiting time d1 is t dmax 15%. Maximum open-circuit voltage applied for time t dmaxFor example, 100µs means that d0 is 10µs and d1 is 15µs.

[0043] However, increasing the open-circuit voltage U0 results in a wider machining kerf and poorer surface roughness. Simply varying the open-circuit voltage will also reduce the reproducibility of the machining results. For this reason, the setpoint value used for servo control (i.e., for inter-electrode gap width control) is then adjusted.

[0044] In the case of servo control based on average pulse voltage, the setpoint value used for servo control is the servo reference voltage, while in the case of servo control based on ignition delay time, the setpoint value is the servo reference delay.

[0045] The method according to the invention further includes forming the average product sum W. step For example, as follows:

[0046] - For each pulse, the partial open-circuit voltage U of the current pulse oi Multiply by the corresponding ignition delay time t di ;

[0047] - Multiply the part of the actual pulse by U oi * t di Add, for example:

[0048] W step_pulse = (U o0 * t d0 )+(U o1 * t d1 )+(U o2 * t d2 );

[0049] - Finally, by using multiple n d Partial product of pulses W step_pulse Add and divide by the number of pulses n under consideration d The number of n to form the plurality of n d The average sum of pulses W step .

[0050]

[0051] For example, if the breakdown occurs during the second waiting time d1, then the sum of the products is:

[0052] W step_pulse = (U o0 * t d0 )+(U o1 * t d1 )+(U o2 * t d2 )

[0053] t d0 = d0; t d2 = 0

[0054] Here, the first product includes the partial ignition delay t corresponding to the entire first latency d0. d0 However, the second product includes a partial ignition delay t corresponding only to a portion of the second latency d1. d1 Furthermore, in this example, the third product is zero because breakdown has already occurred, i.e., t. d2 The result is zero. For this reason, a partial ignition delay is used to calculate the sum of the products W. step_pulse .

[0055] Furthermore, the method according to the invention includes: forming an average product W for the number of pulses under consideration, nd. tec The average product W tec It is the previously set reference open-circuit voltage U 0_tec Multiply by total ignition delay t dT The average product, for example, is as follows:

[0056] - For each pulse, by referencing the open-circuit voltage U 0_tec Multiply by the corresponding total ignition delay t dT To calculate the product W tec_pulse The reference open-circuit voltage U 0_tec It is the open-circuit voltage based on a suitable standard processing technical parameter set, and the total ignition delay t dT Equal to the sum of partial ignition delays t dT = (t d0 + t d1 + t d2 );

[0057] - By multiplying the portions of multiple pulses nd by W tec_pulse The average W of the multiple pulses is obtained by summing and dividing by the number of pulses considered, nd. tec .

[0058] Furthermore, the method according to the present invention includes: summing the average products W step With the average product W tec Compare them. More specifically, from the average product W tec Subtract the sum of the average products W step That is, W step -W tec .

[0059] Finally, as the value W step -W tecThe function is used to adjust the servo setpoint value used for gap width control, as follows:

[0060] - Adjust the servo setpoint value so that the average product sum W step Greater than the average product W tec This reduces the gap width, which means reducing the gap between electrodes;

[0061] - Adjust the servo setpoint value so that the average product sum W step Less than the average product W tec This increases the gap width, which means increasing the gap between electrodes.

[0062] In essence, the sum of the average products W step With the average product W tec The difference is used to adjust the servo setpoint value for gap width control, and thus ultimately adjust the gap between the electrode and the workpiece.

[0063] For example, servo compression COMP is used as a servo setpoint value to set the servo reference delay or servo reference voltage. Servo compression COMP is a parameter used to adjust the gap width; a high COMP value sets a small gap, while a low COMP value indicates a large gap.

[0064] Here, the function used as the value of Comp_adj is used to adjust the servo compression COMP:

[0065] - If the average product sums W step Greater than the average product W tec If Comp_adj is positive, then servo compression COMP is increased;

[0066] - If the average product sums W step Less than the average product W tec If Comp_adj is negative, then the servo compression COMP is reduced.

[0067] In the previous example, by using the average product W tec Subtract the sum of the average products W step To calculate the value of Comp_adj,

[0068] Comp_adj = W step - W tec

[0069] In an alternative to the previous example, the proportional value Comp_adj_U can be calculated as the average open-circuit voltage U. 0_step_avg With reference open-circuit voltage U 0_tec The difference:

[0070] Comp_adj_U = U 0_step_avg – U 0_tec

[0071] Among them, U 0_step_avg It is the quantity n d The average open-circuit voltage of the stepped open-circuit voltage pulses of the present invention, which is obtained by summing the average products W step Divide by total ignition delay t dT To calculate, where t dT It is all the pulses n considered d The sum of partial ignition delays t dT = (t d0 + t d1 + t d2 As mentioned earlier, U 0_tec It is the reference open-circuit voltage, that is, the open-circuit voltage according to the appropriate standard processing technical parameter set, which provides a well-defined processing result, including the defined undercut.

[0072] In this variant, Comp_adj_U—that is, the average open-circuit voltage U— 0_step_avg With reference open-circuit voltage U 0_tec The difference is used to adjust the servo setpoint value for gap width control. All steps except the Comp_adj_U calculation remain unchanged.

[0073] The servo setpoint value used for gap width control is the servo reference voltage or servo reference delay, or a value derived from them.

[0074] Reference open-circuit voltage U 0_tec Typically determined and provided by the machine tool manufacturer, and stored in the control unit's technical database where it is available for use. Therefore, according to the present invention, the reference open-circuit voltage U... 0_tec Previously stored in a standard machine database.

[0075] According to an embodiment of the present invention, the axis position (X, Y, Z, C) is continuously acquired, and the change in gap width is derived from the change in the axis position.

[0076] This axis position is the immediate axis position of the tool and / or workpiece, which is obtained individually for each of the axes (Z, X, Y, C). For example, an exemplary die-cutting EDM machine carries the workpiece on a stationary stage resting on a machine base, and performs translation of the tool electrode by means of three linear axis sliders mounted in series on the base, plus a rotating C-axis placed in series with the Z-axis on the machine head. For simplicity, we only consider the Z-axis movement performed by the machine head carrying the tool electrode. The change in Z-axis position is dZ = Zi – Z i-1 This corresponds to the same change in gap width along the same axial direction.

[0077] like Figure 7 (Prior art) and Figure 8 As shown in this invention, the Z-axis position of the machine head is preferably sampled continuously by means of a glass scale LS or by means of an erosion axis rotary encoder E. This axis position (e.g., referred to as the machine vertical axis position Z) does not show absolute clearance width information because the bottom of the cavity is not precisely known and is continuously eroded. However, the succession of the axis position outlines the relative oscillation (change in axis position) around the immediate axis position setpoint value. This is very useful information because variations in the machining results are related to this oscillation, such as… Figure 11 As illustrated in the diagram. There, you can see how the axis position oscillates around the white line, which represents an ideal, stable feed.

[0078] The immediate axis position setpoint value is given, for example, by the actual servo compression value COMP, and the effective axis position is continuously acquired. The relative clearance width is derived from the difference between the axis position setpoint value and the acquired axis position. In this way, large and correspondingly small clearance widths can be safely determined, and countermeasures can be formulated. Specifically, certain electrical parameters are adjusted in real time as a function of the actual clearance width, thereby enhancing the process.

[0079] Figure 4 The illustration is excerpted from W. König, F. Klocke, Fertigungsverfahren Abtragen and Generieren, 4.Auflage, Springer, p. 40, illustration 2.20, 2007, ISBN 3-540-23492-6. As shown, the relative electrode wear υ and material removal rate V... W It is the gap width S EK The function is . Therefore, if the gap width changes and no action is taken, both of these results will change, thus affecting the machining quality.

[0080] In one embodiment, one or more of the following parameters are adapted as a function of axis position:

[0081] - Open circuit voltage U o ,

[0082] - Pulse pause t o ,

[0083] - Current pulse amplitude,

[0084] - Current pulse shape.

[0085] These actions are designed to counteract the effects of head oscillations around the optimal operating point (the attached diagram is not to scale):

[0086] - As a function of the current machine head position to adapt the open-circuit voltage U0, i.e., the partial open-circuit voltage U o0 U o1 U o2 At least one of the following: as a function of the machine head position, if the gap becomes larger, then U0 is increased; and if the gap becomes smaller, then U0 is decreased. Figure 5 :U o0'' > U o0' );

[0087] - Used as a function of the current machine head position to adapt the pause time t between continuous processing pulses. o By reducing the pause time t when the gap is large. o And when the interval is small, increase t o This can increase the productivity of the process without increasing the risk of arc discharge. Figure 5 :t 0'' < t 0' );

[0088] - As a function of the machine's current head position to adapt the current pulse shape I shape And / or current amplitude I, as follows: When the gap is small, the tendency to wear towards higher electrodes increases, and wear mainly occurs at the beginning of discharge; therefore, a moderate initial current slope and a slightly reduced current are applied. When the gap is large, a steeper slope and nominal current are applied ( Figure 5 :I'< I'').

[0089] Through the action of the present invention, process performance is maintained and enhanced independently of machine head oscillation. For this reason, according to embodiments of the invention, controlled oscillations are superimposed on the axis position. This superimposed micro-oscillation comprises relative movement of the workpiece and electrode at frequencies well above the intrinsic frequency of the machine, with amplitudes ranging from 1 to 30 µm, depending on the gap width. The superimposed micro-oscillation can be implemented using the machine axis, for example, movement of the machining head along the Z-axis, or using auxiliary equipment, such as an electrode holder actuated by a piezoelectric motor. In this way, a continuous flushing action is induced in the gap, which promotes the mixing of dielectrics in the gap and the removal of eroded particles, thereby further improving process performance.

[0090] For electrical parameters—such as the initial open-circuit voltage U0, the pulse pause time t between consecutive pulses— o ( Figure 5 :t0'' < t 0' ), the current I of the normal discharge pulse Figure 5 (I'< I'') and the shape of the current pulse I shape The adaptation can be completed in real time within a control loop, a process that typically takes microseconds.

[0091] Accordingly, in another embodiment, one or more of the following parameters are adapted in real time as a function of waiting times d0 and d1:

[0092] - Pulse pause t o ,

[0093] - Pulse current amplitude I,

[0094] - Pulse current shape I shape .

[0095] These parameters can be adjusted with each pulse or pulse; therefore, parameter fitting is much faster than it can be done by sampling the axis position.

[0096] The following function is constructed for each pulse or as an average of multiple pulses:

[0097] Gap_rel = k a *t d0 + k b *t d1 + k c *I*T

[0098] Where Gap_rel is a value proportional to the gap width, t d0 and t d1 It's ignition delay, k a k b and k c is the fitting coefficient, I is the pulse current amplitude, and T is the pulse length, i.e., the applied pulse on-time.

[0099] According to this embodiment, as explained above, electrical parameters, such as the pause time t between consecutive pulses, are adapted in real time pulse by pulse as a function of Gap_rel. o The amplitude of the pulse I and the shape of the current pulse I shape (That is, if Gap_rel increases, then the pause time t decreases.) o etc).

[0100] Now for reference Figure 7 The diagram illustrates a clearance control loop based on state-of-the-art technology. For simplicity, only the Z-axis is shown here.

[0101] The gap voltage U is measured using the gap acquisition unit 64. Gap Then, the 60 ignition delay t is extracted from the gap voltage signal in a known manner. d The ignition delay value is then filtered according to, for example, EP 0 333 170 61. The filtered and averaged ignition delay value is subtracted from the value derived from the compression value COMP 62, and then further processed, for example, by a proportional-integral-derivative transfer function PID 63, and its output ꜫo is fed as a new position vector input to a machine interpolator (IPO), which assigns the new position vector to the axis involved.

[0102] In this way, the actual value of the ignition delay is controlled based on the servo compression COMP.

[0103] The simplified control rules are:

[0104] Delay [%] = (k offs -COMP*k comp )*100 / k offs

[0105] For example, if k offs = 128 and k comp =2.56

[0106] If COMP = 0, then delay = 100% (open-circuit gap).

[0107] If COMP = 50, then delay = 0% (minimum gap).

[0108] Where delay [%] is the ignition delay expressed as a percentage, k offs It is the offset coefficient, k comp It is an adjustment factor, which is used to set the servo compression COMP (servo setpoint value) with respect to the ignition delay.

[0109] refer to Figure 9 The diagram illustrates an alternative gap width control loop according to the present invention. Figure 8 The control loop shown forms a control, which uses the extracted average pulse voltage as a feedback signal from the process. The gap voltage U is measured by means of the gap acquisition unit 64. Gap Then, the average pulse voltage 58 is extracted from the gap voltage signal, and then filtered 59 in a known manner.

[0110] If, instead of ignition delay, average pulse voltage is used to control the gap width, then a similar control rule applies: Average pulse voltage [%] = Average pulse voltage * 100 / U0

[0111] Average pulse voltage [%] = (k offs -COMP*k comp )*100 / k offs

[0112] If COMP = 0, then Voltage_ave% = 100% (open gap = U0).

[0113] The third method for controlling gap width (very similar to...) Figure 9 The method illustrated in the figure utilizes the average gap voltage feedback signal from the process, rather than the average pulse voltage. This method is advantageous when the pulse duty cycle (the ratio of on time to off time) is small. In these cases, measuring the total average gap voltage (i.e., including the off time as well) conveys more information about gap contamination.

[0114] However, as mentioned above, in practical applications, such as Figure 3 Increasing the open-circuit voltage as shown can also increase the undercut, leading to geometric inaccuracies and reduced reproducibility. Therefore, it is necessary to try using the same average open-circuit voltage as the one in the standard machine database (i.e., the reference open-circuit voltage U). 0_tec The process can be modified to be compatible with existing application technologies and achieve the same undercut, or even work with a smaller gap, thereby increasing the copy accuracy of the EDM process.

[0115] According to the present invention, the initial open-circuit voltage level U o0 Set to a lower reference open-circuit voltage U for processing according to the standard machine database. 0_tec This will generate the known undercut. As mentioned earlier, the initial open-circuit voltage U o0 Then gradually increase the amount to promote breakdown.

[0116] In other words, when the method of the present invention replaces a reference open-circuit voltage U having, for example, 100V... 0_tec In the standard processing mode, an initial open-circuit voltage U of, for example, 75V is selected. o0 What we see is that a trade-off between increasing voltage (and thus increasing processing speed) and maintaining the desired small undercut (side clearance) is the priority.

[0117] Now for reference Figure 6 and Figure 8 In a preferred embodiment, for each voltage pulse, the compression adjustment processor unit 80 constructs a partial open-circuit voltage level U. o0 U o1 U o2 Multiplied by partial ignition delay t d0 t d1 td2 The product W step :

[0118] W step = t d0 *U o0 + t d1 *U o1 + t d2 *U o2

[0119] Here, td0 and t d1 This can correspond to predetermined waiting times d0 and d1. This means that, for example, if breakdown occurs after the second waiting time d1, then t... d0 = d0, and t d1 = d1 (see Figure 3 ).

[0120] Similarly, for each voltage pulse, the compression adjustment processor unit 80 constructs the ignition delay t. dT Multiply by the open-circuit voltage U 0_tec The product W tec :

[0121] W tec = t dT * U 0_tec

[0122] Among them U 0_tec This is the expected open-circuit voltage, for example, from a standard application technology database. Ignition delay t dT It is the sum of the partial ignition delays of the pulse, t. dT = t d0 + t d1 + t d2 .

[0123] These values ​​are filtered using a low-pass filter to match the control loop time constant, and then Comp_adj is calculated.

[0124] Comp_adj = k1*(W step -W tec )

[0125] Finally, add Comp_adj to COMP, such as Figure 8 The control loop of this invention is illustrated in the diagram.

[0126] In this way, if the open-circuit voltage is too high, servo compression is increased, and this brings that voltage to U. 0_tec If the open-circuit voltage is too low, then for the same purpose, reduce servo compression.

[0127] k1 is a factor that is optimized to ensure that the undercut generated by using the pulse of the present invention is equal to the reference open-circuit voltage U according to standard technology. 0_tec The generated undercuts remain the same.

[0128] Furthermore, according to an alternative embodiment of the invention, by using the reference open-circuit voltage U 0_tec Reducing it to a very small value minimizes the undercut.

[0129] By applying the actions disclosed above (adapting electrical parameters), process control operates simultaneously in both the mechanical frequency range (gap width control loop) and the discharge frequency range (which is approximately 1000 times faster).

[0130] Comp_adj processing unit 80 adjusts the open-circuit voltage, i.e., the partial open-circuit voltage U. o0 U o1 U o2 And / or waiting times d0, d1, so that the average value is equal to the reference open-circuit voltage U. 0_tec .

[0131] Additionally, to optimize machine performance, another aspect of the method of the present invention includes: adapting the servo setpoint value in real time by means of the Comp processing unit 62, i.e., servo compression COMP.

[0132] These two algorithms can work simultaneously: Comp_adj controls the open-circuit voltage, and the adaptation of the setpoint value of the servo compression COMP keeps the EDM process at its optimal operating point. This optimal operating point is where the expected machine performance is highest. Due to the stochastic nature of the process and the micro and macro geometry of the cavity, this point changes continuously during processing.

[0133] The algorithms applicable here are those known in the most advanced techniques of optimal control, such as hill climbing, simplex algorithm, binary search, reinforcement learning, etc. Figure 10 The figure illustrates an exemplary, very simplified embodiment of such an optimization algorithm designed to maximize the discharge frequency.

[0134] The algorithm assigns a value to the setpoint COMP and waits for a settling time t. s To stabilize the process, for example, by adjusting the discharge amount N during this period. d Perform counting, or determine the discharge frequency f d = N d / t s .

[0135] Then, it will discharge N d Compare with previously obtained discharge amounts, and if N dIf the value is higher, then COMP is changed in the same direction; otherwise, COMP is changed in the opposite direction. Of course, more complex algorithms can be applied, as mentioned above.

[0136] The discharge quantity N can be obtained in this way. d :

[0137] - Current pulses are measured by a current sensor;

[0138] - For example, a voltage comparator converts current pulses into digital information;

[0139] - Use known methods (e.g., FPGA, digital counter integrated circuit, microcontroller, etc.) to digitally count pulses over a period of time.

[0140] EP1918054 discloses the use of signal 32 to obtain N. d Alternative methods, in Figure 6 The diagram in the middle is shown.

[0141] Assume the generator operates in isopaque mode (meaning the current in each pulse has the same duration t). e If it is necessary to optimize the material removal rate, it is preferable to use the discharge frequency f. d As a preferred parameter, this is because, as disclosed in EP1918054, material removal is related to the discharge frequency f. d Proportional.

[0142] Therefore, according to an embodiment of the present invention, the discharge frequency f is obtained. d And adjust the servo setpoint value so that the process is kept at the operating point where the discharge frequency fd is the highest.

[0143] Assuming the tool electrode wear rate is known, an alternative might be to use the tool electrode's advance speed v into the workpiece. adv As a parameter to be optimized, the advance speed reflects the removal at both electrodes (i.e., material removal at the workpiece and wear at the tool electrode), so by knowing the wear rate, the material removal in the workpiece can be inferred. The second factor also includes tool electrode wear.

[0144] Therefore, according to a further embodiment of the present invention, the forward velocity v of the electrode entering the workpiece is obtained. adv And adjust the servo setpoint value so that the process is maintained at the aforementioned forward speed v. adv At the largest work site.

[0145] The method of the present invention has been exemplarily described using first and second waiting times d0 and d1 with corresponding increases in open-circuit voltage. The method can be extended to additional waiting times d if required by the application. i and voltage step U oi Or it can be extended to a single element. The latter can be beneficial when processing materials with poor conductivity. Therefore, according to one embodiment, if there is an additional waiting time d... i d (i+1) If no discharge occurs, the open-circuit voltage will be further increased to the open-circuit voltage level U. oi U o(i+1) ...or add a single time after a single wait time.

[0146] Furthermore, instead of voltage steps, voltage ramps or other forms of gradually increasing open-circuit voltages can be applied. Here, W is calculated by summing the products of the partial open-circuit voltage and the partial ignition delay time, taking into account the actual open-circuit voltage pulse shape. step_pulse The other steps for adjusting the servo setpoint value remain unchanged.

[0147] Waiting times d0, d1, d i Preferably, it is the maximum open-circuit voltage applied for time t. dmax The predetermined value or percentage. However, optimizing the waiting time is complex. This is especially true for the partially open-circuit voltage U. o0 U o1 U o2 The same applies. Preferably, for example, to maximize the discharge frequency f. d To identify the predetermined optimal waiting time, machine learning algorithms (such as, for example, multi-objective black-box optimization) are used. Here, a similar type of experiment is conducted with varying processing conditions, including varying waiting times and / or varying partial open-circuit voltage U. o0 U o1 U o2 Furthermore, the discharge frequency or other parameters indicating the material removal rate are determined, and the ML algorithm is trained by feeding varying processing conditions, such as varying waiting times and / or varying partial open-circuit voltages and the resulting discharge frequency.

[0148] refer to

[0149] s gap width

[0150] t d Ignition delay time, or ignition delay

[0151] t d0 t d1 td2 t di Partial ignition delay time

[0152] t dT Total ignition delay

[0153] t dmax Maximum open-circuit voltage application time

[0154] t o Pulse pause

[0155] d0, d1, d i Waiting time

[0156] U o Disconnect voltage, open circuit voltage, idle voltage

[0157] U o0 Initial open-circuit voltage level

[0158] U o1 Second open-circuit voltage level

[0159] U o2 Third open-circuit voltage level

[0160] U 0_tec Reference open circuit voltage

[0161] U e Processing voltage, discharge voltage, combustion voltage

[0162] I e Processing current

[0163] W step Partial open-circuit voltages U0, U1, U2 multiplied by partial ignition delay t d0 t d1 t d2 The average product;

[0164] W tec Set the reference open-circuit voltage multiplied by the total ignition delay t. d product

[0165] I shape Pulse current shape

[0166] COMP compression, servo setpoint value

[0167] I Current Amplitude

[0168] The pulse on-time applied by T.

Claims

1. A method for electrical discharge machining of a workpiece, wherein a plurality of discrete electrical discharge machining pulses are applied to a gap between an electrode and the workpiece, and wherein an open circuit voltage is applied between the electrode and the workpiece to induce electrical discharge, characterized in that, - the open circuit voltage of the current pulse is initially applied at an initial open circuit voltage level U o0 and - if no discharge has occurred within the first waiting time do, increasing the open circuit voltage to a second open circuit voltage level U o1 and - if no discharge has occurred within a second waiting time d1, further increasing the open circuit voltage to a third open circuit voltage level U o2 and - Calculate the open-circuit voltage level U based on multiple pulses o0 U o1 U o2 Multiplied by partial ignition delay t d0 t d1 t d2 The average sum of products W step And the sum of the average products W step With the average product W tec The average product W is compared. tec It is the previously set reference open-circuit voltage U 0_tec Multiply by total ignition delay t dT The product of, and - adjusting a servo setpoint value for gap width control such that if the average product sum W step is greater than the average product W tec , the gap width is decreased and the servo setpoint value is adjusted such that if the average product sum W step is less than the average product W tec , the gap width is increased, wherein the servo setpoint value is a servo reference voltage or a servo reference delay or a value derived from a servo reference voltage or a servo reference delay.

2. The method for electric discharge machining of a workpiece according to claim 1, characterized by, The reference open circuit voltage U 0_tec Previously stored in the standard machine database.

3. The method for electric discharge machining of a workpiece according to any one of claims 1-2, characterized in that, - the axis positions (X, Y, Z, C) are continuously acquired, and the variation of the gap width is derived from the variation of the axis positions.

4. The method for electric discharge machining of a workpiece according to claim 3, wherein, - one or more of the following parameters are adapted as a function of the axis positions: - open circuit voltage level U o , - pulse pause t o , - the current pulse amplitude I, - current pulse shape I shape .

5. The method for electric discharge machining of a workpiece according to claim 3, wherein, - a controlled oscillation is superimposed to the axis positions.

6. The method for electric discharge machining of a workpiece according to any one of claims 1-2, wherein, as a function of the partial ignition delay t d0 , t d1 , and t d2 to adapt in real time one or more of the following parameters: - pulse pause t o , - the pulse current amplitude I, - Pulse current shape I shape .

7. The method for electric discharge machining of a workpiece according to any one of claims 1-2, wherein, acquiring a discharge frequency f d and adjusting the servo setpoint value so as to maintain the electrical discharge machining process at a maximum operating point in which the discharge frequency f d is at a maximum.

8. The method for electric discharge machining of a workpiece according to any one of claims 1-2, wherein, - the advance speed of the electrode into the workpiece is acquired, and the servo setpoint value is adjusted so as to maintain the electrical discharge machining process at an operating point where the advance speed is maximum.

2. The method of claim 1, wherein the advance speed of the electrode into the workpiece is derived from the axis positions.

3. The method of claim 1, wherein the advance speed of the electrode into the workpiece is derived from the axis positions and from the open circuit voltage.

4. The method of claim 1, wherein the advance speed of the electrode into the workpiece is derived from the axis positions and from the open circuit voltage and from the current pulse amplitude.

5. The method of claim 1, wherein the advance speed of the electrode into the workpiece is derived from the axis positions and from the open circuit voltage and from the current pulse amplitude and from the pulse current amplitude.

6. The method of claim 1, wherein the advance speed of the electrode into the workpiece is derived from the axis positions and from the open circuit voltage and from the current pulse amplitude and from the pulse current amplitude and from the controlled oscillation.

7. The method of claim 1, wherein the advance speed of the electrode into the workpiece is derived from the axis positions and from 9. The method for electric discharge machining of a workpiece according to any one of claims 1-2, wherein, If no discharge occurs within a further waiting time d i , d (i+1) …, the open circuit voltage is further increased to an open circuit voltage level U oi , U o(i+1) …, or is increased by a single step after a single waiting time.

Images