Pulse and gap control for electrical discharge machining equipment

Abstract

A method of detecting the state of a gap between an electrode and a workpiece in electrical discharge machining (EDM) equipment, the method including the steps of: prior to generating an electrical discharge to remove material from the workpiece, applying a low energy checking pulse across the gap during a checking phase period (T.sub.c); and inferring a short circuit gap state when the gap current exceeds a current threshold (I.sub.T).

Claims

1. A method of controlling electrical discharges in EDM equipment, the method including the steps of: detecting the state of a gap between an electrode and a workpiece in the EDM equipment by, prior to generating an electrical discharge to remove material from the workpiece, applying a low energy checking pulse across the gap during a checking phase period (T.sub.c), and inferring a short circuit gap state when the gap current exceeds a current threshold (I.sub.T), and selectively generating the electrical discharges according to the detected gap state; wherein the detected gam state comprises at least one of the short circuit gap state, an arcing gap state, and a normal gap state; wherein when the short circuit gap state is inferred there will be no discharge pulse applied into the gap, the timing of discharge phase T.sub.on will be interrupted, and the deionise phase timing is extended; wherein when the arcing gap state is inferred the deionise phase will be further extended after the discharge phase; wherein when the normal gap state is inferred, the deionise phase timing will be reset to its predetermined thine; and wherein discharge phase timing is controlled according to the current pulse shape, and the current pulse shape is determined by computing the integration of current pulse during discharge phase.

2. A method according to claim 1, wherein the integrated current pulse is compared with a current integration threshold, and where the integrated current is larger than its threshold, then the discharge phase will cease.

3. A method of generating a smooth servo feed command that is capable of controlling and maintaining the optimised gap between an electrode and a workpiece in an electrical discharge machining (EDM) equipment in which an electrical discharge is generated across the gap to remove material from the workpiece, the method including the step of: if an normalised average build-up phase time (T.sub.d%) exceeds a maximum threshold, setting the relative feed-rate of workpiece towards electrode to a constant maximum rate.

4. A method according to claim 3, and further including the step of: if the normalised average build-up phase time (T.sub.d%) exceeds a lower threshold (T.sub.dref) and is less than a max threshold, then setting the relative feed-rate of workpiece towards electrode as a function of the error between build-up phase time (T.sub.d) and lower threshold (T.sub.dref).

5. A method according to claim 3, and further including the step of: if the normalised average build-up phase time (T.sub.d%) is less than the lower threshold (T.sub.dref) and if the average gap energy during the build-up phase the exceeds an Energy threshold (Eng.sub.ref), then setting the relative feed-rate of workpiece towards electrode as a function of the error between average gap energy during the build-up phase and Energy threshold (Eng.sub.ref).

6. A method according to claim 3, and further including the step of: if the normalised average build-up phase time (T.sub.d%) is less than the lower threshold (T.sub.dref) and if the average gap energy during the build-up phase is less than an Energy threshold (Eng.sub.ref), setting the relative feed-rate of workpiece away from electrode as a function of the average gap energy during the build-up phase and Energy threshold (Eng.sub.ref).

7. EDM equipment for generating an electrical discharge to remove material from the workpiece, the equipment including: a controller for controlling the feed-rate of a workpiece according to claim 3; and servo system means for displacing the workpiece in response to signals from the controller; wherein the electrode is arranged to he rotated and coolant flushed into the gap during the electrical discharge; and wherein the average gap energy and average build up phase time is used as a feedback signal for the gap controller.

8. A gap controller for use in EDM equipment, the gap controller including circuitry configured to: receive average gap energy and average gap voltage signals, derived from gap voltage and gap current feedback signals, from a pulse controller; generate feed-rate control signals from the average gap energy and average gap voltage signals; and send the feed-rate control signals to a servo system means to control the feed-rate of a workpiece.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) The invention will now be described in further detail by reference to the accompanying drawings. It is to be understood that the particularity of the drawings does not supersede the generality of the preceding description of the invention.

(2) FIG. 1 is schematic diagram showing basic components of an electrical discharge machining (EDM) equipment;

(3) FIG. 2 is a graph showing idealised gap voltage and gap current occurring during application of a DC voltage pulse across the gap between a workpiece and a tool in EDM equipment;

(4) FIG. 3 is a schematic diagram showing elements of a pulse controller and gap controller forming part of the EDM equipment shown in FIG. 1;

(5) FIG. 4 depicts gap voltage and current thresholds used by the pulse controller and gap controller shown in FIG. 3;

(6) FIG. 5 shows gap voltage, gap current and a series of switching signals from the EDM equipment depicted in FIG. 1 during application of a checking pulse across the workpiece/tool gap;

(7) FIG. 6 shows the gap voltage, gap current and switching signals present in the EDM equipment shown in FIG. 1 during the application of a DC voltage pulse across the gap between the workpiece and the tool during operation of the equipment;

(8) FIG. 7 is a schematic diagram showing computational blocks of the gap controller shown in FIG. 3; and

(9) FIG. 8 is a flowchart depicting computational steps performed by the pulse controller shown in FIG. 3 during application of a DC voltage pulse across the gap between the workpiece and the tool during operation of the equipment;

(10) FIG. 9 shows different current pulse waveform that can be expected when eroding materials with different conductivity.

DETAILED DESCRIPTION

(11) Referring now to FIG. 1, there is shown generally electrical discharge machining (EDM) equipment 10. The equipment 10 includes a multi-axis machine 12 including a number of machine axes 14 driven by corresponding axis drives 16 in order to position and rotate an electrode 18 in relation to a workpiece 20. The multi-axis machine 12 is controlled by a computer numerical controller (CNC) 22 which acts to automate the various machining processes implemented by the multi-axis machine 12. An example of a CNC controlled multi-axis machine is provided in U.S. Pat. No. 5,604,677 in the name of the present Applicant.

(12) The EDM equipment 10 further includes a power module 24 for applying rapidly recurring current discharges to the gap between the electrode 18 and the workpiece 20 in order to remove material there between. Operation of the power module 24 is governed by an EDM controller 26 providing pulse discrimination, pulse timing control, gap control, gap controller optimisation and parameter optimisation for the EDM process.

(13) The EDM equipment 10 further includes a series of sensors 28 for transmitting gap current and gap voltage signals to a signal conditioning unit 30 which in turn provides feedback current and voltage signals to the EDM controller 26.

(14) As shown in FIG. 2, several phases can be identified during an EDM process. During checking phase a low energy pulse is applied across the gap in detecting the condition of the gap before initiating a high voltage pulse. During an initial build-up phase 40, a no-load voltage V.sub.o is applied across the workpiece/tool gap. Typically, this no-load voltage V.sub.o has a range of values between 60 volts and 400 volts.

(15) Upon establishment of the no-load voltage V.sub.o, a strong electric field is established between the electrode 18 and the workpiece 20. Due to the attractive force of the electric field, at the shortest local distance between the tool and the workpiece (the gap) there is a build-up of particles from the machining process which float in the dielectric fluid. This forms the electrical breakdown and the electrons begin to move towards the positively charged electrode in a discharge phase 42. On their way, the associated electrons collide with the neutral particles from the machining process and the dielectric fluid.

(16) An avalanche ionisation process is set off, in which a large number of negative and positive ions are generated in the discharge phase 42. The ionisation initiates creation of an electro-conductive zone between the workpiece and the tool, thus causing electrical discharge. Through electrical discharge, electrical energy is transformed into thermal energy. A discharge zone is formed at temperatures as high as 40,000 Celsius. Such high temperatures cause local heating, melting, evaporation and incineration of the workpiece.

(17) In the discharge phase 42, the gap voltage decreases from the no-load voltage V.sub.o to a discharge voltage V.sub.e. Discharge current increases during the discharge phase from zero to a maximum discharge current value I.sub.e (a typical current range from 0.5 amps to 30 amps) during discharge duration time T.sub.on (typically 500 nanoseconds to 1 milliseconds).

(18) At the end of discharge duration (T.sub.on), the MOSFET is switched off, causing a disruption of the gap current supply. This results in the annihilation of the discharge zone, causing abrupt cooling which leads to an explosive flushing of melted matter and solid particles of the workpiece surface in the deionise phase 44.

(19) The EDM controller 26 acts to control the feed rate of the electrode 18 with respect to the workpiece 20 in order to maintain an appropriate gap there between, as well as causing application of a series of rapidly recurring DC voltage pulses between electrode 18 and the workpiece 20 in order to remove material from the gap. The EDM controller 26 also acts to monitor various machine parameters in order to optimise the machining process and prevent application of the DC voltage pulses under undesirable gap states.

(20) FIG. 3 shows various elements forming part of the EDM controller 26 including a pulse timing controller 50, a gap controller 60, an Analogue-to-Digital (AD) converter 62, Direct Memory Access (DMA) 64, pulse discrimination unit 66, a computation unit 68 and an EDM optimisation unit 70. As can be seen from this figure, several of these elements are implemented in Digital Signal Processing (DSP) technology whilst other elements are implemented by a Field Programmable Gate Arrays (FPGA). In operation, the feedback voltage and current signals from the signal conditioner 30 representative of the gap current and voltage measured by the sensors 28, are provided to the analogue to digital converter 66 for digitisation. The digitised current and voltage signals are then transferred to FPGA memory via direct memory access channels without interrupting DSP's CPU 64.

(21) The pulse discrimination unit 66 accesses the digitised current and voltage values stored in the DMA 64 in order to infer the state of the gap between the electrode and the workpiece. To assist in this determination, the pulse discrimination unit 66 stores a series of current and voltage thresholds, represented in FIG. 4. In this example, the pulse discrimination unit 66 stores a current threshold I.sub.T, an open circuit voltage threshold V.sub.TOC, a normal gap state voltage threshold V.sub.TN, an arc gap state voltage threshold V.sub.TA and a short circuit gap state voltage threshold V.sub.TSC. The pulse discrimination unit 66 relies upon the values of the voltage and current read from the DMA and compute the average discharge voltage and current 64 The comparison made to each other and to the stored voltage and current thresholds in order to infer the state of the gap between the workpiece 20 and the tool 18.

(22) Together with timing control information from the EDM optimisation unit 70, the pulse controller 50 relies upon the gap state information from the pulse discrimination unit 66 to selectively generate switching commands to control operation of the power module 24 and thus generating the high voltage pulses used in the EDM process. The computation unit 68 computes average voltage and average pulse energy from a series of pulses applied across the electrode/workpiece gap.

(23) In that regard, the pulse controller 50 and gap controller 60 run at different software cycles. The pulse discrimination 66 acts to check the gap state (pulse type) every single discharge pulse (which may typically be between 0.5 and 300 microseconds). The gap controller 60 runs at a much slower cycle. In a single gap control cycle (which may typically be from 1 to 4 milliseconds long), many discharge pulses can take place. In order that feedback signal provided from the computation unit 68 to the gap controller 60 is optimised, average values of voltage and pulse energy are computed and stored by the computation unit 68 for use by gap controller 60.

(24) The EDM optimisation unit 70 provides reference timing information to the pulse timing controller 50 as well as providing a series of control signals to the gap controller 60 in order for a feed rate command signal to be generated and provided to the computer numeric controller 22 which controls the advancement of the tool 18 at the desired rate.

(25) The EDM operations performed by the pulse discrimination unit 66 and the pulse logic controller 50 will now be explained with reference to FIGS. 5 and 6.

(26) FIG. 5 depicts the application of a checking pulse 80 across the workpiece/tool gap. In this example, a low energy checking pulse having a value of 48 volts is applied across the gap, but current flow 82 is immediately detected across the gap. The pulse discrimination unit 66 determines that the current detected during this checking phase exceeded the current threshold I.sub.T and thereby infers that a short circuit exists across the gap. In this case, application of high voltage pulse across the gap would be harmful to the building up of debris and electrode wear. The counter for checking phase period (T.sub.c) and discharge phase period (T.sub.on) will be interrupted. It can also be seen from this figure that whilst a 48V low current switching signal is applied during the checking phase period, the current flow across the gap during this checking phase results in no high voltage switching signal or high current switching signal being generated. The timing for deionise phase (T.sub.off) is further extended from the predefined timing depending on the gap state of previous pulse. If the gap state of previous pulse is short circuit, the T.sub.off is further extended from the previous stored T.sub.off value. The T.sub.off value will reset to its predefined timing after the gap state is recovered from short circuit to normal state.

(27) Referring now to FIG. 6, a low energy pulse 90 having a value of 48 volts is applied across the gap, but no current flow is immediately detected across the gap, which means the gap is filled with fresh dielectric. Since a short circuit gap state is not detected during the checking phase period, the power module 24, under control of the pulse timing controller 50, causes establishment of a high voltage pulse 92 (in this example, having a value of 300 volts) across the gap. A counter in the pulse timing controller 50 for counting the total ignition delay time (T.sub.d) is initiated. Depending on the gap distance, gap current 94 starts to flows in the discharge phase of the EDM process once the plasma channel is established. During the discharge duration time T.sub.on, the gap voltage falls (in this example) to a value of 25 volts.

(28) The discharge duration (T.sub.on) is determined by the current pulse shape. Referring to FIG. 9, the current pulse shape can varies according to the conductivity of workpiece material. To achieve a better surface quality of the workpiece, the total energy per pulse has to be controlled to prevent overheating of localise area by checking the current pulse shape. The current pulse shape is determined by computing the integration of the gap current feedback signal ( I dt) during the discharge phase in the pulse timing controller 50. The duration of T.sub.on is determined by comparing ( I dt) with a current integration threshold which is provided by the EDG optimisation block 70. If ( I dt) is larger than the current integration threshold, the discharge duration will expire, or else the T.sub.on will continues till its reached its threshold. At the end of the discharge phase, both the gap voltage and gap current are switched off and are therefore expected to fall to zero during the deionisation phase.

(29) As has been shown in FIG. 8, during the idealised EDM process depicted in FIG. 6, the EDM controller 26 acts to measure various parameters in order to detect the state of the gap between the electrode 18 and workpiece 20 and if necessary take a corrective action.

(30) For example, if after the high gap voltage is applied during the build-up phase T.sub.d, an open circuit gap state is inferred by the EDM controller 26 when the gap voltage exceeds the open circuit threshold V.sub.TOC shown in FIG. 4 for longer than a predefined build-up phase period T.sub.d after the high voltage is applied, and if the gap current is less than the current threshold IT. Having detected an open circuit gap state, the EDM controller continue to apply high voltage to the gap until gap current is larger than current threshold IT.

(31) However, if the gap voltage is determined to exceed the normal voltage threshold V.sub.TN which is less than the open circuit threshold V.sub.TOC after the build-up phase period T.sub.d and during the discharge phase period T.sub.on and if the gap current exceeds the current threshold I.sub.T, then the controller 26 infers a normal gap state requiring no corrective action to take place.

(32) A short circuit gap state is inferred by the controller 26 after the build-up phase period T.sub.d and during the discharge phase period T.sub.on when the gap voltage is less than the short circuit threshold V.sub.TSC which is less than both the open circuit threshold V.sub.TOC and the normal threshold V.sub.TN, and the gap current exceeds the current threshold I.sub.T having inferred that a short circuit gap state exists, the discharge duration time (T.sub.on) will be interrupted and sends 0V MOSFET command signals from the controller 26 to the power module 24 in order to switch off the discharge pulse applied to the gap. The deionise phase duration (T.sub.off) will be extended to provide extra time for removing debris from the gap.

(33) Similarly, an arc gap state is inferred by the controller 26 after the build-up phase period T.sub.d during the discharge phase period T.sub.on when the gap voltage exceeds the short circuit threshold V.sub.TSC but is less than the normal threshold V.sub.TN, and the gap current exceeds the current threshold. Once again detection of the arc gap state will further extend the deionise phase duration.

(34) It will be appreciated from the foregoing that following detection of the gap state between the electrode 18 and workpiece 20 according to the foregoing techniques, the EDM controller 26 acts to selectively generate electrical discharges across the gap according to the detected gap state. In this manner application of harmful pulses across the gap during short circuit, arc and open circuit gap states is minimised.

(35) In conjunction with gap state detection so as to avoid of the application of harmful pulses to the gap, the EDM controller 26 generates feed rate commands for use by the computer numerical controller 22 to control the feed rate of the electrode 18 as a function of feedback signals.

(36) An overview of elements forming part of the gap controller 60 is depicted in FIG. 7. The gap controller 60 includes three separate control components 100 to 104 for applying different feedback control algorithms to the electrode feed rate according to the size of the electrode/workpiece gap.

(37) If the normalised average build-up phase time (T.sub.d%) over the pulse cycles occurring in between consecutive gap control cycles exceeds a maximum threshold, it shows the workpiece distance from the electrode is very large. The large gap state control component 100, acts to set the relative feed-rate of workpiece towards electrode to a constant maximum rate.

(38) If the normalised average build-up phase time (T.sub.d%) over the pulse cycles occurring in between consecutive gap control cycles exceeds a lower threshold T.sub.dref but is less than a max threshold, it shows the eroding process starts to take place, then the medium gap state control component 102 acts to set the relative feed-rate of workpiece towards electrode as a function of the error between build-up phase time T.sub.d and lower threshold T.sub.dref. This state is to allow the gap controller to reduce the gap between the workpiece and electrode to an optimum distance for erosion.

(39) The small gap state control component 102 applies one of two different feedback control algorithms depending upon the gap state. If the average build-up phase time T.sub.d is less than the lower threshold T.sub.dref and if the average gap energy during the build-up phase exceeds an energy threshold Eng.sub.ref, then a normal gap state is inferred. In this state, an optimum gap distance is achieved; the feedrate controller will feed the workpiece towards the electrode according to the speed of change in gap distance. The small gap state control component 102 sets the relative feed-rate of workpiece towards electrode as a function of the error between average gap energy during the build-up phase and Energy threshold (Eng.sub.ref).

(40) However, if the normalised average build-up phase time (T.sub.d%) is less than the lower threshold (T.sub.dref) and if the average gap energy during the build-up phase is less than an Energy threshold Eng.sub.ref, then a short circuit gap state is inferred. In this case, the small gap state control component 104 sets the feed-rate of workpiece away from electrode as a function of the average gap energy during the build-up phase and Energy threshold Eng.sub.ref.

(41) A method that uses average gap voltage as feedback signal does not accurately represent the actual gap condition and the speed of change in gap distance. In this invention the objective of gap controller is to control the rate of change in gap distance as opposed to the normal perception of controlling the gap distance. Erosion rate that determines the change in gap distance will vary depending on the actual gap distance, thus giving the objective of controlling the gap distance a challenging task. Average gap voltage signal is shown to be stochastic, it's varies under various uncontrolled physical condition such as dielectric condition, electrode condition, the condition of workpiece regardless of the change in gap distance. The gap controller will not be required to respond to such chaotic and unpredictable phenomena, responding to such phenomena will again cause jerking in servo feed command and unstable erosion process is thus the results. In this invention, average gap energy that represents the change in gap distance is used as feedback signal for gap controller. For a fixed eroding area, the rate of change in gap distance is shown to be proportional to the total energy supplied to eroding gap. By calculating the average energy input into the gap, the rate of change in gap distance can be estimated. A smooth servo feed command can be generated from the gap controller and the workpiece is constantly feed towards the electrode without jerking.

(42) In the implementation shown in FIG. 7, each of the threshold values and feedback values are normalised. That is, the threshold values have a unitary value and feedback values have value less than 1. Such an arrangement enables the three control components 80 to 82 to produce a feed rate gain coefficient which is used to multiply a maximum feed rate value.

(43) Although in the above described embodiments the invention is implemented primarily using FPGA and DSP techniques, in other embodiments the invention may be implemented primarily in software, firmware or hardware using, for example, hardware components such as an application specific integrated circuit/s (ASICs). Implementation of a hardware state machine so as to perform the functions described herein will be apparent to persons skilled in the relevant art. In other embodiments, the invention may be implemented using a combination of both hardware and software.

(44) While the present invention has been described in conjunction with a limited number of embodiments, it will be apparent to those skilled in the art that many alternatives, modifications and variations in light of the foregoing description are possible. Accordingly, the present invention is intended to embrace all such alternatives, modifications and variations as may fall within the spirit and scope of the invention as disclosed.