Five-axis dental milling machine dual-mode layer compensation calibration method and five-axis dental milling machine

Abstract

The present application relates to the technical field of five-axis numerical control milling machine precision calibration, and discloses a five-axis dental milling machine double-mode layered compensation calibration method and a five-axis dental milling machine. The method first calls the probe to collect the original motion reference of the five-axis machine tool, generates compensation parameters of the first basic workpiece coordinate system G54; then obtains the jig reference size data obtained by cutting the standard test cutting tool, calculates the independent offset deviation of each translation axis and rotation axis and adds the incremental correction to the G54 parameters as an incremental correction, to obtain the updated G54 coordinate system; finally, the updated G54 is used as the homologous reference, the exclusive axial correction of each station of the titanium column is superimposed, the coordinate system G55 suitable for the titanium column is generated, and the G55 reuses all the basic compensation parameters of the G54. The scheme can consider the geometric reference of the machine tool body and the actual cutting process deviation, improve the coincidence degree of the compensation parameters and the real processing deviation, ensure the reference consistency of the multi-billet switching processing, and adapt to the batch processing demand of the dental blank.

Description

Technical Field

[0001] This invention relates to the field of coordinate measurement and error correction control technology for five-axis CNC milling machines, specifically applied to a five-axis dental milling machine for digital processing of dentures, suitable for milling dental blanks such as zirconia, titanium alloys and glass ceramics. Background Technology

[0002] During long-term operation, factors such as spindle wear, high-frequency reciprocating oscillation of the A / B rotary axis, micro-plastic deformation of the fixture, and fixture clamping offset will continuously cause the machine tool's machining reference to drift, which in turn leads to dimensional deviations in denture workpieces and machining marks on the occlusal mating surface. This directly affects the dimensional accuracy and clinical fit of the finished product. Therefore, it is necessary to perform calibration operations frequently to ensure cutting accuracy.

[0003] Currently, various automated precision adjustment solutions have been developed for five-axis dental prosthesis milling equipment.

[0004] One type of calibration scheme is based on visual recognition. For example, the invention patent application with publication number CN116974240A discloses a calibration method and coordinate system setting method for a five-axis CNC equipment. This scheme uses image acquisition to carry out calibration; controls the turntable and the oscillating head to return to the origin of the equipment coordinate system, sets a marker point on the rotating axis to be measured, and acquires the rotation trajectory image of the marker point during the rotation of the rotating axis to determine the rotation center of the rotating axis to be measured; based on the relative position of the rotation center and the reference point of the oscillating head, calculates the offset vector of the rotating axis to be measured on the target translation axis, and establishes the workpiece coordinate system based on the offset vector.

[0005] Another type is the offline compensation scheme based on mathematical modeling. For example, the invention patent application with publication number CN117724400A discloses a geometric error analysis and compensation method for a five-axis dental prosthesis machining center. This scheme is an offline compensation scheme at the CNC G-code file level. First, the G-code information is read, then the tool pose is calculated with the help of the ideal kinematic forward solution model, and then the actual motion of each axis is derived by considering the actual kinematic inverse solution model of geometric error. The CNC instructions are then rewritten to achieve geometric error compensation.

[0006] Another type of calibration scheme is based on physical geometric detection, which is closest to the technical scenario of this application. For example, the invention patent application with publication number CN116372609A discloses a method and mechanism for adjusting the accuracy of a five-axis dental prosthesis milling machine. This application is mainly used for the overall accuracy calibration of the machine tool during the installation and debugging stage. First, the machine tool's preliminary external offset parameters are obtained by manual inspection using a dial indicator and an edge finder. Then, a standard correction block is cut and its size is measured. The formula is applied to completely recalculate the entire set of accurate external offset coordinates. The calculated set of accurate external offset coordinates is then input into the tool setting interface of the device to replace and update the system offset parameters. If the verification fails, the entire set of external offset coordinates is recalculated and input into the system again.

[0007] The aforementioned existing solutions achieve machine tool accuracy adjustment through different paths such as visual inspection, offline G-code modeling, and physical geometry tuning. However, none of them can simultaneously take into account the inherent geometric reference of the machine tool body and the comprehensive process deviation caused by fixture deformation and cutting tool deflection under actual cutting conditions. In the case of alternating processing of blank types, especially in the scenario of multi-station processing of titanium pillars, it is also difficult to maintain a stable and unified processing reference, which can easily lead to dimensional fluctuations in alternating production.

[0008] In summary, the existing calibration scheme is still insufficient in taking into account both the inherent geometric reference of the machine tool body and the comprehensive process deviation under actual cutting conditions. In scenarios of multi-station machining of titanium pillars and alternating production of multiple blanks, it is also difficult to maintain a uniform and stable machining reference, which is prone to dimensional fluctuations, and there is still considerable room for improvement. Summary of the Invention

[0009] In view of the shortcomings of the prior art, the purpose of this invention is to provide a dual-mode layered compensation calibration method and equipment for a five-axis dental milling machine, which aims to improve the shortcomings of existing calibration schemes in simultaneously taking into account the inherent geometric reference of the machine tool body and the comprehensive process deviation under actual cutting conditions; and further improve the problems of poor reference consistency and easy dimensional fluctuation in multi-station machining of titanium pillars and alternating production of multiple blanks.

[0010] The technical solution of this invention is as follows: A calibration method for a five-axis dental milling machine includes the following steps: S1. Automatic reference calibration: Call the probe to perform point detection on the standard calibration disk, collect the original motion reference of the five-axis machine tool, and generate and store the compensation parameters of the first basic workpiece coordinate system G54. S2. Deviation data acquisition: Acquire the tooling reference dimension data obtained after cutting with the standard trial cutting tooling; Based on the acquired tooling reference dimension data, the system calculates the independent offset deviation of the X-axis, Y-axis and Z-axis translation axes, as well as the A-axis and B-axis rotation axes, respectively. The offset deviation is used as an incremental correction amount and superimposed on the compensation parameters of the G54 to obtain the updated G54 coordinate system. S3. Layered Compensation: Using the updated G54 coordinate system as the same reference, superimpose the exclusive axial correction amount of each station of the titanium column to generate a second layered workpiece coordinate system G55 adapted to multi-station machining of titanium columns. The G55 reuses all the basic compensation parameters of the G54.

[0011] The effect of the above solution is: I. Multi-source data fusion achieves a leap in compensation accuracy The G54 compensation parameters output by the automatic datum calibration, based on multi-point redundant acquisition and least-squares identification of the standard calibration disk by the probe, can stably reflect the geometric datum state of the machine tool body. However, its limitation is that it cannot cover dynamic deviations under actual machining conditions such as fixture deformation, tool wear, and cutting deflection. This solution obtains the fixture datum dimension data after cutting with a standard trial cutting fixture, and calculates the independent offset deviation of each axis. This deviation naturally includes comprehensive process errors such as spindle load deformation, tool runout, and fixture elastic deformation under actual cutting conditions.

[0012] The offset deviation obtained from the actual cutting is used as an incremental correction amount and superimposed on the initial compensation parameters of G54. Instead of replacing the system offset value with the manual measurement results as an independent parameter as in existing technologies (such as CN116372609A), the essential difference is that: overall replacement means discarding the machine tool body geometric reference obtained by automatic calibration and only retaining the limited data points of a single manual measurement as the basis for all compensation. The reference information source is singular and easily affected by the random error of a single measurement. In contrast, the incremental superposition method of this scheme fully retains the full-axis geometric reference information of automatic calibration, while absorbing the process deviation information of manual measurement. The two form a complementary rather than substitutive relationship, so that the final compensation parameter has both the stability of geometric reference and the sensitivity of process deviation. The dimensions of compensation information are richer, the data redundancy is higher, and the consistency between the correction value and the actual machining deviation is significantly improved.

[0013] II. A coherent, layered coordinate system ensures consistent references when switching between multiple billets. This solution uses the updated G54 as the common datum, and generates the G55 series coordinate system by simply superimposing the specific axial corrections for each station of the titanium column. G55 fully reuses all the basic compensation parameters of G54. The deeper technical effect of this design is that the universal denture blank and the multi-station machining of the titanium column share the same set of machine tool body geometric datums. When the machining task switches between the two types of blanks, it is not necessary to re-acquire the full-axis datum; only the corresponding coordinate system parameters need to be called to complete the datum switching. Since G54 and G55 originate from the same datum source, the datum deviation during the machining of different blanks will not introduce additional systematic errors due to different coordinate system sources, and the dimensional fluctuations of alternating blank production are effectively suppressed.

[0014] In a further preferred embodiment, the dedicated axial correction amount includes offset compensation data for the corresponding titanium column station under different rotational postures, and the compensation data for each station is stored independently.

[0015] The effect of the above solution is that during the machining of titanium pillars, the force state, gravity direction, and offset direction of the clamping contact point between the fixture and the spindle are different at the same station under different rotational postures (such as A0°, A90°, A180°). The difference in the cumulative machining time at different stations will also lead to different degrees of fixture wear and deformation. If the compensation data of multiple stations and postures are stored together or only a single correction value is retained, the deviation change of any station or posture will contaminate the correction data of other stations and postures, resulting in a synchronous deterioration of the machining accuracy of all stations.

[0016] This solution stores the compensation data for each workstation and each rotational posture independently in different address segments of the CNC system parameter area, with isolated read / write paths between data blocks to prevent interference. Each set of compensation data corresponds only to the specific correction amount for the current workstation under the current rotational posture. When the system calls the data, it can directly read the corresponding workstation and posture data block to complete accurate compensation. This design completely eliminates compensation crosstalk between workstations and postures at the data structure level, ensuring dimensional consistency in the machining of titanium pillars in multiple postures and workstations. It also provides a data isolation basis for subsequent independent retesting and correction of individual workstations.

[0017] In a further optimized scheme, the titanium column station supports both positive and negative clamping modes; for different clamping modes, a suitable calculation formula is used to solve for the specific axial correction amount of each station.

[0018] The advantages of the above scheme are as follows: In the forward clamping mode (i.e., forward mounting), the titanium column is inserted into the fixture along the spindle axis, and the clamping force is distributed along the Z-axis. The clamping deviation is mainly manifested as Z-axis height offset and A-axis angle sway. In the reverse clamping mode (i.e., reverse mounting), the titanium column is inserted in the opposite direction, and the clamping force is opposite to that in the forward clamping mode. The clamping deviation is mainly manifested as tool entry point offset and B-axis angle sway. The spatial distribution characteristics of the deviations of the two types of fixtures are completely different. If a unified formula is used to solve the axial correction amount of the two types of fixtures, since the formula parameters cannot simultaneously match the two deviation distributions, the calculation deviation of the correction amount will inevitably be systematically amplified on one type of fixture.

[0019] This solution constructs solution models and matches corresponding calculation formulas for different fixture modes, ensuring that the solution logic for the correction amount accurately corresponds to the actual deviation distribution characteristics of the fixture. This design enables the calibration method to be compatible with two mainstream clamping fixtures in dental titanium column machining. When switching fixtures, there is no need to redevelop the calibration process or adjust system parameters; the corresponding mode can be directly selected to complete the solution of the specific correction amount, significantly expanding the process applicability of the equipment and reducing the calibration complexity of switching between multiple fixture modes.

[0020] In a further preferred embodiment, when acquiring the tooling reference dimensions, the system sets upper and lower limit verification rules for the acquired dimension values ​​to filter out invalid measurement data that exceeds the reasonable deviation range.

[0021] The above solution is effective in the following ways: when manually entering measurement values, errors in reading and input misoperation are prone to occur; by checking the upper and lower limits, abnormal values ​​that deviate from the reasonable range can be directly filtered out, preventing erroneous data from entering the deviation calculation process, avoiding coordinate system compensation disorder caused by incorrect input, and improving the fault tolerance and system reliability of the manual calibration process.

[0022] In a further preferred embodiment, when calculating the independent offset deviations of the X-axis, Y-axis, and Z-axis translation axes, as well as the A-axis and B-axis rotation axes, the two five-axis machine tool configurations, Type_RW_AB and Type_RW_BA, are distinguished, and the analytical calculation formulas for the corresponding configurations are used to solve for the offset deviations.

[0023] The above scheme has the following effect: In the Type_RW_AB configuration, the A-axis is a rotary table and the B-axis is a oscillating spindle. The motion transmission path of the rotation axis is that the tool pose change is jointly determined by the oscillation of the B-axis and the rotation of the A-axis. The oscillation deviation of the B-axis will be coupled to the calculation of the offset of the rotation center of the A-axis. In the Type_RW_BA configuration, the B-axis is a rotary table and the A-axis is a oscillating spindle. The motion transmission path is completely opposite, and the oscillation deviation of the A-axis will be coupled to the calculation of the offset of the rotation center of the B-axis. The axis kinematic transfer functions of the two configurations are different. If the calculation formula of the AB configuration is applied to the BA configuration, the solution results of the deviation of the Y-axis and B-axis will have systematic errors. At best, it will lead to a decrease in the consistency of the machining dimensions of the titanium pillar; at worst, it will cause the risk of tool collision.

[0024] Before calculating the offset deviation, this solution first identifies the machine tool configuration through system parameters and automatically matches the corresponding analytical calculation formula, fundamentally eliminating calculation distortion caused by formula mismatch. This ensures that the calibration method can be adapted to two mainstream five-axis configurations, eliminating the need to develop separate calibration software for different machine models, thus reducing the maintenance costs for equipment manufacturers and the usage threshold for users.

[0025] In a further optimized scheme, before the automatic reference calibration is performed, the probe tool setting verification is completed; if the probe fails to capture, the automatic reference calibration process is terminated and an alarm message is output.

[0026] The above solution is effective in the following ways: when the detector is bumped, dirty, or misaligned, it cannot complete the point detection normally; by adding a capture verification before formal calibration, the process will be terminated and an alarm will be triggered immediately if the capture fails, which can avoid using the erroneous reference parameters obtained from abnormal detection to carry out all subsequent compensation, and prevent the erroneous reference from being passed down from the source, thus protecting the validity of the entire calibration result.

[0027] In a further optimized scheme, after the G54 coordinate system update is completed, the system automatically calls the probe again to perform calibration comparison, and checks the deviation between the comparison result and the updated G54 compensation parameters; configure two levels of out-of-tolerance handling logic: if the single check is out of tolerance, output a calibration reminder signal; if the check is out of tolerance twice in a row, lock the machine tool processing start permission and trigger the recalibration process.

[0028] The above solution is effective in the following ways: After completing the incremental update of the G54 coordinate system, the system is recalibrated using a probe, and the newly generated G54 compensation parameters are verified using actual measurement results. This verifies the correctness of the entire dual-mode calibration output. At the same time, two levels of out-of-tolerance handling logic are set up. If an out-of-tolerance error occurs during a single verification, only an alert signal is output to avoid interrupting production due to momentary anomalies. Only if two consecutive verifications result in out-of-tolerance errors will the processing permission be locked and recalibration triggered, thus balancing production continuity and processing quality control.

[0029] In a further preferred embodiment, when the offset deviation is superimposed on the G54 compensation parameters, the weighting coefficient of the A / B rotation axis configuration is greater than the weighting coefficient of the X / Y / Z translation axis.

[0030] The advantages of the above solution are as follows: In dental milling, the A / B rotary axes bear the feed motion for complex shapes such as the occlusal surface and curved surfaces of dentures. The frequency of reciprocating oscillations in a single machining cycle is typically 3 to 5 times that of the translation axis. Therefore, the backlash accumulation rate and wear rate of the rotary axes are significantly higher than those of the translation axes. If a uniform weighting coefficient is used, the rapid accumulation of deviations on the rotary axes cannot be adequately compensated, leading to surface defects such as seams and steps in the machined curved areas. On the other hand, the translation axes, due to their slower deviation accumulation, are given excessively high correction ranges, which may result in both undercompensation and overcompensation, reducing overall accuracy.

[0031] This solution sets weighting coefficients for the differentiated wear patterns of translation and rotary axes: the translation axis weighting coefficient ranges from 0.8 to 1.0 (optimal 0.9), aiming to filter out high-frequency noise introduced by temporary process factors such as tool deflection in manually measured deviations, while retaining stable geometric reference information; the rotary axis weighting coefficient ranges from 1.2 to 1.5 (optimal 1.3), aiming to amplify the effective signal reflecting the accumulation of rotary axis clearance in manually measured deviations, compensating for dynamic deviations that automatic geometric calibration cannot capture. The two types of weighting coefficients, one decreasing and one increasing, work synergistically to ensure that the compensation amount for each axis precisely matches its actual deviation magnitude. Compared to a uniform weighting correction method, the dimensional deviation of the corresponding machining area of ​​the rotary axis can be reduced by approximately 40%, with better dimensional consistency in key machining parts such as mating surfaces and curved surfaces.

[0032] In a further optimized scheme, a dynamic calibration cycle scheduling step is also included: the system continuously records three types of variables: the type of processed material, the average number of processed parts per day, and the dimensional deviation of each calibration, based on the formula... Adaptively adjust the timing of the next calibration reminder: (The following is a partial translation of the original text.) For the preset basic calibration cycle, This is the material correction factor. This is the production correction factor. This is the deviation correction factor; the corresponding hard metal billet Smaller than zirconia and glass ceramics, the higher the daily processing capacity, the better. The smaller the value, the greater the historical calibration deviation. The smaller.

[0033] The effectiveness of the above solution lies in the fact that existing technologies generally employ fixed-duration calibration reminder mechanisms, which implicitly assume that the machine tool reference drift rate is constant. However, in actual working conditions, the cutting resistance of high-hardness blanks such as titanium alloys is much greater than that of non-metallic blanks such as zirconium oxide. When machining hard metal blanks, the shaft system bears a greater cutting load and a faster wear rate; in high-volume production modes, the frequency of shaft reciprocating motion is higher and the clearance accumulation is more rapid; and the historical deviation growth rate directly reflects the acceleration of reference drift in the current stage. These three types of variables change independently, and the fixed-cycle calibration mode cannot adapt to the above dynamic working conditions, inevitably leading to the contradiction of excessive calibration occupying production time in low-load scenarios and calibration lag causing batch defects in high-load scenarios.

[0034] This solution continuously records three variables: the type of material being processed, the average daily number of parts processed, and the dimensional deviations from previous calibrations. It then uses a formula to calculate the timing of the next calibration reminder. Smaller correction coefficients result in shorter reminder cycles, and vice versa. Essentially, this mechanism links calibration frequency to the reference drift rate—faster drift leads to more frequent calibration, and slower drift leads to sparser calibration—ensuring that calibration intervention always occurs before deviations exceed the tolerance range. This shifts the calibration strategy from "remedial" to "preventative," avoiding both over-calibration that wastes production capacity and the risk of batch scrap due to calibration delays.

[0035] As a further optional improvement, after each automatic benchmark calibration and deviation data collection is completed, the system automatically archives the calibration operation log. The log records at least include the operation type and the calculated axis offset deviation, which is used for traceability of machining dimension quality.

[0036] The above solution is effective in the following ways: dental processing is a medical device-related production process that requires traceability; by archiving key information of calibration operations, when dimensional abnormalities occur in subsequent finished products, the operations and axis offset data of previous calibrations can be traced back to help locate whether the quality problem originated from the calibration process, thus meeting the traceability requirements of production quality control.

[0037] As a further optional improvement, the standard test cutting fixture is a test cutting disc made of replaceable material, and the fixture is equipped with multiple sets of standard measurement points.

[0038] The advantages of the above solution are: dry and wet cutting processes have different requirements for the material of the test cutting fixture, and the replaceable material test cutting disc can be adapted to different cutting processes; multiple sets of standard measurement points can provide multiple effective sampling positions for manual measurement, enrich the sources of measurement data, and reduce the impact of random errors of a single measurement point on manual deviation calculation.

[0039] As a further optional improvement, when collecting the tooling reference dimensions, the same reference point is measured repeatedly, and the arithmetic mean is taken as the effective offset deviation data.

[0040] The advantages of the above solution are as follows: Conventional manual measurement typically involves directly inputting and calculating data from a single measurement. During manual measurement using measuring tools, differences in reading angles and slight shifts in the measuring point position introduce random errors, directly affecting the accuracy of subsequent compensation calculations. This solution involves repeated measurements at the same reference point, using the arithmetic mean as the valid data. Because random errors exhibit bidirectional fluctuations, the mean of multiple measurements converges towards the true size value, offsetting the random fluctuations introduced by a single measurement. This makes the manually collected deviation data closer to the actual offset, thereby improving the accuracy of subsequent compensation calculations.

[0041] A five-axis dental milling machine includes a five-axis machining host, a spindle tool assembly, a tool magazine, a human-machine interface touch screen, and a CNC control module; the CNC control module includes a processor and a memory, the memory stores a calibration program, and when the processor calls and executes the calibration program stored in the memory, it implements the five-axis dental milling machine calibration method described above.

[0042] The advantages of the above solution are as follows: The five-axis dental milling machine has a pre-installed calibration program in its CNC control module. The processor can complete the entire process of benchmark calibration, deviation calculation, incremental overlay correction of the coordinate system, and generation of the layered coordinate system simply by running the calibration program. This eliminates the need for manual cross-system parameter transfer, avoiding human error during parameter transfer, achieving seamless integration between the calibration process and the machine tool processing process, reducing timing errors and input deviations caused by manual operation, and improving the consistency and repeatability of calibration results. Calibration-related prompts can be directly displayed through the human-machine interface touchscreen, eliminating the need for an additional independent operation and maintenance management system, and adapting to the management needs of continuous batch production in dental processing plants.

[0043] Compared with the prior art, the five-axis dental milling machine calibration method and the five-axis dental milling machine provided by the present invention have the following beneficial effects: First, the accuracy and stability of the calibration are improved synergistically. Compared with automated calibration schemes that rely solely on a single path, this invention uses the offset deviation obtained from actual cutting as an incremental correction amount and adds it to the G54 reference parameters obtained from automatic calibration. The compensation parameters also incorporate the geometric deviation of the machine tool body and the comprehensive process deviation caused by actual cutting. The calibration values ​​have a higher degree of agreement with the actual machining deviations, and the long-term accuracy and stability of the finished product dimensions are better under continuous machining conditions with high-frequency oscillation and multiple material switching.

[0044] Secondly, the consistency of the reference is stronger when switching between multiple billets. Compared with the conventional method of using a single reference or recalibrating the entire axis when machining titanium pillars, this invention reuses all the basic compensation parameters of G54 through the same source layered coordinate system. It can adapt to the multi-station machining scenario of titanium pillars by only superimposing the axial correction amount specific to the titanium pillar station. This can ensure the uniformity of the reference when machining different billets and reduce the dimensional fluctuations when alternating production of billets.

[0045] Third, it has strong adaptability to multiple working conditions and machine models. This solution can be adapted to both positive and negative clamping titanium column clamps, and is also compatible with two mainstream five-axis configurations, Type_RW_AB or Type_RW_BA, expanding the actual application range of the equipment.

[0046] Fourth, the calibration process offers enhanced reliability and quality control. Through supporting designs such as input value upper and lower limit verification, probe capture failure alarms, re-verification, and two-level out-of-tolerance handling, it reduces the risk of calibration errors caused by manual input while automatically verifying calibration output results. This balances production continuity with processing quality control, reduces the experience requirements for operators, and is suitable for mass production use in dental processing plants.

[0047] Fifth, it has the capability of adaptive calibration reminders based on operating conditions. Relying on dynamic calibration cycle scheduling logic, and combining the processing material type, average daily number of processed parts, and dimensional deviations from previous calibrations, it outputs calibration reminders in a linked manner, ensuring that the calibration timing matches the actual rate of deviation accumulation and avoiding calibration delays or over-calibration.

[0048] This solution differs from existing technologies in the following essential ways: Compared with vision-based solutions (such as CN116974240A), which determine the rotation center offset vector through image acquisition, the calibration accuracy of vision-based solutions is limited by image resolution, lighting conditions, and the installation accuracy of marker points. In contrast, this solution obtains the comprehensive deviation through actual cutting, and the deviation data comes directly from the actual processing state, is not affected by optical environmental factors, and has a higher degree of agreement between the correction value and the actual processing deviation.

[0049] Compared with offline compensation schemes based on mathematical modeling (such as CN117724400A), which calculate errors based on ideal kinematic models and do not rely on actual cutting data for verification, this scheme uses actual cutting data as the source of deviations, so that the compensation parameters naturally include all the influencing factors of the actual machining conditions, and the gap between the model and reality is greatly reduced.

[0050] Compared with calibration schemes that focus on physical geometric inspection (such as CN116372609A), which replace the system offset value with the manual measurement results as independent parameters and discard the geometric reference information obtained by automatic calibration, this scheme integrates the manual deviation into the automatic reference in an incremental superposition manner. It retains the full-axis geometric reference information while absorbing the process deviation information, resulting in richer dimensions of compensation information and higher data redundancy. Attached Figure Description

[0051] Figure 1 This is a flowchart illustrating the dual-mode layered compensation calibration method for a five-axis dental milling machine in this invention. Detailed Implementation

[0052] This invention provides a dual-mode layered compensation calibration method and device for a five-axis dental milling machine. To make the purpose, technical solution and effects of this invention clearer and more explicit, the invention will be further described in detail below with reference to the accompanying drawings and examples.

[0053] This embodiment is applied to a five-axis dental milling machine for digital processing of dentures, and is suitable for milling dental blanks such as zirconia, titanium pillars, and glass ceramics. The dual-mode layered compensation calibration method for the five-axis dental milling machine in this application is executed by the machine tool CNC control module; the entire software system is divided into three layers: the upper layer is a human-machine interface module, responsible for calibration process triggering, status prompts, parameter result reading and table refreshing; the middle layer is the CNC system common macro variables and shared memory data exchange layer, responsible for data routing and parameter transmission between the three layers; the bottom layer is the CNC system G-code macro program execution layer, responsible for driving the probe to perform point detection and coordinate acquisition; the three layers rely on the CNC system common macro variables and shared memory to complete parameter transmission.

[0054] The overall business process includes four core steps: automatic benchmark calibration, deviation data collection, dual-mode linkage hierarchical compensation, and cross-verification of different sources. It can also selectively execute dynamic calibration cycle scheduling.

[0055] I. Hardware verification and tool setting verification of the detector component In this embodiment, a contact-triggered probe is preferably used as the detection device for performing reference acquisition. This type of device is a conventional and mature device in the field of five-axis CNC equipment accuracy calibration, and its structure and working principle have been fully disclosed in existing technology. The probe mainly consists of a carbide probe rod, a ruby ​​ball, a contact sensing unit, and a signal output unit. The ball diameter is typically 3mm or 5mm, and the repeatability accuracy can reach ±0.001mm. When the ball contacts the workpiece reference surface and produces a small displacement, the contact sensing unit immediately triggers a position latch signal, and the CNC system synchronously latches the current coordinate values ​​of each axis, completing single-point data acquisition. It should be noted that the detection device in this invention is not limited to the aforementioned contact-triggered probe. Those skilled in the art can use non-contact laser probes, vision imaging probes, or other devices with equivalent point-position coordinate acquisition functions as substitutes, depending on the machine tool's range, accuracy level, and actual working conditions. As long as accurate acquisition of reference point coordinates can be achieved, they fall within the protection scope of this invention.

[0056] Before formally executing automatic reference calibration, the system first performs a probe tool setting verification: the underlying G-code macro calls the tool setter, and drives the probe to touch the reference surface of the tool setter along the Z direction via the G37.1 instruction, obtaining the actual tool setting height of the probe and storing it in the common variable #2079; if the status bit @10152 returns 0 (capture failure), the alarm number @9999=1551 is output and the process jumps to the abnormal exit; the upper-level human-machine interface reads the alarm flag, terminates the current automatic reference calibration process, and outputs the alarm information in a pop-up window. Only when the probe capture is normal will the subsequent full-axis reference acquisition process begin.

[0057] S1, Automatic Reference Calibration This step is used to obtain the original motion reference of the machine tool body and generate the compensation parameters of the first basic workpiece coordinate system G54.

[0058] After the user triggers the CalibStart() process by clicking the axis calibration button on the human-machine interface, the system first calls the PrepareL8013MacroFile() function to decrypt the L8013.NC macro program file, verify its validity, and synchronize its path. Upon successful verification, a command is issued to trigger the underlying G-code macro to execute full-axis reference acquisition. The probe touches the reference features point by point along a preset standardized acquisition path to complete the full-axis reference data acquisition. In this embodiment, the preferred configuration is as follows: 12 sets of evenly distributed straightness reference points are set for each of the X, Y, and Z translation axes, arranged at equal intervals along the entire stroke of each axis, used to acquire the positioning deviation, straightness deviation, and inter-axis perpendicularity deviation of the translation axes; 8 sets of rotation reference points are set for each of the A and B rotation axes, evenly distributed along the circumference of the rotation circle, used to acquire the rotation center offset, axial runout, and oscillation deviation of the rotation axes. Those skilled in the art can adjust the number and arrangement of reference points according to actual accuracy requirements without affecting the implementation of the core method of this invention. During the data acquisition process, the CNC system pre-stores the theoretical coordinate values ​​of each point. After the probe touches the reference surface, the system automatically records the measured coordinate values ​​of the corresponding axis, forming a pair of point data: "theoretical coordinates - measured coordinates." After all points are acquired, multiple sets of valid reference point data are obtained, forming a redundant observation dataset. The design of multiple redundant points allows subsequent algorithms to offset random fluctuations from single-point acquisitions, improving the overall reliability of the reference data.

[0059] Based on the acquired redundant reference point data, the CNC kernel performs a solution according to the following error identification model to generate the initial compensation parameters for G54:

[0060] in, Let be the deviation vector between the measured coordinates and the theoretical coordinates of the i-th point; This is the error Jacobian submatrix corresponding to the i-th point, reflecting the contribution of a single geometric error term to the coordinate deviation at that point. Let be the geometric error vector to be solved; n be the total number of reference points. The optimal estimates of each geometric error term can be obtained by solving this optimization problem using the least squares method.

[0061] Finally, the geometric errors of each axis obtained from the solution are converted into coordinate system compensation values, generating the initial all-axis compensation parameters for the first basic workpiece coordinate system G54, and writing the parameters into the dedicated storage area of ​​the G54 coordinate system of the CNC system. Users can read the G54 compensation parameters in the shared memory through the human-machine interface and view them in the UI table; at the same time, the system automatically completes the parameter compensation confirmation process after calibration.

[0062] S2, Deviation Data Acquisition This step is used to obtain the comprehensive deviation under actual cutting conditions, including process errors such as fixture deformation, tool wear, and cutting deflection, to provide measured data support for subsequent fusion compensation.

[0063] Pre-process external steps (not software-executed actions): The standard test cutting fixture is clamped and the standard cutting process is completed. The operator uses measuring tools to measure the fixture reference dimensions. The measured dimensions are entered into the machine tool through the human-machine interface. The software obtains the entered fixture reference dimension data.

[0064] 2.1 Trial cutting fixture clamping and cutting As a preliminary step in the implementation process, the standard test cutting fixture is clamped in the standard clamping position of the machine tool table. After clamping, the system calls the preset standard cutting program and uses the same cutting parameters as actual production to process the test cutting disc, machining all the reference features. Using the same cutting parameters as mass production can ensure that the process factors such as tool deformation and cutting deflection during the test cutting process are completely matched with the actual processing, making the measured deviation closer to the real production state.

[0065] 2.2 Dimension Acquisition and Input Verification After processing, the operator collects the dimensions of each reference point using measuring tools. In this embodiment, an outside micrometer and an inside dial indicator are preferably used as measuring tools, as these are conventional dimensional measuring devices in the machining field. During the measurement process, the human-machine interface simultaneously displays a measurement diagram corresponding to the point, marking the measurement position and the direction of force application, guiding the operator to align the measurement position. It should be noted that this invention does not limit the specific type of measuring tool; those skilled in the art can also use digital micrometers, coordinate measuring machines, or other devices with equivalent dimensional measuring functions, as long as the measured dimensional data of the reference points can be obtained.

[0066] When the software acquires the tooling reference dimensions, the system performs upper and lower limit checks on the entered dimension values. The checks cover six measured dimension parameters: B, C, D, E, F, and G. The reference values ​​correspond to the Y-axis height reference and the X / Z-axis width reference, respectively. The preset maximum deviation threshold is 2.0 mm. When the deviation between the entered value and the reference value exceeds the threshold, the system automatically records the operation log, displays an error message, and terminates subsequent calculations, filtering out invalid measurement data. Upper and lower limit checks directly filter out abnormal values ​​that deviate from the reasonable range, preventing erroneous data from entering the deviation calculation process. This avoids coordinate system compensation errors caused by incorrect input, improving the fault tolerance and system reliability of the data acquisition process.

[0067] 2.3 Differentiate machine tool configurations to solve offset deviations Among them, the Type_RW_AB configuration refers to a five-axis layout where the A-axis is a rotary table and the B-axis is a oscillating spindle; the Type_RW_BA configuration refers to a five-axis layout where the B-axis is a rotary table and the A-axis is a oscillating spindle. The motion transmission paths of the rotary axes differ between the two configurations, therefore the analytical calculation formulas for offset deviations also differ. The system reads the machine tool type parameters to identify the configuration, distinguishes between the Type_RW_AB and Type_RW_BA five-axis machine tool configurations, and calls the analytical calculation formulas for the corresponding configuration branch to complete the offset deviation calculation. The calculation logic is executed directly at the human-machine interface layer.

[0068] Formula for calculating Type_RW_AB configuration:

[0069] in, ,and These are the initial offset values ​​of the X-axis, Z-axis, and A-axis in the original offset coordinate system; b, c, d, e, f, and g are the six measured dimension data entered. The reference height parameter is 8. The reference spacing parameter is 73.

[0070] Type_RW_BA configuration: The system dynamically retrieves the B-axis index number, and the calculation formula corresponds to the AB configuration, which applies to the Y-axis, Z-axis, and B-axis.

[0071] The rotary axis arrangement of five-axis dental milling machines with different configurations varies, and the motion transmission relationship of the axis system is different. Matching the corresponding analytical calculation formulas according to the configuration can adapt to the deviation solution logic of machine tools with different axis system layouts, prevent the distortion of calculation results caused by formula mismatch, and improve the adaptability and universality of this calibration method to different machine models.

[0072] 2.4 Differentiated weight overlay update G54 The offset deviation is used as the superposition correction amount. Weighting coefficients are set according to the wear characteristics of the shafts. The weighting coefficients of the A-axis and B-axis rotation axes are greater than the weighting coefficients of the X-axis, Y-axis and Z-axis translation axes. These are superimposed on the initial compensation parameters of G54 to obtain the updated G54 coordinate system:

[0073] in, These are the final compensation parameters for the i-th axis after fusion; The initial compensation parameters for the i-th axis in the G54 coordinate system; The offset deviation of the i-th axis is obtained by decomposing the deviation data acquisition. is the weight coefficient corresponding to the i-th axis.

[0074] In this embodiment, the weighting coefficients are set based on the statistical law of shaft wear in dental milling conditions: the weighting coefficient for translation axes ranges from 0.8 to 1.0, with 0.9 being preferred in this embodiment; the weighting coefficient for rotary axes ranges from 1.2 to 1.5, with 1.3 being preferred in this embodiment. Weighting coefficients less than 1 correspond to translation axes because the automatic geometric calibration of translation axes has high reference stability, and the manually measured deviations include temporary process factors such as tool deflection, which do not need to be fully superimposed. Weighting coefficients greater than 1 correspond to rotary axes because rotary axes wear quickly and accumulate significant clearances; the automatic calibration of static geometric deviations is insufficient to cover the actual offsets in dynamic machining, requiring a weight greater than 1 to strengthen the correction. By setting differentiated weights, the compensation amount of each axis can be precisely matched to its actual deviation magnitude. Compared to the uniform weight correction method, the dimensional accuracy consistency of the machining area corresponding to the rotary axis is better, reducing the probability of joint marks and steps appearing on the finished product mating surface.

[0075] After the superposition is complete, the updated G54 parameters are written to shared memory; the human-machine interface refreshes the table to display the updated G54 values.

[0076] S3, Layered Compensation: Generation of Multi-Station Layered Coordinate System for Titanium Column This step is used to fuse the two types of calibration data and generate a hierarchical coordinate system, and it is the core calculation step of the entire scheme.

[0077] From a business logic perspective, the updated G54 coordinate system is used as the common reference, and the exclusive axial corrections for each station of the titanium column are superimposed to generate the second-layer workpiece coordinate system G55 series, which is adapted to multi-station machining of titanium columns. G55 reuses all the basic compensation parameters of G54.

[0078] Among them, the X-axis and Y-axis In other words, the X and Y axes fully reuse the compensation parameters of G54 without adding additional corrections; the Z, A, and B axes allow non-zero correction values. The design is based on the fact that the titanium column is clamped using a dedicated axial fixture, and the X / Y axis positioning shares the same worktable reference as the general denture blank. The positioning reference is completely consistent, and the deviation difference is only concentrated in the Z-axis clamping height and rotation clamping posture; therefore, G55 fully reuses the full-axis basic compensation of G54, only adding axial-specific corrections, without the need to re-acquire the full-axis reference.

[0079] From a code implementation perspective, the compensation parameters of the G55 series coordinate system are independently detected and calculated by the dedicated G-code macro for titanium column calibration and then written into the common macro variables @6320~@6519 (0° attitude) and @6560~@6625 (90° attitude). Its business logic reuses the A-axis and B-axis references of G54, but its storage and calling are independent of G54. The specific implementation is as follows: Multi-clamp mode: The system supports multiple clamp modes, including Arum, Medentika, R_Clamp reverse clamp, R_Clamp 90°, etc. Different clamp modes trigger different underlying G-code detection branches, calling the appropriate calculation formula to solve for the station-specific axial correction. The installation force state and deviation generation mechanism of positive clamp and reverse clamp are different. By using the appropriate calculation formula, the deviation distribution characteristics of the two types of clamps can be matched separately, avoiding the solution error caused by a single calculation formula to be adapted to two types of clamps. It is compatible with many mainstream clamping fixtures in dental titanium column processing, broadening the applicable scenarios of the equipment.

[0080] Multi-rotational attitude independent storage of compensation data: The underlying G-macro supports cyclic detection at multiple workstations; each workstation performs multi-point probe contact acquisition in the A0° and A180° attitudes (A90° is also included in reverse clamping). The G-code macro first identifies the diameter of the titanium column workpiece, and the probe probes at multiple points along the X-positive and Z-positive directions, triggering probe capture; if capture fails, an alarm signal is output and the system exits abnormally. After detection, the offset results of each axis in each attitude are written to an independent common macro variable area. The upper-level Qt function reads these macro variables, overlays the correction parameters input from the UI page, and writes them back to the common macro; coordinate system parameters are persistently saved to the storage medium. The HMI table displays the compensation data for each workstation in each rotational attitude, and the compensation data for each workstation and each rotational attitude are stored independently; during machining, the CNC program directly calls the corresponding coordinate system. Compensation data for different rotational postures are stored independently, and the data between each station does not interfere with each other. It can provide accurate correction for each rotational posture at each station, avoid crosstalk between compensation data at different stations and postures, and improve the dimensional consistency of multi-post titanium column processing.

[0081] The G54 and G55 coordinate systems are stored independently in the parameter area of ​​the CNC system. During machining, the corresponding coordinate system parameters can be directly called according to the type of blank being machined. The homogeneous layered design ensures the consistency of the reference for the two types of machining scenarios, avoiding the reference offset introduced by repeated calibration, and can also be specifically adapted to the clamping characteristics of titanium pillars. When switching between multiple materials, there is no need to re-perform the full process calibration, and the dimensional fluctuations of blanks produced alternately are smaller.

[0082] S4, Heterogeneous Cross-Verification After completing the G54 coordinate system update, the system triggers the automatic benchmark calibration process again for secondary detection, and verifies the deviation between the calibration results and the updated G54 compensation parameters; and configures two-level out-of-tolerance handling logic.

[0083] After completing the coordinate system parameter fusion and update, the system again drives the probe to start the automatic calibration process, performing secondary detection on the reference points of the standard test cutting fixture to obtain re-inspection coordinate data, and verifying the deviation with the updated G54 compensation parameters. The re-inspection points correspond one-to-one with the manually measured points, ensuring the consistency of the comparison benchmark and avoiding comparison distortion caused by differences in points.

[0084] The system performs graded handling based on the re-inspection deviation results. In this embodiment, the preset acceptable tolerance range is ±0.008mm. If a single verification exceeds the tolerance, the system will only output a calibration reminder signal to the operator and will not forcibly interrupt production. This design can avoid unnecessary downtime caused by temporary factors such as instantaneous fluctuations in ambient temperature and occasional errors in single measurements, thus maintaining production continuity.

[0085] If two consecutive checks exceed the tolerance, it is determined that the reference drift has persisted. The system automatically locks the machine tool's processing start permission and triggers a dual-mode recalibration process for the entire machine. This design can promptly prevent batch defects caused by systematic deviations and ensure processing quality.

[0086] By comparing two sets of heterogeneous data—automatic detection and manual measurement—a preliminary determination of the source of deviation can be achieved: if the deviation from the automatic re-inspection exceeds the tolerance but the deviation from the manual measurement is within the normal range, it indicates that the actual cutting state is normal, and the deviation only occurs in the detection and acquisition stage. This suggests the deviation originates from problems in the detection system, such as wear of the detection component or abnormal probe signal. If both the automatic re-inspection and manual measurement deviations exceed the tolerance, it indicates that both paths detect a reference offset. This suggests the deviation originates from problems such as mechanical structure deformation or increased shaft clearance, indicating a drift in the body reference. This differentiation logic can quickly narrow down the scope of fault diagnosis, significantly reducing the time cost of precision maintenance compared to a general, indiscriminate inspection approach.

[0087] S5, Dynamic Calibration Cycle Scheduling In a further preferred embodiment, this calibration method also includes a dynamic calibration cycle scheduling step: The system continuously records three types of variables: the type of material being processed, the average number of parts processed per day, and the dimensional deviation from each calibration. Based on these three types of variables, the system adaptively adjusts the timing of the next calibration reminder.

[0088] in, This is the preset basic calibration period, with a default value of 7 days; The value is 1.0, which is a material correction factor. Zirconia and glass ceramic blanks have low cutting resistance and slow shaft wear. Titanium and other hard metal blanks have high cutting resistance and increased shaft wear rate. The value is 0.6 to 0.8. This is the output correction factor. When the daily output is below 30% of the rated capacity, it is 1.2; when it is between 30% and 70% of the rated capacity, it is 1.0; and when it is above 70% of the rated capacity, it is 0.7 to 0.9. This is the deviation correction factor. It is 1.1 when the deviation of the previous calibration is within 30% of the tolerance range, 1.0 when it is between 30% and 70%, and 0.8 when it is above 70%.

[0089] Among them, the processing material type is automatically identified by the blank parameters of the current processing task, the average number of processed pieces per day is automatically counted by the system processing log, and the dimensional deviation of each calibration is automatically extracted from the historical calibration record; all three types of data are automatically recorded and updated by the CNC back-end business module, without the need for manual input.

[0090] By using a three-tiered coefficient linkage correction, the calibration cycle is shortened for scenarios involving hard metal billets, high production volumes, and rapid deviation growth, while the cycle is extended for lower-volume scenarios. This ensures that the calibration timing precisely matches the deviation accumulation rate under actual working conditions. Compared to a fixed-cycle calibration mode, this reduces unnecessary calibration operations in low-load scenarios and avoids dimensional deviations caused by calibration lag in high-load scenarios. It keeps the calibration process synchronized with the actual deviation accumulation rate, continuously guaranteeing the consistency of finished product dimensional accuracy.

[0091] The scheduling logic is executed by an independent business module in the CNC backend, while the human-machine interface is responsible for receiving reminder signals and displaying calibration prompts in a pop-up window.

[0092] As a further optional improvement: Calibration Log Archiving: After each automatic benchmark calibration and deviation data collection, the system automatically archives the calibration operation log. The log records must at least include the operation type and the calculated axis offset deviation, for traceability of machining dimensional quality. In this embodiment, CSV format is preferred for easy reading by mainstream spreadsheet software. This invention does not limit the specific format of the log file; any storable and exportable data format such as TXT or XLSX can be selected according to actual needs. Operators can query historical calibration records or export log files through the human-machine interface touch screen. When machining quality problems occur, the calibration status of the corresponding period can be traced back through the log to locate the root cause of the dimensional deviation, adapting to the quality traceability management needs of dental machining scenarios.

[0093] The arithmetic mean of multiple measurements is used: the arithmetic mean of multiple probe results at the same reference point is taken as the valid data. This calculation is performed by the underlying G-code macro (e.g., in the reverse clamp calibration macro, the center coordinates are calculated by averaging the results of multiple probes). The human-computer interaction layer does not perform this operation. In conventional measurement, single measurement data is usually used directly in the calculation. During the measurement process using measuring tools, differences in reading angles and slight offsets in the position of the measuring point will introduce random errors, directly affecting the accuracy of subsequent compensation calculations. This solution performs multiple repeated measurements at the same reference point and takes the arithmetic mean as the valid data. Since random errors have the characteristic of fluctuating in both positive and negative directions, the mean of multiple measurements will converge to the true size value, which can offset the random fluctuations introduced by a single measurement, making the collected deviation data closer to the actual offset, thereby improving the accuracy of subsequent compensation calculations.

[0094] Replaceable Material Test Cutting Fixture: The standard test cutting fixture can adopt a replaceable material test cutting disc structure. The disc body can be replaced with a corresponding material depending on the dry or wet cutting processing mode. The disc diameter is consistent with the size of conventional dental blanks. The fixture surface has multiple sets of standard measurement points preset, each corresponding to different axis sensitivities. Dry and wet cutting processing conditions have different requirements for the test cutting fixture material; the replaceable material test cutting disc can adapt to different cutting processes. Multiple sets of standard measurement points provide multiple effective sampling positions for manual measurement, enriching the measurement data sources and reducing the impact of random errors at a single measurement point on manual deviation calculation. It should be noted that the standard test cutting fixture of this invention is not limited to a disc structure. Those skilled in the art can adjust the fixture structure and the number of points according to the machine tool clamping method. As long as the fixture contains standard measurement features that can distinguish deviations of each axis, the same effect can be achieved, and all fall within the protection scope of this invention.

[0095] This embodiment also provides a five-axis dental milling machine, including a five-axis machining host, a spindle tool assembly, a tool magazine, a human-machine interface touch screen, and a CNC control module; the CNC control module includes a processor and a memory, the memory stores a calibration program, and when the processor calls and executes the calibration program stored in the memory, the aforementioned five-axis dental milling machine calibration method is implemented.

[0096] The five-axis dental milling machine has a pre-installed calibration program in its CNC control module. The processor runs this program to complete the entire process, including probe verification, datum calibration, deviation calculation, incremental overlay correction of the coordinate system, generation of layered coordinate systems, deviation verification, and dynamic calibration reminders. Among these: Automatic calibration subunit: Triggers the underlying G-code probe calibration macro, reads common macro variables, and obtains the initial compensation parameters of G54; Deviation data acquisition subunit: Receives measured dimension input, performs analytical calculation of configuration matching, completes weighted superposition, and outputs updated G54; Layered compensation subunit: triggers the underlying G-code macro to complete multi-station, multi-rotational attitude probe detection, reads and writes common macro variables, generates and persists the G55 series layered coordinate system; Cross-check subunit: Reuses the automatic calibration process to perform secondary detection, realizing the interlocking of reminders and processing permissions; Periodic dynamic scheduling subunit: statistically analyzes operating conditions and outputs reminder signals.

[0097] No manual parameter transfer across systems is required, avoiding human error during parameter transfer. This enables seamless integration of the calibration process with the machine tool processing process, reducing timing errors and data entry deviations caused by manual operation, and improving the consistency and repeatability of calibration results. Calibration-related prompts can be directly displayed through the human-machine interface touchscreen, eliminating the need for a separate operation and maintenance management system, and adapting to the management needs of continuous batch production in dental processing plants.

[0098] Those skilled in the art should understand that the above embodiments are merely illustrative examples. Without departing from the core principles of the present invention, various modifications, substitutions, or combinations can be made to the parameters, structures, and processes. All equivalent technical solutions should fall within the protection scope of the present invention.

[0099] Furthermore, those skilled in the art will understand that although some embodiments herein include certain features included in other embodiments but not others, combinations of features from different embodiments are meant to be within the scope of the invention and form different embodiments; for example, any one of the embodiments disclosed in this invention can be used in any combination.

Claims

1. A dual-mode layered compensation calibration method for a five-axis dental milling machine, characterized in that, Includes the following steps: S1. Automatic reference calibration: Call the probe to perform point detection on the standard calibration disk, collect the original motion reference of the five-axis machine tool, and generate and store the compensation parameters of the first basic workpiece coordinate system G54. S2. Deviation Data Acquisition: Acquire the tooling reference dimension data obtained after cutting with a standard trial cutting fixture; based on the acquired tooling reference dimension data, the system calculates the independent offset deviations of the X-axis, Y-axis, and Z-axis translation axes, as well as the A-axis and B-axis rotation axes. The offset deviations are used as incremental corrections and superimposed on the compensation parameters of the G54 coordinate system to obtain the updated G54 coordinate system; when the offset deviations are superimposed on the G54 compensation parameters, the weighting coefficients of the A / B rotation axes are greater than the weighting coefficients of the X / Y / Z translation axes. S3, Layered Compensation: Using the updated G54 coordinate system as the same source reference, superimpose the exclusive axial correction amount of each station of the titanium column to generate a second layered workpiece coordinate system G55 adapted to multi-station machining of titanium columns. The G55 reuses all the basic compensation parameters of the G54. The G55 reuses all the basic compensation parameters of the G54 in the following ways: the X-axis and Y-axis fully reuse the compensation parameters of the G54, and only the Z-axis, A-axis and B-axis are superimposed with the exclusive axial correction amount; The dedicated axial correction amount includes offset compensation data for the corresponding titanium column station under different rotational postures, and the compensation data for each station is stored independently.

2. The dual-mode layered compensation calibration method for a five-axis dental milling machine according to claim 1, characterized in that, The titanium column station supports both positive and negative clamping modes; for different clamping modes, a corresponding calculation formula is used to solve for the specific axial correction amount of each station.

3. The dual-mode layered compensation calibration method for a five-axis dental milling machine according to claim 1, characterized in that, When acquiring the tooling reference dimensions, the system sets upper and lower limit verification rules for the acquired dimension values ​​to filter out invalid measurement data that exceeds the reasonable deviation range.

4. The dual-mode layered compensation calibration method for a five-axis dental milling machine according to claim 1, characterized in that, When calculating the independent offset deviations of the X-axis, Y-axis, and Z-axis translation axes, as well as the A-axis and B-axis rotation axes, the two five-axis machine tool configurations, Type_RW_AB and Type_RW_BA, are distinguished, and the analytical calculation formulas for the corresponding configurations are used to solve for the offset deviations.

5. The dual-mode layered compensation calibration method for a five-axis dental milling machine according to claim 1, characterized in that, Before the automatic reference calibration is executed, the probe tool setting verification is completed first; if the probe fails to capture, the automatic reference calibration process is terminated and an alarm message is output.

6. The dual-mode layered compensation calibration method for a five-axis dental milling machine according to claim 1, characterized in that, After the G54 coordinate system update is completed, the system automatically calls the probe again to perform calibration comparison, and checks the deviation between the comparison result and the updated G54 compensation parameters; it is configured with two levels of out-of-tolerance handling logic: if the single check is out of tolerance, a calibration reminder signal is output; if the check is out of tolerance twice in a row, the machine tool processing start permission is locked and the recalibration process is triggered.

7. The dual-mode layered compensation calibration method for a five-axis dental milling machine according to any one of claims 1 to 6, characterized in that, It also includes a dynamic calibration cycle scheduling step: the system continuously records three types of variables: the type of processed material, the average number of processed parts per day, and the dimensional deviation of each calibration, based on the formula. Adaptively adjust the timing of the next calibration reminder: (wherein) For the preset basic calibration cycle, This is the material correction factor. This is the production correction factor. This is the deviation correction factor; the corresponding hard metal billet Smaller than zirconia and glass ceramics, the higher the daily processing capacity, the better. The smaller the value, the greater the historical calibration deviation. The smaller.

8. A five-axis dental milling machine, characterized in that, The system includes a five-axis machining center, a spindle tool assembly, a tool magazine, a human-machine interface touchscreen, and a numerical control module. The numerical control module includes a processor and a memory. The memory stores a calibration program. When the processor calls and executes the calibration program stored in the memory, it implements the dual-mode layered compensation calibration method for a five-axis dental milling machine as described in any one of claims 1 to 7.

Images