The present disclosure relates to the monitoring of milling tools. In particular, the present disclosure relates to the monitoring of one or more tools of a dental milling machine for indicators of tool breakage and/or tool wear during a milling process.
Dental milling machines are used to mill dental restorations. The milling machines grind or cut away portions of a block of material (e.g., ceramic, gold, porcelain, etc.) in order to create the dental restoration. The milling machines utilize one or more tools to shape the restoration, including cutting tools and/or grinding tools. During the course of the milling process, the tools wear down and/or break because of the high amount of friction and/or loading associated with forming the dental restoration. A tool breakage results in a significant delay in completion of the dental restoration because the milling machine will continue to try to mill the restoration after the breakage, resulting in an improper restoration as the broken tool cannot properly prepare the block. This requires the dental restoration to be completely remilled in the event of damage to the block and, therefore, the restoration caused by the broken tool.
Accordingly, there is a need for improved methods, systems, and devices for monitoring milling tools and, in particular, monitoring tools of a dental milling machine for indicators of tool breakage and/or tool wear.
Methods, systems, and devices for monitoring tool breakage and/or wear are provided.
In one embodiment, a dental milling system is provided. The milling system includes a milling tool for milling a dental prosthetic and a spindle operable to receive, fixedly engage, and rotate the milling tool. A first accelerometer is positioned adjacent to the spindle and is operable to detect vibrations associated with rotation of the milling tool. A processor is in communication with the first accelerometer to receive data sets representative of the vibrations detected by the first accelerometer. The processor processes the data sets to identify changes in one or more harmonics of the detected vibrations indicative of a break of the milling tool.
In another embodiment, a method of detecting a tool break in a dental milling machine is provided. The method includes monitoring vibrations associated with rotation of a milling tool in an unloaded state; monitoring vibrations associated with rotation of the milling tool in a loaded state; identifying one or more harmonics associated with rotation of the milling tool in the loaded state; and monitoring vibrations associated with rotation of the milling tool for the one or more identified loaded-state harmonics during a milling process to detect a break of the first milling tool.
In another embodiment, a method of milling a dental prosthetic is provided. The method includes detecting vibrations of a milling machine during a milling process with an accelerometer, where the milling machine comprises a spindle for rotating a tool engaged with the spindle. The method further includes analyzing data sets representative of the detected vibrations for changes in amplitude of one or more harmonics of the tool rotation indicative of tool breakage. In some instances, the method further comprises stopping the milling process upon detection of a change in amplitude indicative of tool breakage, replacing the broken tool with a replacement tool, and resuming the milling process.
Additional aspects, features, and embodiments of the present disclosure are described in the following detailed description.
For the purposes of promoting an understanding of the principles of the present disclosure, reference will now be made to the embodiments, or examples, illustrated in the drawings and specific language will be used to describe the same. It will nevertheless be understood that no limitation of the scope of the present disclosure is thereby intended. Any alterations and further modifications in the described embodiments, and any further applications of the principles as described herein are fully contemplated as would normally occur to one skilled in the art to which the disclosure relates.
Referring to
The milling system 10 further includes an accelerometer 20. Generally, the accelerometer 20 is configured for monitoring and detecting vibrations of the milling system 10 and, in particular, vibrations associated with rotation of the milling tool 14 and/or spindle 12. Accordingly, the accelerometer 20 is positioned within the milling system 10 at a location adjacent to the milling tool 14 and/or spindle 12 in some instances. In the illustrated embodiment, the accelerometer 20 is positioned on the housing of the spindle 12. However, in other embodiments the accelerometer is positioned elsewhere adjacent to the milling tool 14 and/or spindle 12. In that regard, the accelerometer 20 is not necessarily in contact with the spindle 12 and is spaced from the spindle and the milling tool 14 in some instances. In some instances, analog accelerometers are utilized. Examples of suitable accelerometers include, without limitation, Analog Devices ADXL322 and Analog Devices ADXL001.
The accelerometer 20 is in communication with an analog-to-digital converter 22. The analog-to-digital converter 22 receives analog signals output by the accelerometer 20 indicative of the detected vibrations and converts the analog signals to corresponding digital signals. In the illustrated embodiment, the accelerometer 20 is shown connected to the analog-to-digital converter 22 via line 24, which may be any type of suitable communication line between the accelerometer and converter. An example of a suitable analog-to-digital convertor is the ARM microcontroller NXP LPC2146, which has two built-in 10 bit ADCs.
The analog-to-digital converter 22 is in communication with a processing system 26. The processing system 26 receives the digital signals output by the analog-to-digital converter 22 and processes the digital signals to detect indicators of tool wear and/or tool breakage. Exemplary processing methods are described in greater detail below. However, in general, the processing system 26 analyzes the digital signals for changes in amplitude of harmonics of the tool rotation speed that are indicative of tool wear and/or tool breakage. In the illustrated embodiment, the processing system 26 is shown connected to the analog-to-digital converter 22 via line 28, which may be any type of suitable communication line between the processing system and converter. In some instances, the processing system 26 controls other aspects of the milling system 10, such as spindle/motor speed, positioning of the spindle 12, positioning of the block 16, and/or other features of the milling system. Accordingly, as described below, in some instances the processing system 26 does not continuously monitor for tool wear and/or tool breakage, but rather intermittently analyzes data from the accelerometer 20 to detect tool wear and/or tool breakage. In such instances, the processing system does not need to be a real time system, but rather simply have available processing time to process the data for detecting tool wear and/or tool breakage. Accordingly, in some instances the system does not require a separate dedicated processor or processing system for the wear/break detection, but rather utilizes a general processor or processing system of the milling system. Accordingly, a processor suitable for running an Operating System, such as the Microsoft Windows XP or Linux, is used in some instances.
Referring to
The analog-to-digital converter 42 is in communication with a processing system 46. The processing system 46 receives the digital signals output by the analog-to-digital converter 42 and processes the digital signals to detect indicators of tool wear and/or tool breakage. In some instances, the processing system 46 analyzes the digital signals for changes in amplitude of harmonics of the tool rotation speed that are indicative of tool wear and/or tool breakage. In the illustrated embodiment, the processing system 46 is shown connected to the analog-to-digital converter 42 via line 48, which may be any type of suitable communication line between the processing system and converter.
The milling system 30 also includes a spindle 52 that receives and fixedly engages a tool 54 for milling block 36 in combination with tool 34. The spindle 52 and the tool 54 are substantially similar to the spindle 32 and the tool 34 in some instances. In that regard, in some instances, the spindle 52 includes a motor for providing power to rotate the tool 54. The rotating tools 34 and 54 are utilized to grind, machine, cut, and/or otherwise remove material from the block 36 to shape a dental restoration in some embodiments. Accordingly, the tools 34 and 54 are moved with respect to the block 36 and/or the block is moved with respect to the tools to facilitate removal of specific portions and amounts of the block to appropriately shape of the block into the desired dental restoration. In some instances, the processing system 46 controls or directs movement of one or more of the spindle 32, tool 34, spindle 52, spindle 54, and block 36.
The milling system 30 further includes an accelerometer 60 associated with the spindle 52 and the tool 54. Generally, the accelerometer 60 is configured for monitoring and detecting vibrations of the milling system 30 and, in particular, vibrations associated with rotation of the milling tool 54 and/or spindle 52. Accordingly, the accelerometer 60 is positioned within the milling system 30 at a location adjacent to the milling tool 54 and/or spindle 52 in some instances. In the illustrated embodiment, the accelerometer 60 is positioned on the housing of the spindle 52. The accelerometer 60 is in communication with an analog-to-digital converter 62. The analog-to-digital converter 62 receives analog signals output by the accelerometer 60 indicative of the detected vibrations and converts the analog signals to corresponding digital signals. The accelerometer 60 is shown connected to the analog-to-digital converter 62 via line 64, which may be any type of suitable communication line between the accelerometer and converter.
The analog-to-digital converter 62 is also in communication with the processing system 46. The processing system 46 receives the digital signals output by the analog-to-digital converter 62 and processes the digital signals to detect indicators of tool wear and/or tool breakage. In that regard, the processing system 46 analyzes the data from the accelerometers 40 and 60 separately in some instances, such that the data associated with accelerometer 40 is indicative of tool wear or breakage of tool 34 and the data associated with accelerometer 60 is indicative of tool wear or breakage of tool 54. In other instances, the processing system 46 analyzes the collective data from both accelerometers 40 and 60 to detect indicators of tool wear and/or tool breakage. In that regard, in some instances, the processing system 46 analyzes the digital signals for changes in amplitude of harmonics of the tool rotation speed that are indicative of tool wear and/or tool breakage. In some such instances, harmonics are associated with the tools 34 and 54 such that changes in amplitude of a particular harmonic is indicative of tool wear or tool breakage of one of the tools 34 or 54. In the illustrated embodiment, the processing system 46 is shown connected to the analog-to-digital converter 62 via line 68, which may be any type of suitable communication line between the processing system and converter.
Referring to
The milling system 70 further includes an accelerometer 78. Generally, the accelerometer 78 is configured for monitoring and detecting vibrations of the milling system 70 and, in particular, vibrations associated with rotation of the milling tools 34, 54 and/or spindles 32, 52. Accordingly, the accelerometer 78 is positioned within the milling system 70 at a location adjacent to the milling tools 34, 54, spindles 32, 52, and/or mandrel subassembly 72 in some instances. In the illustrated embodiment, the accelerometer 78 is positioned on the mandrel socket 74 of the mandrel subassembly 72. However, in other embodiments the accelerometer is positioned elsewhere adjacent to the milling tools 34, 54, spindles 32, 52, and/or mandrel subassembly 72.
The accelerometer 78 is in communication with a processing system 80. The processing system 80 receives signals output by the accelerometer representative of the detected vibrations. In some instances, the signals output by the accelerometer 78 are digital. In other instances, the signals output by the accelerometer 78 are analog. In some instances, the analog signals output by the accelerometer 78 are passed through an analog-to-digital converter (not shown), which then sends a digital signal representative of the detected vibrations to the processing system 80. The processing system 80 processes the signals received from the accelerometer to detect indicators of tool wear and/or tool breakage. In some instances, the processing system 80 analyzes the signals for changes in amplitude of harmonics of the tool rotation speeds that are indicative of tool wear and/or tool breakage. In some such instances, harmonics are associated with the tools 34 and 54 such that changes in amplitude of a particular harmonic is indicative of tool wear or tool breakage of one of the tools 34 or 54. In the illustrated embodiment, the processing system 80 is shown connected to the accelerometer 78 via line 82, which may be any type of suitable communication line between the processing system and accelerometer.
In some embodiments, the processing system 80 controls other aspects of the milling system 10, such as spindle/motor speed, positioning of the spindles 32, 52, positioning of the block 36, and/or other features of the milling system. Accordingly, in some instances the processing system 80 is in communication with spindles 32, 52 and/or the mandrel subassembly 72. In the illustrated embodiment, the processing system 80 is in communication with the mandrel subassembly 72 via line 84, which may be any type of suitable communication line. In some instances, the processing system 80 provides instructions to the mandrel subassembly 72 regarding positioning of the block 36. In that regard, the mandrel subassembly 72 controls the position of the block 36 in one or more axes in some instances. In one particular embodiment, the mandrel subassembly 72 controls the position of the block 36 in two substantially perpendicular axes defining a plane. The processing system is also shown in communication with the spindle 32 and the spindle 52 via lines 86 and 88, respectively. The lines 86, 88 are any type of suitable communication line for sending signals between the processing system 80 and the spindles 32, 52. In some instances, the processing system 80 provides instructions to the spindles 32, 52 (or devices controlling the spindles) regarding positioning of the tools 34, 54, respectively. In that regard, the spindles 32, 52 control the position of the tools 34, 54, in one or more axes in some instances. In one particular embodiment, the spindles 32, 52 control the position of the tools 34, 54 along an axis substantially perpendicular to the plane of movement of the block 36 as controlled by the mandrel subassembly 72. In some instances, the tools 34, 54 extend substantially parallel to one another, but offset with respect to one another.
Accordingly, in some instances the processing system 80 does not continuously monitor for tool wear and/or tool breakage, but rather intermittently analyzes data from the accelerometer 78 to detect tool wear and/or tool breakage. In some instances, the processing system 80 processes the data as processing time becomes available. In that regard, in some instances one or more processes of the processing system are prioritized such that the highest priority process that is still outstanding is performed first. In some instances, the monitoring of tool wear and/or tool breakage has a lower priority than executing milling commands, such as positioning of the tools 34, 54 and/or block 36, rotation speed of the tools 34, 54, and/or other milling commands. In other instances, the processing system 80 includes a timing circuit for controlling when the data from the accelerometer 78 is analyzed. In some instances, the data is analyzed at a fixed interval. In some instances, the data is analyzed between approximately 4 and 10 times per second, or approximately once every 100 to 300 milliseconds. In other instances, the data is analyzed between approximately 1 and 100 times per second, or once every 1000 to 10 milliseconds. The data is analyzed more or less frequently in other instances. In some instances, the processing system includes memory for storing the data from the accelerometer, the analysis of the data output by the processing system, and/or other information. In that regard, in some instances a series of data points are analyzed and compared to one another to detect tool wear and/or tool breakage. In some instances, use of the series of data points reduces or eliminates the number of false positives of tool wear or tool breakage generated by the system 70.
In some instances, the milling systems and methods of the present disclosure are utilized with the milling systems and/or methods disclosed in U.S. Pat. No. 7,270,592 filed Feb. 22, 2005 and titled “Milling Machine,” hereby incorporated by reference in its entirety. Further, in some instances the milling systems and methods of the present disclosure are utilized in combination with the laser-based systems and methods of optical scanning described in U.S. Pat. No. 7,142,312 filed Dec. 30, 2003 and titled “Laser Digitizer System for Dental Applications,” U.S. Pat. No. 7,184,150 filed Mar. 19, 2004 and titled “Laser Digitizer System for Dental Applications,” U.S. Pat. No. 7,355,721 filed May 5, 2004 and titled “Optical Coherence Tomography Imaging,” and U.S. Pat. No. 7,342,668 filed Sep. 17, 2004 and titled “High Speed Multiple Line Three-Dimensional Digitalization,” each of which is hereby incorporated by reference in its entirety,
Referring to
Referring again to
Similar to the unloaded state, in some instances, the amplitude spectrum is obtained for a plurality of rotation speeds. Generally, the rotation speeds considered are those that will be used in the milling process. Further, in some instances, the tool is subjected to different types of loading during the milling process. For example, the loading is dependent on such factors as the tool rotation speed, tool path parameters (e.g., feed rate, depth of cut, stepover, etc.), the type of material being milled (e.g., ceramic, composite, zirconia, gold, etc.), tool type (e.g., cutter, grinder), tool shapes (e.g., profile, number of cutting flutes for cutters, grit size of diamonds for grinders, etc.) and/or other factors. Accordingly, in some instances the amplitude spectrum is obtained for a plurality of loading situations based on one or more of these variables.
Referring again to
Referring again to
Generally, if the amplitude of one or more of the identified harmonics varies from the established threshold for the corresponding state of the tool, then it is an indication of tool breakage. For example, in some instances the amplitude of one or more of the identified harmonics is less than the amplitude associated with the loaded state and approaches the baseline amplitude associated with the unloaded state, which is indicative of tool breakage. In some instances, the reduced harmonics are a result of the fact that the broken tool does not engage the block (or only partially engages the block) when an unbroken tool would engage the block. Accordingly, when the increased amplitude of the harmonics associated with the loaded state are not detected it is an indicator that the tool is broken.
Unlike many previous systems and methods for monitoring tool breakage, the method 90 does not require continuous monitoring of the vibrations for tool breakage. Rather, in some instances the vibrations are monitored intermittently. For example, in some instances, the processing unit controlling the milling machine and/or analyzing the data received from the accelerometer requests a sample of vibration data. The vibration data measured by the accelerometer is collected and stored in a data set. In some instances, the data set includes 2048 measured values. In other instances, the data set includes greater or fewer values. In some instances, the size of the data set is dependent on the available memory of the system and/or the available processing power of the system. The vibration data set is stored in a buffer or other memory device for retrieval by the processing unit. In some instances, the buffer or memory device is associated with the processing system. The processing system in turn analyzes the vibration data set to detect variations in the amplitudes of the harmonics indicative of tool breakage. In some instances, the processing system requests the next sample of vibration data upon the completion of the analysis of the previous data set.
Generally, the rate of data sampling is chosen such that a break of the tool is identified within 5 seconds or less after the break occurs. In some instances, the rate of data sampling is selected to be a multiple of the frequency of tool rotation in order to minimize effects of spectral leakage. In some instances, the data sets are collected and analyzed between approximately 4 and 10 times per second, or approximately once every 100 to 300 milliseconds. In other instances, the data is analyzed between approximately 1 and 100 times per second, or once every 1000 to 10 milliseconds. In one particular embodiment, the data sets are collected and analyzed approximately every 200 milliseconds. The exact timing of the data collected is checked against the stored toolpath data in order to determine the corresponding stage (e.g., loaded or unloaded) of the milling process at the time of data collection. In some instances, the stage of the milling process determines what threshold the amplitudes of the harmonics of the data sets are compared against or whether the data set should even be considered. For example, in some instances the data sets associated with non-loaded stages of the milling process are not analyzed.
In some instances, the data sets obtained from the accelerometer are processed in the following manner. First, the data sets are zero-leveled and normalized. Second, a Kaiser window is applied to reduce spectral leakage. In some instances, the Kaiser window has a beta of 7. Next a Fourier transform is performed in order to obtain a power spectrum associated with the data set. A median filter is applied to the power spectrum, and the resulting data are smoothed with a gaussian window having a width equivalent to that of the median filter. In some instances, the median filter and gaussian window have a width of 25. The smoothed power spectrum reduces the effects of the noise present in the signal, while the contributions of harmonic vibrations are excluded by the median filtering. The difference between the log of the original spectrum and the log of smoothed power spectrum is computed and the variance of this quantity is computed over the range of the relevant frequencies. Accordingly, the amplitudes of the identified harmonics are obtained from the power spectrum and normalized according to the formula: An=(Log(A)−Log(N))/Var. Whenever this value exceeds a specified threshold, the corresponding harmonic is deemed to be present in the vibrations. In some instances, the thresholds for each harmonic are determined based on the initial testing of the milling machine where the vibrations associated with the unloaded tool are monitored.
In some instances, the data set is given a binary score for each of the harmonics associated with rotation of the tool in a loaded state. If the harmonic is determined to be present, then a variable associated with the state of the tool for that harmonic is given the value of 1, otherwise the variable is given a value of 0. In some instances, each of the harmonics is monitored such that the lack of presence of a single harmonic is an indicator of tool breakage. In some instances, the sum of the variable values for all of the identified harmonics provides an overall variable score, S. In some instances, a change in the overall variable score, S, during similar milling states (e.g., loaded states) is an indicator of tool breakage. In some instances, a running average of the variable score S and/or each of the harmonics is performed over the successive data sets. Averaging over multiple data sets reduces the possibility of a false positives and false negatives. In some instances, between 10 and 20 successive data sets are utilized for detecting changes indicative of a tool break. Depending on the sample rate of the data sets, the time delay between the actual tool break and detection by the system is between 1-5 seconds and, in some instances, less than 1 second. Further, if the data set was collected during the idle or layering stage of the milling process (i.e., non-loaded stage), then the tool state variable is not included in the running average and is discarded in some instances.
Referring to
Referring to
In some embodiments, a fluid stream emits from the spindle ports 126 to wash and/or cool the blank and tools 128, 130 during the milling process. This effluent exits to the reservoir 108 where particulate matter can settle. For example, referring to
Accurate milling requires knowledge of the exact location of the tips of the tools and the x, y, z coordinates of the blank. Accordingly, very precise motors are used to move the frames 114 and 116 of the milling subassembly 110 to adjust the positions of the tools 128, 130 along the x-axis relative to the blank. Referring to
Referring to
Where the tool break is confirmed, or in situations where no confirmation is performed, the broken tool is replaced with a substitute tool by the automatic tool changer 150. The ability to engage and disengage the tools is shown in
Referring now to
Upon detecting harmonics indicative of a tool break, the method 200 continues at step 206 where the tool break is verified. In some instances, the tool break is verified using an optical sensor. In other instances, the tool break is verified using a camera. In general, the tool break is verified by comparing the current dimensions of the tool with the original, known dimensions of the tool. If the dimensions of the tool remain unchanged or within an acceptable range of the original dimensions, then the tool is not verified as being broken. Upon verifying that the tool is broken, the method 200 continues at step 208 where the broken tool is replaced with a replacement tool. In some instances, the broken tool is replaced automatically by a tool changer of the milling machine. In other instances, the milling machine sounds an alarm (visually, audible, or other human intelligible signal) that notifies a user that the tool needs to be replaced. In some instances, the alarm is sounded when no suitable replacement tool is available in the milling machine for the automatic tool changer to replace the broken tool. Accordingly, in some instances the tool replacement is performed at least partially by manual action of a user.
After replacement of the broken tool at step 208, the method 200 continues at step 210 where the milling process is resumed. In that regard, the milling process resumes at a time point earlier than the detection of the break in some instances. In some instances, the milling process resumes at a point in the milling process approximately 10 seconds prior to the point in the milling process when the break was detected. In other instances, the milling process resumes at a point in the milling process closer or further away in time to the point where the break was detected.
After the milling process has resumed at step 210, the method 200 continues again at step 204 with the monitoring of vibrations for harmonics indicative of tool breakage. In some instances, the monitoring is delayed until the milling process at least reaches the point in the milling process when the break was detected. For example, if the milling process resumes at a point 10 seconds prior to the break detection point, then monitoring of the vibrations is delayed for at least 10 seconds. Delaying the monitoring prevents a false positive of tool breakage immediately following the resumption of the milling process. In some instances, the delay goes beyond the point where the break detection occurred. The milling process continues with monitoring of the vibrations and replacement of broken tools as necessary until the milling process is completed at step 212.
Referring now to
The method 220 continues at step 224 where vibrations associated with rotation of the milling tool in a first loaded state are monitored. In that regard, in some instances the milling tool is considered to be rotating in a loaded state when it is subjected to external resistance or friction, such as that caused by engaging a milling block. In other words, the tool is loaded if it is subjected to a load outside of the inherent loads present in the milling system itself. Similar to the unloaded state, in some instances, the vibrations are monitored for a plurality of rotation speeds. Generally, the rotation speeds monitored are those that will be used in the milling process. Further, in some instances, the tool is subjected to different types of loading during the milling process. For example, the loading is dependent on such factors as the tool rotation speed, tool path parameters (e.g., feed rate, depth of cut, stepover, etc.), the type of material being milled (e.g., ceramic, composite, zirconia, gold, etc.), tool type (e.g., cutter, grinder), tool shapes (e.g., profile, number of cutting flutes for cutters, grit size of diamonds for grinders, etc.) and/or other factors. Accordingly, in some instances the first loaded state of the tool includes one or more of these variables.
The method 220 continues at step 226 where the harmonics associated with rotation of the tool in the first loaded state are identified. Specifically, in some instances the harmonics having a sufficient amplitude difference between the unloaded state and the loaded state are identified as harmonics associated with rotation of the tool in the first loaded state. In that regard, the amount of amplitude difference that is considered to be sufficient is dependent on such factors as the noise in the system, the sensitivity and/or accuracy of the accelerometer, the available processing power of the processing system, the amount of available memory, and/or other factors.
The method 220 continues at step 228 where vibrations associated with rotation of the milling tool in a second loaded state are monitored. Similar to the first loaded state, in some instances the second loaded state of the tool includes one or more of these variables associated with the different types of loading present in the milling process. The method 220 continues at step 230 where the harmonics associated with rotation of the tool in the second loaded state are identified. Specifically, in some instances the harmonics having a sufficient amplitude difference between the unloaded state are identified as harmonics associated with rotation of the tool in the second loaded state.
Finally, the method 220 continues at step 232 where vibrations associated with rotation of the tool are monitored for the harmonics identified at steps 226 and 230 to detect tool breakage. For example, in some instances the harmonics associated with the rotation of the tool in the first loaded state are monitored when the tool is supposed to be in the first loaded state in accordance with the planned milling pattern and the harmonics associated with the rotation of the tool in the second loaded state are monitored when the tool is supposed to be in the second loaded state. In some instances, the accelerometer monitors the vibrations for the harmonics associated with both loaded states when the tool is to be milling in any loaded state. In some instances, the accelerometer constantly monitors the vibrations of the system during the milling process, but the processing system only analyzes those vibrations that occur during the time periods when loading of the tool is expected (e.g., during times of cutting and/or grinding of the block). In some instances, the processing system analyzes the vibrations only when the tool is expected to be in the first or second loaded states. In that regard, the first and second loaded states have the highest probability of tool breakage in some instances. For example, in some embodiments the first and second loaded states have the largest amount of load and/or friction placed on the tool. In other instances, the first and second loaded states are two identifiable, but not necessarily high-load states of the tool. Generally, if the amplitude of one or more of the identified harmonics varies from the established threshold for the corresponding state of the tool, then it is an indication of tool breakage.
In some instances the vibrations are monitored intermittently for tool breakage. In that regard, the rate of data sampling is chosen such that a break of the tool is identified within 5 seconds or less after the break occurs. In some instances, the rate of data sampling is selected to be a multiple of the frequency of tool rotation in order to minimize effects of spectral leakage. In some instances, the data sets are collected and analyzed between approximately 4 and 10 times per second, or approximately once every 100 to 300 milliseconds. In other instances, the data is analyzed between approximately 1 and 100 times per second, or once every 1000 to 10 milliseconds. In one particular embodiment, the data sets are collected and analyzed approximately every 200 milliseconds. In some instances, the exact timing of the data collection is checked against the stored toolpath data for the milling process in order to determine the corresponding expected state of the tool (e.g., unloaded, first loaded state, second loaded, state, etc.) of the milling process at the time of data collection.
Referring now to
The method 240 continues at step 244 where vibrations associated with rotation of the spindle in a loaded state are monitored. In that regard, in some instances the spindle is considered to be rotating in a loaded state when it is subjected to external resistance or friction, such as that caused by a tool extending from the spindle engaging a milling block. In other words, the spindle is loaded if it is subjected to a load outside of the inherent loads present in the milling system itself. Similar to the unloaded state, in some instances, the vibrations are monitored for a plurality of rotation speeds. Generally, the rotation speeds considered are those that will be used in the milling process. Further, in some instances, the spindle is subjected to different types of loading during the milling process. For example, the loading is dependent on such factors as the spindle rotation speed, tool path parameters (e.g., feed rate, depth of cut, stepover, etc.), the type of material being milled (e.g., ceramic, composite, zirconia, gold, etc.), tool type (e.g., cutter, grinder), tool shapes (e.g., profile, number of cutting flutes for cutters, grit size of diamonds for grinders, etc.) and/or other factors. Accordingly, in some instances the vibrations are monitored for a plurality of loading situations based on one or more of these variables.
The method 240 continues at step 246 where the harmonics associated with rotation of the spindle in the loaded state are identified. Specifically, in some instances the harmonics having a sufficient amplitude difference between the unloaded and loaded states are identified as harmonics associated with rotation of the spindle in the loaded state. In that regard, the amount of amplitude difference that is considered to be sufficient is dependent on such factors as the noise in the system, the sensitivity and/or accuracy of the accelerometer, the available processing power of the processing system, the amount of available memory, and/or other factors. Finally, the method 240 continues at step 248 where vibrations associated with rotation of the spindle are monitored for the harmonics identified at step 246 to detect spindle wear and/or breakage. In some instances, the monitoring is performed intermittently as discussed above with respect to other embodiments.
Referring now to
In the case where a tool has become worn, the tool becomes less efficient in its grinding or cutting operation. This loss of efficiency is manifested as a decrease in the actual rotation speed of the tool in some instances. In that regard, the rotation speed of the tool becomes lower than the desired or commanded rotation speed due to the decrease in cutting efficiency. In other instances, the tool speed does not decrease but the power required to spin to the tool at the desired speed is increased because of the less efficient cutting or grinding state of the worn tool. Further, in some instances there are additional detectors in the milling machine, for example strain gauges, that are used to measure the mechanical loading of the tool on one or more axes. Accordingly, in some instances, the actual rotation speed of the tool, the instantaneous power consumed by the spindle, and/or the mechanical loading of the tool are monitored as indicators of tool wear. In some instances, these factors are monitored in addition to the harmonics to detect tool wear. In other instances, these factors are utilized to select which harmonics should be monitored for signs of increased tool wear. In that regard, the tool rotation speed, power consumption, and mechanical loading are factors associated with the loading of the tool. Accordingly, these factors are considered as part of a loading state of the tool, in some instances, and are utilized to select the harmonics that should be monitored for detecting tool wear.
Referring now to
Upon detecting harmonics indicative of tool wear, the method 260 continues at step 265 where the milling process is paused. That is, the cutting and grinding of the block is temporarily stopped in order to address the tool wear detection. The method 260 continues at step 266 where the amount of tool wear is determined and/or verified. In some instances, the tool wear is determined or verified using an optical sensor. In other instances, the tool wear is verified using a camera. In general, the tool wear is verified by comparing the current dimensions of the tool with the original, known dimensions of the tool. If the dimensions of the tool remain unchanged or within an acceptable range of the original dimensions, then the tool is not verified as being too worn for use and the tool is kept at step 268. However, upon verifying that the tool is too worn down for suitable milling use, the method 260 continues at step 270 where the worn tool is replaced with a replacement tool. In some instances, the worn tool is replaced automatically by a tool changer of the milling machine. In other instances, the milling machine sounds an alarm (visually, audible, or other human intelligible signal) that notifies a user that the tool needs to be replaced. In some instances, the alarm is sounded when no suitable replacement tool is available in the milling machine for the automatic tool changer to replace the broken tool. Accordingly, in some instances the tool replacement is performed at least partially by manual action of a user.
After the tool has either been kept at step 268 or replaced at step 270, the method 260 continues at step 272 where the milling process is resumed. In that regard, in some instances the milling process resumes at a point in the milling process earlier than the point where the milling process was paused upon detection of the wear. In some instances, the milling process resumes at a point in the milling process approximately 10 seconds prior to the pause point. In other instances, the milling process resumes at a point in the milling process closer or further away in time to the pause point.
After the milling process has resumed at step 272, the method 260 continues again at step 264 with the monitoring of vibrations for harmonics indicative of tool wear during the milling process. In some instances, the monitoring is delayed until the milling process at least reaches the point in the milling process when the milling process was paused. For example, if the milling process resumes at a point 10 seconds prior to the pause point, then monitoring of the vibrations is delayed for at least 10 seconds. Delaying the monitoring prevents a false positive of tool wear immediately following the resumption of the milling process. In some instances, the delay goes beyond the paused point where the break detection occurred. The milling process continues with monitoring of the vibrations and replacement of broken tools as necessary until the milling process is completed at step 274.
Referring now to
After the tool has either been kept at step 288 or replaced at step 290, the method 280 continues at step 292 where the milling process is resumed. In that regard, in some instances the milling process resumes at a point in the milling process earlier than the point where the milling process was paused upon detection of the wear and/or breakage. After the milling process has resumed at step 292, the method 280 continues again at step 284 with the monitoring of vibrations for harmonics indicative of tool wear and/or breakage during the milling process. In some instances, the monitoring is delayed until the milling process at least reaches the point in the milling process when the milling process was paused. The milling process continues with monitoring of the vibrations and replacement of worn down and broken tools, as necessary, until the milling process is completed at step 294.
The present disclosure has been set forth with reference to specific exemplary embodiments and figures. It will, however, be evident that various modifications and changes may be made thereto without departing from the broader spirit and scope of the present disclosure. For example, the various components, features, or steps of the different embodiments described herein may be combined with the components, features, and steps of the other embodiments described herein. Accordingly, the specification and drawings of the present disclosure are to be regarded in an illustrative sense rather than a restrictive sense.
The present application claims priority to U.S. Provisional Patent Application No. 60/988,199 filed Nov. 15, 2007 and titled “Break Detection of Grinding Tools in a Dental Milling Machine”, which is hereby incorporated by reference in its entirety.
Number | Date | Country | |
---|---|---|---|
60988199 | Nov 2007 | US |