JP2015208812A - Grinding processing device and method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a grinding processing device and a method for exactly performing grinding processing in short time without depending on an experience value, and measuring an actually ground amount.SOLUTION: A grinding processing device comprises a normal resistance detection part which detects normal resistance when grinding processing is performed, in which, when a drive part 5 is controlled so as to calculate grinding trace depth on the basis of the normal resistance detected by the normal resistance detection part and a set notch amount and to repeat feed for a plurality of paths until the grinding trace depth becomes the set notch amount when the grinding processing is performed with the set notch amount, a control part 7 calculates target grinding trace depth to calculate the number of times of paths until the grinding trace depth becomes the set notch amount, and performs the grinding processing by passing through processes of calculating grinding trace depth and specific grinding force by a path at that time by using normal resistance detected at the path, calculating normal resistance at the next path from the specific grinding force, and calculating a grinding trace depth from the calculated normal resistance.

Description

  The present invention relates to a grinding method and apparatus capable of predicting a machining time.

During grinding, the grinding machine, grindstone, and workpiece are elastically deformed by the grinding resistance. Although these elastic deformation amounts are small, they occupy several tens of percent of the depth of cut and cannot be ignored from the viewpoint of machining accuracy. Therefore, the workpiece can be removed with a desired depth of cut by grinding the same portion many times until the elastic deformation amount disappears. Here, the cycle time until the elastic deformation amount disappears becomes a problem, but at present, it often depends on the intuition of the site.

On the other hand, Non-Patent Document 1 and Patent Document 1 exist.
In Non-Patent Document 1, by inputting the grinding conditions and the type of grinding wheel as parameters, the grinding cycle is calculated from the grinding theory and experience values, and the work is performed based on the calculated values. Reductions are being made.
In “Grinding Method and Grinding Machine” of Patent Document 1, a method is proposed in which the rigidity of the system is obtained from the grinding resistance at the initial stage of grinding and the actual grinding amount, and the grinding time until the grinding resistance becomes zero is proposed. ing.

Non-Patent Document 1 uses a grinding theory, but depends on experience values, and therefore has a problem that it cannot cope with new grindstones and workpieces that have never existed.
On the other hand, since the “grinding method and grinding machine” of Patent Document 1 measures the grinding resistance and the amount actually ground, it can cope with any new combination of a grindstone and a workpiece. be able to.
However, it is currently difficult to measure the actual amount of grinding, particularly in surface grinding. Therefore, in practical use, how to measure the amount actually ground becomes a problem.

JP2013-116435A

Mechanical Technology December 2011, pages 74-75 (published by Nikkan Kogyo Shimbun)

  The problem to be solved is that methods and equipment that use grinding theory depend on experience, so it is not possible to deal with new grindstones and workpieces that have never existed, while the amount actually ground Measurement and processing methods and apparatuses are difficult to measure.

The device of the present invention supports the grindstone table so that it can be rotated and driven accurately in a short time without depending on experience values or measuring the actual amount of grinding. A grinding wheel for grinding the workpiece, a drive unit for performing the grinding by performing cutting and feeding between the grinding wheel and the workpiece, and setting the cutting depth of the cutting to perform the grinding And a control unit that controls the drive unit so as to perform a grinding apparatus including a normal resistance detection unit that detects a normal resistance when performing the grinding process, and the control unit includes the setting In consideration of the contact rigidity between the grindstone and the workpiece based on the normal resistance detected by the normal resistance detection unit and the set cut amount when the grinding is performed with the cut amount set. Calculate the grinding mark depth and feed the feed until the set cutting depth is reached. When the drive unit is controlled to repeat several passes, the grinding trace depth and the specific grinding resistance by the current pass are calculated using the normal resistance detected in the pass, and the next grinding pass is calculated from this specific grinding resistance. The normal resistance is calculated, the target grinding mark depth is calculated through the process of calculating the grinding mark depth from the calculated normal resistance, and the path of the path until the set cutting depth is reached. The number of times is obtained and the grinding is performed.

  The method of the present invention is a grinding method in which grinding is performed by cutting and feeding between a grindstone that is rotationally driven and a workpiece, and the grinding is performed by setting a cutting amount of the cutting. The grinding trace depth is calculated by taking into account the contact rigidity between the grindstone and the workpiece based on the detected normal resistance and the set cutting depth until the set cutting depth is reached. When repeating multiple passes, the normal resistance detected in the pass is used to calculate the grinding mark depth and specific grinding resistance of the pass at that time, and the normal resistance in the next pass is calculated from this specific grinding resistance. The target grinding trace depth is calculated through the process of calculating the grinding trace depth from the calculated normal resistance, and the number of passes until the set cutting depth is reached is obtained. And

  Since the grinding device of the present invention has the above-described configuration, the number of passes is calculated based on the calculation of the grinding trace depth and the specific grinding resistance by the pass at that time using the normal resistance detected by the pass. Since it can be processed and does not depend on experience values, it can handle new grinding stones and workpieces, eliminates the need to measure the amount of grinding, and sets the cut amount according to the calculated number of passes. Grinding can be performed easily and accurately until it becomes.

  Since the grinding method of the present invention has the above-described configuration, the number of passes can be obtained based on the calculation of the grinding trace depth and specific grinding resistance of the pass at that time using the normal resistance detected by the pass. In addition, the number of passes until reaching the set cutting depth can be easily and accurately determined without depending on the empirical value and making it unnecessary to measure the amount of grinding.

It is a schematic diagram of a surface grinder. (Example 1) A grinding process model is shown, (a) is an explanatory view at the start of grinding, and (b) is an explanatory view at the time of grinding. (Example 1) It is a graph which shows the relationship between a table load and a table flying height. (Example 1) It is a schematic diagram of the contact state of a grindstone and a workpiece. (Example 1) It is explanatory drawing which shows the contact arc length at the time of contact with a grindstone and a workpiece. (Example 1) It is a graph which shows the measurement result of the number of abrasive grains of a grindstone surface. (Example 1) It is a graph which shows the measurement result of abrasive grain support rigidity. (Example 1) It is a graph which shows the relationship between a particle size and the Young's modulus of a grindstone. (Example 1) It is explanatory drawing which shows a contact state with the workpiece | work in the case of accompanying the elastic deformation of a grindstone. (Example 1) (A) is a schematic front view of an apparatus, (b) is the schematic side view. (Example 1) (A) is a perspective view of a workpiece showing a grinding position, and (b) is an end view thereof. (Example 1) It is explanatory drawing which shows the measurement of the setting cutting amount. (Example 1) It is a graph of measurement data. (Example 1) It is explanatory drawing which shows the example of a measurement of the workpiece shape after grinding. (Example 1) It is a chart which shows grinding conditions. (Example 1) It is a graph which shows the measurement result and calculation result in a resinoid grindstone. (Example 1) It is a chart following FIG. 16 which shows the measurement result and calculation result with a resinoid grindstone. (Example 1) It is a graph which shows the comparison with the theoretical value and experimental value of the grinding trace depth in a resinoid grindstone. (Example 1) It is a graph which shows the measurement result and calculation result in a vitrified grindstone. (Example 1) FIG. 20 is a table following FIG. 19 showing measurement results and calculation results with a vitrified grindstone. (Example 1) It is a chart following FIG. 20 which shows the measurement result and calculation result with a vitrified grindstone. (Example 1) FIG. 22 is a chart following FIG. 21 showing measurement results and calculation results with a vitrified grindstone. (Example 1) It is a graph which shows the comparison with the theoretical value and experimental value of the grinding trace depth in a vitrified grindstone. (Example 1) It is explanatory drawing which shows the calculation method of specific grinding resistance. (Example 1) It is explanatory drawing which shows the measurement of the grinding resistance by a motor load current. (Example 1) It is a chart which shows the specification of a whetstone axial force variation meter. (Example 1) It is a calibration figure of a load sensor. (Example 1) It is a graph which shows the measurement of normal resistance by a load sensor. (Example 1) It is a graph which shows the normal resistance measured using the force sensor, and the grinding trace depth. (Example 1) It is a chart which shows the calculation result of normal resistance using specific grinding resistance, and the calculation result of grinding mark depth. (Example 1) FIG. 31 is a chart subsequent to FIG. 30 showing a calculation result of normal resistance using a specific grinding resistance and a calculation result of grinding mark depth. (Example 1) It is a graph which shows the calculation result of normal resistance using specific grinding resistance. (Example 1) It is a graph which shows the calculation result of the grinding mark depth using specific grinding resistance. (Example 1)

The device of the present invention can be driven to rotate on the grindstone base for the purpose of enabling grinding to be performed accurately in a short time without relying on experience values or measuring the actual amount of grinding. A grindstone for grinding a workpiece supported by a grinding wheel, a drive unit for performing the grinding by performing cutting and feeding between the grindstone and the workpiece, and setting the cutting depth of the incision for the grinding A grinding device including a control unit that controls the drive unit to perform machining, and includes a normal resistance detection unit that detects a normal resistance when performing the grinding process, and the control unit includes: Based on the normal resistance detected by the normal resistance detector when the grinding is performed with the set cutting depth and the set cutting depth, the contact rigidity between the grindstone and the workpiece is taken into account. And calculate the grinding mark depth until it reaches the set depth of cut. When controlling the drive unit so as to repeat a plurality of passes, the grinding path depth and the specific grinding resistance are calculated using the normal resistance detected in the pass, and the next pass is calculated from the specific grinding resistance. The normal resistance in the above is calculated, the target grinding trace depth is calculated through the process of calculating the grinding trace depth from the calculated normal resistance, and the above-described cutting amount is reached. This was realized by obtaining the number of passes and performing the grinding process.

  The said control part may consider the thermal expansion amount estimated at the time of the said grinding process between the said grindstone and a workpiece to the calculation of the said grinding trace depth.

  The method of the present invention is a grinding method in which grinding is performed by cutting and feeding between a grindstone that is rotationally driven and a workpiece, and the grinding is performed by setting a cutting amount of the cutting. The grinding trace depth is calculated by taking into account the contact rigidity between the grindstone and the workpiece based on the detected normal resistance and the set cutting depth until the set cutting depth is reached. When repeating multiple passes, the normal resistance detected in the pass is used to calculate the grinding mark depth and specific grinding resistance of the pass at that time, and the normal resistance in the next pass is calculated from this specific grinding resistance. Realized by calculating the target grinding trace depth through the process of calculating the grinding trace depth from the calculated normal resistance and determining the number of passes until the set cutting depth is reached. did.

The calculation of the grinding mark depth may take into account the amount of thermal expansion predicted during the grinding process between the grindstone and the workpiece.
In grinding processing, factors that cause uncut parts in the workpiece include a grinding machine, a grindstone, the amount of elastic deformation of the workpiece, and the amount of wear of the grindstone. On the other hand, the amount of thermal expansion between the workpiece and the grindstone can be cited as a factor for cutting the workpiece more than the set cutting amount. Thus, in the grinding process, since there are many factors that cause uncut and cut, it is difficult to quantitatively predict the grinding amount with respect to the set cutting amount.

  Here, the main factors that cause the uncut and cut as described above are the elastic deformation amount of the grinder and the grindstone and the thermal expansion amount of the workpiece. Other factors have a smaller impact than the main factors, so it may be possible to obtain them using the relational rules with the main factors. Therefore, the elastic deformation amount of the grinding machine can be calculated from the rigidity and grinding resistance of the grinding machine. The amount of thermal expansion of the workpiece can be calculated from other studies.

On the other hand, the elastic deformation amount of the grindstone can be calculated from the contact rigidity between the grindstone and the workpiece. Until now, the value measured when the grindstone is stationary has been used for the contact rigidity of the grindstone. However, even if the inventors calculated the uncut amount of the workpiece based on the contact rigidity at rest, the experimental results did not match. Therefore, we thought that this was one of the factors that could not predict the grinding amount until now.
Therefore, the inventors have been conducting research to theoretically calculate the contact stiffness of the grinding wheel during grinding, and found that the contact stiffness during grinding is higher than the contact stiffness measured at rest. . In addition, it is now possible to calculate the amount of grinding in 1-pass grinding during surface grinding using the calculated contact stiffness. In calculating this theoretical contact stiffness, it is necessary to consider the factors that cause the uncut and cut as mentioned above, and what was proposed at that time is a grinding process model. In other words, by using this theoretical contact stiffness and grinding process model, we thought that it would be possible to calculate the grinding amount during grinding, and predict the grinding time, number of grinding times, spark out time, etc. to reach the set depth of cut. .

Therefore, the purpose is to calculate the grinding amount using a grinding process model that takes into account the factors that cause uncutting and incision during grinding, and to predict the grinding time. If this can be systematized and mounted on a grinding machine, it will be possible to develop a high-value-added grinding machine that can calculate the optimum grinding time and grind itself.
Since it is difficult to measure the actual amount of grinding, a method of theoretical calculation is proposed rather than measurement. The actual amount of grinding can be obtained from the set amount of cut and the difference from the amount of elastic deformation of the grinding machine, grindstone, and workpiece.
Therefore, by measuring the rigidity of the grinder or the grindstone in advance and measuring only the grinding resistance at the time of grinding, it is possible to obtain the respective elastic deformation amounts from this rigidity and resistance.

Further, as the grinding continues, the amount of grinding gradually increases. If the grinding amount in this process can be calculated, the grinding time until the grinding amount reaches the set cutting amount can be obtained.
Therefore, by calculating the specific grinding resistance from the grinding resistance at the initial stage of grinding and using this to calculate the grinding resistance from the depth of cut in the next process, the amount of elastic deformation and grinding amount of the grinding machine and grinding wheel in each process Can be obtained. Then, by continuing this in each process, the process until the grinding amount finally reaches the cutting amount can be obtained, and the feed is repeated a plurality of passes until the grinding time, that is, the set cutting amount is reached. It becomes the number of times. One pass means that the grinding process is performed once by passing between the grindstone and the workpiece.

[Calculation of grinding amount using grinding process model]
(Grinding process model)
FIG. 1 shows a surface grinder 1 as a grinding device used for grinding. The grinding device is not limited to the surface grinding machine 1 and can be applied to various types such as a cylindrical grinding machine, a universal grinding machine, and an internal grinding machine.

The surface grinding machine 1 includes a grindstone 3, a drive unit 5, and a control unit 7.
The grindstone 3 is attached to the main shaft 11 of the grindstone base 9, and the grindstone 3 that is a grindstone wheel can be driven to rotate by the rotation of the main shaft 11. Grinding is performed by rotating and cutting the grindstone 3 and feeding the workpiece W fixed to the table 13.

  The drive unit 5 is composed of various motors built in the surface grinding machine 1. For example, the drive of the grindstone 3 is performed by the spindle motor that drives the spindle 11, and the cutting movement to the workpiece W is performed in the first way. This is done by moving the grinding wheel base 9 with one servo motor, and feeding between the grinding wheel 3 and the workpiece W, for example, by moving the table 13 in the feeding direction and the feeding orthogonal direction by the second and third servo motors. The grinding process is performed.

The control unit 7 is configured by a computer, and includes, for example, an MPU, a ROM, a RAM, and the like, and controls the driving unit 5 so as to perform the grinding by setting a cutting amount of the cutting. In this embodiment, the main shaft motor and the first, second and third servo motors are controlled.
The grinding apparatus 1 includes a normal resistance detecting unit which will be described later with reference to FIG. 25 for detecting a normal resistance when performing grinding, and the control unit 7 performs grinding with a set cutting amount. Based on the normal resistance detected by the normal resistance detection unit and the set cutting depth, the grinding mark depth is calculated by taking into account the contact rigidity between the grindstone 3 and the workpiece W, and is set as described above. When the drive unit 5 is controlled so as to repeat the feeding a plurality of passes until the cutting depth is reached, the grinding trace depth and the specific grinding resistance by the pass at that time are calculated using the normal resistance detected in the pass. The normal grinding resistance in the next pass is calculated from the specific grinding resistance, and the target grinding trace depth is calculated through the process of calculating the grinding trace depth from the calculated normal resistance. Find the number of passes until the depth of cut is reached and perform the grinding process That.
Here, the contact rigidity between the grindstone and the workpiece is a concept corresponding to the elastic deformation of the grindstone base 9 and the grindstone 3, and in consideration of this, the elastic deformation amount of the grindstone base 9 and the grindstone 3 in this embodiment. Means that

  The control part 7 can also add the thermal expansion amount estimated at the time of the said grinding process between the said grindstone 3 and the workpiece W to calculation of the said grinding trace depth as mentioned later.

  The grinding method according to the embodiment of the present invention is a grinding method that performs grinding by cutting and feeding between the grindstone 3 and the workpiece W that are rotationally driven. The grinding mark depth is calculated by taking into account the contact rigidity between the grindstone 3 and the workpiece W based on the normal resistance detected when the grinding process is performed by setting and the set cutting depth. When the feed is repeated a plurality of passes until the set cutting depth is reached, the grinding trace depth and the specific grinding resistance by the pass at that time are calculated using the normal resistance detected in the pass, and from this specific grinding resistance Calculate the normal resistance in the next pass, and calculate the target grinding trace depth by going through the process of calculating the grinding trace depth from this calculated normal resistance until the set cutting depth is reached. This can be realized by obtaining the number of passes.

  Such a grinding apparatus 1 and a processing method are not limited to so-called spark-out grinding, but the cutting amount is set in a plurality of stages, and grinding is performed by the apparatus 1 at each cutting amount. By obtaining the number of times, it is possible to accurately perform grinding of a target cutting amount in a short time without depending on experience values or measuring the amount actually ground.

More specifically, at the time of grinding, the grindstone 3 is cut with a set cutting amount, and the workpiece W fixed to the table 13 is moved while moving left and right and back and forth (feeding direction and feeding orthogonal direction). To do. The relative feed between the grindstone 3 and the workpiece W may be configured to move the grindstone 3 side in the feed direction as in the cylindrical grinding machine 1 or the like.
Here, FIG. 2 shows a grinding process model obtained by modeling this grinding process. (A), the at the time of grinding starts machining of the workpiece W by setting depth of cut a _p using grindstone 3 of diameter d _e. Then, as shown in (b), since the grinding resistance generated during grinding causes elastic deformation of the grinding machine 1 and the grindstone 3 and thermal expansion of the workpiece W and the grindstone 3, the actual grinding depth is set to a predetermined level. It is different from the loading amount a _p . The respective deformation amounts at this time can be obtained as follows.
First, the table 13 on which the workpiece W is fixed is supported by a sliding guide as shown in FIG. 1, and lubricating oil is supplied to the guide surface to improve lubrication. Therefore, when the table 13 is moved left and right, the table 13 is slightly lifted by the influence of the lubricating oil. Therefore, at the time of grinding shown in FIG. 2B, first, the table 13 flying height _ht is generated by the oil film thickness of the lubricating oil. In this embodiment, the initial contact between the grindstone 3 and the workpiece W, that is, the reference of the cutting direction during grinding is performed with the table 13 stationary.
Next, the grinding resistance is generated at the time of grinding, as shown in FIG. 1, since the grinding wheel and the workpiece is supported by the support rigidity K _s and K _w, the elastic deformation depending on the magnitude of the supporting rigidity Arise. The amount of elastic deformation of the elastic deformation of the grinding wheel of the grinding machine of this time, the [delta] _m and [delta] _w of FIG. 2 (b). Moreover, when a grindstone and a workpiece contact, a grindstone mainly deforms elastically. The rigidity at this time is called the contact rigidity between the grindstone and the workpiece. The elastic deformation amount depending on the contact rigidity is δ _con . Furthermore, the grinding wheel being ground gradually wears. Therefore, this amount of wear becomes δ _s . These elastic deformation amounts δ _m , δ _w , δ _con , and wear amount δ _s are factors that reduce the depth of cut.

On the other hand, the grinding wheel rotates at high speed during grinding, so that grinding heat is generated when it comes into contact with the workpiece. The workpiece and the grindstone thermally expand due to the influence of the grinding heat. The thermal expansion amounts at this time are h _th-w and h _th-g . The thermal expansion amount and the table flying height h _t described above is a factor to increase the cutting depth. Regarding the thermal expansion amount h _{th-w of} the workpiece, although it is thermally expanded during grinding, it is cooled and contracted after grinding, so the true depth of cut a during grinding a as shown in Fig. 2 (b) grinding traces depth a _e after grinding against _re consists deeply only h _th-w.

Therefore, considering the factors that reduce and increase the depth of cut with respect to the set depth of cut, the grinding mark depth a _e after grinding can be obtained from the following equation:

Here, since the amount of elastic deformation δ _w of the workpiece and the amount of wear δ _{s of the} grindstone are small compared to other factors, if ignored in this study, Equation (1) becomes the following equation.

  The amount of elastic deformation in equation (2) can be calculated from the grinding resistance and rigidity acting on each part. Further, regarding the amount of thermal expansion, since it is considered that the grinding heat is proportional to the grinding resistance, it can also be calculated from the grinding resistance.

(Calculation method of grinding amount)
Table flying height h _t varies by the load i.e. grinding resistance acting on the table. It also depends on the table feed speed. Therefore, the table flying height was experimentally obtained by changing the load applied to the table. Fig. 3 shows the measured table flying height. From the figure, the relationship between the applied load and the table flying height at table feed speeds of 3.3 m / min, 4.2 m / min, and 6.0 m / min can be expressed by the following equation.

The elastic deformation amount δ _m of the grinder can be calculated from the following equation from the rigidity K _m of the grinder and the grinding resistance F _n .

As a result of measuring the stiffness K _m of the grinding machine 1, it was K _m = 42.3N / μm.
The elastic deformation amount δ _con of the grindstone can be calculated using the contact rigidity K _{con of the} grindstone. The contact rigidity of the grindstone during grinding depends on the number of abrasive grains in contact with the workpiece and the abrasive support rigidity supporting one abrasive grain. Therefore, as shown in FIG. 4, it can be calculated from the product of the abrasive grain support rigidity _kgs and the number of abrasive grains in contact with the workpiece.

FIG. 5 shows the contact state between the grindstone and the workpiece. As first (a), the grinding wheel diameter d _e, when starting the grinding setting depth of cut a _p, produces a table flying height h _t by the lubricating oil of the table guide surface as shown in (b). Further, the grinding resistance causes an elastic deformation amount δ _m of the grinding machine. Therefore, true depth of cut a _'re the grinding initial stage may be represented by the following equation.

Therefore, the contact arc length l _g at this time can be obtained geometrically from the following equation.

Grinding the state of FIG. 5 (b), but shows a state before the grinding wheel 3 is elastically deformed, the contact area contact stiffness of the wheel 3 in this case, consisting of contact arc length l _g and the grindstone 3 width b Can be calculated by the product of the number of abrasive grains present in γ and the abrasive support stiffness _kgs .

Here, in order to obtain the number of abrasive grains acting on the contact area, the number of abrasive grains _ng per unit area is required. The number of abrasive grains was determined by measuring the surface shape of the grindstone 3 using a laser displacement meter and directly counting the number of abrasive grains in a unit area. FIG. 6 shows the measurement results of the number of abrasive grains per unit area. It can be seen that the number of abrasive grains increases depending on the depth from the top surface of the grindstone 3. Therefore, since the number of abrasive grains to be contacted varies depending on the depth of cut, the number of abrasive grains according to the depth of cut is formulated by formulating as shown in the relationship diagram between the depth from the surface of the grindstone 3 and the number of grains. n _g can be obtained.

Further, the abrasive grain support rigidity _kgs can also be obtained from an abrasive grain indentation test. FIG. 7 shows the measurement results of the abrasive support rigidity. This is a result obtained by applying a load to the tip of the abrasive grains and measuring the amount of displacement at that time. FIG. 7 shows the results of measurement with respect to the resinoid grindstone 3, but the vitrified grindstone 3 can be obtained with reference to the ratio of the Young's modulus of the resinoid grindstone 3 in FIG.
Therefore, the contact rigidity K ′ _con at the initial _stage of grinding can be obtained from the following equation as the product of the contact area between the grindstone 3 and the workpiece W, the number of abrasive grains per unit area, and the abrasive grain support rigidity.

The contact arc length l _g obtained so far has been based on the premise that the grindstone 3 is not elastically deformed. However, as the grinding wheel 3 that started the grinding shown in FIG. 5 (c), the resulting elastic deformation by grinding force F _n during grinding. Therefore, the contact arc length in the case where the grinding wheel 3 is elastically deformed is longer than the contact arc length obtained previously. Therefore, the contact arc length when the grindstone 3 is elastically deformed can be obtained as follows.
The elastic deformation amount δ ′ _con of the grindstone 3 is obtained by the following equation from the contact rigidity and the grinding resistance previously obtained.

Next, FIG. 9 shows a schematic diagram of a method for calculating the contact arc length l _c in consideration of elastic deformation of the grindstone 3 proposed by Rowe. The equivalent grindstone diameter in consideration of the diameter of the workpiece W and the grinding wheel 3 and d _e, the diameter of the contact arc in the case of the grinding wheel 3 is in contact with the workpiece W by elastically deformed and d _cu. Here, as with the general method of calculating the equivalent grindstone diameter during grinding using the workpiece W diameter as the grinding wheel diameter, assuming the diameter before deformation of the grinding wheel 3 and the effective whetstone diameter d _ef, d _ef can is by using the d _e and d _cu, expressed by the following equation.

Further, the grinding wheel 3 equivalent grindstone diameter d _e is in contact with the workpiece W in depth of cut a _'re, when the theoretical contact arc length at this time is l _g, these relationships are represented by the following expression.

Similarly, if the grindstone 3 is in contact with the workpiece W with the contact arc diameter d _cu with the contact arc length l _c , these relationships are as follows.

Further, when the grindstone 3 having an effective grindstone diameter _def generates an elastic deformation amount δ ′ _con and is in contact with the contact arc length l _c , these relationships are geometrically expressed by the following expression.

Therefore, when the grinding wheel 3 is ground with the true cutting amount a ′ _re and further elastic deformation δ ′ _con is generated, the contact arc length l _c between the grinding wheel 3 and the workpiece W is expressed by the equations (12), ( 13) and (14) can be obtained from the following equation by substituting them into equation (11)).

Therefore, the contact stiffness K _con of the grindstone 3 at this time can be calculated using the following equation, similarly to the equation (9).

However, depending on the contact rigidity, the elastic deformation amount and the contact arc length of the grindstone 3 are further different again. Therefore, the theoretical contact stiffness can be obtained by repeatedly calculating the equations (10), (15), and (16) until the contact stiffness becomes constant.
Therefore, the elastic deformation amount [delta] _con grindstone 3 can than contact stiffness K _con and the grinding resistance F _n of the wheel 3, is calculated using the following equation.

The thermal expansion amount h _th-g of the grindstone 3 is calculated from the following equation using the equivalent thermal expansion coefficient α ′ _g (= 0.005), assuming that it is proportional to the grinding point temperature T _eq. I made it.

Here, the grinding point temperature T _eq is required, but it is difficult to measure this, and it is not practical to measure at the processing site. Therefore, in the embodiment of the present invention, another study of the grinding point temperature in wet grinding (Heiji Yasui: Effect of grinding conditions on wet grinding temperature (2), Precision Machinery, 51, 9 (1985) 1718-1724.) As a reference, when normal grinding was performed, the grinding point temperature was assumed to be about 100 ° C., which is lower than the boiling point of the grinding fluid. Furthermore, since the grinding resistance at this time is about 100 N, it was decided to calculate from the following equation using the grinding point temperature coefficient α _g (= 1).

The thermal expansion amount h _th-w of the workpiece W can be calculated using the following formula proposed by Okuyama et al.

Where α is the thermal expansion coefficient of the workpiece W (= 1.3 × 10 ⁻⁶ ° C. ⁻¹ ), ν is the Poisson's ratio (= 0.3), and K _td-m is the temperature conductivity of the workpiece W ( = 11.4 × 10 ⁻⁶ m ² / s), v _w indicates the table 13 feed speed. The contact arc length l _cth between the grindstone 3 and the workpiece W in the _formula is a length that _takes into account all the elastic deformation amount and thermal expansion amount of the grindstone 3. This l _cth is the provisional contact arc lengths l _g and l _c obtained by considering only the flying height of the table 13 and the elastic deformation amount of the grinding machine 1 when _calculating the contact rigidity of the previous grindstone 3. Are different.
Therefore, the true contact arc length l _cth is calculated as follows. In the grinding process model shown in FIG. 2 (b), the final true cutting amount a _re is the thermal expansion amount h _th-g of the grindstone 3 that is an increase factor from the set cutting amount a _p and the table 13 flying height. Since h _t is added and the elastic deformation amounts δ _m and δ _w of the grinder 1 and the grindstone 3 as the uncut factors are subtracted, it can be expressed by the following equation.

Therefore, the true contact arc length l _cth between the grindstone 3 and the workpiece W can be obtained from the following equation in the same manner as the equation (15).

By substituting this l _cth into equation (20), the thermal expansion amount h _th-w of the workpiece W can be obtained.

By calculating the elastic deformation amount of the grinding machine 1 and the grindstone 3 and the thermal expansion amount of the grindstone 3 and the workpiece W shown above and substituting them into the equation (2), the set cutting depth a _p and grinding The grinding mark depth a _e in the first pass can be calculated from the resistance F _n .

  Next, when spark-out grinding is performed as it is, that is, the grinding mark depth after the second pass is calculated as follows.

First, if grinding is continued without changing the set cut amount, the uncut amount in the first pass becomes the cut amount in the second pass. In the previous calculation of the grinding mark depth, the set depth of cut a _p was used. However, in order to calculate the uncut length, the flying height of the table 13 under no load must be taken into consideration. During spark-out grinding, the flying height of the table 13 increases as the normal resistance decreases, and after spark-out grinding, the normal resistance becomes zero, so that the flying height is maximized. Therefore, the total cut amount a ′ _p is the sum of the set cut amount a _p and the table 13 flying height h _t0 at no load as shown in FIG.

In addition, _{ht0 at} this time can be calculated | required by making normal resistance _Fn into zero in Formula (3)-(5) which is a measurement result of the table 13 flying height.

Therefore, the uncut amount Δ _res1 of the first pass can be obtained by subtracting the grinding mark depth a _e1 of the first pass from the cut amount a ′ _p . For this reason, the cutting depth a ′ _p2 in the second pass is _Δres1 and can be expressed as the difference between the total cutting depth a ′ _p and the grinding mark depth a _e1 by the following equation.

Thus since considering table 13 flying height of the no-load to depth of cut, the truth depth of cut a _'re during grinding initial equation (7), the table 13 drop relative to the table 13 flying height h ' _t must be taken into account. Here, the table 13 lowering amount is obtained as the following equation as a difference from the table 13 flying height when there is no load.

Accordingly, the true cutting amount a ′ _re at the initial stage of grinding in the equation (7) can be expressed as follows in consideration of the total cutting amount a ′ _p and the table 13 descent amount h ′ _t .

The grinding mark depth of the second pass can be calculated by performing the calculation after the equation (8) using the true cutting amount a ′ _re at the initial stage of grinding. In addition, if the calculation formula of the grinding mark depth of the formula (2) is expressed by the above a ′ _p and h ′ _t , the following formula is obtained.

The total grinding mark depth a _E in the second pass can be obtained as the sum of the grinding mark depths in the first pass and the second pass.

  For the third and subsequent passes, the grinding mark depth can be obtained by repeating the calculation method for the second pass.

(Experimental method and experimental apparatus)
In order to confirm whether the calculation of the grinding mark depth using the proposed grinding process model is appropriate, spark-out grinding was performed during surface grinding, and the calculated grinding mark depth and the actual grinding mark depth were calculated. I decided to compare.

  FIG. 10 shows a schematic diagram of the experimental apparatus. The workpiece W was attached to the table 13 on the surface grinding machine 1 via the force sensor S. Thereby, the grinding resistance at the time of spark-out grinding can be measured. FIG. 11 shows a grinding part on the workpiece W. First, the upper surface of the workpiece W is all ground and a reference surface is produced. The workpiece W was subjected to one-pass grinding with a certain set depth of cut, followed by spark-out grinding, and the grinding mark depth at that time was measured and compared with the calculation result.

(a) Set cutting amount A method for measuring the set cutting amount a _p will be described.
As shown in FIG. 12, the electric micrometer 17 is fixed to the spindle head on the back side of the grinding machine 1 using a magnet stand 15. And the L-shaped metal plate 21 is fixed using the C clamp 19, and the terminal of the electric micrometer 17 is made to contact there. The measurement is performed once by the electric micrometer 17 in the state as it is. Next, a cut is given to the grindstone 3 and measurement is performed again.
The measurement result obtained by the electric micrometer 17 is shown in FIG. FIG. 13 shows the measurement results when the set depth of cut by eye measurement is 10 μm.
As shown in FIG. 13, the average difference between the measurement data before and after cutting is the set cutting amount. The set cutting depth a _p was measured as described above.

(b) Grinding mark depth The actual grinding mark depth was measured by measuring the depth of the grinding groove formed in the workpiece W by grinding using a shape measuring instrument. FIG. 14 shows a measurement example. The depth of the groove portion subjected to grinding was measured as a difference from both end portions which are reference surfaces.

(c) Influence of grinding fluid (coolant) When the surface grinding machine 1 performs grinding, the grinding fluid is supplied to the grinding part so that grinding burn does not occur. It has been found that the normal resistance measured by the force sensor increases due to the influence. Under this experimental condition, the appropriate amount of grinding fluid that does not cause grinding burn and does not affect grinding is about 13 l / min. When the influence on the normal resistance at that time was measured, the resinoid grinding wheel 3 was about 5 N, vitrified grinding stone. 3 was about 1N. Therefore, in this experiment, the difference between the measured normal resistance and the influence of the grinding fluid (5N for the resinoid grinding wheel 3 and 1N for the vitrified grinding wheel 3) is used as an actual normal resistance for the calculation of the calculation of the grinding mark depth described later. I will decide.

(d) Experimental procedure Table 1 shows the grinding conditions. As the grindstone 3, vitrified and resinoid grindstone 3 were used, and the peripheral speed of the grindstone 3 was 1800 m / min. As the workpiece W, steel NAK55 for plastic mold made by Daido Special Steel Co., Ltd. was used.
The experiment was performed in the following order.
(1) The workpiece W is ground with the surface grinder 1, and the grinding mark depth is actually measured with the shape measuring instrument. Further, the theoretical grinding mark depth is obtained from the depth of cut and normal resistance measured during the experiment using the above theoretical formula.
(2) Increase the number of grindings once, twice and three times and grind until the spark-out grinding is completed. This is performed with the resinoid grindstone 3 and the vitrified grindstone 3.
(3) The actual grinding trace depth is compared with the theoretical grinding trace depth for each number of grindings. Further, the grinding time given by the difference in contact rigidity due to the difference in the grindstone 3 until the end of the spark-out grinding is evaluated.

(Measurement result and calculation result of grinding amount)
The measurement results and calculation results of the grinding mark depth with the resinoid grinding wheel 3 are shown in FIGS. 16 and 17, and the comparison of the grinding mark depths is shown in FIG. Similarly, the results with the vitrified grindstone 3 are shown in FIGS. 19 to 22 and FIG.

  The contact stiffness of the grindstone 3 was calculated using the equation (16) from the measurement results of the set cutting depth and normal resistance at the time of each grinding. Then, the amount of elastic deformation and the amount of thermal expansion were calculated from the normal resistance, and the grinding mark depth after grinding was calculated using Equation (2).

  It can be considered that the grinding mark depth can be calculated by using a theoretical formula because the experimental value and the calculated value are comparable in any of the grindstones 3. It appears that the spark-out grinding has been completed in the sixth pass of the resinoid grindstone 3 and the fifth pass of the vitrified grindstone 3. In addition, some of the results of the vitrified grinding wheel 3 may be considered to have finished the spark-out grinding with the result of the third pass, but since this was only one out of several experiments, the fifth pass was finished with the spark-out grinding. It is considered.

  Next, comparing the experimental values of the vitrified grindstone 3 and the resinoid grindstone 3 until the end of spark-out grinding, it can be seen that there is a large difference in the grinding mark depth of the first pass between the resinoid grindstone 3 and the vitrified grindstone 3. It is done. There is also a difference in the increase in grinding mark depth from the first pass to the end of spark-out grinding. Vitrified grinding wheel 3 has a large amount of grinding in the first pass, and gradually goes out to the spark-out after the second pass. In contrast, in the resinoid grinding wheel 3, the grinding amount in the first pass is small, but the increase in the grinding depth is larger in the second and subsequent passes than in the vitrified grinding stone 3. As a cause of such a difference, it is considered that a difference in contact rigidity during grinding greatly influences.

  First, in the first pass, the maximum depth of cut and normal resistance in this grinding process occur. The cutting amount in the first pass is almost the same because it does not depend on the difference in the grindstone 3. Further, the normal resistance in the first pass is not significantly different from the experiment due to the difference in the grindstone 3. On the other hand, with respect to the above two points, the contact rigidity is greatly different due to the difference in the binder. From this, it can be considered that a large difference occurred in the first pass because the difference in contact rigidity greatly affects the elastic deformation amount of the grindstone 3. However, the number of spark-out grindings is not significantly different because the resinoid grinding wheel 3 has 6 passes and the vitrified grinding wheel 3 has 5 passes. Further, the amount of increase in the grinding mark depth after the second pass increases in a linear manner and reaches a spark-out. From this, it is considered that the difference in the grinding amount and the uncut amount in the first pass due to the difference in the contact rigidity of the grindstone 3 greatly affects the number of spark-out grindings.

[Prediction of grinding time using grinding resistance in the first pass]
(Calculation method of grinding resistance)
In [Calculation of grinding amount using grinding process model], it was clarified that the grinding mark depth in the spark-out grinding process in surface grinding can be calculated based on the normal resistance. However, since the normal resistance is measured using a force sensor, it is not practical. Further, if the force in each grinding pass is measured, it is only necessary to determine whether or not the grinding process has been completed using this force.

  Therefore, if the grinding mark depth in the grinding process can be obtained without measuring the grinding resistance, it is possible to predict the time required for the grinding process in advance before grinding, which is wasted in the spark-out grinding process. It can be expected to save time. The time required for the grinding process is proportional to the number of passes when the feed speed is constant, and also means the number of passes.

  Here, as a method for calculating the grinding resistance, a method using specific grinding resistance can be cited. If the specific grinding resistance is known, the grinding resistance can be obtained by calculation by multiplying the contact area between the grindstone 3 and the workpiece W.

  However, the problem here is the specific grinding resistance. It is difficult to theoretically determine this specific grinding resistance, and it is necessary to actually perform grinding to obtain this specific grinding resistance. Furthermore, it varies depending on the combination of the grindstone 3 and the workpiece W, and also varies depending on the grinding conditions. Therefore, in order to obtain the specific grinding resistance, there is no choice but to calculate the specific grinding resistance by actually measuring the grinding resistance using the grindstone 3 used on the spot, the workpiece W and the grinding conditions.

  Therefore, at the initial stage of grinding, for example, when the grinding resistance of the first pass is measured and grinding is continued, if the specific grinding resistance is calculated by NC using this grinding resistance, the grinding mark depth until the grinding is completed I thought it was possible to predict. That is, the specific grinding resistance is calculated from the grinding resistance in the first pass, the grinding resistance in the subsequent pass is predicted from the specific grinding resistance, and the grinding mark depth is obtained from the grinding resistance. Then, while performing the grinding after the next pass, the number of passes to reach the desired grinding mark depth is calculated instantaneously, and when the number of passes is reached, the grinding is terminated. It becomes possible to obtain the workpiece W with appropriate machining accuracy in the grinding time.

  At present, the grinding resistance is measured by attaching a force sensor S between the workpiece W and the table 13. Although it is possible to measure with high accuracy, it is difficult to attach the force sensor S to each workpiece W at the site, which is not practical.

  Therefore, the grinding resistance during grinding is changed to that using a wattmeter that can be easily measured from the power consumption of the motor. As a result, the grinding resistance can be measured relatively easily on site. If the grinding time can be predicted from this grinding resistance, it is most practical.

  Here, a method for obtaining the specific grinding resistance from the grinding resistance in the first pass and predicting the grinding time will be described below.

First, the grinding resistance in the first pass is measured and the specific grinding resistance K is calculated. The specific grinding resistance K is calculated from the normal resistance F _n1 of the first pass and the grinding area _{Ag 1} . Grinding area A _g1, as shown in FIG. 24 can be obtained as the product of the first pass of the truth depth of cut a _re1 and the grinding width b.
The true cutting amount a _{re1 in} the first pass can be obtained from equation (21). Next, the grinding area A _g1 during grinding can be calculated by the following equation from the product of truth depth of cut a _re1 grindstone 3 width b.

Therefore, the specific grinding resistance K can be obtained from the following equation by dividing the normal resistance F _n1 of the first pass by the grinding area _Ag1 at the initial _stage of grinding.

Subsequently, the normal resistance F _n2 of the second pass is calculated from the specific grinding resistance K obtained by Expression (31). The cutting amount a ′ _{p2 in} the second pass can be obtained from Expression (24).

The provisional grinding area _A′g2 is calculated from the following equation from the cutting amount _a′p2 and the grindstone 3 width b in the same manner as the equation (30).

Then, based on the calculated grinding area A ′ _g2 and the specific grinding resistance K, a provisional normal resistance f _n2 of the second pass is calculated from the following equation.

This provisional normal resistance is used to determine the amount of table 13 descent and the amount of elastic deformation of the grinding machine 1, and finally the amount of descent of the table 13 and the amount of elastic deformation of the grinder 1 as will be described later. After calculating the depth of cut taking into account, the final normal resistance is calculated.

This normal resistance f _n2, since the table 13 drop h '' _t2 and the elastic deformation of the grinding machine 1 [delta] '' _m occurs, the depth of cut a '' _p2 of the second pass in consideration of these following formula Can be calculated.

Therefore, the grinding area A _g2 in the depth of cut a '' _p2 is

It can be expressed as.

Then, using this grinding area _Ag2 , the final normal resistance _Fn2 of the second pass is calculated from the following equation.

Accordingly, the grinding mark depth a _e2 of the second pass is calculated from the calculated normal resistance F _n2 of the second pass using the grinding process model described in [Calculation of grinding amount using grinding process model]. It becomes possible.

  From the third pass, the same calculation can be performed to calculate the normal resistance and the grinding mark depth. Then, the calculation is repeated until the desired number of grindings, and the number of passes at the end of spark-out grinding is calculated.

(Measurement of grinding resistance using a power meter)
At present, the grinding resistance is measured using a force sensor installed under the workpiece W. Since this is not practical in the field, the grinding resistance can be measured using a load sensor that can measure the power consumption from the output current of the motor.
As shown in FIG. 25, the load current of the spindle motor 11a changes according to the grinding resistance. Therefore, a load sensor 23 was attached to measure the motor load current. FIG. 26 shows the specifications of the load sensor 23 used. Since the output is a current value manufactured by Elphi Co., Ltd., the load current can be measured using an AD conversion board by converting the voltage value to a voltage value via a resistor.

  Since the output of the load sensor 23 is power consumption, this power consumption must be calibrated to the grinding resistance. Therefore, it was decided to perform grinding using the current force sensor and calibrate the power consumption to the grinding resistance.

  The grinding resistance obtained by the load sensor 23 is tangential resistance. In the embodiment of the present invention, the amount of elastic deformation of the grinding machine 1 and the grindstone 3 and the amount of thermal expansion of the workpiece W and the grindstone 3 are calculated using the normal resistance. Therefore, the normal resistance must be calculated by multiplying the tangential resistance by the grinding dichotomy force ratio. However, since the grinding two-component force ratio varies depending on the grinding conditions, it is difficult to identify a certain value. Therefore, the normal resistance is determined directly from the power consumption by calibrating the current consumption obtained from the load sensor 23 with the normal resistance.

  FIG. 27 shows the result of calibrating the voltage value of the current consumption of the load sensor with the normal resistance. It can be seen that it can be calibrated linearly with some variation. When the normal resistance is small, the load cell cannot measure small power consumption. Therefore, the calibration line does not pass through the origin. However, in this measurement method, it is only necessary to measure a large normal resistance in the first pass at the start of grinding, and there is no problem even if a small grinding resistance cannot be measured. Therefore, the normal resistance may be measured using this calibration line.

  FIG. 28 shows an example of measurement results of normal resistance. The horizontal axis shows the number of samplings, but corresponds to the grinding time. It can be seen that the normal resistance increases with the start of grinding, and the normal resistance is almost constant during grinding. Thus, it was found that the normal resistance can be measured by using the load sensor 23 without installing a force sensor on the workpiece W.

(Experimental method and experimental apparatus)
In the experiment, without using a load sensor, the specific grinding resistance was calculated using the measurement result of the first pass using the force sensor described in [Calculation of grinding amount using grinding process model], and the normal resistance was calculated. Obtained and examined whether the grinding mark depth can be calculated. The reason why the load sensor was not used is that it is difficult to measure with a load sensor in the region where the normal resistance is small as described above. This is because it cannot be confirmed whether the result is valid. This experiment method uses a force sensor, but if this result is reasonable, a similar result can be obtained by measuring the normal resistance of the first pass with a large depth of cut with a load sensor. Conceivable.

  Therefore, the present experimental method and experimental apparatus are the same as described in [Calculation of grinding amount using grinding process model].

(Calculation result of grinding resistance and prediction of grinding time)
FIG. 29 shows the results of measuring the normal resistance at the time of 10-pass grinding measured by using a force sensor and the grinding trace depth generated at that time with a shape measuring instrument. The specific grinding resistance was obtained from the normal resistance of the first pass at this time, the normal resistance in each pass was calculated using the specific grinding resistance, and compared with the measurement results. Further, the grinding mark depth was calculated from the calculated normal resistance and compared with the measurement result.

FIG. 30 and FIG. 31 show the calculation results of normal resistance using specific grinding resistance and the grinding mark depth calculated using this normal resistance. First, the specific grinding resistance was calculated by dividing 37.5 N, which is the normal resistance of the first pass of grinding, by the grinding area. The grinding area at this time is determined by the product of the true cutting amount a _re = 10.2 μm and the grinding width b = 25 mm in the first pass. Therefore, the specific grinding resistance is 147.1 N / mm ² from the equation (31).

The grinding mark depth in the first pass can be calculated as 10.7 μm using the normal resistance measured by the force sensor. Therefore, the cutting depth of the first pass is subtracted from the cutting depth of 12.7 μm (= set cutting depth of 10.3 μm + table 13 flying height of 2.4 μm) in the first pass. 2.0 μm. Using this grinding mark depth and grinding width and the calculated specific grinding mark depth, the provisional normal resistance F ′ _n2 in the second pass is calculated to be 7.24N. Accordingly, when the amount of elastic deformation of the temporary grinding machine 1 and the amount of descent of the table 13 are obtained using this normal resistance, δ ″ _m = 0.17 μm and h ″ _t = 0.23 μm. Therefore, the provisional cut amount is 1.57 μm, and the normal resistance to be obtained is 5.76 N from the equation (35). Using this normal resistance, the amount of elastic deformation and the amount of thermal expansion are sequentially calculated, and when these are used to determine the grinding mark depth, it is 11.3 μm. The results of calculating the above for each path are shown in the charts of FIGS.

  FIG. 32 shows the normal resistance in each pass calculated using the specific grinding resistance. The dotted line is the result of measurement using a force sensor, and the solid line is the result of calculating the normal resistance after the second pass. From the figure, it is qualitatively consistent, and it was found that a certain amount of normal resistance can be calculated by using specific grinding resistance.

  Note that the normal resistance can be calculated with higher accuracy by considering the up-cut and the down-cut, which are grinding conditions, and considering the difference in specific grinding resistance according to the cutting depth.

Next, FIG. 33 shows the result of obtaining the grinding mark depth in each pass using the calculated normal resistance. From the figure, it can be seen that there is no change in the grinding mark depth after the second pass, and it is a substantially constant value. This is probably because the normal resistance shown in FIG. 32 did not change so much, and the calculation result of the grinding mark depth did not change.
In addition, the calculation result of the grinding scar depth can be made more consistent with the measurement result by considering the influence of up-cut and down-cut in the calculation of normal resistance and the calculation method of specific grinding resistance due to the dimensional effect. .

  The prediction of the grinding time can be obtained as follows.

The end of grinding can be defined as when the grinding mark depth reaches the depth of cut. Therefore, when the calculation result of the grinding mark depth in each pass at the time of spark-out grinding reaches the cutting amount, it can be regarded as the end of grinding, and the time until that point can be predicted as the grinding time.
However, in grinding, the grinding mark depth does not necessarily reach the depth of cut. For example, the cutting depth in the experimental result shown in FIG. 33 is 12.7 μm which is the set cutting depth 10.3 μm plus the floating height 2.4 μm when the table 13 is not loaded. Grinding ends when the time reaches. However, the grinding mark depth does not actually reach the depth of cut, and in the measurement results using the force sensor shown in the chart of FIG. 29, the normal resistance at the 10th pass is 1 N or less. Regardless, the amount of uncut remains, and even if grinding is continued thereafter, it is difficult to think that the grinding mark depth reaches the depth of cut.

  Therefore, paying attention to the geometric tolerance at the time of machining, for example, if it is considered that the grinding should be performed so that the shape fits within a certain tolerance with respect to a certain dimension, It is thought that time can be calculated.

  Therefore, in FIG. 33, looking at the transition of the calculation result of the grinding mark depth using the force sensor, for example, the grinding is completed when the depth of cut is within a tolerance of ± 1 μm with respect to the cutting depth of 12.7 μm. Then, the grinding mark depth is 11.9 mm in the ninth pass, and the grinding is finished at this point. As described above, when it is considered that the grinding end time should be within a certain tolerance with respect to the shape, it is possible to predict the time (number of passes) until the grinding is finished even when a grinding error occurs.

  As described above, a grinding process model was proposed for the purpose of quantitatively calculating the grinding mark depth during grinding, and it was examined whether it could be applied to a spark-out grinding process.

  In the grinding process model, the elastic deformation amount of the grinder 1 and the grindstone 3 generated during grinding, the thermal expansion amount of the grindstone 3 and the workpiece W, and the flying height of the table 13 are taken into account, and these are calculated based on the normal resistance. Proposed. In particular, the contact rigidity of the grindstone 3 could not be theoretically calculated so far, but this has been made possible.

  To determine whether the proposed grinding process model is valid, calculate the amount of deformation using the normal resistance measured during grinding, calculate the grinding mark depth using them, and compare with the actual grinding mark depth. did. As a result, it is possible to determine the grinding mark depth in each pass using the normal resistance during spark-out grinding, and the grinding mark at the end of spark-out grinding, which is the cumulative grinding mark depth in each pass. It was clarified that the depth can also be obtained quantitatively.

Therefore, an attempt was made to calculate the normal resistance using the specific grinding resistance from the viewpoint that if the normal resistance can be calculated in advance, the grinding mark depth can be calculated before grinding. In actual grinding, the specific grinding resistance varies depending on the combination of the grinding wheel 3 and the workpiece W and the grinding conditions. Therefore, the specific grinding resistance is calculated from the normal resistance of the first pass at the initial stage of grinding. The normal resistance in each pass was calculated using the grinding resistance. As a result, it was clarified that it qualitatively agrees.
However, since the grinding direction repeats up-cutting and down-cutting during surface grinding, the specific grinding resistance varies depending on this difference. Moreover, since the form in which the workpiece W is removed differs from cutting, prowing, and rubbing due to the difference in the cutting depth, the specific grinding resistance also differs.
Therefore, the calculated normal resistance is different from the measured results, but if the normal resistance can be calculated in consideration of the above, the normal resistance and the grinding mark depth during spark-out grinding can be obtained more quantitatively. Showed the possibility of being.

  Further, regarding the prediction of the grinding time, generally, the time when the grinding mark depth reaches the cutting depth is regarded as the end of grinding. However, in actual grinding, the cutting depth may not reach the cutting depth due to a grinding error, so the grinding trace depth does not always reach the cutting depth. Therefore, focusing on geometric tolerances, it was proposed that if the grinding mark depth falls within a certain intersection, it is considered that the grinding is finished. Therefore, the specific grinding resistance is obtained using the normal resistance measured in the first pass, the normal resistance is calculated from the specific grinding resistance during grinding, and the grinding mark depth in each pass is calculated to be within the geometric tolerance. The grinding time can be predicted during grinding by comparing whether it fits.

  Thus, since it is difficult to measure the amount actually ground, the embodiment of the present invention proposes a grinding apparatus and method for theoretically calculating and grinding instead of measuring this. The actual amount of grinding is obtained from the difference between the set cutting amount and the elastic deformation amount of the grinding machine 1, the grindstone 3, and the workpiece W. Therefore, by measuring the rigidity of the grinding machine 1 and the grindstone 3 in advance and measuring only the grinding resistance at the time of grinding, it is possible to obtain the respective elastic deformation amounts from this rigidity and resistance.

Further, as the grinding continues, the amount of grinding gradually increases. If the grinding amount in this process can be calculated, the grinding time until the grinding amount reaches the set cutting amount can be obtained. Therefore, if the specific grinding resistance is calculated from the grinding resistance at the initial stage of grinding, and the grinding resistance is calculated from the depth of cut in the next process, the elastic deformation amount of the grinding machine 1 and the grindstone 3 in each process and It is possible to obtain the grinding amount. And by continuing this in each process, the process until the grinding amount finally reaches the cutting amount can be obtained, and this is the grinding time.
In the embodiment of the present invention, since only the grinding resistance at the time of grinding needs to be measured, there is no need to provide another measuring device such as a grinding amount.
In addition, as an advantage of measuring only the grinding resistance during grinding, regardless of the combination of the grindstone 3 and the workpiece W that are actually used, the actual grinding amount and grinding time regardless of the experience value. Can be calculated and does not have to depend on the experience value.

1 Surface grinding machine (grinding equipment)
3 Grinding wheel 5 Drive unit 7 Control unit 23 Load sensor (normal resistance detection unit)

Claims (4)

A grindstone that is supported by a grindstone table so as to be rotationally driven and grinds a workpiece;
A drive unit that performs the grinding process by cutting and feeding between the grindstone and the workpiece;
A control unit that controls the drive unit so as to perform the grinding by setting a cutting amount of the cutting;
A grinding apparatus comprising:
A normal resistance detector for detecting normal resistance when performing the grinding,
The control unit, between the grindstone and the workpiece based on the normal resistance detected by the normal resistance detection unit and the set cut amount when performing the grinding with the set cut amount When the drive unit is controlled so that the feed is repeated a plurality of passes until the set cutting depth is reached, the normal resistance detected in the pass is calculated. The process of calculating the grinding trace depth and specific grinding resistance by the pass at that time, calculating the normal resistance in the next pass from this specific grinding resistance, and calculating the grinding trace depth from this calculated normal resistance. By calculating the target grinding mark depth by passing through and obtaining the number of passes until the set cutting depth is reached, the grinding process is performed.
A grinding apparatus characterized by that. The grinding apparatus according to claim 1,
The control unit takes into account the amount of thermal expansion predicted during the grinding process between the grindstone and the workpiece in the calculation of the grinding mark depth,
A grinding apparatus characterized by that. A grinding method for performing grinding by cutting and feeding between a rotationally driven grindstone and a workpiece,
Grinding marks taking into account the contact resistance between the grindstone and the workpiece based on the normal resistance detected when performing the grinding process by setting the depth of cut and the set depth of cut When calculating the depth and repeating the feed a plurality of passes until the set cutting depth is reached, the grinding trace depth and the specific grinding resistance by the current pass are calculated using the normal resistance detected in the pass. The normal grinding resistance in the next pass is calculated from the specific grinding resistance, and the target grinding trace depth is calculated through the process of calculating the grinding trace depth from the calculated normal resistance. Find the number of passes until reaching the depth of cut,
A grinding method characterized by that. A grinding method according to claim 3, wherein
The calculation of the grinding mark depth takes into account the amount of thermal expansion predicted during the grinding process between the grindstone and the workpiece,
A grinding method characterized by that.

Images