JPH0241848A - Correcting method for thermal displacement of machine tool - Google Patents

Abstract

PURPOSE:To approximate the calculation value of a thermal deformation and an observed value by correcting the thermal displacement in the axial direction of a main shaft with the estimated calculation of a thermal deformation from temperature by separating to the thermal deformation of the nose part of the main shaft, the thermal deformation of a tool part and the thermal deformation of a base part. CONSTITUTION:A nose thermal deformation means the deformation in a Z axial line direction of the main shaft 2 part projected from a column 1, a tool thermal deformation the projection quantity in the Z axial line direction of the tool 3 part projected from the main shaft 2 and a base thermal deformation the deformation in the Z axisl direction of the space from the column 1 side of the main shaft 2 to a table 4. The outputs of temp. sensors 12, 16, a nose temp. sensor 13, base temp. sensor 15 are input to a computing element 18 via an A/D converter 17 and the thermal variation amt. in the above three Z axial directions due to temp. is operated according to logic herewith. The correction timing corresponding to work working conditions is then set(at each fixed time at tool changing time) by a correction conditions setter 19. The output of the setter 19 is output to a numerical value control device 20 and a machine Z axial original point is corrected.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a thermal displacement correction device for a machine tool. 1.
? The present invention relates to a thermal displacement correction device for a machine tool having a function of correcting elongation of a spindle in the axial direction.

[Prior art] Machine tools are equipped with heat sources in each part of the machine.

For example, there are many causes such as friction heat caused by rolling of the main shaft bearing, heat generated from the cutting part, etc. These heat sources are transmitted to various parts of the fuselage and deform the fuselage. This deformation of the machine body affects machining accuracy. Various proposals have been made in the past regarding prevention of thermal deformation of the body of a machine tool and correction devices after deformation. As a correction device for a machining center, a correction device in the X and Y axis directions is known. Furthermore, a correction device for the elongation in the Z-axis direction, that is, the elongation in the direction of the main axes, has also been proposed.

Japanese Patent Publication No. 6m-59860 creates an experimental formula for the elongation of the spindle, stores this experimental formula in the program memory, detects the temperature with a sensor installed in the spindle head and body of the machine, and calculates the elongation of the spindle. The amount of deformation is corrected by calculating using an experimental formula stored based on the detected value.

[Problems to be Solved by the Invention] However, in the correction device η, it is not always clear how each element of the main shaft deforms. The amount of deformation is simply modeled on an approximate straight line. For this reason, an error occurs in the actual deformation amount.In order to accurately correct the deformation error, it is necessary to analyze in detail which elements are thermally deformed and how, and make decisions based on the results of this analysis. Must be. This invention focuses on these problems.

The purpose of this invention is to create an empirical formula that breaks down the thermal deformation in the axial direction of the spindle into each element that makes up the spindle, and to propose a thermal displacement correction method for work fi[ that accurately predicts and corrects the thermal deformation. It is something.

Another object of the present invention is to provide a method for correcting thermal displacement of a machine tool in which the calculated value and the measured value of thermal deformation are very close to each other.

[Means and effects for solving the above problems] A method for correcting thermal displacement in the axial direction of a main spindle of a machine tool, which corrects thermal displacement in the axial direction of a main spindle of a machine tool, which includes thermal deformation of the nose portion of the main spindle, and tool heat deformation of the tool portion. This is a thermal displacement F411 positive method for machine tools that corrects thermal displacement in the axial direction of the main spindle by predicting and calculating the thermal deformation from the temperature by separating the deformation and the base thermal deformation, which is the thermal deformation of the base portion.

By decomposing the nose thermal deformation into an immediate response element, the tool thermal deformation into an immediate response element, and the base thermal deformation into an immediate response element and a first-order lag element, the correction can be made with higher accuracy.

Furthermore, if the thermal deformation of the base is calculated by linearly approximating the room temperature change, the correction can be made closer to the actually measured value.

[Principle of the Invention] FIG. 1 shows a column 1, a spindle 2, a tool 3, and a workpiece table 4 of a machining center. The nose here refers to the part where the main shaft 2 protrudes from the =1 ram 1. The tool 3 refers to a tool portion such as a cutting tool attached to the spindle 2. The workpiece table 4 is provided on the bed and is used to place a workpiece during machining and perform machining. In both cases, these sideways have a structure known as a machining center. Furthermore, although the illustrated example is a horizontal type machining center with a horizontal type main axis, a vertical type machining center with a straight type main axis is also the same, except that the names of the axes are different.

The thermal deformation of the nose hereinafter refers to the deformation of the portion of the main shaft 2 protruding from the column 1 in the ZIllI line direction. The tool thermal deformation refers to the amount of protrusion of the tool 3 portion protruding from the main shaft 2 in the Z-axis direction. Base thermal deformation refers to deformation in the Z-axis direction from column 2 +11J of main shaft 2 to table 4 t:'d.Hereinafter, actual measured values for each thermal deformation will be shown. Experiments were conducted and data were collected using a machining center manufactured by Hitachi Seiki Co., Ltd. (Hr'-300). Figure 2 shows the experimental setup.

The specifications of the experimental equipment are as follows.

Thermometer *Manufacturer: Takara Kogyo Co., Ltd. *Model: E641-2 *Person Crab element compatible thermistor sensor 5 points ('T'
1-""5) Measuring range -9,18~+79.28°C
(Usage range 0 to +40.64°C) *Output + TI-T
5 (ON/OFF of each output possible: OFF point is data O output) Serial data (Binary 16-bit parallel: 7 bits of these are used) Strobe (negative ethics) Data update: Approximately 1 second, 71 times open collector *Resolution: 0.01°C (Used resolution: 0.32° (E)) Intelligent module *Manufacturer: Hitachi Seiki Co., Ltd., Hokkaido University Crab general-purpose parallel input boat with 8 points /IV changed to L2V) *Out Crab PA quasi RS 232 C (MAX 9600 nPs
) zK Language: SE I K r-52r3 A S
I <2*Others: RAM 851 K bytes (battery backup) C> Non-contact displacement meter (capacitance type) *Manufacturer: Iwasaki Tsushinki Co., Ltd. Model: Meter section ST-3501 Probe 31
D *Measurement range: l, 0:j=0, 5 m
+n*Out crab top 5V, /±0.5…rn *Linearity: ±1% 2) Amber magnetic stand *Manufacturer: Hitachi F17 Machine Co., Ltd. E) Data logger *Manufacturer: 1" Hondenki Sanei ( Co., Ltd. *Model Nia V13 % Formula % *Out Crab Data 9! , New approx. 0.1 seconds/1 point (to PB *Resolution: 0.0OIV (resolution used: 0゜01■)) PC Manufacturer: NEC *Formula: Main unit PC-98XA Color display N5923 (1120X750 dots) ) *Human power; QP-T B, R8232C;
35:MS-oos N88 BAS[C
(1) Nose thermal deformation (a) Figure 3 shows the relationship between nose temperature and nose thermal deformation when the main shaft rotational speed is changed at a constant room temperature. The figure shows the nose temperature and nose thermal deformation when the change is saturated during continuous operation at each 1Ii1 rotation number with Cold Star I (the machine is turned off and the machine is in a sufficiently stable state). The following three points are clear from the figure showing a comparison of Hot Star 1 to (based on the state in which the machine is sufficiently stable after the hydraulic power is turned on).

b) At the point when mechanical changes are saturated, there is a linear relationship between nose temperature and nose deformation.

Mouth) In a hot start, the level of nose temperature is approximately 4.5°C higher than in a cold start.

This can be considered to be a phenomenon that depends on the capacity of the spindle cooling device.

This is a means of evaluating the spindle cooling system of a machining center based on machining with the machine.

C) The spindle rotation speed is basically not 1 < Rammeter -1.

Next, FIG. 11 shows the relationship between the nose temperature and the nose deformation from the opening of rotation fj to the stop at a constant rotation speed. The next °11 can be seen from the same IA.

2) When measuring at 10 minute intervals, fli
There is no delay in nose deformation. There is the same relationship between the nose temperature and gloss and the nose deformation as in -1- at the time of saturation.

B) If the total length of the nose is 105 m rn, the value of the linear expansion coefficient α of the nose deformation is 02×1♂r/°C1, which almost matches the linear expansion coefficient of the cast iron forming the nose.

That is, the relationship between nose temperature and nose deformation is expressed by the following equation.

ΔL=αXLXΔT (1) (ΔL: Nose deformation, α: Linear expansion coefficient, L: Nose total length, Δ]゛: Nose temperature change) (1)) Room temperature change Relationship between nose temperature and nose deformation when room temperature changes It is shown in Fig. 5 of Golden Brother. In Figure 5, as conditions for changing the nose temperature, the hydraulic power is turned on from the mechanical power OFF state (■) to circulate cooling oil in the jacket part, and the room temperature, which is the installation environment, is changed according to the pattern shown in the figure. Ta. Even when the room temperature changes, it can be said that the same theoretical formula for linear expansion expressed by equation (1) as when the room temperature is constant holds between the nose temperature and the nose deformation.

(2) Tool thermal deformation (:1) Figure 6 shows the relationship between nose temperature and tool deformation when the spindle rotation speed is changed at a constant room temperature.9. In FIG. 6, the position of the relationship between the nose temperature and tool deformation when the change is saturated during continuous operation at each rotation speed is indicated by a circle. In addition, each saturation position is drawn by a straight line from the origin of the figure. This is because although there is no direct relationship between spindle rotation speed and nose temperature [see (C) above)], the relationship between nose temperature and tool deformation at a constant rotation speed is on this line. will be explained in detail in FIG. 9).

Further, a broken line is drawn parallel to the vertical axis, crossing each line from the position where the nose temperature is 8° C. (arbitrary) and reaching the intersection point 1 with the Orpm line. From FIG. 0, the air cooling effect predetermined value of tool deformation at each rotation speed can be determined. The length of the broken line in the figure from point P to the point where it intersects with each rotation speed line is
It means the air cooling effect amount of each rotation speed with respect to the air cooling effect of Orpm. The relationship between rotation speed and tool air cooling effect is shown in the seventh section.
As shown in the figure. FIG. 7 shows the tool air cooling effect at each rotation speed when the nose temperature of FIG. 6 is 8°C.

It can be seen that the value of the air cooling effect is almost the same when the rotation speed is 4,0OOrp+n or higher, but at a rotation speed below that, the value of the air cooling effect is strongly influenced by the rotation speed. The solid line in the figure is a straight line approximation of the relationship between the rotation speed and the air cooling effect. The actual tool deformation is obtained by subtracting the value of the air cooling effect shown in FIG. 7 from the tool deformation Q r p tn shown in FIG. 6. FIG. 8 shows the tool deformation at each rotation speed using the nose temperature as a parameter. It is shown that in the region of OOOrpm or more, the amount of tool deformation is constant regardless of the rotation speed.

This means that there is no need to incorporate the rotational speed in this region when predicting tool deformation. In addition, there is no close relationship between the nose temperature and the rotational speed [see 1. Or
)Orpm, the nose temperature does not exceed 6°C (see Figure 6).

This shows that the large tool deformation below 4,000 rpm in FIG. 8 does not actually occur. therefore,
The relationship between nose temperature and tool deformation over the entire range of rotational speeds is calculated within an error of 4,00 m in Fig. 6.
It can be represented by the Orpm line.

That is, the relationship between nose temperature and tool deformation is expressed by the following equation.

Δ[, 5, 5, ΔT (Δ1..: tool deformation, Δ′ "two-nose temperature change)" Next, the nose temperature and tool deformation from the start (cold) to the stop (cold) at a constant rotation speed. The relationship between is shown in No. 9 and 1. After the start of rotation, the nose temperature increases and the tool deformation progresses as time passes.The point at which the nose temperature and tool deformation are determined every 10 minutes is 1λ
Move on a straight line drawn with a certain slope from the origin of 1. If the rotation is stopped when the change is almost saturated, approximately 1
After time, this also has a slope of I/, 3 from the origin of the figure ↑: 1
The point moves along the straight line -L drawn with , and returns to the origin.

The following can be said from the figure.

b) When measuring at 10 minute intervals, there is no delay in tool deformation due to changes in nose temperature. Nose temperature and tool deformation, including the time of saturation, have a relationship shown by a straight line drawn from the origin with a certain slope.

r7) The relationship between nose temperature and tool deformation at a constant rotation speed changes when the rotation is stopped.
・Striped! It takes about 1 hour for the transition to take place.

The point P L-P5 in 1'4 is 10,000 r
It includes the process of transitioning to the balanced state of P Ing) y F or I') Orp m. This l) 1-
FIG. 10 shows P 5 as the saturation rate with respect to the value obtained by subtracting the value on the Or p rn line from the value on the 10,000 r p In line at each nose temperature, and is calculated by 10 points for each elapsed time.

Saturation rate (nose temperature) - [Tool elongation (Pi5)
10. OOOrpm reading] / [Orpm reading - 1,00
0 rpm reading] That is, FIG. 10 means the step ripening response at Q r p m of the tool used for Test 1 (tool root main shaft end). From the figure, the temperature saturation time constant of the tool system at Or p m is approximately 13 min.

Also, contrary to Fig. 9, the balance 4JQ at Or p m
By examining the process of transition from 1 g to the balanced state during rotation, the step heat input response during rotation can be obtained. However, as shown in Figure 9, 10.000r
Since the tool deformation is already on the 10,000 rprn line at t & 10 min from the start of rotation at pm, it can be expected that the displacement saturation 11 time constant at this time will be less than a few minutes later.

b) Room temperature change When the room temperature changes, the nose temperature changes as a result. The relationship between the nose temperature and tool deformation during the process of change has already been clarified (Figure 9). Here, we consider the direct influence of room temperature changes on tool deformation. Therefore, the component caused by a change in nose temperature due to a change in room temperature is excluded. FIG. 11 shows a relationship between room temperature and tool deformation when the room temperature is changed with the spindle rotation stopped and the deformation is saturated. The following can be said from the figure.

b) At the point when tool deformation is saturated, there is a linear relationship between room temperature change and tool deformation. The tool length is 110 nm, and the linear expansion of the material (system IJa is 1.0 nm).
2X 1.0-5E/°C] -('XI- This is about twice the calculated value.

As in the case of the nose temperature in FIG. 9, it can be seen that when the room temperature changes, not only the tool itself is deformed, but also the fη position at the end of the spindle changes greatly.

Next, FIG. 12 shows the time-dependent response of tool deformation when room temperature changes. The following can be said from the figure.

Mouth) Tool deformation at Qrpm responds to changes in room temperature with a delay of about 20 minutes.

The delay in tool deformation with respect to room temperature change during Orpm in FIG. 12 is approximately equal to the result shown in FIG. 10 above. This can also be expected from the fact that there is no difference in the basic boundary conditions between the two tests. From this, it can be predicted that the tool deformation time constant due to room temperature changes during rotation is several minutes or more. That is, the relationship between room temperature change and tool deformation is expressed by the following equation.

ΔL=2.2XΔT (3) (Δ1.:
Tool deformation, ΔT: room temperature change) (3) Base thermal deformation thermistor The sensor of the thermometer was attached to the base (base of the nose) with adhesive to detect the temperature, and this was taken as the base temperature. At the same time, the sensor part of another thermistor thermometer was wrapped in clay to detect room temperature with a somewhat slower response. Furthermore, using an electric micrometer from the table, the base displacement was defined as the base' deformation.

(a) Figure 13 shows the relationship between base temperature and base deformation when the spindle rotational speed is varied at a constant room temperature. In Fig. 13, the position of the relationship between the base temperature and base deformation when the change is saturated after continuous operation at each rotation speed is indicated by a circle. From the figure, we can say the following.

b) At the point when the change is saturated, the base temperature and base deformation have a linear relationship.

Similarly, at the point when the change was saturated, a somewhat loose first-order relationship between the base deformation and the nose temperature was observed. However, if the base deformation is subtracted from the nose temperature, there will be an error of 4zm, including reproducibility. When considering the displacement between the table and the base, is it better to set a representative temperature measurement point within the path, which is both a structural path and a thermal path? 1).

Next, FIG. 13 shows the relationship between base temperature and base deformation when the rotation speed is changed. The following can be said from the figure.

(Note) When measuring at 10 minute intervals, 115% of the base temperature changes.
There is no delay in base deformation.

That is, the relationship between base temperature and base deformation is expressed by the following equation.

Δ1,=2.5XΔ'r<71>
(Δ1-two base deformation, ΔT two base temperature change) (1)
) Room temperature change When the room temperature changes, the base temperature changes as a result. The relationship between base temperature and base deformation during this deformation process has already been clarified (Fig. 14).Here, we will consider the direct influence of room temperature changes on base deformation.Therefore,
Components due to changes in base temperature due to changes in room temperature are excluded.

Figure 15 shows the state in which the main shaft rotation has stopped! The room temperature was changed in step 5, and the relationship between the room temperature change and the base deformation when the deformation was saturated is shown. In addition, in the figure, the column is 100 r in the Z-axis direction.
Data is stored when moving by nm. The following can be said from the figure.

b) At the point when base deformation is saturated, there is a linear relationship between room temperature change and base deformation. However, when the room temperature rises by 42 degrees, the space between the table and the base shifts in the direction of widening, which has the effect of offsetting the thermal deformation of the nose and the tool deformation, which is 1.15 times the room temperature.

The relationship between room temperature and base deformation is influenced by the Z-direction position of the column with respect to the daple, the Z-axis ball position, the HL of the Britten stain 1 applied to the dimple, the presence or absence of a scale feedback function, etc.

C) Lo from setting the representative column position for thermal displacement correction
At the position where the om+n column is moved, the room temperature changes by 1 degree.
An error of about 1 μm occurs in C.

This approximately corresponds to the theoretical value of thermal expansion of a 100 mm ball screw.

Next, FIG. 17 shows the response over time of base deformation such as changes in room temperature. Controllers 1 to 1 of the constant temperature equipment were operated every 15 minutes according to a pattern in which the room temperature rose by 6°C in 3 hours. It is difficult to set various room temperature change patterns using Control 1'74, which was originally intended for constant temperature equipment. However, the 17th
The pattern of chamber changes (ramp input) as given in the figure can be produced relatively easily, although unsatisfactorily.

The following can be said from the figure.

2) The base deformation responds to the ramp input of room temperature change with a delay of several hours, and it takes a long time for the base deformation to finally reach saturation even after the room temperature change has stopped.

B) At the point when the base deformation is saturated, the relationship between the room temperature change and the base deformation is along the straight line 1- shown in FIG. 15 above.

Now, the base deformation can be caused by a thermal element that has relatively no time lag in response to room temperature changes, such as the ball screw shown in section 3), or a first-order lag element, where the deformation saturates over a long period of time, as shown in section 2). Assume that In other words, there are many thermal elements that actually make up a machine, but these are represented by two characteristic thermal elements. By formulating the response of these two thermal elements to room temperature changes, it becomes possible to estimate the base displacement from the momentary room temperature. The input format for room temperature changes is as follows: (1) Because it is necessary to make the coefficients used in the formula sound like experimental values,
It should be easy to set up experimentally.

■It must be possible to approximate the pattern of actual room temperature changes.

For this reason, the lamp was powered by humans. That is, the relationship between room temperature change and base deformation is expressed by the following equation.

(Δl-: Base deformation, α: No, 1-element linear expansion coefficient,
K: temperature gradient, β: N o, two-element linear expansion coefficient, 1
・Elapsed time, τ: N0, two-element time constant) The function 1 system equation shown in equation (5) contains 3 (14 unknowns (α, β, τ
). The procedure for determining these unknowns will be explained below. Point A in FIG. 17 indicates the time when the room temperature change stopped.

Equation (5) shows the base deformation when the room temperature change 1 is used as the lamp input, so it only shows the relationship at the time below point A in FIG. 17. The base deformation after point A where the chamber 711-1-hi stopped is expressed by the following equation.

Δshi−(α+β)K・1. +β・K・τ(11/τ e )−(α+β)K(−φ)−β・K・(L
-φ/τ) τ(1-e)(6) (Δ1-two base deformation, α: NO1 element linear expansion coefficient, B
: NO2 element linear expansion coefficient, K: temperature gradient: elapsed time, τ: NO2 element time constant, φ: time to point A) The first and third terms of equation (6) are expressed as follows: These are the determined deformation components, and the second and fourth terms are components that affect the deformation due to the time delay caused by room temperature change.
It is also a component that can be ignored. That is, the base deformation expressed by equation (6) becomes the following equation after a long time.

ΔL-(α+β)K・φ (α+β)ΔT (7) <ΔI-: Base deformation, α: NOI required! i: coefficient of linear expansion, β: coefficient of linear expansion, 1 (: temperature gradient, φ: time to point A, ΔT: change in room temperature) The values of α and β in equation (7) are as shown in Figure 15. It can be determined from the slope of the straight line.

α + β 2 −4, 3[MeL 10 /′°C 1 Also, the base deformation in FIG. 17 is shifted parallel to the lamp manual force at room temperature, and the time constant τ becomes the value of the time constant τ. In FIG. 17, in response to the lamp input of room temperature change, the room temperature change stops when the base deformation is in a transient state, so an accurate value of τ is not obtained. Originally, the values of α and β can be uniquely determined by substituting this value of τ and the value of (Z+β) into the right side of equation (5) at any point in time, and substituting the actual measured displacement value from the test into the left side. Here, the value of τ was determined by trial and error method.

In FIG. 17, a point P indicates a position where a line drawn parallel to the vertical axis from point A intersects with the locus of base deformation. This point P
Set an appropriate value of τ and substitute it into equation 5). (
5) Calculate tentative values of α and β from the equation, substitute these (Z, β, and τ values into equation (6), and calculate the degree of agreement with the actual displacement value at any point in time. The correlation between the lamp input of the room temperature change and the base displacement determined as follows is as follows.

Δ1. =-・l 3XK-t+16XK (1-t-4
, 3xK (t-1φ) 16XK (1-(Δ1, two-base deformation, K: temperature gradient, 1.: elapsed time, φ: time to point A) Room temperature change is approximated by a straight line in Figure 17. FIG. 18 shows the difference between the calculated value using equation (8) and the actual deformation measurement when represented.

(11) Total thermal deformation Nose thermal deformation, tool thermal deformation, and base thermal deformation were each calculated from the data of nose temperature, base temperature, and room temperature using equations (L)-(8). Furthermore, the sum of these thermal deformations was taken as the comprehensive calculation value. In addition, the displacement of the tool tip measured from the table with a non-contact displacement meter was defined as the actual measured displacement of the tool tip.

(a) Temperature deformation of various parts at constant room temperature, and the total &31 calculated value,
FIG. 19 shows a comparison with the actual tool tip displacement 11jll. From the figure, the following ', 11 can be said.

(2) At a specific rotation speed, the calculated value is within an error of several μm from the actual Jjll displacement.

(1)) Room temperature change When there is a room temperature change, the total calculated value of thermal deformation predicted using a linear approximation of this change is shown in Fig. 20 in comparison with the measured displacement of the tool tip. In other words, ■For a thermal deformation of 40-50 μm, the calculated and actual value is approximately 5
This results in an error of μmn.

(5) Spindle elongation corrector for machining center” Through test 1-2, the relationship between machine temperature and thermal deformation of each part is determined.
This was clarified by showing each formula. The simulation mechanism shown below is a machining center spindle elongation hli positive system based on a method of predicting thermal deformation from temperature and correcting it.

(6) Simulation mechanism The relational expressions between mechanical temperature and thermal deformation obtained for nose thermal deformation, tool thermal deformation, and base thermal deformation can be basically classified here as response expressions to the following two types of thermal elements. .

B) Immediate response element: Base thermal deformation (nose temperature number) ΔL-1,02
xlOxi-xΔT (9) (ΔL; nose deformation, I
-Connose length, ΔT: Nose temperature change) Single tool thermal deformation (nose temperature number) AL=5. ”;
/s:<Δ”r (10) (Δ1-: tool deformation, ΔT two-nose temperature change)・Mig tool thermal deformation (room temperature dependence f/: min) (11) ΔL=2.2XΔT (Δ1-: tool Deformation, ΔT: Room temperature change>・Knee base thermal deformation (base temperature dependence) ΔL=2.5XΔT
(12>(ΔL two base deformation, T two base temperature change) mouth) - next lag element *Base thermal deformation (room temperature dependence) ΔL word -4,3xK-t+16xK (1-(a) Linear approximation of model room temperature Fig. 21 is an example of a linear approximation of the SIN curve, which is likened to a change in room temperature.The following method was used for the linear approximation.

(ΔI-: base deformation, K: temperature gradient, shi: elapsed time) (7) Linear approximation of room temperature change For the immediate response element, the equation can be easily programmed. However, for the -order d lag element, it is necessary to solve some problems with the software. (13
) means the response of the first-order lag element to the lamp human power. In order to actually determine the value of ΔL in the clonic state, it is necessary to know the value of the slope of temperature change with respect to time and the elapsed time since the temperature change started. That is, it is required to approximate the momentary change in room temperature by the slope from an arbitrary point in the past and convert it into a ramp input (linear approximation).

Next predicted temperature = current temperature x ΔtΔT > (next actual temperature - next predicted temperature) (N: Karan! number, Δt
a1TIl constant interval ΔT:fi value) This method means that a group of straight lines having the same reference temperature are treated as one block. When the difference between the next actual temperature and the next predicted temperature exceeds a threshold value (for example, twice in a row)
The temperature at that time (the next actual temperature) is adopted as the new Buddhist temperature, and the count number is reset. The sensor sensitivity is controlled by taking into account the resolution of the thermometer that detects room temperature and room temperature disturbances (installed on the sheet metal cover,
By appropriately selecting the threshold value, it is possible to automatically decompose and approximate changes in room temperature using a group of straight lines. FIG. 22 is an example of automatic linear approximation of the room temperature change shown in FIG. 17 using this software. The divisions of each straight line group are shown by lines parallel to the vertical axis.

(B) Accumulation of Equation (13) Equation (13) is a response equation for a group of straight lines when there is only one reference time for room temperature change, and as shown in Figures 21 and 22IyI, there are several If the room temperature is divided into a group of straight lines, it is necessary to add response equations for the second and subsequent groups of straight lines. Here, we do not show the structure of the equation when it is assumed that a maximum of five straight line groups are created for normal room temperature changes.

8L=-4,3XT4-16XK1 (L cx+)
(-L7'3.13)1 -16XK1 [1-exp (-(shi-φ1),/'
3.1311 +16 X K 2 [1-e x p (-(-φ
1)/3.131] -16xK2 [1-exp (-(-φ1-φ2>
/3.1311 +16XK3 [1exp < (t-φ1-φ2
>, /3.13)] 16XK3 [1-exp +-(t--φic/
, 2-cI) 3)/3. 131]+1.6XK4
[1-exp (-(t-φl −φ2-$3>/3
．． 13) 1 16xK4 [1-exp (-(t-!/)1-
φ2- , /, 3-φ4), /3.13)] +16x
K5 [1-exp I-(t-φl-φ2-φ3-φ
4>, /3.1311 (Δ1-two-base deformation, T: temperature potential, 1: elapsed time, K1-5: temperature gradient of each straight line group, φ1-4: duration of K1-4 6th straight line When a group is created, the influence of the temperature change on the first straight line Iff is assumed to be saturated and is ignored.In other words, only the influence of the latest five straight lines 1rt is always accumulated.

Yoshio Hegami 1 or 1 - What is shown in FIG. 23 is a functional block diagram of a control circuit for carrying out the method for correcting thermal displacement of a machine tool according to the present invention. The machining center 10 includes a room temperature sensor 12.10.
, a nose temperature sensor 13, and a base temperature sensor 15 are housed. These sensors may be of any type, but a thermistor temperature sensor is preferred. The room temperature sensor 12 detects the temperature inside the splash cover, and the room temperature sensor 16 detects the temperature of the machine installation environment.

The outputs of the room temperature sensor 12 , 16 , nose temperature sensor 13 , and base temperature sensor 15 are input to the D/D converter 17 . The input analog signal is converted to a digital signal. The output of the Δ/D converter 17 is input to the arithmetic unit 18. The calculator 18 calculates the amount of thermal displacement in the Z-axis direction according to the theory described in 1iif. The output of the operation 2t18 is output to one hli correct event setter. Amendment article f'
1. One general ruler sets the hli positive timing according to the workpiece machining conditions (every fixed time, when changing the tool). The output of the correction condition setter 19 is output to the numerical control device 20, and the machine Z-axis origin is corrected.

[Effects of the Invention] As described in detail above, the present invention can accurately correct the elongation of the main shaft in the axial direction.

[Brief explanation of the drawing]

Figure 1 is a diagram showing the main shaft displacement measurement target, Figure 2 is a diagram showing the experimental equipment, Figure 3.4.5 is a diagram showing the relationship between nose temperature and nose deformation, and Figure 6 is a diagram showing the relationship between nose temperature and tool. Figure 7 shows the relationship between deformation, Figure 7 shows the air cooling effect of the tool, Figure 8 shows the effect of air cooling on the tool.
Figure 9 shows the relationship between spindle rotation speed and tool deformation, Figure 9 shows the history of nose temperature and tool deformation, Figure 10 shows changes in the tool after rotation has stopped, and Figure 11 shows the relationship between the spindle rotation speed and tool deformation. Figure 12 is a diagram showing the relationship between room temperature change and tool deformation, Figure 12 is a diagram showing tool deformation when room temperature is changed, Figures 13 and 14 are diagrams showing the relationship between base temperature and base deformation, and Figure 15 is room temperature change. Figure 16 is a diagram showing the positional relationship between the table and the column, Figure 17 is Ia showing the temporal response of base deformation when room temperature changes, and Figure 18 is the calculated value and the measured value of deformation. Figure 19 is a diagram showing the actual measured value and total value of each thermal deformation, and Figure 20 is a diagram showing the relationship between the calculated P3 loaf value of thermal deformation and the actual measured displacement of the tool tip when there is a change in room temperature. Figure 21 is an example of a linear approximation of a sine curve based on room temperature changes, Figure 22 is an example of automatic linear approximation of the room temperature changes shown in Figure 17, and Figure 23 is an example of the invention. It is a figure showing an example of an A displacement control device. 1...Column, 2...Spindle, 3...Tool, 11...
Workpiece table, 10... machining center, 12.
16...room temperature sensor, 13...nose l,'! Degree sensor, 15...Base temperature sensor! , 17...A/D
Converter, 18...Arithmetic unit, 19...Correction condition setter, 20...Numerical controller No. 1r7゜3rd!

Claims (1)

[Claims] 1. In a method for correcting thermal displacement in the axial direction of the main spindle of a machine tool, nose thermal deformation that is thermal deformation of the nose portion of the main spindle, tool thermal deformation that is thermal deformation of the tool portion, and thermal deformation of the base portion A thermal displacement correction method for a machine tool that corrects thermal displacement in the axial direction of the main shaft by predicting and calculating the thermal deformation from the temperature separately from the base thermal deformation that is deformation. 2. In item 1, the nose thermal deformation is corrected by decomposing it into an immediate response element, the tool thermal deformation into an immediate response element, and the base thermal deformation into an immediate response element and a first-order lag element. A thermal displacement correction method for machine tools. 3. The method for correcting thermal displacement of a machine tool according to item 2, characterized in that the base thermal deformation is corrected and calculated by linearly approximating a change in room temperature.