JP2575780B2 - Positioning error correction device for numerically controlled machine tools - Google Patents

Description

The present invention relates to a numerically controlled machine tool that performs positioning control and includes a D / A converter in a transmission system of a speed command value, and particularly a D / A with a high bit number. The present invention relates to a positioning error correction device for a numerically controlled machine tool that automatically corrects a positioning error caused by a converter.

(Prior art) Numerical control (hereinafter simply referred to as NC) axes positioning errors can be roughly classified into the following: (a) Control loop gain due to sliding surface condition of NC shaft (b) Control system The above factors (a) are dominant in general NC lathes and machining centers, although there are some factors due to the error factors of the components. The positioning error caused by the above factor (a) can be improved by increasing the control loop gain, but is limited by the problem that micro vibration occurs when the machine tool moves or stops.

By the way, the prior art in the present invention includes, in particular, a shaft composed of a short stroke and a hydrostatic sliding surface instead of a sliding sliding surface and a rolling sliding surface with a long stroke like a general NC lathe and a machining center shaft. Is the target condition.
The positioning error in this case is dominated by the above factor (b). Since the conventional NC axis employs a completely digital control system, the D / A converter is not provided in the speed command value transmission system. Therefore, the factor (b) is also assumed to be very small. However, analog control systems have better frequency response than complete digital control systems, and are particularly advantageous in applications that follow complex command shapes at high speed. Is widely used. Therefore, in the present invention, the NC axis using the analog control having a D / A converter in the transmission system of the speed command value satisfies the above-described conditions, has a good followability even when a complicated command shape is input, and is a conventional technology. Think as an object.

In the NC axis having a D / A converter in the transmission system of the speed command value, a positioning error based on the factor (b) becomes a problem due to the above condition. In the above factor (b), the drift characteristics of the D / A converter greatly affect positioning errors due to analog control. In particular, with a high-bit D / A converter, the positioning error increases in accordance with the number of bits. However, a D / A converter with a high bit number is required to follow a complicated commanded shape at high speed and achieve positioning accuracy on the order of a digitally controlled ordinary axis.

Here, an example of a block configuration of a control system according to the related art using analog control will be described with reference to FIG. This conventional device includes a linear motion mechanism 1, a position detector 2 for detecting a moving position of the linear motion mechanism 1, a function generator 3 for generating a position command value PC, and a position detection device for detecting the position detected by the position detector 2. A subtractor 4 for subtracting the value DP from the position command value generated by the function generator 3, a D / A converter 5 for converting a position deviation value PE output from the subtractor 4 into an analog signal, and a D / A converter A servo amplifier 6 for amplifying the analog signal output APE of the converter, and the servo motor 7 is driven by the amplified signal SA to move the linear motion mechanism 1.

In such a positioning control system, the feed rate v:
^mm / _sec ], position deviation PE [mm], and position loop gain
It is known that the relationship with Kp [sec ^-1 ] is generally as follows.

PE = v / Kp (1) D / D required to represent the position deviation value PE [mm] at this time
Assuming that the number of bits of the A converter 5 is n [bit] and the positioning resolution is x [mm], PE = x (2 ⁿ -1) (2), and from the above equations (1) and (2) Becomes Here, assuming that the position loop gain Kp is a constant, to increase the positioning resolution x (x → small) and increase the feed speed v (v → large), use the D / A converter This corresponds to increasing the bit number n of 5. Therefore, it is understood that the adoption of the D / A converter having a high bit number is indispensable for positioning control capable of high-resolution and high-speed feeding.

(Problems to be Solved by the Invention) However, adoption of a high-bit-number D / A converter having a resolution higher than the practical resolution has been meaningless for the following reasons. For example, using a unipolar (unipolar) type n / bit D / A converter (corresponding to (n + 1) bits when including positive and negative polarities), this output voltage width V _FSR [v] and drift coefficient Δd [ ^FSR / ° C.], putting the temperature change [delta] T [° C.], the drift voltage VΔ _T [v] is due to temperature change, a _{V Δ T = Δd × V FSR} × ΔT ...... (4), 1 unit of positioning resolution Output voltage V _X corresponding to
[V] is V _X = V _FSR / (2 ⁿ −1) (5) Since the positioning error ER [mm] is Becomes Here, assuming that a commercially available unipolar 16-bit D / A converter is used, Δd = 15 [ppm of ^FSR / ° C]
= 1.5 × 10 ^-5 [ ^FSR / ° C], ΔT = 30 [° C], and the number of bits corresponding to the resolution of a practical D / A converter such that the positioning resolution is 1 unit or less even with the above temperature change is n '. [B
it] Next, become the 'equation ER / x = 1 Distant, n = n' (6) and replace the 1 (= ER / x) ≧ Δd · ΔT · (2 n '-1) ...... (7) . Thus, solving for n ', Δd = 1.5 × 10 ^-5 [ ^FSC / ° C], ΔT = 30 [° C]
Under these conditions, the practical resolution of the unipolar 16-bit D / A converter was found to be about 11 bits. Since the speed command voltage is bipolar, if there is a polarity bit outside the D / A converter, the above unipolar 16-bit D / A
The converter has 17 bits, and the practical resolution is about 12 bits.

Here, the drift characteristics of the analog amplifiers other than the D / A converter are considered to be not a problem because the drift characteristics of the drift characteristics of the D / A converter are about 1/10 to 1/100. This analog amplifier is used for inverting the polarity of the output voltage of the unipolar D / A converter and setting the position loop gain.

In short, the configuration of the NC axis is short stroke, the movable part equivalent mass is lighter than a general NC lathe and machining center axis (several hundred mm to several tens m → 20 mm, several hundred kg to several t → 2 kg), and the sliding surface The analog axis is controlled by D / A conversion of the speed command value to respond to a complicated command shape with the NC axis manufactured so that the occurrence of the slip stick is small (static pressure guide). High resolution with special NC axis
To achieve high-speed feeding, a high-bit-number D / A converter is required. However, if a high-bit D / A converter is used due to the drift characteristics of the D / A converter itself, a positioning error occurs. For example, if the above-mentioned commercially available unipolar 16-bit D / A converter is used under the above conditions, the equation (6) gives Δd = 1.5 × 10 ⁻⁵ [ ^FSR / ° C.], ΔT = 3
Substituting 0 [° C.], n = 16 [bit], x = 0.1 × 10 ⁻³ [mm] results in ER ≒ 2.9 × 10 ⁻³ [mm]. Therefore, an error occurs as much as 29 times the positioning resolution.

The present invention has been made under the circumstances described above,
An object of the present invention is to provide a high-bit-number D / A converter when a high-bit-number D / A converter is adopted for an NC machine tool having the above-mentioned special NC axis.
Provide a device that corrects positioning errors due to the influence of the drift characteristics of the A / A converter, and provides a positioning error correction device that enables high-speed feeding while maintaining high positioning resolution while using a high-bit-number D / A converter. It is in.

(Means for Solving the Problems) The present invention relates to a positioning error correction device for an NC machine tool that performs positioning control and includes a D / A converter in a transmission system of a speed command value.
A correction value calculation start signal generating means for detecting that the control target is not moving and waiting for a predetermined time to elapse, and then generating a timing of the positioning error correction; and a correction value calculation start signal generating means. Correction operation means for calculating a positioning error by a start signal, correction value storage means for storing the correction value calculated by the operation means, and addition means for adding the correction value stored in the storage means to the position command value. This is achieved by providing According to the present invention, the positioning error is automatically corrected for each of the above timings, and the high bit D /
Even when an A converter is used, the positioning error can be reduced.

(Operation) In the present invention, a correction value calculation start signal is generated at a timing at which a positioning error can be corrected, and a correction value obtained by performing a predetermined correction calculation is added to a position command value from a function generator. As a result, even if a high-bit D / A converter is used for the speed command value transmission system, the positioning error due to the drift characteristic is removed, and the positioning error can be reduced.

(Embodiment) The block diagram of FIG. 1 shows an embodiment of the present invention corresponding to FIG. 5, and a position command value PC from a function generator 3 is shown.
Is input to the correction value calculator 9 and also to the adder 11. The position detection value DP from the position detector 2 is input to the correction value calculator 4 and also to the subtractor 4,
The start signal ST generated from the correction value calculation start signal generator 8 at the timing when the positioning error can be corrected is input to the correction value calculator 9 and the calculated value obtained by the correction value calculator 9 is calculated.
COMP (N + 1) is input to the correction value storage 10, and the storage value COMP is input to the adder 11.

The operation of such a configuration will be described with reference to the flowchart of FIG.

The correction value calculation start signal generator 8 has various combined functions. First, it is determined whether or not the axis for performing automatic correction is moving based on a program command (step S1). It proceeds to measure the time from the movement and stop, waits for the elapse of a predetermined time, determines that the positioning is completed, and gives the correction value calculator 9 an activation signal ST. The determination of the completion of the positioning in the step S2 is based on the assumption that the elapsed time after the movement speed or the position of the axis after the completion of the axis movement command in the step S1 becomes equal to or less than each set value is equal to or more than the set value. Is also feasible. Next, the correction value calculator 9 calculates the position detection value DP based on the input start signal ST.
[Mm] is averaged, and the average value of the current position is ▲ ▼ [m
m] (step S3), and a positioning error ER with respect to the position command value PC from the function generator 3 is obtained from the following equation (9) (step S4).

ER = PC- ▲ ▼ (9) Here, assuming that the previous correction value is COMP [mm] and the new correction value is COMP (N + 1) [mm], COMP (N + 1) = COMP (N ) + ER (10) (step S5). Where N is the number of corrections N = 0, 1, 2,
, And COMP (0) = 0. The correction value COMP (N + 1) calculated by the correction value calculator 9 as described above is stored in the correction value storage 10 (storage value COMP = COMP (N + 1)) (step S6), and it is determined whether to continue the automatic correction. Then, if continuing, the above operation steps S1 to S6 are repeated (step S7).
If not continued, the automatic correction ends. The correction value storage 10 holds the current correction value COMP until the next correction value calculation, and the stored value COMP is added to the position command value PC by the adder 11.

In the NC axis employing the above-mentioned fully digital control, the positioning error caused by the D / A converter, which is a problem in the present invention, does not occur. Therefore, when trying to follow the trajectory of the command shape that these NC axes cannot follow, it is necessary to add a machining axis using analog control. Next, a specific example applied to lead surface processing of a VTR or a digital audio tape (hereinafter abbreviated as DAT) will be described. That is, FIGS. 3A and 3B show the plane and side surfaces of the apparatus to which the present invention can be applied, where the extension direction of the rotation center line of the spindle motor 12 is the Z axis, and the vertical direction is the X axis. The spindle motor 12 and the center stand 17 are held on the NC lathe bed via the sliding surface, and the Z-axis motor 18 and the X-axis motor 19 are directly fixed on the NC lathe bed. It has been omitted. In this NC lathe, movement in the Z-axis direction is performed by moving the spindle motor 12 with respect to the bed, and movement in the X-axis direction is performed by moving the center table 17 with respect to the bet. Z axis movement is Z
The X-axis movement is performed by an X-axis motor 19 and an X-axis ball screw 22. In the processing of the lead surface of the VTR and the DAT, the outer diameter, the inner diameter, the end face, and the like of the work 13 mounted by the chuck 21 are processed by fixing several cutting tools 15 to the tool rest 14 corresponding to each processing. The processing of the lead surface is performed by synchronously controlling the displacement of the additional one-axis unit 16 and the feed of the Z-axis motor 18 with respect to the rotation angle of the spindle motor 12. Here, since the additional single-axis unit 16 places importance on responsiveness, a normal NC axis does not use a D / A converter for speed command transmission, and a positioning error that does not occur is a problem. In this lead surface processing, the position of the lead surface is also ± 2μ.
Dimensional accuracy within m is required. Therefore, if the positioning resolution is 0.1 μm, a positioning error of 2.9 μm will be generated from the above equation (6) ′, and the required accuracy will not be satisfied. Specifically, the positioning error can be improved to within ± 0.1 μm.

According to the present invention, according to the present invention, a D / A converter with a high number of bits is used to achieve both high positioning resolution and high-speed feeding, and the positioning error, which has been a conventional problem, to be within ± 1 unit of the positioning resolution. I can do it. That is, the left side of the equation (6) 'is defined as ER' [1 / positioning resolution], and Δd ·
If (2 ⁿ -1) is k, then ER '= k ＝ ΔT (11), and the positioning error ER ′ is proportional to the temperature change ΔT.
Assuming that k = 1 for the sake of simplicity, FIG.
The graph is as shown in FIG. However, the axis is stopped.
FIG. 4 (A) shows the temperature change, and FIG. 4 (B) shows (11).
It is a graph showing the positioning error amount (V positioning resolution) when k = 1 in the equation, and FIG. 4C is a graph showing the positioning error amount according to the present invention. In FIG. 4, performs offset adjustment of the D / A converter at time t _0, but the present invention conventional in this time t ₀ because it also amounts positioning error is zero. Then, when the temperature rises by 1 ° C. at time t ₁ after a lapse of 6 minutes from time t _0, the positioning error amount without correction is 1 [1 / positioning resolution]. When the positioning correction is performed, the effect of the correction appears only when the positioning error amount exceeds the resolution of the position detector. After that, the time t ₂ , t ₃ , and t _4, and the positioning error amount increases with the temperature change when there is no correction, but it can be seen that the correction always falls within ± 1 unit of the resolution when there is correction. However, in actual positioning, since the position loop is digitally controlled, there is at least a fluctuation of ± 1 unit in the resolution with respect to the command value, but here, it is considered that ideal positioning is possible.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention.
FIG. 3 is a flowchart showing an operation example thereof, FIG. 3 (A) is a plan view showing an example of a machine tool to which the present invention can be applied, FIG. 3 (B) is a side view thereof, and FIGS. 4 (A) to 4 (C). FIG. 5 is a diagram for explaining the effect of the present invention, and FIG. 5 is a block diagram showing a conventional control system. 1. Linear motion mechanism 2. Position detector 3.
Function generator, 4 ... Subtractor, 5 ... D / A converter, 6 ...
... Servo amplifier, 7... Servo motor, 8... Correction value calculation start signal generator, 9... Correction value calculator, 10... Correction value storage device, 11.

Claims (1)

(57) [Claims] 1. A numerically controlled machine tool that performs positioning control and includes a D / A converter in a transmission system of a speed command value, detects that a control target is not moving, and waits for a predetermined time to elapse. And a correction value calculation activation signal generating means for generating a timing of positioning error correction;
Calculating means for calculating a positioning error based on a start signal of the correction value calculation starting signal generating means; correction value storing means for storing the correction value calculated by the calculating means; A positioning error correction device for a numerically controlled machine tool, further comprising an adding means for adding to a command value, wherein a positioning error is automatically corrected at each timing.