JPS5913681B2 - In-process measurement method for rotary axis motion - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a method for in-process measurement of the motion of a rotating shaft.

The technology to precisely track the motion of a rotating axis is fundamental in mechanical engineering, and a variety of methods have been developed that correspond to the size of the object, rotation speed, surrounding environment, etc. . In machine tools as well, as the demand for high-precision machining has recently increased, there has been a need for in-process measurement of spindle rotational motion for both research and practical purposes. However, at present, only a few methods have been proposed to measure it in a laboratory setting, and there are no methods that can measure the rotational accuracy of a shaft, including translational movement, during actual machining, that is, in-process. Not found. The present invention primarily aims to provide a method for easily in-process measuring the rotational motion of a machine tool spindle using an optical method.

Generally, the rotation center line of a rotating body (the line segment that is the rotation center of the rotating body at a certain moment) is constantly changing.
To understand this accurately, it is necessary to consider the fluctuation as a relatively long-period fluctuation with respect to the reference line ιo, that is, the average center line j! m (a line segment representing the average position of the rotational centerline observed at a certain time interval) and short-period fluctuations (i.e. rotations relative to the average centerline tm). There must be.

The displacement of the average center line tm is, as shown in FIG.
It is decomposed into D2 and Dθ (vector).

In the same figure, 3 is a line that passes through one end A of the average center line tm and is parallel to the z-axis, DrA (or HrB) is in a plane perpendicular to the z-axis, and 6z is on the Z-axis. . Also,
Each symbol for one end A of the average center line Tm has a subscript A.
, and the other end B is given a subscript B. Furthermore, as shown in Fig. 2, the short-period error motion of the rotation center line t is also caused by three components ErA (or FrB), Fz, and fθ (vector) in the radial direction, axial direction, and inclination with respect to the average center line Tm. It is expressed as follows.

In the same figure, T2 is a line passing through one end of the rotation center line t and parallel to the average center line Tm, and is 100 RA (or NrB).
lies in a plane perpendicular to the Z axis. In order to obtain the displacement of the average center line Tm with respect to the reference line mentioned above and the error movement of the rotation center line t with respect to the average center line, in the present invention, the rotation center line - with respect to the reference line TO is determined by the method detailed below. } Measure the radial variation Vr and the axial variation → Vz at both ends of t.

Note that since the measurement method for both ends of the rotational center line t is exactly the same, hereinafter, when there is no need to distinguish between them, they will be simply referred to as Vr. Therefore, the measured V
Each component described above can be obtained from r and Vz as follows.

-C -1+ ★Here, Vi and Vj indicate the averages of Vr and Vz, respectively, and L is the distance between A and B.

FIG. 3 shows the principle for optically detecting the radial variation Vr(t) of the rotation center line t with respect to the reference line TO, where P is a quadrant photocell (P1 to P4
), S is a light spot such as a laser beam, and the photocell P is attached to the rotation axis with its center 0' aligned with the center line of the rotation axis. Parallel light whose optical axis is aligned with the reference line so as to be located on the reference line is projected onto the photocell P from a stationary system. By rotating the rotating shaft in this state, an output V1 is generated in proportion to the difference in the amount of light received by the four-split photocells P1 and P2,
And when we take out the output V2 that is generated in proportion to the difference in the amount of light received by the 4-split photocells P3 and P4, the center 0' of the photocell P is V from the center 0 (reference line) of the spot S, as shown in the figure.
r(t)=Vr (The above-mentioned V, , V2 in the case where there is a deviation by t degree is expressed by the following equation.

What we need to find now is Vr(t), so (1)
ω●t must be eliminated from equation (2).

Therefore, a rotation angle detector such as a rotary encoder may be attached to the rotating shaft, and the following calculations may be performed.

V1・3+V2・V4=r(t)・μsα(t)−Δx
(t) ... (5) V2・V3-V4・-Vr(t)
- Shlα(t)−Δy(t) (6) Since ΔX and Δy are determined by this calculation, Vr(t) is determined as follows.

Alternatively, if the variation in the radial direction of the rotation center line t with respect to the reference line TO is measured in this way at at least two locations on the rotation axis, such as both ends of the rotation axis, the inclination of the rotation center line can be determined based on the measurement. The fluctuation of can be easily determined.

FIG. 4 shows the configuration of an interferometric length measuring instrument for measuring the axial variation Vz(t) of the rotating shaft by light wave interference, in which a corner cube prism 2 is installed at the end of the rotating shaft 1. Attach the laser beam from laser light source 3 to interferometer 4
The beam splitter 5 splits the beam into two, and one of the beams is inverted at the corner cube prism 2, and then interfered with the other beam at the interferometer 4, and the output from the detector 6 is the axial fluctuation Vz(t).
It is configured to obtain .

FIG. 5 shows the fluctuations VrA, VrB and V of the center line of rotation.
It shows an optical system that can measure z simultaneously.

In this optical system, the laser beam from the laser light source 10
The beam is first passed through the beam splitter 1 in the interferometer 11.
2 splits the beam into two for light wave interference, and sends one beam to the rotation axis 1.
A beam splitter 14 attached to one end of the photocell 3 further divides the beam into two, and one of the two is projected onto the light receiving surface PA of a photocell 15 at the end of the rotating shaft, and VrA is obtained based on the output of this photocell 15. The other beam split into two by the beam splitter 14 is split into two by a corner cube prism 16.
The beam is projected onto the light receiving surface PB of the photocell 17, which is then split into two by the beam splitter 14 and projected onto the light receiving surface PB of the photocell 17.
Based on the output of this photocell 17, Vr is obtained.

The other beam split into two by the beam splitter 14 causes light wave interference in the interferometer 11, and Vz is obtained as the output of the detector 18.

The arrangement of the photocells 15 and 17 in the optical system described above is such that they are separated by the distance between them and the corner cube prism (
This corresponds to arranging them on the center line at an interval twice as large as L/2). Although not shown, it goes without saying that a rotation angle detector may be attached to an appropriate position on the rotation shaft.

FIG. 6 shows how the electrical outputs obtained from the photocells 15, 17 and the sensor 11 shown in FIG. The processing circuit is shown. As is clear from the above detailed description, according to the present invention, various fluctuation components during the rotational movement of the shaft can be easily measured in-process using an optical method.

Although the above description assumes that a laser beam is used, it is not necessarily necessary to use a laser beam in this measurement method, and a well-collimated parallel beam is sufficient for measuring Vr. For measuring both Vr and Vz, it is sufficient to use light that is both collimated and coherent.

[Brief explanation of the drawing]

Figures 1 and 2 are explanatory diagrams of the average center line and rotation center line, Figure 3 is a principle diagram for explaining the principle of detecting radial fluctuations in the rotation center line, and Figure 4 is the rotation center. Figure 5 is a configuration diagram of an interferometric length measuring device that measures variations in the axial direction of a line. FIG. 3 is a block diagram of a processing circuit that performs arithmetic processing on the electrical output obtained from the device shown in the figure.

Claims (1)

[Scope of Claims] 1. For a 4-split photocell installed at both ends of a rotating shaft or at a position corresponding thereto so that the center coincides with the center line of the rotating shaft, the optical axis is arranged so as to match the reference line. Based on the output of the 4-split photocell obtained by this and the output of the rotation angle detector attached to the rotation axis, the fluctuation of the rotation center line on the rotation axis with respect to the reference line is determined as a standard. 1. An in-process measuring method for rotational axis motion, comprising optically measuring radial and inclination displacements of an average center line with respect to a line, and radial and inclination error movements of a rotation center line with respect to the average center line. 2 Coherent parallel light beam arranged so that the optical axis coincides with the reference line for a 4-split photocell installed at both ends of the rotation axis or at positions corresponding to the axis so that the center coincides with the center line of the rotation axis. An interferometric length measuring device that measures the output of a four-part photocell obtained by projecting a 4-split photocell, the output of a rotation angle detector attached to its rotation axis, and the axial fluctuation of the rotation axis by light wave interference. Based on the output of An in-process measurement method for rotational axis motion, characterized by optically measuring error motion in direction and inclination.