JPH03270040A - Theta mechanism - Google Patents

Abstract

PURPOSE:To improve the accuracy of a mechanism, and to conduct positioning at a high degree by converting a linear motion into a rotary motion, estimating a parameter after manufacture from the input value of the linear motion in rotation and the output value of the rotary motion to the input value and correcting the motion precision of rotation. CONSTITUTION:A command value is input to a theta drive mechanism in response to the result of the detection of a camera for upper alignment, thus turning a motor 512, then rotating a ball screw 508. Consequently, a guide member 510 is moved along a guide rail 505, a roller 545 abutting against the guide member 510 is pushed, an arm 544 is shifted, and a chuck section 54 is revolved. The quantity of the ball screw 508 turned is detected by a rotary encoder 506, and a command stopping the revolution of the motor 512 is output when the quantity of rotation corresponding to said command value is detected. Accordingly, the theta drive mechanism moves the arm by the linear displacement of the ball screw, and revolution theta is acquired. As a result, the values of theta are measured at three points or more, and the precision of the mechanism is calibrated by using a nonlinear type least square method, thus improving the prevision of theta.

Description

[Detailed description of the invention] [Purpose of the invention] (Industrial application field) The present invention relates to a θ mechanism.

(Prior Art) Conventionally, a θ mechanism that converts linear motion into rotational motion has been used as a means for positioning.

However, in conventional θ mechanisms, the parameters of each part of the mechanism use dimensions based on design values.
There is a problem that an error occurs in the conversion from linear motion to rotational motion (X→θ conversion).

(Issue to be Solved by the Invention 'Jf3) As described above, the conventional θ mechanism has a problem in that an error occurs in the conversion from linear motion to rotational motion.

The present invention is intended to solve the above-mentioned conventional problems,
The object of the present invention is to provide a θ mechanism that can improve viscosity by transferring the manufacturing dimensions of the mechanism and using these dimensional values to convert linear motion to rotational motion.

[Structure of the Invention] (Means for Solving the Problems) The present invention provides a rotation means for converting a linear motion into a rotational motion to perform θ positioning, an input value of the linear motion in the rotation means, and a rotation with respect to the input value. The apparatus is provided with a correction means for estimating the parameters after manufacture from the measured value and the output value of the movement, and correcting the movement accuracy of the rotating means based on the estimated value.

(Function) In the present invention, while converting linear motion into rotational motion,
The parameters after manufacture are estimated from the measured values of the input value of the linear motion in this rotation and the output value of the rotational motion relative to the input value, and the precision of the rotational motion is corrected based on the estimated values.

Therefore, it is possible to improve the precision of the mechanism and achieve advanced positioning.

Ru.

(Example) Hereinafter, an example of the present invention will be described with reference to the drawings.

FIG. 1 is a diagram showing the configuration of an inspection apparatus to which a θ mechanism according to an embodiment of the present invention is applied.

As shown in the figure, the apparatus main body 1 includes an inspection section 10 that performs a predetermined inspection of a workpiece 3 in which a semiconductor module 2, which is an object to be inspected, can be transported and inspected in a non-conductive liquid; A plurality of inspection contact terminals corresponding to a plurality of semiconductor elements (hereinafter referred to as chips) mounted on the semiconductor device 2 are accommodated therein, and a contact terminal supply unit 20 is provided for exchanging and aligning these contact terminals. Further, a work loader part 30 for loading and unloading the work 3 is detachably installed at the end of the apparatus main body 1 on the inspection section 10 side. On the apparatus main body 1 and the work loader part 30, there are workpiece conveyors that are movable along the parallel direction (hereinafter referred to as the X direction) of the work loader part 30, the inspection part 10, and the contact terminal supply part 20, respectively. A mechanism 40 and a contact terminal moving mechanism 50 are mounted.

Note that, as shown in FIG. 2, the above-mentioned waste 3 is located at the bottom of a semiconductor module 2 configured by bonding, for example, 36 chips 2a of multiple types onto a multilayer ceramic substrate 2b having internal wiring. , a socket 4 and a socket board 5 are arranged in electrical contact with each input/output bin (not shown) of the semiconductor module 2.

Further, a support board 8 serving as a clamp portion in which a liquid-tight seal is formed by a holding plate 6 and an O-ring 7 is disposed on the outer peripheral side of the semiconductor module 2.

Also, although not shown, there is a 7TPI around the chip 2a.
A regular electrode pad is provided.

The inspection section 10 has a liquid tank 13, and the workpiece 3 is placed on a workpiece mounting table 12 by a clamp 14 and tightly fixed to form a liquid-tight seal. And liquid tank 1
The workpiece 3 is placed in a non-conductive inert liquid 15 injected into the workpiece 3.
is immersed, and the test is carried out in that state. Workpiece mounting table 1
2, a work set base unit 17 has a number of protruding pogo pins (not shown) that are inserted into the socket board 5 of the work 3 for electrical connection.
is provided.

Further, the contact terminal supply unit 20 individually supplies a plurality of bin blocks 21 in which probe bins are implanted according to the shape and pitch of the measurement electrode pads formed around the chip 2 mounted on the semiconductor module 2. Hold multiple e.g. 9
The camera 23 includes a bottle block changer 22, and a bottle block correction camera 23 that confirms the position of the bottle block 2 passed to the contact terminal moving mechanism 50 side.

The bin block changer 22 includes a bin block 21
and a cylinder mechanism 25, for example, which raises the holding part 24 toward the contact terminal moving mechanism 50 side, and individually raises the bottle blocks 21 to be used according to the inspection program. It is supplied to the contact terminal moving mechanism 50.

The work loader part 30 includes a loader part base 31, a work setting table 32 that accommodates the work 3 and is capable of moving between workpiece exchange positions, and a workpiece that transfers the work 3 to the work transfer mechanism 40. It is composed of a lifting mechanism 33.

The contact terminal moving mechanism 50 includes an X stage 51, a Y stage 52, and a Z stage 53, and includes a chuck part 54 fixed to the Z stage 53 and a measuring part 56 in which an upper alignment camera 55 is installed. but,
It is movable in the x-y-z directions. Furthermore, the chuck portion 54 is rotatable by a θ drive mechanism (not shown) which will be described later. Then, the bottle block 21 held by the chuck section 54 is lowered into the liquid tank 13 of the inspection section 10 by the Z stage 53, and comes into contact with the workpiece 3 immersed in the liquid tank 13.

The measuring section 56 is provided with a non-conductive inert liquid introduction pipe 57 that is close to the chuck section 54 and connected to a cooling mechanism (not shown). Further, a touch plate connected to the bottle block 21 and a tester (not shown) is arranged on the upper part of the chuck part 54, and electrical connection between the bottle block 21 and the tester is established by this touch plate.

The inspection procedure in the inspection apparatus having the above configuration will be explained below with reference to FIG.

First, the work 3 is placed on the work loader part 30, and the work 3 is moved to a delivery position above the work lifting mechanism 33. Next, while positioning the workpiece 3, it is raised to the position of the holding part 41 of the workpiece transport mechanism 40. After confirming that the workpiece 3 has risen, the pair of pawls 4 of the holding part 41
2 to hold the workpiece 3 (Fig. 3-a).

Next, the workpiece 3 is transported to the workpiece delivery position.

The contact terminal moving mechanism 50 then moves toward the contact terminal supply section.

Then, the workpiece 3 is positioned on the workpiece mounting table 12.
A workpiece 3 is placed on top and fixed liquid-tightly to the workpiece mounting table 12 by a clamp 14, and electrical connection between the socket board 5 of the workpiece 3 and the workpiece set base unit 17 on the inspection section 10 side is established. (Fig. 3-b)
).

Before and after setting workpiece 3, set bin block 2.
1 is installed. First, the contact terminal moving mechanism 50 is attached to the bin block changer 22 holding the bin block 21 corresponding to the chip 2a to be inspected first, automatically according to the inspection program.
The X stage 51 and the Y stage 52 are driven so that the chuck portion 54 is positioned. When the chuck part 54 reaches the delivery position of the bottle block 21, the cylinder mechanism 25 of the bottle block changer 22 rises, and the bin block 21 is held by the chuck part 54 (
Figure 3-b).

Thereafter, the bin block correction camera 23 captures an image of the holding state of the bin block 21, and after confirming the position with the upper alignment camera 55, the contact terminal moving machine tg50 moves above the inspection section 10.

Next, alignment between the workpiece 3 and the bin block 21 is performed. Then, alignment between the measurement electrode pad and the probe bin is performed by driving the X stage 51, the Y stage 52, and the θ drive mechanism described later.

Then, when the alignment is completed, a fluorine-based inert liquid is supplied into the liquid tank 13 from the non-conductive liquid introduction pipe 57,
The workpiece 3 is now immersed.

After this, lower the IN-] fixing section 6, and lower the bin block 2.
1 probe bottle is brought into contact with the measurement electrode pad. Then, a test voltage is supplied to the semiconductor module 2 to test the chip 2a (FIG. 3-C).

The above operations complete the inspection of one chip 2a. Next, if there is a chip 2a of the same standard in the semiconductor module 2, the next chip 2a is inspected in the same way after positional correction is performed as necessary.

After checking the chips of the same standard, if there are any untested chips 2a, replace the bin block 21 and test the chips 2a in the same way.

After the inspection of all the chips 2a is completed by repeating the above steps, the non-conductive inert liquid 15 is discharged and the workpiece 3 is transferred. Then, the workpiece transport mechanism 4
0 is moved to the work loader part 30, the work 3 is carried out, and a series of inspection steps is completed.

Next, details of the θ drive mechanism for rotating the chuck portion 54 in the contact terminal moving mechanism 50 described above will be described.

FIG. 4 is a plan view showing the θ drive mechanism, FIG. 5 is a side sectional view of FIG. 4, and FIG. 6 is a front view of FIG. 5.

In these figures, 501 indicates a base plate that rotatably supports the chuck portion 54 on which the arm 544 is formed. A support block 504 having support pieces 502 and 503 at predetermined intervals is disposed on the base plate 501. Support piece 502 on support block 504.
A guide rail 505 is formed between the guide rails 503 . A rotary encoder 506 is disposed on the support piece 502, a bearing 507 is disposed on the support piece 503, and a ball screw 508 is rotatably disposed between them. A nut 509 is screwed onto the ball screw 508, and a guide member 510 that is guided by the guide rail 505 is fixed to the nut 509. Also,
A force/knob ring 511 is disposed on the support piece 503, and one end of a ball screw 508 is connected to the force/knob ring 511, and a rotating shaft 513 of a motor 512 provided on the support piece 503 is connected to the force/knob ring 511. has been done.

Further, a roller 545 is rotatably disposed on the above-mentioned arm 544, and this roller 545 is in contact with the guide member 510. Further, the arm 544 and the guide member 510 are connected by coil springs 546 and 546.

Therefore, a command value (input value) is input to the θ drive mechanism according to the detection result of the upper alignment camera 55,
This causes the motor 512 to rotate and the ball screw 50 to rotate.
8 is rotated. As a result, the guide member 510 is moved along the guide rail 505, the roller 545 in contact with the guide member 510 is pushed, and the arm 544 is moved, thereby rotating the chuck portion 54. The amount of rotation of the ball screw 508 is detected by the rotary encoder 506, and when the amount of rotation corresponding to the above-mentioned command value is detected, a command to stop the rotation of the motor 512 is output.

Next, the calibration of the command value-input value (linear displacement X) and output value (rotational displacement θ) in the θ drive mechanism described above will be explained.

<1> Specification stroke of rotational displacement θ...±4 deg (±144
00 seconds) Positioning accuracy...±6 x 10-O-3de
(±21.6 seconds) (2) Relationship between X and θ The θ drive mechanism moves the arm by linear displacement X of the ball screw to obtain rotation θ. A schematic diagram thereof is shown in FIG.

In the figure, the relationship between X and θ is expressed by the following equation.

A sin θo +r S s n 2θ0...
(2-7) As a result, the relational expression between X and θ was determined.

Next, the conversion formula for X-θ is (2-7 Consider the inverse transformation formula for X.

sin (θ+θo)-d cos (θ+θ)-Jl-d” tan (θ10)-d/J1-62,°, θ-1
an -' (d/J1-d”)-〇... (2-
9) This equation allows the simulation of the θ mechanism and the calculation of the angular q power.

Note that the theoretical value of resolution is shown in FIG.

(3) Calibration method Relational expression between X and θ % formula %) (31) ) ) To make θ rotate the same, calculate equation (2-7) with the control software, and use the X as a command to the ball screw. Value (input value).

Then, consider a case where the dimensions after assembly are (10 Δ°C, r+Δ', θo 1 Δθ) for the design value of the parameter (, r5θ.

The required rotation angle is θ-f (XS, 9, rl θ0) However, the actual rotation angle is θ--f (X,, 9+Δ℃, 10Δr1θ 10Δθ0).In other words, θ'- An error occurs in θ.This error may result in a value that does not satisfy the specifications.

Therefore, by determining the value of Δr1Δθ0 at 6℃, the error θ′-〇 can be eliminated by using can.

This operation is called calibration.

Therefore, we measured the value of θ at three or more points and used the 11 linear least squares method to calculate the parameter error Δ(, Δr1Δθ
We measured 0 and tried to improve the accuracy of θ by calibration.

Let J2S rsθ0 be the design value of the machine parameter. Let this be the most probable approximate value.

The dimensions after assembly, that is, the most probable value, are (10ΔA, r
+Δr1θ +Δθ0. From n measured values of θ, Δ
(, Δr, Δθ. are estimated.

First, in the mechanism, θ, -f, i, r1θo) (i-1~n)1
Give n command values that are 1. In contrast, the actual rotation angle is measured.

θ　, -f, i+Δ℃, r+Δr1 1 1 θ0+Δθ0) (i-1~n) n measured values are obtained.

Here, the following 1 exists between the command value θ and the measured value θ'.
The relationship 1 holds true.

(Margin below) Δ　θ　, −〇　−, −〇　.

l l 1=f1 (
[+6℃, r+Δr1θ0+Δθ)−f, i, rl
θ0) l*C9f,/9℃)Δ℃+ (i9f−/9r) Δ“+C9f,/9θ.)Δθ.

(i -1~n) ... (3-3) (6°C, Δr1Δθ0 is a minute amount, so perform Teller expansion and omit terms of second order or higher.) 6°C, Δr1Δθ. We were able to obtain n linear equations with unknowns. If we put it in the form of a matrix, we get (below the margin) (below the margin) (3-4), and let this be b-to・X.

next, 9f, /lJ2.9fi/9r, 9f-/9θ.

Find a specific formula for l. Relational expression (2-7) between X and θ
To reproduce the formula, i + r sin 2 θ - Il sin θQ O Since θ cannot be differentiated, it is expressed in the form of an implicit function and is denoted by g((, r, θ., θ,).

g. −(Qg i/[1)ΔA+(9fi/9r)Δr0(
9gi/Qθ0)Δθo + (9g i”θ1)Δθ
i −〇 Transformed to Δθi −(-1)/C9g/9θ, )([Qgi/[1)Δ
i+ (c)gi/lr)Δr+(9gt/c)θ0)
Δθo)) ... (3-6 Reproducing equation (3-3), Δ θ i -C9f, /912) ΔJl+(fQf i/lr)
Δr+<9f,/9θ0)Δθ.

...(3-7) From (3-6)-(3-7), (,9f,/912>-(-1)/<9g-/Qθi) (9g,/cl) (fQf,,' 9r) −(−1)/(Qg−/lθ,)1
1 (Qg, /9 r) <9f, /, Jθ. ) −(-1)/ (lag, /9θ)1 (Qg, /lθ.) ...... (3-8) The right side of (3-8) is the partial differentiation of (3-5). It can be easily determined by

(9g, /f9θ,) 1 1 =-J2cos(θ, +θo) +2 rs+n (
θi ten θ) cos (θ, +00) low 11) g, / [1) uniform sin (θ, ten θo) + sin θO (Qg,
/9r) ・2・2”Sln (θ・tenθo) −sin θO(9
g, /f9θ. ) 1-,9cos(θ, 10θo)+Ccosθ0+2rs
ln(θ, 10θ) cos(θ, 10θ0)1 0
l -2r sin θoCO8θ0 Here, the matrix of equation (3-4) b-A-% has been found, so 1. T −<6°C, Δr, Δθ0) are determined.

b (lxn)-A (3Xn) φX (I
X3) The number of equations (n) is greater than the number of unknowns (3). Find the least squares solution that minimizes the equation error.

Find X that minimizes ε-Ax-b εTE, and O(ε”E)
Find X to make /9x-0.

X-(A"A)-'A" b...(3-9
) (3-9), Δ(, Δr1Δθ.

can be found.

As a result, the machine/parameter is corrected in equation (2-7).

Bu-(+6℃ r → r + Δ r θ →θ0+Δθ0 Then, the command value X to the ball screw is X-((+Δf
f1) sin (θ100+Δθ0) (r1Δr) sin
2 (θ+θ0+Δθ0)((+Δ()sln(θ0+
Δθ0) 2 + (r+Δr) sin (θ0+Δθ0).

FIG. 9 shows the positioning accuracy when the machine parameters are corrected using the above-described calibration method, and FIG. 10 shows the positioning accuracy when the machine parameters are set to design values as a comparative example.

As shown in FIG. 9, the accuracy of the θ drive mechanism satisfies the specification of 21.6 seconds over the entire range. Furthermore, as shown in FIG. 10, in the comparative example without calibration, the angles of 3.5° and 4° deviate from the specification of 21.6 seconds.

Therefore, in this embodiment, the positioning accuracy can be improved by calibration.

[Effects of the Invention] As explained in detail above, the present invention provides a θ mechanism that can improve accuracy by transferring the fabrication dimensions of the mechanism and using these dimension values to convert linear motion to rotational motion. can be provided.

[BRIEF DESCRIPTION OF THE DRAWINGS] FIG. 1 is a partially sectional perspective view showing a semiconductor inspection apparatus in an embodiment of the method of the present invention, FIG. 2 is a side sectional view showing the workpiece in FIG. 1, and FIG. , ) to (C) are diagrams for explaining the inspection process of the device in FIG. 1, FIG. 4 is a plan view for explaining the θ drive mechanism in the device in FIG. 1, and FIG.
The figure is a side sectional view of Fig. 4, Fig. 6 is a front view of Fig. 5, Fig. 7 is a schematic diagram of the θ drive mechanism of Fig. 4, and Fig. 8 shows the resolution of the θ drive mechanism of Fig. 7. Figure 9 shows the theoretical values, Figure 9 shows the positioning viscosity when the machine parameters are corrected using calibratability, Figure 10 shows the positioning accuracy when the machine parameters are set to the design values as a comparative example of Figure 9. FIG. 2... Semiconductor module, 2a... Chip, 2b.
...Ceramics substrate, 50...Contact terminal moving mechanism,
54... Chuck part, 504... Support block, 5
05... Guide rail, 506... Rotary encoder, 508... Ball screw, 509... Nut,
510...Guide member, 512...Motor, 544
...Arm, 545...Roller, 546...Coil spring. Tonoki 2': ゛Mini 3

Claims (1)

[Claims] A rotating means for converting linear motion into rotary motion, and estimating the error between the input value of the linear motion in this rotating means, the measured value of the output value of the rotary motion with respect to the input value, and the design value of each of the motion systems. and a correction means for correcting the motion accuracy of the rotation means based on the value.