JP2000340620A - Probe device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a probe device, with which the electric characteristics, circuit function, etc., of a semiconductor chip mounted on a wafer can be tested surely by uniformly maintaining the contacting state between pads formed on the semiconductor chip and probe pins, even when the number or arranging density of the probe pins is increased or the lengths of the pins are reduced. SOLUTION: A probe device 10 is used for testing the electrical characteristics of a semiconductor chip 21 mounted on a wafer 20, by bringing a plurality of probe pins 32 into contact with a plurality of pads formed on the chip 21. The probe device 10 is provided with placing means 11 and 12 on which the wafer 20 is placed, a support means 13 which supports the placing means 11 and 12 in inclinable states, and inclination detecting means 17, 18, 27, and 28 which detect the inclinations of the placing means 11 and 12. The device 10 is also provided with an inclination control means 60, which maintains the placing means 11 and 12 in prescribed inclined states by driving the support means 13, based on the detected results of the detecting means 17, 18, 27, and 28.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a probe device for testing the electrical characteristics and circuit functions of a semiconductor chip in a wafer state.

[0002]

2. Description of the Related Art As is well known, a large number of semiconductor chips are formed on a wafer through various kinds of wafer processing (forming elements such as transistors and wiring between elements). And
These many semiconductor chips are subjected to tests such as electrical characteristics and circuit functions in a wafer state before being divided by a dicing operation. Incidentally, a semiconductor chip having a good test result is divided and then selected as a non-defective product and sent to an assembling process.

As shown in FIG. 16, a probe device 70 for testing a semiconductor chip includes a wafer holder 73 for holding a wafer 72 by suction, a sample table 74 on which the wafer holder 73 is mounted, and a sample table 74. Stage 7 for moving 74 in XY and Z directions
5, a probe card 76 mounted above the wafer holder 73, and a tester 77.

The probe card 76 includes a card board 81 on which a plurality of wiring patterns (not shown) are formed, and a plurality of probe pins 82 connected to one end of the wiring pattern and fixed to the card board 81. 82,... Note that a tester 77 is electrically connected to the other end of the wiring pattern (not shown) formed on the card substrate 81.

Therefore, a plurality of probe pins 82, 8
Are electrically connected to the tester 77 via a wiring pattern (not shown). FIG. 16 shows only two probe pins 82, 82, but the actual number is several hundred to several thousand. The arrangement of the tips of the plurality of probe pins 82, 82,.
A plurality of pads 8 formed on each semiconductor chip 71 above
3, 83,... Are arranged.

Here, each probe pin 82 is formed by bending a thin metal wire into a "C" shape, and its total length is 2 m.
m to 3 mm. In order to perform a test on the semiconductor chip 71 using such a probe device 70, first, the sample stage 74 is moved in the XY directions by the stage 75,
A plurality of pads 83, 83,... Formed on the semiconductor chip 71 to be tested are replaced with a plurality of probe pins 82, 82,.
To the tip of

[0007] Next, the sample stage 74 is raised by a predetermined amount in the Z direction by the stage 75, and as shown in FIG. 17 (a), is placed on a semiconductor chip 71 to be tested (hereinafter referred to as a "test chip 71a"). A plurality of pads 83, 8 formed
Are brought into contact with the tips of the plurality of probe pins 82, 82,. Then, in the probe device 70, the pad 83 and the tip of the probe pin 82 come into contact with each other (the state shown in FIG.
7 (b)), the sample stage 74 is pushed up.

As a result, each probe pin 82 is largely bent by the elastic deformation, and its tip is pressed against the pad 83 (the state shown in FIG. 17B). in this way,
By increasing the contact pressure between the pad 83 and the tip of the probe pin 82, the contact resistance is reduced, and the circuit in the test chip 71a and the tester 77 (FIG. 16) are reliably electrically connected via the pad 83 and the probe pin 82. Can be connected.

Therefore, while maintaining the state in which the tip of the probe pin 82 is in strong contact with the pad 83 (FIG. 17 (b)), a predetermined electric signal is given from the tester 77 to the circuit in the test chip 71a, By examining the response from 71a, tests such as electrical characteristics and circuit functions can be accurately performed. By the way, the probe device 70
, When the tip of the probe pin 82 is pressed against the pad 83 (FIG. 17B), the load W applied by this contact is about several g to 10 g per pin, and the total of the probe pins 82, 82,. It is about several kg to 50 kg.

The size of the entire area where the tips of the probe pins 82, 82,... And the pads 83, 83,.
a), which is very small compared to the size of the wafer 72 and the sample stage 74. For this reason, when the tip of the probe pin 82 is pressed against the pad 83 (FIG. 17B), that is, during the process of pushing up the sample stage 74 or during the test on the test chip 71a, the wafer 72 and the sample stage 74 have a small size. This means that a large load W due to contact is applied to a part of the region.

Therefore, when the test chip 71a is located at the periphery of the wafer 72, the wafer 72 and the sample stage 74 are inclined at a maximum of about 15 μm as shown in FIG. As a result, the test chip 71a also tilts. However, in this probe device 70, the amount of pushing up of the sample stage 74 is set to about 100 μm, and the probe pins 82 are largely bent.
As shown in FIG. 8B, the inclination of the test chip 71a can be absorbed by the deflection of the probe pin 82. As a result, the contact state between the tips of all the probe pins 82, 82,... And the pads 83, 83,.

Further, in the probe device 70, the temperature of the wafer 72 is set to -50.degree. C. to 150.degree.
The test is performed with different values set in the range. During the high-temperature test, if the wafer 72 is heated to, for example, 150 ° C., the wafer 72 itself and the wafer holder 73 may be distorted due to thermal deformation. At this time, as shown in FIG. 19, the surface 72a of the wafer 72 undulates. As a result, each semiconductor chip 71 on the wafer 72 has a different inclination.

However, in this probe device 70, the amount of pushing up of the sample table 74 is set to about 100 μm, and the probe pins 82 are largely bent. Therefore, as in the case of FIG. The deflection of the test chip 71a can be absorbed by the deflection. And
The tips of all the probe pins 82, 82,.
The contact state with 3, 83,... Can be favorably maintained.

Further, when the surface 72a of the wafer 72 is undulating (FIG. 19), the heights of the semiconductor chips 71 vary. Height difference is up to 2
It is about 0 μm. However, in the probe device 70, the amount of pushing up of the sample stage 74 is set to about 100 μm, and the probe pins 82 are largely bent, so that FIG.
The test chip 71a having the height H1 shown in FIG.
Similarly, the contact state between the tips of all the probe pins 82, 82,... and the pads 83, 83,... can be favorably maintained in the test chip 71a having the height H2 (> H1) shown in FIG.

Therefore, according to the probe device 70,
By using the deflection of the probe pins 82, a test can be performed on all the semiconductor chips 71 on the wafer 72 with high accuracy.

[0016]

However, in the conventional probe device 70, if the number of probe pins 82, 82,... Is increased (multiple pins) or the mounting density is increased (density), , Probe pins 82,
, And the inclination of the sample table 74 and the wafer 72 generated by the contact between the tips of the pads 83, 83,.
The deflection of the probe pins 82 makes it impossible to absorb the inclination of the test chip 71a.

The inclination of the test chip 71a is set to the probe pin 8
If it cannot be absorbed by the deflection of the probe pin 8
Cannot maintain good contact with the tips of pads 83, 83,. At this time, the test cannot be performed on the test chip 71a. Also,
In order to increase the density of the probe pins, the size of the pad formed on the semiconductor chip 71 must be reduced. Therefore, it is required to further improve the positioning accuracy between the probe pin and the pad.

Further, in the conventional probe device 70, since the long probe pins 82 having a total length of 2 mm to 3 mm are used, the high frequency region (about 200 mm) of the semiconductor chip 71 is used.
MHz or higher). In recent years, there has been an increasing demand for testing the semiconductor chip 71 in a high frequency range.

For this purpose, the total length of the probe pins must be 10
It must be shortened to about 0 μm (shortened pin). Further, when the probe pins are shortened, the outer shape becomes vertical, so that the conventional "U" -shaped probe pins 8 are used.
2, the allowable amount for the deflection of the probe pin becomes extremely small.

Further, the push-up amount after the contact between the tip of the probe pin and the pad must be set to a small value of about 10 μm. Here, suppose a probe pin 86 (see FIG. 21) having a total length of about 100 μm is used, and
It is assumed that a test of the semiconductor chip 71 is performed with the setting of about μm. As described above, during the process of pushing up the sample stage 74 and the test on the test chip 71a, a large load W due to contact is applied to the wafer 72 and the sample stage 74 in a small partial area. Tilt may occur.

When the test chip 71a is inclined, for example, as shown in FIG. 18, the contact between the probe pins 86, 86,... And the pads 83, 83,. Would. This is the probe pin 86
This is because the inclination of the test chip 71a cannot be absorbed. At this time, the test on the test chip 71a cannot be performed.

Further, as described above, the wafer 72 itself and the wafer holder 73 are distorted by thermal deformation,
When undulation occurs on the surface 72a of No. 2 (see FIG. 19)
But the same is true. That is, the surface 72a of the wafer 72
If the inclination of each semiconductor chip 71 is different due to the undulation, the probe pins 86, 86,.
The contact state with the test chips 71a becomes uneven,
Cannot be tested.

If the height of each semiconductor chip 71 varies due to the undulation of the surface 72a of the wafer 72, the test chip 7 having a height H2 shown in FIG.
1a, even if the contact state between the probe pin 86 and the pad 83 can be maintained well, the height H1 shown in FIG.
In the test chip 71a (<H2), all the probe pins 8
May not be able to contact pads 6, 86,... And pads 83, 83,. The heights H1 and H2 are based on the upper surface of the sample table 74.

As described above, if the pins are shortened in the probe device 70, all the semiconductor chips 71 on the wafer 72 cannot be tested satisfactorily. An object of the present invention is to maintain a uniform contact state between a pad formed on a semiconductor chip on a wafer and a probe pin even when the number of pins of the probe pin is increased, the density of the probe pin is shortened, or the like. It is an object of the present invention to provide a probe device capable of reliably performing a test such as an electrical characteristic and a circuit function of the probe device.

[0025]

According to a first aspect of the present invention, a plurality of probe pins are brought into contact with a plurality of pads formed on a semiconductor chip on a wafer to test the electrical characteristics of the semiconductor chip. A mounting device on which a wafer is mounted, a supporting device for tiltably supporting the mounting device, a tilt detecting device for detecting a tilt of the mounting device, and a detection result by the tilt detecting device. And a tilt control means for controlling the support means to drive the support means on the basis of the control means to maintain the inclination of the mounting means at a predetermined inclination.

The invention described in claim 2 is the first invention.
Wherein the inclination control means performs control to make the inclination of the surface formed by the plurality of pads of the wafer mounted on the mounting means coincide with the inclination of the surface formed by the tips of the plurality of probe pins. It is constituted in. The invention according to claim 3 is the invention according to claim 1 or claim 2.
The tilt detecting means described in (1) has a reflecting mirror surface provided on the mounting means, and a collimator for measuring a tilt angle of the reflecting mirror surface.

The invention described in claim 4 is the first invention.
Alternatively, the tilt detecting means according to claim 2 has a reflecting mirror surface provided on the mounting means and an interferometer for measuring a tilt angle of the reflecting mirror surface. According to a fifth aspect of the present invention, in the probe device according to any one of the first to fourth aspects, the moving means for moving the mounting means in the direction of contact and separation between the pad and the probe pin; Load detecting means for detecting a load applied by the contact of the probe pin with the load, calculating a shift amount between the load detected by the load detecting means and a predetermined load, and driving the moving means based on the shift amount, Load correction means for correcting the load applied by the contact.

The invention described in claim 6 is the same as the invention in claim 5
The load detecting means described in (1) is configured to detect a load at at least three places of the placing means. According to a seventh aspect of the present invention, there is provided the load correcting means according to the sixth aspect of the present invention, wherein the load correcting means detects the load by the load detecting means in accordance with a relative position between the position at which the load is detected and the position of the semiconductor chip to be tested. It is configured to control the driving amount.

[0029] The invention described in claim 8 is the same as in claim 5.
8. The probe device according to claim 7, wherein the load detecting unit has a plurality of strain gauges. 9. According to a ninth aspect of the present invention, there is provided a probe device for testing electrical characteristics of a semiconductor chip by contacting a plurality of probe pins with a plurality of pads formed on a semiconductor chip on a wafer, The mounting means on which the wafer is mounted, the in-plane moving means for moving the mounting means in a plane perpendicular to the direction of contact and separation between the pad and the probe pin, and the displacement amount of the semiconductor chip according to the elongation of the wafer. And a positioning control means for positioning the semiconductor chip at a point facing the plurality of probe pins by driving the in-plane moving means in consideration of the calculated and the amount of displacement.

[0030]

Embodiments of the present invention will be described below in detail with reference to the drawings. This embodiment is defined by claims 1 to
This corresponds to claim 9. Probe device 10 of the present embodiment
FIG. 1 shows an apparatus for testing electrical characteristics and circuit functions of a semiconductor chip in a wafer state.

Before describing the probe device 10, a semiconductor chip to be tested will be described. FIG.
As shown in FIG.
A plurality are arranged on 0. Each semiconductor chip 21 has a predetermined circuit (not shown) formed thereon.

Further, each semiconductor chip 21 has a structure shown in FIG.
Are formed, a plurality of pads 22, 22,... Are formed. The pads 22 are external terminals connected to circuits in the semiconductor chip 21.
In such a test on the semiconductor chip 21, a predetermined current is applied to the circuit through the plurality of pads 22, 22,.
This is done by detecting the response.

FIG. 2B shows 16 pads 22,
22,... Are shown, but the actual number is several tens to several thousand. Now, the probe device 1 of the present embodiment
As shown in FIG. 1, the wafer holder 11 is provided with a wafer holder 11 for holding the wafer 20 on which the semiconductor chips 21 are formed by vacuum suction. This wafer holder 11
In the test, the temperature of the wafer 20 is set to a high temperature (for example, 10
0 ° C.) is built in.

The probe device 10 is provided with a card holder 31 above the wafer holder 11. The probe card 30 is mounted on the card holder 31. The probe card 30 has a plurality of probe pins 32, 32,... Attached to a card board 33, as shown in FIG. Each of the probe pins 32 is formed to have a short length of about 100 μm and a substantially vertical shape so that a test in a high frequency range of the semiconductor chip 21 can be performed.

Further, a plurality of probe pins 32, 32,.
As shown in FIG. 3B, a plurality of pads 22, 2 formed on the semiconductor chip 21 (FIG. 2B) are arranged as shown in FIG.
It corresponds to the arrangement of 2,. That is, when testing the semiconductor chip 21, the probe pins 32 and the pads 22 can be brought into one-to-one contact. Further, a tester 34 is provided in the probe device 10 (FIG. 1). The tester 34 is electrically connected to the plurality of probe pins 32, 32,... When the probe card 30 is mounted on the card holder 31.

The tester 34 is provided with a plurality of probe pins 32, 3 according to a test program stored in advance.
A predetermined electric signal is output to 2,.
Further, the tester 34 exchanges an electric signal with a circuit in the semiconductor chip 21 in a state where the probe pin 32 is in contact with the pad 22 of the semiconductor chip 21 (details will be described later), Is determined.

The probe device 10 further includes a sample table 12 on which the wafer holder 11 is mounted,
The sample table 12 is moved in the Z direction and the tilt direction (θx, θy).
Leveling stage 13 for positioning the sample stage 1
An X stage 14 for positioning the sample stage 2 in the X direction;
A Y stage 15 for positioning 2 in the Y direction is provided on a stage base 16.

A .theta.z stage (not shown) for positioning the wafer holder 11 in the rotation direction is provided between the wafer holder 11 and the sample table 12. The θx direction, θy direction, and θz direction correspond to rotation directions about the X axis, Y axis, and Z axis, respectively. The above-described wafer holder 11 and sample table 12 correspond to the “mounting means” in the claims.

Here, the Y stage 15 is driven by a linear motor (not shown) and is movable in the Y-axis direction. The position of the Y stage 15, that is, the Y position of the sample stage 12, is detected using an encoder (not shown). The X stage 14 is driven by a linear motor (not shown),
It is movable in the X-axis direction. The position of the X stage 14, that is, the X position of the sample stage 12 is detected using an encoder (not shown).

Further, the Z leveling stage 13 is arranged as shown in FIG.
As shown in (a), three Z drive units 41, 42, 4
3 These Z drive units 41, 42, 43 are respectively provided at portions 41 a, 42 a,
At 43a, the sample stage 12 is supported from below. As shown in FIG. 5A, the Z drive unit 41
4 and a lower inclined member 47 mounted on a slope 46a of the lower inclined member 46.
A servo motor 48 for sliding the upper inclined member 47 along the inclined surface 46a;
and a roller 49 that rotates in accordance with the sliding on a.

The rotation shaft 49a of the roller 49 is shown in FIG.
As shown in (b), it is attached to a portion 41 a of the lower surface 12 a of the sample table 12. In this Z drive unit 41,
By sliding the upper inclined member 47 along the inclined surface 46a, the upper inclined member 4 is moved as shown in FIGS.
7, the height H1 of the upper surface 47a can be changed by ΔH1. Then, as a result of the change of the height H1 of the upper surface 47a, the Z position (Z1) in the portion 41a of the sample table 12
Also changes by ΔH1.

That is, according to the Z drive unit 41, the Z position (Z1) (see also FIG. 4B) for supporting the sample table 12 at the portion 41a of the lower surface 12a of the sample table 12 is set freely. be able to. Further, the other Z drive units 42 and 43
(FIG. 4A) has the same configuration as the Z drive unit 41 described above.

Therefore, according to the Z drive unit 42, the Z position (Z2) (see FIG. 4B) for supporting the sample stage 12 at the portion 42a of the sample stage 12 can be set freely. Further, according to the Z drive unit 43 (FIG. 4A), the Z position (Z3) (see FIG. 4B) when the sample stage 12 is supported at the portion 43a of the sample stage 12 is freely set. be able to.

The Z at the portion 41a of the sample table 12
Position (Z1), Z position (Z2) at site 42a, site 4
Each Z position (Z3) in 3a is detected using an encoder (not shown). In the Z leveling stage 13 configured as described above, the Z driving units 41 to 43 are individually driven, and the Z positions (Z positions) in the parts 41a, 42a, 43a are set.
By making at least one of (1, Z2, Z3) different, a tilt in the θy direction or the θx direction can be generated on the sample stage 12 or the tilt can be freely changed.

In the Z leveling stage 13, the Z level
By driving the driving units 41 to 43 in synchronization with each other, the sample stage 12 can be moved in the Z direction while keeping the inclination of the sample stage 12 in the θy direction and the θx direction constant. Further, in the probe device 10 (FIG. 1) of the present embodiment, in order to detect the inclination of the sample stage 12 in the θy direction, an X-axis movable mirror 17 is attached to the side surface of the sample stage 12, and the X-axis movable mirror 17 An X-axis collimator 27 is provided to face the same.

As shown in FIG. 6A, the X-axis moving mirror 1
7 is orthogonal to the upper surface 12b of the sample stage 12. Therefore, when the upper surface 12b of the sample stage 12 is parallel to the X axis, the reflection surface 17a of the X axis moving mirror 17 is kept parallel to the YZ plane. Then, as shown in FIG. 6B, when the upper surface 12b of the sample stage 12 is inclined by the angle θys in the θy direction, the reflecting surface 17a of the X-axis movable mirror 17 is also inclined by the angle θys with respect to the YZ plane. Become. This angle θys
Corresponds to the “tilt angle” in the claims.

As shown in FIG. 6A, the X-axis collimator 27 provided to face the X-axis moving mirror 17
A lens 51 arranged with the optical axis aligned with the axis;
, A half mirror 53 arranged between the light source 52 and the lens 51, and a detector 54 (for example, a CCD line sensor) arranged on a reflection optical path of the half mirror 53. Have been.

In the X-axis collimator 27, the light source 5
The light from 2 becomes parallel light via the lens 51. Then, the parallel light enters the reflection surface 17a of the X-axis moving mirror 17. Light reflected from the reflection surface 17a passes through the lens 51, is reflected by the half mirror 53, and enters the detector 54. Here, when the reflection surface 17a is kept parallel to the YZ plane (the state of FIG. 6A), the reflected light from the reflection surface 17a is parallel to the X axis. The reflected light is transmitted to the detector 54 via the lens 51 and the half mirror 53.
And an image is formed at the reference point X0.

On the other hand, when the reflecting surface 17a is inclined by an angle θys with respect to the YZ plane (the state shown in FIG. 6B), the reflected light L1 from the reflecting surface 17a has an angle of 2 · (θys) with respect to the X axis. Will only lean. The reflected light L1 is transmitted to the lens 5
An image is formed on the point X1 of the detector 54 via the first mirror 1 and the half mirror 53. The shift amount ΔX between the point X1 and the reference point X0 is proportional to f · sin [2 (θys)] when the focal length of the lens 51 is represented by f.

Therefore, in the X-axis collimator 27, the deviation ΔX between the image forming point X1 of the reflected light L1 and the reference point X0.
Is detected by the detector 54, the inclination angle θys of the reflection surface 17a of the X-axis movable mirror 17, that is, the inclination angle θys of the upper surface 12b of the sample stage 12 can be detected. When the sample stage 12 moves in the X direction, the X-axis moving mirror 17 also moves in the X direction, and the distance D from the X-axis collimator 27 to the X-axis moving mirror 17 changes. Deviation Δ detected by detector 54
X does not change with distance D. For this reason, the sample stage 12
Irrespective of the X position, the tilt angle θy of the sample stage 12 is similarly calculated.
s can be detected.

In the probe device 10 (FIG. 1), the inclination of the sample table 12 in the θx direction is detected.
The Y-axis movable mirror 18 is attached to the side surface of the optical disk 1 and a Y-axis collimator 28 is provided to face the Y-axis movable mirror 18. The Y-axis moving mirror 18 has a reflecting surface orthogonal to the upper surface 12b of the sample stage 12, similarly to the reflecting surface 17a of the X-axis moving mirror 17 described above. Therefore, the upper surface 12 of the sample stage 12
When b is parallel to the Y axis, the reflecting surface of the Y axis moving mirror 18 is XZ
Kept parallel to the plane. When the upper surface of the sample stage 12 is inclined by θxs in the θx direction, the reflecting surface of the Y-axis moving mirror 18 is also inclined by θxs with respect to the XZ plane.

The Y-axis collimator 28 has the same configuration as the X-axis collimator 27 described above, is rotated by 90 ° in the XY plane, and is arranged such that the optical axis of the lens 51 is aligned with the Y-axis.

Therefore, according to the Y-axis collimator 28, parallel light is made incident on the reflecting surface of the Y-axis moving mirror 18, and the deviation amount ΔY (∝f · sin [2 (θx
s)]), the inclination angle θxs of the reflection surface of the Y-axis movable mirror 18, that is, the inclination angle θxs of the upper surface 12b of the sample stage 12 can be detected. The sample table 12
Moves in the Y direction, the Y-axis moving mirror 18 also moves in the Y-direction, and the distance from the Y-axis collimator 28 to the Y-axis moving mirror 18 changes. However, the displacement ΔY detected by the Y-axis collimator 28 changes. do not do. Therefore, the inclination angle θxs of the sample stage 12 can be detected in the same manner regardless of the Y position of the sample stage 12.

Further, in the probe device 10 of the present embodiment, a microscope 25 for observing an alignment mark (not shown) on the wafer 20 is provided on the stage base 16. The microscope 25 captures an image of an alignment mark of the wafer 20 located within a field of view of an objective lens (not shown), and obtains an XY image of the obtained alignment mark image.
This is to detect coordinates.

The microscope 25 detects the Z coordinate of the wafer 20 at the position of the alignment mark based on the amount of autofocus driving when the contrast of the image is maximized when the image of the alignment mark is captured. Further, the probe device 10 includes a probe card 3
A microscope 35 for observing the probe pins 32 attached to the sample stage 12 is attached to the side surface of the sample table 12. The microscope 35 captures an image of the probe pins 32 located in the field of view of an objective lens (not shown), and detects the XY coordinates of the obtained probe pin image.

The microscope 35 detects the Z coordinate of the tip of the probe pin 32 based on the auto-focus driving amount at the time of capturing the probe pin image. Further, a reference mark 19 is attached to the upper surface of the sample table 12 of the probe device 10. This fiducial mark 19
The positional relationship between the camera and the microscope 35 is predetermined as a device constant. It should be noted that the device constant is measured by measuring the positional relationship between the index and the reference mark 19 with the microscope 25 in which an index indicating the optical axis of the microscope 35 is attached to the optical member of the microscope 35 in advance. You may do it.

On the other hand, the positional relationship between the reference mark 19 and the microscope 25 changes depending on the positions of the X stage 14 and the Y stage 15, that is, the X position and the Y position of the sample table 12. However, the reference mark 19 is observed with the microscope 25, and the X stage 14 and the Y stage 15 at that time are observed.
Is detected, the coordinate system of the microscope 25 and the coordinate system of the microscope 35 can be matched.

As shown in FIG. 7A, the probe device 10 of the present embodiment
a between the portion 41 a of the “a” and the roller 49 of the Z drive portion 41,
A load sensor 55 is provided. This load sensor 55
Is the holding frame 5 attached to the portion 41a of the sample table 12.
8 and a strain gauge 59 attached to the inside of the holding frame 58. The resistance value of the strain gauge 59 changes according to the load W applied by the contact between the pad 22 of the semiconductor chip 21 and the probe pin 32.

Therefore, the load sensor 55 can detect the load W applied by the contact by detecting a change in the resistance value of the strain gauge 59 by a bridge circuit (not shown). 7B, the probe device 10 is provided with a load sensor 56 between the portion 42a of the lower surface 12a of the sample table 12 and the roller 49 of the Z drive unit 42. The load sensor 56 has the same configuration as the load sensor 55, and can detect the load W applied by the contact.

Further, the probe device 10 has the structure shown in FIG.
As shown in FIG.
a and the load sensor 5 between the roller 49 of the Z drive unit 43.
7 are provided. The load sensor 57 has the same configuration as the load sensor 55, and can detect the load W applied by the contact.

As described above, the probe device 10 is provided with the three load sensors 55 to 57. In the present embodiment, the pads 22 of the semiconductor chip 21 and the probe pins 3 are provided.
One of the three load sensors 55 to 57 is used to detect the load W applied by the contact with the second 2 (described later). The probe device 10 (FIG. 1) of the present embodiment includes the tester 34, the X stage 14, the Y stage 15, and the Z leveling stage 13 (the Z driving units 41 to 41).
43), a not-shown θz stage, and an X-axis collimator 2
7, a Y-axis collimator 28, microscopes 25 and 35, and load sensors 55 to 57, and a control unit 60 for controlling these components is provided.

Incidentally, the Z leveling stage 13 (Z driving sections 41 to 43) of the above-described embodiment corresponds to “supporting means” and “moving means” in the claims. Further, the X stage 14 and the Y stage 15 of the embodiment correspond to “in-plane moving means” in the claims. Further, the X-axis moving mirror 17, the Y-axis moving mirror 18, the X-axis collimator 27, and the Y-axis collimator 28 of the embodiment correspond to "tilt detecting means" in the claims. The reflecting surface 17a of the X-axis moving mirror 17 corresponds to a "reflecting mirror surface" in the claims. Further, the load sensors 55 to 57 of the embodiment correspond to “load detecting means” in claims. In addition, the control unit 60 includes a “tilt control unit”,
It corresponds to "load correction means" and "positioning control means".

Next, an operation of testing the semiconductor chip 21 on the wafer 20 using the probe device 10 configured as described above will be described. The control unit 60 controls the card holder 3 of the probe card 30 prior to the actual test.
1 and the state of attachment of the wafer 20 to the wafer holder 11, and also the surface state of the wafer 20 due to elongation, undulation, warpage, bending, etc. of the wafer 20 or the wafer holder 11. .

The observation of the mounting state of the probe card 30 is performed by observing the probe pins 32 themselves at three or more places using a microscope 35 (FIG. 1). Microscope 3
5 detects the XY coordinates of the probe pin 32 to be observed and the Z coordinate of the tip of the probe pin 32. The detection results at three or more locations by the microscope 35 are input to the control unit 60 as XY coordinate information and Z coordinate information.

The control unit 60 calculates the mounting position in the XY direction and the mounting position in the θz direction of the probe card 30 based on the XY coordinate information of the probe pins 32.
The control unit 60 also determines a surface formed by the tips of the probe pins 32, 32,... Based on the Z coordinate information of the probe pins 32 (hereinafter, referred to as a “pin tip surface 32a” in FIG. 8).
) With respect to the horizontal are calculated.
FIG. 8 shows only the inclination θyp in the θy direction.

The probe card 30 calculated as described above
Is stored in the memory in the control unit 60. On the other hand, the observation of the mounting state and the surface state of the wafer 20 is performed by observing the alignment marks (not shown) on the wafer 20 at three or more places using the microscope 25 (FIG. 1). At this time, the upper surface 12b of the sample stage 12 is kept horizontal (parallel to the XY plane).

The microscope 25 detects the XY coordinates of the alignment mark to be observed and the Z coordinate of the wafer 20 at the position of the alignment mark. The detection results at three or more locations by the microscope 25 are input to the control unit 60 as XY coordinate information and Z coordinate information. The controller 60 controls the XY coordinate information of the alignment mark and the probe card 3
0 based on the mounting position in the θz direction
The rotation deviation amount in the θz direction with respect to the entire probe card 30 is calculated. Then, based on the rotation deviation amount, θ
Drive control of the z stage (not shown)
The mounting position in the z direction is matched with the mounting position of the probe card 30 in the θz direction.

The control unit 60 calculates the amount of displacement in the XY direction of each semiconductor chip 21 formed on the wafer 20 based on the XY coordinate information of the alignment mark. The displacement of each semiconductor chip 21 in the XY directions is caused by a displacement of the mounting position of the entire wafer 20 in the XY directions, the elongation of the wafer 20 itself, and the like. Incidentally, a high temperature test (the temperature of the wafer 20 is, for example, 100 ° C.)
At the time of, 150 μm
Some elongation occurs.

Further, the control unit 60 controls the Z
By performing processes such as data fitting and interpolation on the coordinate information, the inclination θx of each semiconductor chip 21
c and θyc are calculated (see FIG. 9, in which only the inclination θyc in the θy direction is shown). Note that the calculated inclinations θxc and θyc of each semiconductor chip 21 are calculated based on the sample table 1
2 shows the inclination with respect to the upper surface 12b. The inclinations θxc and θyc are equal to the inclination of the surface formed by the plurality of pads 22, 22,... Of each semiconductor chip 21.

The inclination θxc, θyc of each semiconductor chip 21
Is caused by the overall inclination related to the mounting state of the wafer 20 and the undulation, warping, bending, etc. of the wafer 20 and the wafer holder 11. Information on the mounting state and surface state of the wafer 20 calculated in this way is also stored in the memory in the control unit 60.

The actual test for the semiconductor chip 21 is as follows.
Probe card 30 stored in a memory in control unit 60
Is set as a reference XY position, and the inclinations θxp and θyp (FIG. 8) of the pin tip surface 32a with respect to the horizontal are set as reference inclination angles in accordance with the procedure shown in FIGS. 10 to 12 show a plurality of semiconductor chips 21, 21,... Formed on the wafer 20.
FIG. 3 shows a procedure for testing one of the test pieces (FIG. 2). All the semiconductor chips 21 and 21 on the wafer 20
.. May be sequentially performed by repeating the procedures in FIGS. Here, one semiconductor chip 21
The operation when testing is described.

First, as shown in FIG. 10, the control unit 60 aligns the semiconductor chip 21 to be tested (hereinafter referred to as "test chip 21a") in the X and Y directions (S11). For this reason, the control unit 60
The position shift amount a in the XY directions is read from the memory, and the X stage 14 and the Y stage 15 are drive-controlled in consideration of the position shift amounts in the XY directions.

As a result, the test chip 21a is attached to the probe card 30 in the XY direction (reference XY position).
Move up to. At this time, as shown in FIG. 13, each pad 22 of the test chip 21a is located immediately below each probe pin 32 of the probe card 30 and faces the tip thereof. Next, the controller 60 corrects the inclination of the test chip 21a (S12 in FIG. 10).

For this reason, the control unit 60 controls the test chip 21a
The inclinations θxc and θyc with respect to the upper surface 12b of FIG. 13 are read out from the memory, and the difference (θxc−θxp, θyc−θyp) between the inclinations θxc and θyc and the inclinations θxp and θyp (reference inclination angles) of the pin tip surface 32a with respect to the horizontal is read. ), That is, the inclination shift amount of the test chip 21a with respect to the pin tip surface 32a is calculated.

The inclination deviation of the test chip 21a is corrected, and the inclinations θxa and θy of the test chip 21a with respect to the horizontal are corrected.
a (see FIG. 13) to match the inclinations θxp and θyp of the pin tip surface 32a with respect to the horizontal, the sample table 12
Of the angles θxs, θys of the above-mentioned difference θxc−θxp, θy
What is necessary is just to make it coincide with c-θyp. Therefore, the control unit 60
Is the difference θxc−θx prior to S12 in FIG.
Let p, θyc−θyp be the reference inclination θxs of the sample stage 12.
o and θyso are stored in the memory.

Then, the control unit 60 executes the inclination correction processing of the test chip 21a shown in S12 of FIG. That is, the current inclinations θxs and θys of the sample stage 12 are acquired from the Y-axis collimator 28 and the X-axis collimator 27 (S31 in FIG. 12A), and the acquired inclinations θxs and θys.
Is shifted from the reference inclinations θxso and θyso stored in the memory (S32), the drive of the Z leveling stage 13 (Z driving units 41 to 43) is controlled in accordance with the amount of the deviation, The inclinations θxs and θys are made to match the reference inclinations θxso and θyso (S33).

As a result, as shown in FIG. 13, the inclinations θxa and θya of the test chip 21a with respect to the horizontal are equal to the inclinations θxp and θyp of the pin tip surface 32a with respect to the horizontal, and the test chip 21a and the pin tip surface 32a are separated. Be parallel. When the above-described positioning and tilt correction are completed, the control unit 60 starts to raise the sample stage 12 in the Z direction (S13 in FIG. 10).

At this time, the Z leveling stage 13
By controlling the driving while synchronizing the driving units 41 to 43, the inclinations θxs and θys of the sample table 12 are set to the reference inclination θ.
The sample stage 12 can be raised in the Z direction while keeping xso and θyso. As a result, the test chip 21a
Rises in the Z direction while keeping parallel to the pin tip surface 32a.

Next, the control unit 60 monitors whether the pad 22 of the test chip 21a has contacted the probe pin 32 while raising the sample stage 12 in the Z direction (S14 in FIG. 10). In the present embodiment, one of the load sensors 55 to 57 (FIG. 7) attached to the lower surface 12a of the sample stage 12 is used as a reference load sensor for monitoring whether or not the contact has been made.

This reference load sensor is used as load sensors 55-5
In order to select one of the test chips 21a from the test chip 21a on the wafer 20, as shown in FIG.
, And the relative positions R1, R2, R3 between the mounting positions of the load sensors 55, 56, 57 are calculated. Then, the control unit 60 determines the load sensor having the smallest relative position as the reference load sensor. In FIG. 14, since the relative position R1 of the load sensor 55 is the smallest, the load sensor 55 is hereinafter referred to as the reference load sensor 55.

The reference load sensor 55 thus determined
Is detected by the pad 22 (FIG. 13) of the test chip 21a.
Is changed according to the load W applied by this contact. Therefore, the control unit 60
Monitors whether the detected value of the reference load sensor 55 has changed while raising the sample stage 12 in the Z direction in S14 of FIG. It is determined that 22 has come into contact with the probe pin 32.

At this time, as shown in FIG. 15, all the pads 22 of the test chip 21a contact the tips of the probe pins 32 equally. Then, in order to ensure the contact between the pad 22 and the tip of the probe pin 32, the controller 60 further pushes up the sample table 12 by a predetermined amount (about 10 μm) in the Z direction.

The push-up amount (the above-mentioned predetermined amount) after the contact between the probe pin 32 and the pad 22 is strictly controlled so that the contact pressure between the tip of the probe pin 32 and the pad 22 becomes an optimum value. There is a need to. Therefore, immediately after the tip of the probe pin 32 and the pad 22 come into contact with each other, the control unit 60 starts measuring the push-up amount (S1 in FIG. 10).
5).

The measurement of the push-up amount is performed by moving the portion 41a provided with the reference load sensor 55 (see FIG. 14) to Z
Z drive unit 41 (hereinafter referred to as “reference Z drive unit 4”)
1 ") (not shown).

When the sample stage 12 is being pushed up, the load suddenly increases in proportion to the pushing amount. Therefore, the inclinations θxs and θys of the sample table 12 (FIG. 15) change from the reference inclinations θxso and θyso, and the pin tip surface 3
There is a risk that the parallelism between the test chip 2a and the test chip 21a may not be maintained. Therefore, the control unit 60 determines that the pushing amount (measured by the reference Z driving unit 41) is a predetermined value (1
The inclination of the test chip 21a is corrected until it becomes about 0 μm (the determination in S17 in FIG. 10 becomes yes) (S16 in FIG. 10).

Step S16 is the same as step S12 described above. That is, the inclinations θxs and θys of the sample stage 12 are obtained from the Y-axis collimator 28 and the X-axis collimator 27, and if the inclinations θxs and θys deviate from the reference inclinations θxso and θyso, the inclinations are determined according to the deviation amounts. The drive of the Z leveling stage 13 is controlled and the inclination θx of the sample stage 12 is controlled.
s and θys are made to match the reference inclinations θxso and θyso.

In the case of S16, since the reference Z drive unit 41 is used for measuring the pushing amount, the drive control of the other Z drive units 42 and 43 is performed, and the test chip 2 is controlled.
It is preferable to perform the inclination correction of 1a. As described above, since the sample stage 12 is pushed up in the Z direction while correcting the inclination of the test chip 21a, even if the load suddenly increases in proportion to the pushing amount, the inclination θxs, θys of the sample stage 12 is used as a reference. The inclination can be maintained at θxso and θyso. As a result, the test chip 21a is pushed up in the Z direction while always keeping parallel to the pin tip surface 32a.

Therefore, all the pads 22 of the test chip 21a are equally pressed against the tips of the probe pins 32. Then, when the push-up amount measured by the reference Z drive unit 41 reaches a predetermined value (about 10 μm) (FIG. 1).
0, the determination in S17 is yes), the control unit 60
The push-up in the Z direction of Step 2 is stopped (S18).

At this time, since the pin tip surface 32a is kept parallel to the test chip 21a, the test chip 2
In this state, all the pads 22a are pressed against the tips of the probe pins 32 equally and with the optimum contact pressure. In this state, since the probe pins 32 and the pads 22 are pressed with optimal contact pressures, respectively,
The circuit in a is reliably electrically connected to the tester 34 (FIG. 1) via the pad 22 and the probe pin 32.

The control unit 60 outputs a control signal to the tester 34 to start the test on the test chip 21a (S19 in FIG. 11). The tester 34 executes a test such as an electrical characteristic or a circuit function by applying a predetermined electric signal to a circuit in the test chip 21a and examining a response from the test chip 21a according to a test program stored in advance. I do.

The time required for the test by the tester 34 varies depending on the type of test, but is about several seconds to 30 minutes. Here, in order to complete the test by the tester 34 to the end, it is necessary to maintain a state in which all the pads 22 of the test chip 21a are in contact with the probe pins 32 at the same optimal contact pressure.

During the test, the probe pin 32 and the pad 2
If the contact state with the probe 2 is deteriorated and the optimal contact pressure cannot be maintained, the probe pin 32 and the pad 22 start to separate, causing a problem such as an increase in contact resistance, and making it impossible to continue a good test. . However, since the probe pins 32 and the pads 22 are strongly pressed by the above-described push-up of about 10 μm, the sample table 12 and the wafer 20 may be bent at the test chip 21a during the test. The probe pin 32 has a short overall length (1
00 μm), and the tolerance for deflection is small. For this reason, the optimal contact pressure may not be maintained.

Therefore, the control unit 60 performs correction for maintaining the contact pressure between the pad 22 and the probe pin 32 at an optimum value until the test is completed (the determination in S22 in FIG. 11 becomes yes). (S20 in FIG. 11). Since the change in the contact pressure between the pad 22 and the probe pin 32 appears as a change in the detection value of the reference load sensor 55, the control unit 60
Stores the value detected by the reference load sensor 55 immediately after the start of the test in the memory as the reference value Wo.

Then, as shown in FIG. 12B, the control unit 60 acquires the load W applied by the contact between the pad 22 and the probe pin 32 from the reference load sensor 55 (S
34), this detected value W is stored in the memory as a reference value Wo
Is changed (S35), the drive of the reference Z drive unit 41 is controlled in accordance with the change amount, and the detection value W of the reference load sensor 55 is returned to the original reference value Wo (S36).

Thus, even if the sample table 12 or the wafer 20 is bent at the test chip 21a during the test, the test can be performed while maintaining the contact pressure between the short probe pin 32 and the pad 22 at the optimum value. Can be continued. Further, during the test, the inclinations θxs and θys of the sample stage 12 change from the reference inclinations θxso and θyso, and the pin tip surface 3
The parallelism between 2a and the test chip 21a may be lost.

Therefore, until the test is completed (Yes in S22 of FIG. 11), the control unit 60 controls the test chip 2
The inclination correction of 1a is performed (S21 in FIG. 11). This S21
Are the same as S12 and S16 in FIG. 10 described above. That is, the inclinations θxs and θys of the sample table 12 are acquired from the Y-axis collimator 28 and the X-axis collimator 27, and the inclination θ
When xs and θys deviate from the reference inclinations θxso and θyso, the drive of the Z leveling stage 13 is controlled based on the deviations, and the inclinations θxs and θ of the sample stage 12 are adjusted.
ys is made to match the reference inclinations θxso and θyso.

In the case of S21, since the reference Z drive unit 41 is used for correcting the contact pressure between the probe pin 32 and the pad 22 (S20), other Z drive units 42 and 43 are used. It is preferable to perform drive control to correct the inclination of the test chip 21a. Thus, the test chip 21
During the test for S (S19 to S22), the contact pressure between the pad 22 and the probe pin 32 is corrected (S20) and the inclination of the test chip 21a is corrected (S21), so that the short probe pin 32 (approximately 100 μm in total length) is used. Is used, the contact pressure between all the pads 22 of the test chip 21a and the pads 22 can be equally maintained at the optimum value.

Accordingly, the test of the test chip 21a by the tester 34 can be performed with high accuracy even in a high frequency region while keeping all the pads 22 of the test chip 21a and the probe pins 32 in a stable contact state.
Then, when a signal notifying the end of the test is input from the tester 34 (S22 in FIG. 11), the control unit 60 sets the sample stage 12
Is lowered in the Z direction (S23 in FIG. 11).

As described above, according to the probe device 10 of the present embodiment, the X
The tilts θys and θxs of the sample stage 12 are always detected by the axis moving mirror 17, the Y axis moving mirror 18, and the X axis collimator 27 and the Y axis collimator 28 provided opposite thereto, and the reference angles θxso and θyso are obtained. It can be corrected so as not to be deviated. Therefore, the inclinations θxa and θya of the test chip 21a with respect to the horizontal can be kept constant.

Further, the reference inclination θxs of the sample stage 12
o and θyso are the differences (θxc−θxp, θy) between the inclinations θxc and θyp of the test chip 21a with respect to the upper surface 12b and the inclinations θxp and θyp of the pin tip surface 32a with respect to the horizontal.
c-θyp), the wafer 20
When the wafer is mounted, the overall tilt may occur or the wafer 2
The test chip 21a and the pin tip surface 32a can be kept parallel even when the wafer 0 itself or the wafer holder 11 has undulation, warpage, bending, or the like.

The amount of change in the inclination θys and θxs of the sample table 12 generated when the probe pins 32 and the pads 22 come into contact with each other depends on the number and arrangement of the probe pins 32 (the number and arrangement of the pads 22) and the test chip 21a. Position (X table 1
4 or the position of the Y table 15), the amount of push-up after the contact, the temperature of the wafer 20, and the like. However, the probe device 10 of the present embodiment directly detects the tilt of the sample table 12 at that time, and detects the tilt in real time. Since the correction value is obtained, all the semiconductor chips on the wafer 20 can be accurately and reliably tested regardless of the type of the semiconductor chip 21 and the test conditions.

Further, in the probe device 10, since the sample table 12 is raised in the Z direction while maintaining the parallel between the test chip 21a and the pin tip surface 32a, the pad 22 is brought into contact with the tip of the probe pin 32. When pushing up,
There is no displacement of the tip of the probe pin 32 on the surface of the pad 22 or partial increase in the contact pressure between the probe pin 32 and the pad 22. Therefore, it is possible to prevent the surface of the pad 22 from being damaged by the tip of the probe pin 32.

The tip of the probe pin 32 is
Since the size of the pads 22 does not protrude, there is no possibility that the tips of the probe pins 32 are short-circuited due to contact between the pads 22 even if the size of the pads 22 is reduced or the distance between the pads 22 is reduced with the miniaturization of the LSI. . Further, in the probe device 10, taking into account the amount of displacement in the XY directions of the test chip 21a obtained in advance before the actual test,
Since the X stage 14 and the Y stage 15 are drive-controlled, even if the entire mounting position of the wafer 20 is shifted in the X and Y directions or the wafer 20 itself is extended, each pad 22 of the test chip 21a and the probe card 30 And the tip of each probe pin 32 can be reliably opposed to each other for positioning.

For this reason, even if the mounting density of the probe pins becomes higher in the future and the size of the pad becomes smaller, the positional relationship between the pad and the probe pin is kept constant.
The pad and the probe pin can be reliably contacted.

In the probe device 10, the Z drive unit provided at the position closest to the test chip 21a among the three Z drive units 41 to 43 is selected as the reference Z drive unit. Even if the semiconductor chip 21 becomes the test chip 21a, the push-up amount after the contact can be strictly measured, and the contact pressure between the probe pin and the pad can be corrected efficiently.

Further, in the probe device 10, since the load sensor provided at the position closest to the test chip 21a is selected as the reference load sensor among the three load sensors 55 to 57, any semiconductor chip on the wafer 20 can be selected. 2
Even if 1 is the test chip 21a, it is possible to accurately detect the Z position when the probe pin comes into contact with the pad and the change in the contact pressure during the test.

In the probe device 10, the sample table 12
Is monitored in the Z direction while the Z position where the probe pins 32 and the pads 22 are in contact with each other is monitored.
Even if the height (Z coordinate) of the semiconductor chip 21 varies due to undulation or the like, the amount of push-up after the contact is μm
It can be strictly controlled by order. Therefore,
For any semiconductor chip 21 on the wafer 20,
The contact pressure between the probe pin 32 and the pad 22 can be unified to an optimum value.

As a result, electrical characteristics and circuit function tests can be performed on all the semiconductor chips 21, 21,... On the wafer 20 with high accuracy. Further, in the probe device 10, since the inclination correction of the sample stage and the correction of the contact pressure between the pad and the probe pin are always performed, the X stage 1
There is no need to use a high precision linearity as the 4 or Y stage 15, and the stage manufacturing cost can be reduced.

In the above-described embodiment, the total length of the semiconductor chip 21 is set so that a test in the high frequency region can be performed.
The case where the test is performed using the short probe pin 32 of about 0 μm (short pin) has been described as an example, but the present invention is effective even when applied to the case of using a long probe pin having a total length of about 2 mm to 3 mm. It is.

That is, a long probe pin (total length 2 mm)
When the number of attachments is increased (increase in number of pins) or the attachment density is increased (increase in density), the inclination of the sample stage caused by the contact between the probe pins and the pads is increased. The probe pin will not be absorbed due to the deflection of the probe pin, so the tilt correction of the sample stage and the correction of the contact pressure between the pad and the probe pin will be performed to maintain the contact state between the probe pin and the pad while maintaining the accuracy. Test can be performed well.

In the above embodiment, the sample table 12
The X-axis moving mirror 17 and the Y-axis moving mirror 1
8, an example has been described in which detection is performed by the X-axis collimator 27 and the Y-axis collimator 28. Instead of the X-axis collimator 27 and the Y-axis collimator 28, an X-axis interferometer capable of measuring pitching, which is the tilt of each movable mirror, Even when the Y-axis interferometer is provided, the inclinations θys and θxs of the sample table 12 can be detected in the same manner.

The X-axis movable mirror 17, the Y-axis movable mirror 18,
Instead of the X-axis collimator 27 and the Y-axis collimator 28, a plurality of non-contact proximity sensors are arranged above the sample table 12,
The inclination θys and θxs of the sample table 12 may be detected by measuring the distance between the proximity sensor and the upper surface 12b of the sample table 12. Further, in the above-described embodiment, when calculating the inclinations θxc and θyc of the respective semiconductor chips 21, the alignment mark on the wafer 20 is observed using the microscope 25, and the Z coordinate of the wafer 20 at the alignment mark position is detected. An example was described, but the wafer 20
Each semiconductor chip 21 based on the undulation of the wafer holder 11 measured before the wafer is held on the wafer holder 11.
May be calculated.

Further, in the above-described embodiment, the sample table 12
Although the example in which the contact position between the probe pin and the pad is detected by the load sensors 55 to 57 attached to the lower surface 12a of the Z drive unit 41 to 43 is described, the load state of the servomotor 48 of the Z drive units 41 to 43 is monitored. Can also be detected. Further, in the above-described embodiment, an example has been described in which one of the Z driving units 41 to 43 is selected according to the test chip 21a and one of the load sensors 55 to 57 is selected and used for control. Alternatively, the reference Z drive unit and the reference load sensor may be fixed regardless of the test chip 21a.

Further, in the above-described embodiment, an example in which a test is performed for each semiconductor chip has been described. However, the present invention can be applied to a case where a plurality of semiconductor chips are simultaneously tested. In this case, as a probe card, a multi-probe having probe pins formed in an arrangement corresponding to pads formed on a plurality of semiconductor chips is attached to a card holder.

[0115]

As described above, according to the first to ninth aspects of the present invention, even if the number of probe pins is increased, the density of the probe pins is increased, or the number of pins is reduced, the semiconductor on the wafer is reduced. Since the contact state between the pads formed on the chip and the probe pins can be kept uniform, it is possible to reliably test the electrical characteristics and circuit functions of the semiconductor chip in a wafer state. As a result, the reliability of the test on the semiconductor chip is improved. In addition, it can contribute to improvement of test efficiency and productivity, and further, miniaturization of a semiconductor chip.

[Brief description of the drawings]

FIG. 1 is a diagram showing a configuration of a probe device 10 of the present embodiment.

FIG. 2 is a diagram illustrating a wafer 20 and a semiconductor chip 21.

FIG. 3 is a diagram illustrating a probe card 30.

FIG. 4 is a diagram illustrating a Z-leveling stage 13;

FIG. 5 is a diagram illustrating a Z drive unit 41.

FIG. 6 is a diagram illustrating an X-axis collimator 27.

FIG. 7 is a diagram illustrating load sensors 55 to 57.

FIG. 8 is a view showing an inclination of a pin tip surface 32a.

FIG. 9 is a diagram showing an inclination of a semiconductor chip 21;

FIG. 10 is a flowchart illustrating a test operation using the probe device 10.

FIG. 11 is a flowchart illustrating a test operation using the probe device 10.

FIG. 12 is a flowchart illustrating a test operation using the probe device 10.

FIG. 13 is a diagram showing a state after correcting the positional relationship between the pad 22 of the test chip 21a and the probe pin 32 and the inclination.

FIG. 14 is a diagram showing a positional relationship between a test chip 21a and load sensors 55 to 57.

FIG. 15 is a diagram showing a state in which a pad 22 of a test chip 21a and a probe pin 32 are in contact with each other.

FIG. 16 is a diagram showing a configuration of a conventional probe device 70.

FIG. 17 is a diagram illustrating push-up after contact between a probe pin 82 and a pad 83;

FIG. 18 is a diagram illustrating a state where a sample table 74 is tilted due to contact between a probe pin 82 and a pad 83.

FIG. 19 is a diagram showing a undulating state of the surface of a wafer 72.

FIG. 20 is a diagram illustrating a state in which the height differs for each semiconductor chip 71.

FIG. 21 is a diagram illustrating a problem when a short probe pin 86 is used.

FIG. 22 is a diagram illustrating a problem when a short probe pin 86 is used.

[Explanation of symbols]

 10, 70 Probe device 11, 73 Wafer holder 12, 74 Sample table 13 Z leveling stage 14 X stage 15 Y stage 16 Stage base 17 X axis moving mirror 18 Y axis moving mirror 19 Reference mark 20, 72 Wafer 21, 71 Semiconductor chip 22, 83 Pad 25, 35 Microscope 27 X-axis collimator 28 Y-axis collimator 30, 76 Probe card 31 Card holder 32, 82, 86 Probe pin 33, 81 Card board 34, 77 Tester 41-43 Z drive section 46 Lower inclined member 47 Upper inclined member 48 Servo motor 49 Roller 51 Lens 52 Light source 53 Half mirror 54 Detector 55-57 Load sensor 58 Holding frame 59 Strain gauge 60 Control unit

 ──────────────────────────────────────────────────続 き Continued on front page F term (reference) 2G011 AA02 AA17 AC06 AC14 AE03 2G032 AA00 AE04 AF02 AL03 4M106 AA01 AA02 BA01 BA14 DD05 DD06 DD12 DD13 DJ01 DJ03 DJ04 DJ05 DJ07 9A001 GG01 HZ34 JZ45 KK37 KZ54 LL05 LZZ

Claims (9)

[Claims] 1. A probe device for testing electrical characteristics of a semiconductor chip by bringing a plurality of probe pins into contact with a plurality of pads formed on a semiconductor chip on a wafer, wherein the wafer is placed thereon. Mounting means, supporting means for tiltably supporting the mounting means, tilt detecting means for detecting the tilt of the mounting means, and driving the supporting means based on a detection result by the tilt detecting means And a tilt control means for controlling the tilt of the placing means to a predetermined tilt. 2. The probe device according to claim 1, wherein the tilt control unit determines a tilt of a surface formed by the plurality of pads of the wafer mounted on the mounting unit.
A probe device for performing control to match the inclination of a surface formed by the tips of the plurality of probe pins. 3. The probe device according to claim 1, wherein the inclination detecting means has a reflecting mirror surface provided on the placing means, and a collimator for measuring a tilt angle of the reflecting mirror surface. A probe device characterized by the above-mentioned. 4. The probe device according to claim 1, wherein the inclination detecting means includes: a reflecting mirror provided on the placing means; and an interferometer for measuring a tilt angle of the reflecting mirror. A probe device comprising: 5. The probe device according to claim 1, wherein: a moving means for moving the placing means in a direction of contact and separation between the pad and the probe pin; Load detecting means for detecting a load applied by the contact of the probe pin; calculating a shift amount between the load detected by the load detecting means and a predetermined load, and driving the moving means based on the shift amount. And a load correcting means for correcting a load applied by the contact. 6. The probe device according to claim 5, wherein the load detecting means detects a load at at least three places of the placing means. 7. The probe device according to claim 6, wherein the load correcting means is configured to move the moving means in accordance with a relative position between a position at which the load detecting means detects a load and a position of a semiconductor chip to be tested. A probe device for controlling a driving amount of a probe. 8. The probe device according to claim 5, wherein said load detecting means has a plurality of strain gauges. 9. A probe apparatus for testing electrical characteristics of a semiconductor chip by contacting a plurality of probe pins with a plurality of pads formed on a semiconductor chip on a wafer, wherein the wafer is placed on the wafer. Mounting means to be mounted; in-plane moving means for moving the mounting means in a plane orthogonal to the direction of contact and separation between the pad and the probe pin; And positioning control means for positioning the semiconductor chip at a point facing the plurality of probe pins by driving the in-plane moving means in consideration of the displacement amount. Probe device.