# US 12,007,413 B2 · Prober controlling device, prober controlling method, and prober

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 12,007,413 B2
- Document ID: US12007413B2
- Status: Granted
- Kind: B2
- Application number: 18/527,013
- Filed: 2023-12-01
- Published: 2024-03-21
- Granted: 2024-06-11
- Earliest claimed: 2021-06-04
- Assignees: Tokyo Seimitsu Co., Ltd.
- Applicants: Tokyo Seimitsu Co., Ltd.
- Inventors: Tetsuo Yoshida; Shunsuke Otake; Tatsuya Matsuoka
- Web page: https://patentgrep.com/patent/US12007413B2

## CPC

- G01R 1/07307: {with individual probe elements, e.g. needles, cantilever beams or bump contacts, fixed in relation to each other, e.g. bed of nails fixture or probe card}
- G01R 1/0675: {Needle-like}
- G01R 31/2887: {involving moving the probe head or the IC under test; docking stations (moving single probes G01R1/06705; moving individual probes in multiple probes G01R1/07392)}
- G01R 31/2889: {Interfaces, e.g. between probe and tester (G01R31/31905 and G01R1/07364 take precedence)}
- G01R 31/2891: {related to sensing or controlling of force, position, temperature (G01R31/2874 takes precedence; sensing of force G01L; sensing of position G01B, G01D; sensing of temperature G01K; controlling in general G05)}
- H10P 72/0602: {Temperature monitoring}
- H10P 72/7602: {the wafers being placed on a robot blade or gripped by a gripper for conveyance}
- H10P 74/00: Testing or measuring during manufacture or treatment of wafers, substrates or devices
- G01R: MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES (indicating correct tuning of resonant circuits H03J3/12)
- H10P: GENERIC PROCESSES OR APPARATUS FOR THE MANUFACTURE OR TREATMENT OF DEVICES COVERED BY CLASS H10

## Abstract

A prober controlling device for bringing probe needles into contact with semiconductor chips, includes an input data acquiring unit configured to acquire input data including temperature data of at least one of a probe card and a card holder, a predicting unit configured to predict a position of a tip end of a probe needle based on the input data acquired by the input data acquiring unit, using a prediction model that receives input of the input data and outputs the position of the tip end of the probe needle, and a determining unit configured to determine whether or not to execute prediction by the predicting unit, based on input data used as teacher data for machine learning of the prediction model and the input data acquired by the input data acquiring unit, before the prediction by the predicting unit.

## Claims

### Claim 1

A prober controlling device for a prober including a wafer chuck configured to hold a wafer on which semiconductor chips are formed, a probe card having probe needles, a card holder configured to hold an outer periphery of the probe card and place the probe card opposite the wafer and a relative moving unit configured to move the wafer chuck relatively to the probe needles, the prober controlling device being configured to drive the relative moving unit to bring the probe needles into contact with the semiconductor chips, the prober controlling device comprising:
an input data acquiring unit configured to acquire input data including temperature data of at least one of the probe card and the card holder;
a predicting unit configured to predict a position of a tip end of a probe needle based on the input data acquired by the input data acquiring unit using a prediction model that receives input of the input data and outputs the position of the tip end of the probe needle; and
a determining unit configured to determine whether or not to execute prediction by the predicting unit, based on input data used as teacher data for machine learning of the prediction model and the input data acquired by the input data acquiring unit, before the prediction by the predicting unit.

### Claim 2 (depends on 1)

The prober controlling device according to claim 1, wherein
the determining unit performs:
processing of calculating, for respective parameters of the input data, differences between the input data acquired by the input data acquiring unit and the input data used as the teacher data; and
processing of calculating root sum squares of the differences for respective parameters, and determining whether or not to execute the prediction by the predicting unit based on whether or not at least one of the root sum squares of the differences for respective parameters is in a predetermined certain range.

### Claim 3 (depends on 1)

The prober controlling device according to claim 1, comprising:
a needle position acquiring unit configured to acquire the position of the tip end of the probe needle in a case in which the determining unit determines not to execute the prediction; and
a relearning unit configured to perform relearning of the prediction model using the teacher data to which the input data acquired by the input data acquiring unit and the position of the tip end of the probe needle acquired by the needle position acquiring unit are added, wherein
the needle position acquiring unit, the relearning unit, the input data acquiring unit and the determining unit are repeatedly operated until the determining unit determines to execute the prediction.

### Claim 4 (depends on 3)

The prober controlling device according to claim 3, wherein
the relearning unit excludes, from the teacher data, an oldest input data and the position of the tip end of the probe needle corresponding to the oldest input data, and then executes relearning of the prediction model based on the teacher data.

### Claim 5 (depends on 1)

The prober controlling device according to claim 1, wherein
in a case in which the determining unit determines to execute the prediction, the predicting unit predicts the position of the tip end of the probe needle, and
the prober controlling device further comprises a movement controlling unit configured to control the relative moving unit based on the position of the tip end of the probe needle predicted by the predicting unit, so as to bring the probe needle into contact with a semiconductor chip.

### Claim 6 (depends on 1)

The prober controlling device according to claim 1, wherein
the input data acquiring unit acquires, in addition to the temperature data as the input data, alignment data including at least one of: a chip size of a semiconductor chip, a position of the wafer, and a positional relation between a first camera used to detect the semiconductor chip and a second camera used to detect the probe needle.

### Claim 7 (depends on 1)

A prober comprising:
a wafer chuck configured to hold a wafer on which semiconductor chips are formed;
a probe card having probe needles;
a card holder configured to hold an outer periphery of the probe card and place the probe card opposite the wafer;
a relative moving unit configured to move the wafer chuck relatively to the probe needles; and
the prober controlling device according to claim 1.

### Claim 8

A prober controlling method for a prober including a wafer chuck configured to hold a wafer on which semiconductor chips are formed, a probe card having probe needles, a card holder configured to hold an outer periphery of the probe card and place the probe card opposite the wafer and a relative moving unit configured to move the wafer chuck relatively to the probe needles, the prober controlling method of driving the relative moving unit to bring the probe needles into contact with the semiconductor chips, the prober controlling method comprising:
an input data acquiring step of acquiring input data including temperature data of at least one of the probe card and the card holder;
a predicting step of predicting a position of a tip end of a probe needle based on the input data acquired in the input data acquiring step using a prediction model that receives input of the input data and outputs the position of the tip end of the probe needle; and
a determining step of determining whether or not to execute the predicting step, based on input data used as teacher data for machine learning of the prediction model and the input data acquired in the input data acquiring step, before the predicting step.