JP2005347944A - Management method for reference vibrator - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a management method for a reference vibrator by which the fluctuations of the reference vibrator connected to a product as a test target are accurately determined. <P>SOLUTION: The measurement of a plurality of reference oscillation circuits 32 is executed by using the reference vibrator 31. The obtained measurement value is recorded in a storage medium 44 of a tester 4 and a plurality of the reference oscillation circuits 32 are stored. The testing of a plurality of continuously produced oscillation circuits is executed by using the reference vibrator 31. After that, the re-measurement of a plurality of the reference oscillation circuits 32 is executed by using the reference vibrator 31. The obtained measurement value and the measurement value recorded in the storage medium 44 are compared with each other. When the characteristic fluctuation of the reference vibrator 31 is detected, the reference vibrator 31 or the testing condition is adjusted so as to cancel the characteristic fluctuation. The testing of a plurality of the continuously produced oscillation circuits is continued after executing the adjustment for canceling the fluctuation. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a method for managing a reference vibrator used for testing a plurality of continuously produced oscillation circuits.

  Crystal oscillator circuit products such as TCXO (temperature compensated crystal oscillator circuit) and VCXO (voltage controlled crystal oscillator circuit) are formed on a semiconductor substrate (wafer), probe test in wafer state, and individual chips Various characteristics are measured by the final test after the division, and those whose measured values are within the required specification range are shipped as non-defective products. A crystal oscillator circuit having a predetermined specification is connected to a crystal oscillator circuit shipped as a non-defective product, and an oscillation signal having a predetermined oscillation frequency is output from the crystal oscillator circuit.

Here, in the probe test, the probe needle is brought into contact with the pad provided on the wafer, and the electrical performance of the crystal oscillation circuit is inspected through the probe needle. If it is bad, the crystal oscillation circuit that should normally operate normally does not operate and is treated as a defective product. Therefore, check whether the operation is possible under predetermined conditions that are difficult to operate compared to the standard operating conditions, and if the operation cannot be confirmed, take measures to improve the contact state and then test the crystal oscillation circuit (this test) Thus, there has been proposed a probe test method capable of accurately selecting non-defective products and defective products (see, for example, Patent Document 1).
Japanese Patent Laid-Open No. 2003-258045

  In the above-described probe test and final test, a reference crystal resonator is connected to the product to be tested, and characteristics including both the product and the reference crystal resonator are measured. Therefore, it is essential that the characteristics of the reference crystal unit be constant in order to perform a reliable test.

  However, when a large number of products are continuously manufactured and tested for a long period of time, the characteristics of the reference crystal unit used for the test may fluctuate, and the measurement result may fluctuate accordingly. This variation progresses gradually and may not be detected at an early stage. In addition, the product characteristics may change due to fluctuations in the state of the manufacturing equipment, so even if fluctuations are found in the measured values, the characteristics fluctuation of the reference crystal unit It is difficult to determine the cause. As a result, it cannot be determined that the characteristic variation of the reference crystal unit has occurred, and the test is continued, resulting in the shipment of a product that does not meet the specifications, or the test of the product is delayed due to the time taken for the determination. Sometimes.

  In addition, if it becomes clear that the characteristics of the reference crystal unit have changed or the reference crystal unit has been damaged, and it is necessary to replace it with another reference crystal unit, the characteristics of the first reference crystal unit It is difficult to obtain a reference crystal resonator having the same characteristics as (the characteristics before fluctuation). That is, even if a reference crystal unit having the same specifications as the first reference crystal unit is purchased, the actual characteristics are different within the range of errors allowed by the specifications. For this reason, the consistency of the test results is lost before and after the replacement of the reference crystal unit. In recent years, specifications for oscillator circuit products have become stricter, and such inconsistencies in test results due to differences in characteristics of the reference crystal resonator cannot be allowed.

  Here, if the difference in characteristics between the first reference crystal unit and the replaced reference crystal unit can be grasped quantitatively, the test conditions can be adjusted based on that and the consistency of the test results can be maintained. Is possible. However, since it is not clear what the difference in actual characteristics is between the reference crystal resonators of the same specification, such adjustment cannot be performed.

  In addition, due to an increase in production volume, it may be necessary to introduce a plurality of test facilities in the same facility and perform tests in parallel or outsource to an external tester. For testing at these multiple locations, a reference crystal resonator is required. For this reason, even if a reference crystal unit with the same specifications is purchased, there is a problem in that there is a difference in the actual characteristics, and there is no consistency between the test results at different test facilities. To do.

  In view of the above circumstances, an object of the present invention is to provide a reference vibrator management method capable of easily detecting characteristic fluctuations of a reference vibrator and ensuring consistency of test results. More specifically, an object of the present invention is to provide a reference vibrator management method capable of accurately determining a fluctuation of a reference vibrator connected to a product to be tested. It is another object of the present invention to provide a reference oscillator management method capable of maintaining consistency between test results before and after replacement when a reference oscillator needs to be replaced. It is another object of the present invention to provide a reference oscillator management method capable of maintaining consistency between test results at different locations when testing at a plurality of locations.

Of the reference oscillator management methods of the present invention that achieves the above object, the first reference oscillator management method is a reference oscillator management method used for testing a plurality of continuously produced oscillation circuits. There,
Measure a plurality of reference oscillation circuits using the first reference oscillator, record the obtained measurement values, and store the plurality of reference oscillation circuits.
Testing the plurality of continuously produced oscillator circuits using the first reference oscillator,
Thereafter, the plurality of reference oscillation circuits are re-measured using the first reference oscillator, the obtained measurement values are compared with the recorded measurement values, and the first reference oscillator is compared. If a characteristic variation is observed, adjust the first reference vibrator or the test condition so as to cancel the characteristic variation,
After the adjustment for offsetting the fluctuation, the test of the plurality of oscillation circuits continuously produced is further continued.

  The first reference resonator management method according to the present invention includes a plurality of reference oscillation circuits (for example, an oscillation circuit manufactured with the same specifications as the product to be tested) before the reference resonator is used for product testing. Is measured, the measured value is recorded, and the plurality of reference oscillation circuits are stored. Then, while the reference vibrator is used for product testing, the stored reference oscillation circuit is remeasured. By comparing the remeasured measurement value with the recorded measurement value, it is determined whether or not the reference vibrator (or the reference oscillation circuit) has a characteristic variation.

  Here, if it is assumed that only one of the reference oscillator and the plurality of reference oscillation circuits is changed by storing two or more reference oscillation circuits and performing remeasurement, Even if a change occurs, it can be determined which has changed. In addition, it is assumed that the reference oscillator and one of the plurality of reference oscillation circuits may fluctuate at the same time by storing three or more reference oscillation circuits and performing re-measurement. Can be determined.

If it is determined that the reference vibrator has changed, for example, one of the following adjustments is performed.
(1_1) Adjust the reference vibrator so that fluctuations in the characteristics of the reference vibrator are offset. Specifically, the variable reactance element (variable capacitance element) connected to the reference vibrator is adjusted so that the resonance frequency can be adjusted, and the resonance frequency of the reference vibrator and the variable reactance element is the value before the fluctuation. Adjust so that
(1_2) The amount of variation is grasped by comparing the measured value obtained by remeasurement with the recorded measured value, and the test conditions are adjusted so as to cancel the variation. For example, an offset value necessary for offsetting is calculated and set for a measured value or a specification value.

The second reference oscillator management method of the reference oscillator management method of the present invention that achieves the above object is the management of the reference oscillator used for testing a plurality of continuously produced oscillation circuits. A method,
Measure a plurality of reference oscillation circuits using the first reference oscillator, record the obtained measurement values, and store the plurality of reference oscillation circuits.
Testing the plurality of continuously produced oscillator circuits using the first reference oscillator,
Thereafter, the plurality of reference oscillation circuits are re-measured using the first reference oscillator, the obtained measurement values are compared with the recorded measurement values, and the first reference oscillator is compared. If the need for replacement is found,
The first reference oscillator is replaced with a new first reference oscillator, the plurality of reference oscillation circuits are measured using the replaced first reference oscillator, and the obtained measurement values are obtained. Comparing the characteristic of the replaced first reference vibrator with the recorded measurement value, and the difference between the characteristic of the first reference vibrator before the change before the change,
Adjust the exchanged first reference vibrator or the test conditions so as to offset the grasped difference,
After the adjustment that cancels out the difference, the test of the plurality of oscillation circuits that are continuously produced is further continued.

  In the second reference vibrator management method of the present invention, there are fluctuations beyond the adjustable range in the above (1_1) and (1_2), or for other reasons (for example, breakage of the reference vibrator). When it is determined that the reference vibrator needs to be replaced, the reference vibrator is replaced.

In this case, before using the replaced reference oscillator for product testing, measure the stored reference oscillation circuit and compare the measured value with the recorded measurement value. From this, the difference between the characteristic of the reference vibrator before replacement (before characteristic change) and the characteristic of the replaced reference vibrator is grasped, and for example, one of the following adjustments is performed.
(2_1) The reference vibrator is adjusted so as to cancel out the difference.
(2_2) Adjust test conditions to offset the difference.

  On the other hand, when it is determined that any of the reference oscillation circuits has a characteristic variation, the reference oscillation circuit is replaced with a new reference oscillation circuit. The measured value of the replaced reference oscillation circuit is recorded and stored together with other reference oscillation circuits and used for subsequent management.

Here, when at least one characteristic variation of the plurality of reference oscillation circuits is recognized by comparing the measurement value obtained by the remeasurement with the recorded measurement value,
Replace the reference oscillation circuit in which the above characteristic variation is recognized with a new reference oscillation circuit,
The new reference oscillation circuit is measured using the first reference oscillator, the obtained measurement value is replaced with the measurement value of the reference oscillation circuit in which the characteristic variation is recognized, and recorded.
The new reference oscillation circuit may be added to the plurality of reference oscillation circuits excluding the reference oscillation circuit in which the characteristic variation is recognized, and stored as a plurality of new reference oscillation circuits.

Here, the test of the oscillation circuit is performed by comparing a measured value obtained by measuring the characteristics of the oscillation circuit with a specification value.
The adjustment may be performed by setting an offset value for the measurement value or the specification value.

Further, the plurality of reference oscillation circuits are stored together with the first reference oscillator on the test board for performing a test for testing the oscillation circuit and a connection between the oscillation circuits. Is done by wearing
The re-measurement of the plurality of reference oscillation circuits may be performed while mounted on the test board.

  The adjustment is performed by supplying a control voltage to a variable capacitance element connected to the reference resonator from a tester for testing the oscillation circuit, and changing the capacitance value of the variable capacitance element, thereby changing the reference vibration. It may be performed by adjusting the resonance frequency of the child.

  Usually, the reference vibrator is mounted on a test board having a probe needle and a socket for connecting to a product to be tested, and is connected to the product via the probe needle and the socket during the test. By storing the reference oscillation circuit on the test board and connecting it to the reference resonator with a switch when necessary, it can be re-measured, so that the test equipment (usually for the product type to be tested) By replacing the test board together, it is possible to re-measure the reference oscillation circuit with the test board still attached to the test board, which is commonly used for testing multiple types of products. The time during which testing cannot be performed can be shortened. Further, the cost associated with storing the reference oscillation circuit can be minimized.

  Furthermore, a variable capacitance element is mounted as a variable reactance element on the same test board, and the capacitance value is changed by a control voltage from the tester, thereby adjusting the test apparatus with the test board attached. Furthermore, the generation of time during which product testing cannot be performed can be reduced.

  Further, the plurality of reference oscillation circuits may be selected from the plurality of oscillation circuits that are continuously produced.

The third reference oscillator management method of the reference oscillator management method of the present invention that achieves the above object is to test a plurality of oscillation circuits that are continuously produced at a plurality of test points. A management method of a reference resonator to be used,
When testing the plurality of oscillation circuits using the first reference oscillator at the first test location, a test board equipped with the first reference oscillator and the plurality of first reference oscillation circuits is tested. Attached to the
Using the switch mounted on the test board, the reference oscillator is connected to each of the plurality of first reference oscillation circuits, the measurement result is recorded,
While the test board is not attached to the test apparatus, a part of the plurality of first reference oscillation circuits is removed from the test board and prepared as a second reference oscillation circuit,
A second reference vibrator for testing the plurality of oscillation circuits at a second test location is connected to the second reference oscillation circuit and measured, and the obtained measurement value and the recorded measurement are measured. Comparing the values, and grasping the difference in characteristics between the first reference vibrator and the second reference vibrator,
The test condition at the second reference vibrator or the second test point is adjusted so as to cancel the grasped difference.

  Here, in the third reference oscillator management method of the present invention, a part of the plurality of first reference oscillation circuits is removed from the test board, and the same number of new reference oscillation circuits as the removed first reference oscillation circuits are provided. A new reference oscillation circuit is mounted on the test board as a part of the plurality of first reference oscillation circuits, the first reference oscillator is connected to the newly mounted reference oscillation circuit, and the new installation is performed. It is preferable to prepare a predetermined number of the second reference oscillation circuits by repeating the operation of measuring the measured reference oscillation circuit and recording the obtained measurement result a plurality of times.

  Further, when removing some of the plurality of first reference oscillation circuits from the test board, at least two of the first reference oscillation circuits may be left.

Further, the second reference oscillation circuit includes a plurality of second reference oscillation circuits,
The measurement value obtained by measuring the second reference oscillation circuit by continuing the second reference oscillator, or after adjusting to cancel the difference, the second reference oscillator is connected again and the second reference oscillator is connected. Recording the measured value of the reference oscillation circuit of the second, and holding the second reference oscillation circuit,
After the test of the plurality of oscillation circuits at the second test location, the second reference oscillator is connected to perform remeasurement of the plurality of second reference oscillation circuits, and the measured values obtained Is compared with the recorded measurement value of the second reference oscillation circuit, and when the characteristic fluctuation of the second reference vibrator is recognized, the second reference oscillator circuit is canceled so as to cancel the characteristic fluctuation. Adjust the test conditions at the reference vibrator or the second test point,
After the adjustment that cancels the fluctuation, the test of the plurality of oscillation circuits at the second test point may be continued.

The third reference vibrator management method of the present invention is a management method for testing at a plurality of test points, and is provided at another test point (other test equipment in the same factory or an external test service company When a new test is performed at a test facility or the like, first, a reference oscillation circuit having a measurement value recorded using a first reference vibrator used at an existing test point is used as a second reference oscillation. Prepare as a circuit. Then, before using the second reference oscillator used for the test at the new test location for the product test, the second reference oscillator is measured using the second reference oscillator, By comparing this measured value with the recorded measured value, the difference in characteristics between the second reference vibrator and the first reference vibrator used at the existing test location is grasped. For example, one of the following adjustments is performed.
(3_1) The reference vibrator is adjusted so as to cancel the difference in characteristics.
(3_2) Adjust test conditions so as to offset the difference in characteristics.

  Here, the first reference oscillator and the reference oscillation circuit (first reference oscillation circuit) are mounted on the test board used for the test at the existing test location, and the reference oscillator is connected to the reference oscillation circuit by the switch. It is assumed that the reference oscillation circuit can be measured by a tester that tests the product. Then, the first reference oscillation circuit mounted on the test board and measured using the first reference oscillator is removed from the test board while the test board is not used for testing the product. The second reference oscillation circuit is prepared. As a result, the second reference oscillation circuit can be prepared only by preparing a test board and a tester for testing a product at an existing test location, without preparing a dedicated test board or a measuring instrument. . In addition, the test board can be kept in a state where it can be used for testing the product at the first test location while the first reference oscillation circuit is measured to prepare as the second reference oscillation circuit. That is, it is not necessary to remove the reference vibrator or attach it to another test apparatus. For this reason, the influence with respect to the test in the existing test location can be suppressed to the minimum.

  As described above, in existing test locations, a plurality of first reference oscillation circuits are mounted on a test board and stored, and used for management of the first reference oscillator. A part of the first reference oscillation circuit can be removed and prepared as a second reference oscillation circuit. If the second reference oscillation circuit prepared in this way is used only for grasping the difference in characteristics between the second reference vibrator and the first reference vibrator, the stage where the difference grasp is completed. Thus, it is possible to return to the existing test location and attach it to the test board again and continue to use it as the first reference oscillation circuit.

  However, when the second reference oscillation circuit is continuously used for the management of the second reference resonator in the new test location, the second reference oscillation circuit is removed from the test board used in the existing test location. Therefore, when a part of the first reference oscillation circuit is removed, it is preferable to add the same number of new reference oscillation circuits as a part of the first reference oscillation circuit and attach them to the test board. Then, when the test board is next attached to the test apparatus, the measurement is performed by connecting to the reference oscillation circuit to which the first reference vibrator is added, and the measurement result is recorded to manage the first reference vibrator. It is preferable to be ready for use.

  Further, only for the purpose of grasping the difference in characteristics from the first reference vibrator, it is sufficient to prepare only one second reference oscillation circuit. However, in order to continuously manage the characteristics of the second reference resonator, a plurality of, for example, the same number of first reference oscillation circuits stored for the management of the first reference resonator. It is preferable to prepare and store two reference oscillation circuits. For this purpose, a part of the first reference oscillation circuit is removed from the test board to prepare as a second reference oscillation circuit, and the same number of new reference oscillation circuits are added and mounted on the test board. It is preferable to repeat the operations of connecting and measuring the reference oscillators and recording the results until the required number of second reference oscillation circuits are prepared.

  According to the reference oscillator management method of the present invention, it is possible to accurately determine the presence or absence of characteristic fluctuation of the reference oscillator by storing a plurality of reference oscillation circuits and performing remeasurement. Then, by grasping the amount of characteristic variation and making an appropriate adjustment, it is possible to maintain the consistency of test results even when the characteristic variation of the reference vibrator occurs. Furthermore, even when the reference vibrator needs to be replaced, the consistency of the test results can be maintained. In addition, by mounting and storing multiple reference oscillation circuits on the test board and enabling connection to the reference oscillator with a switch, re-measurement can be performed with the test board still attached to the test equipment. Thus, the time during which product testing cannot be performed can be shortened.

  In addition, a reference oscillation circuit is mounted on a test board for testing a product at an existing test location, this test board is attached to a test apparatus for product testing, and measured by a tester. By removing from the test board while it is not being used for testing, only the test board and tester for testing the product are used, and the characteristics with the first reference vibrator used at the existing test location A second reference oscillation circuit for grasping the difference can be prepared. In addition, the influence on the test schedule at the existing test points can be minimized.

  Furthermore, a part of the first reference oscillation circuit is removed from the test board to prepare as a second reference oscillation circuit, and the same number of new reference oscillation circuits are added to the test board and attached to the test apparatus. In order to continuously manage the fluctuation of the second reference vibrator after the adjustment by repeatedly performing the operation of connecting the first reference vibrator and measuring and recording the result. A necessary number of second reference oscillation circuits can be prepared.

  Hereinafter, embodiments of the present invention will be described.

  FIG. 1 is a diagram showing a configuration of a test system to which an embodiment of a reference vibrator management method of the present invention is applied.

  A test apparatus 100 for a probe test which is a test system shown in FIG. 1 includes a prober 2 and a tester 4. The prober 2 is provided with a stage 21 that can be raised and lowered. On this stage 21, a wafer 1 on which a plurality of semiconductor chips (product chips) 10 are formed is placed. The prober 2 is provided with a test board 3 (probe test test board) for connecting to the product chip 10 formed on the wafer 1 and for connecting to the tester 4. The test board 3 includes a reference crystal resonator 31 (corresponding to an example of the reference resonator according to the present invention), a plurality of reference oscillation circuits 32, a variable capacitance diode 33, and a switch 35 for setting connection of the reference crystal resonator 31. (See FIG. 2). Further, the test board 3 is provided with a probe needle 34 for performing connection (probing) to the product chip 10 in the wafer state via the pads P1 and P2. The probe card having the probe needle 34 may be divided into a board for connection with the tester 4 (this may be referred to as a “test board”), but here both are included. This is called a “test board”.

  On the other hand, the tester 4 controls the prober 2 and tests the product by program control. The tester 4 measures a current and oscillation frequency, etc., a CPU 41 that is a control processor, a power supply / signal unit 42 that generates a power supply voltage and various signals for operating the product chip 10 to be tested. It has a measuring instrument 43 and a storage medium 44. The storage medium 44 records the test results of the product chip 10 and the reference crystal resonator 31 such as the measurement values obtained by the measurement of the plurality of reference oscillation circuits 32 performed prior to the test of the product chip 10. Information necessary for management is also recorded.

  FIG. 2 is a diagram showing circuit elements in the test board shown in FIG.

  In the test board 3 shown in FIG. 2, the reference crystal resonator 31 is tested by the command from the reference crystal resonator 31, the plurality of reference oscillation circuits 32, the variable capacitance diode 33, and the tester 4. A switch 35 that can be switched between a state connected to the probe needle 34 and a state connected to each of the reference oscillation circuits 32 is mounted. The variable capacitance diode 33 is for changing the resonance frequency of the reference crystal unit 31 by changing its capacitance in accordance with the control voltage Vc from the tester 4, and via the DC cut capacitors 36 and 37. Thus, the reference crystal unit 31 is connected in parallel.

  FIG. 3 is a diagram showing a test board different from the test board shown in FIG.

  A test board 3_1 shown in FIG. 3 is a test board for final test. Compared to the test board 3_1 shown in FIG. 1, in place of the probe needle 34, a socket 38 is mounted to connect each chip 10 formed on the wafer 1 to a semiconductor product housed in a semiconductor package. Is different. In the test apparatus for final test, the test board 3_1 having the socket 38 is used instead of the probe needle 34 in this way, and the prober 2 attaches / detaches the product to be tested to the socket 38 of the test board 3_1. It replaces the test station with the robot to perform, but the basic configuration is common.

  In any case, a mechanism (probe needle 34, socket 38) for connecting to the product to be tested is provided, and the reference crystal unit 31, the plurality of reference oscillation circuits 32, and the reference crystal unit 31 are connected. A test board having a switch 35 for setting the connection is used.

  The plurality of reference oscillation circuits 32 are stored in the test apparatus 100 for testing the product. For example, a reference crystal unit 31, a reference oscillation circuit 32, and a reference crystal according to a command from the tester 4 are mounted on a test board 3 on which a probe needle 34 for electrically connecting the tester 4 and a product to be tested is attached. A switch 35 that can be switched between a state in which the vibrator 31 is connected to the product to be tested and a state in which the vibrator 31 is connected to the reference oscillation circuit 32 is mounted.

  Then, under the control of the program from the tester 4, each reference oscillation circuit 32 is connected to the reference crystal resonator 31 at a timing when the product test is not performed, and remeasurement is performed. Necessary adjustments are made based on the results. Alternatively, an alarm is issued to request replacement of the reference crystal resonator 31 or the reference oscillation circuit 32.

  Further, a variable capacitance element 33 whose capacitance value is changed by a control voltage supplied from the tester 4 is mounted on the test board 3 on which the reference crystal resonator 31 is mounted, and is used for adjusting the reference crystal resonator 31.

  As described above, in the present embodiment, by storing a plurality of reference oscillation circuits 32 on the test board 3, remeasurement of the reference oscillation circuit 32 can be performed by using a little free time during product testing. Do. In this way, the reference crystal unit 31 can be managed without disturbing the product test schedule. Further, the program control makes it possible to manage the reference crystal unit 31 while minimizing the burden on the operator.

  FIG. 4 is a diagram illustrating a state in which the reference crystal resonator and the adjustment variable capacitance element are connected to an oscillation circuit that is a product.

  FIG. 4 shows a reference crystal resonator 31, a variable capacitance diode 33, and DC cut capacitors 36 and 37 mounted on the test board 3. FIG. 4 shows an oscillation circuit of the product chip 10 formed on the wafer 1. This product chip 10 is a product chip to be tested that undergoes a probe test.

  The product chip 10 includes a variable capacitance diode 11 and a CMOS inverter 12 that constitute an oscillation circuit, and a resistor and a capacitor as illustrated. A pad P1 is connected to the input side of the CMOS inverter 12, and a pad P2 is connected to the output side of the CMOS inverter 12 via a resistor Rd. A capacitor Cd and a capacitor C0 are connected in parallel between the pad P2 and the ground GND. A capacitor Cp and a capacitor Cg are connected in series between the pad P1 and the ground GND. A variable capacitance diode 11 is connected between a connection point between the capacitor Cp and the capacitor Cg and the ground GND, and a resistor R 1 is connected between the connection point and the input terminal 13. Further, a resistor Rf is connected to both ends of the CMOS inverter 12.

  When the probe needle 34 is in contact with the pads P1 and P2, an oscillation circuit that outputs a signal having an oscillation frequency determined by the resonance frequency of the reference crystal resonator 31 and the characteristics of the product chip 10 is configured.

A predetermined control voltage Vin is applied to the input terminal 13 of the product chip 10. The applied control voltage Vin is applied as a reverse bias voltage to the variable capacitance diode 11 via the resistor R1. The capacitance of the variable capacitance diode 11 changes depending on the magnitude of the applied control voltage Vin. As the capacitance of the variable capacitance diode 11 changes, the frequency of the oscillation signal output from the output terminal _Vout changes.

  A control voltage Vc is applied from the tester 4 to both ends of the variable capacitance diode 33. The capacitance of the variable capacitance diode 33 changes according to the applied control voltage Vc. Thereby, the resonance frequency of the reference crystal unit 31 combined with the variable capacitance diode 33 can be changed.

  When the product chip 10 is tested, the product 10 to be tested is probed (attached to the socket in the case of a final test) by the control signal from the tester 4 and the switch 35 on the test board 3 is connected. Is set, and the reference crystal unit 31 is connected to the product chip 10 to be tested. Then, the power supply voltage and various input signals (in the example of the illustrated oscillation circuit, the control voltage Vin for frequency adjustment) are supplied from the tester 4 to the product chip 10, and the output signal Vout from the product chip 10 is supplied to the tester 4. Supplied. Various characteristics of the product chip 10 are measured based on the output signal Vout. Then, the measured value and the specification value are compared, and if the specification value is satisfied, it is determined as a non-defective product, and if not, it is determined as a defective product.

  On the other hand, when the measurement of the reference oscillation circuit 32 is performed, the switch 35 on the test board 4 is set so as to connect each reference oscillation circuit 32 to the reference crystal resonator 31 which is a standard resonator in order. . Then, a power supply voltage and various input signals are supplied from the tester 4 to the reference oscillation circuit 32 connected to the reference crystal resonator 31 to perform measurement.

  The tester 4 for testing the product chip 10 can be used as it is by mounting the reference oscillation circuit 32 on the test board 3 and connecting to the reference crystal resonator 31 by the switch 35 for measurement. That is, it is not necessary to provide a measurement facility different from the tester 4 for testing the product chip 10 or to modify the product tester. In addition, by placing the test board 4 on the same side as the reference crystal unit 31, it becomes easy to control the impedance of the wiring between the reference crystal unit 31 and the reference oscillation circuit and to enable accurate measurement. .

  The tester 4 compares the measured value obtained by re-measurement of the reference oscillation circuit 32 with the stored measured value, and determines the presence or absence of characteristic variation of the reference crystal resonator 31. If it is determined that there has been a change, an adjustment is made to offset the change.

  This adjustment is performed by adjusting the control voltage Vc supplied from the tester 4 to change the capacitance of the variable capacitance diode 33 and changing the resonance frequency of the reference crystal resonator 31 combined with the variable capacitance diode 33. The value of the control voltage Vc necessary for the adjustment is recorded in the storage medium 44 of the tester 4 and used in subsequent product tests or remeasurement of the reference oscillation circuit.

  It is also possible to calculate an offset value for canceling the characteristic fluctuation of the reference crystal unit 31 by using the conversion formula recorded in the storage medium 44 of the tester 4. For example, it may be an offset value for a measured value or an offset value for a specification value. The calculated offset value is also recorded in the storage medium 44 of the tester 4 and used in subsequent product testing or re-measurement of the reference oscillation circuit.

  Further, when it is determined that the characteristic fluctuation exceeding the adjustable range has occurred in the reference crystal unit 31 and when it is determined that the characteristic fluctuation of the reference oscillation circuit 32 has occurred, an alarm requesting replacement is issued.

  Next, management of the reference crystal unit 31 at a single test location will be described with reference to FIG. Here, the reference crystal resonator 31 is simply referred to as a reference resonator.

  FIG. 5 is a flowchart of the steps executed to manage the reference transducer at a single test location.

  First, in step S1, a plurality of first reference oscillation circuits using the first reference oscillator (the plurality of reference oscillation circuits are, for example, the same manufacturing equipment as the product to be tested and the same specifications) Are selected from the oscillation circuit manufactured in (1), and the measurement values (for example, the oscillation frequency under a specific measurement condition or a control voltage necessary to obtain a specific oscillation frequency) are recorded. At the same time, store the reference oscillation circuit.

  In the measurement at step S1 and the test at the next step S2, the control voltage Vc is set to a predetermined initial value, for example, a center value within a variable range.

  Next, in step S2, a product test is performed using the first reference vibrator. Tables 1, 2 and 3 show numerical examples (measurement examples in the initial stage using the first reference vibrator).

  Table 1 shows the specification value (27.000 MHz) of the first reference vibrator, the measurement conditions (power supply voltage VDD = 3.3 V, control voltage Vc = 1.65 V), and the specification value (27. 000000MHz to 27.000600MHz).

  Table 2 shows the oscillation frequencies measured (recorded) of the four first reference oscillation circuits 1, 2, 3, and 4 using the first reference oscillator.

  Further, Table 3 shows the measured oscillation frequency and the result of OK / NG determination (good / bad determination) of four mass-produced products (product chips) A, B, C, and D.

  Further, in step S3, the first reference vibrator is used at an appropriate time (when a predetermined period has elapsed, when a predetermined number of products have been tested, product characteristics have changed, etc.). Then, the first plurality of reference oscillation circuits are measured again, and the process proceeds to step S4.

  In step S4, the measured value obtained by re-measurement is compared with the recorded measured value, and the presence / absence of characteristic variation of the first plurality of reference vibrators is determined. For example, in the case where all or a majority of the measured values of the first reference oscillation circuit have the same tendency (for example, increase in oscillation frequency), it is determined that the characteristics of the first reference oscillator have changed. To do. In this case, the first reference vibrator or test condition adjustment 1 is performed in step S5 described later. On the other hand, if it is determined that the characteristics of the first reference vibrator have not changed, the process proceeds to step S6.

  In step S6, it is determined whether or not there is a change in characteristics of the first reference oscillation circuit. For example, when there is a variation in the measured value of one or only part of the first reference oscillation circuit and no variation is observed in the measurement value of the other first reference oscillation circuit, the variation It is determined that the characteristics of the observed first reference oscillation circuit have changed. In this case, in step S7, the first reference oscillation circuit whose characteristics have changed is replaced, and the characteristics of the replaced first reference oscillation circuit are measured and recorded for subsequent management. Then, a new first plurality of reference oscillation circuits are stored together with the first reference oscillation circuit in which no fluctuation is observed, and stored. Table 4 shows numerical examples (including examples in which the first reference oscillation circuit fluctuates).

  Table 4 shows pre-recorded oscillation frequencies of the first reference vibrators 1, 2 (variation), 3, 4. Table 4 shows measured oscillation frequencies of the first reference vibrators 1, 2 (variation), 3,4, 2 ′ (replacement with 2). As can be seen from Table 4, the numerical value of the minimum (1 Hz) digit includes measurement variations. The oscillation frequencies measured for the reference vibrators 1, 3, and 4 coincide with the recorded oscillation frequencies within the range of the measurement variation. However, the oscillation frequency measured for the two reference vibrators shows a fluctuation far exceeding this measurement variation from the recorded oscillation frequency. Accordingly, it was determined that the characteristics of the two reference vibrators had changed, and the reference frequency was replaced with 2 ', and the oscillation frequency was measured and recorded. Thereafter, the process returns to step S2 and the product test using the first reference vibrator is continued.

  The first reference oscillation circuit is stored in a state in which it is mounted on the test board 3, and the re-measurement is performed by the switch 35 provided on the test board 3. In order to test the product, the first reference oscillation circuit can be used with the test board 3 still attached to the test apparatus in order to test the product, using a little free time between the tests of the product. Can be measured. If it is determined that neither the first reference oscillator nor the first reference oscillation circuit has a characteristic variation, the product can be tested again immediately. As described above, the first reference oscillation circuit is stored in the state where it is mounted on the test board 3, so that the first reference oscillator and the first reference oscillation circuit have the first variation for the purpose of determining the presence or absence of characteristic variation. In order to perform re-measurement of the reference oscillation circuit, the time during which the product cannot be tested can be minimized, and the influence on the test schedule of the product can be minimized. Further, by storing the first reference oscillation circuit while being mounted on the test board 3, a measurement value due to a change in the connection state between the first reference oscillation circuit and the first reference oscillator is obtained. Can be prevented from occurring.

  However, it is not essential to store the first reference oscillation circuit with the test board 3 mounted. For example, even when measurement and remeasurement are performed with the first reference oscillation circuit mounted on the test board 3, it may be removed after measurement and stored in a location different from the test board 3. Is possible. However, in order to secure the storage location of the removed first reference oscillation circuit and reduce the cost associated with management, it is preferable to store the first reference oscillation circuit while it is mounted on the test board 3.

  If there is a variation in the measured value, the number of first reference oscillation circuits to be stored is 3 or more in order to determine which of the first reference oscillator and the first reference oscillation circuit has changed. It is preferable that Furthermore, if there are four or more, as will be described later, even if one first reference oscillation circuit is removed for use in grasping the characteristics of the first reference vibrator used in another test location. The management of the first reference vibrator by the remaining three first reference oscillation circuits can be continued.

  FIG. 6 is a flowchart of the adjustment 1 process in step S5 shown in FIG.

  First, in step S51, it is determined whether or not the first reference vibrator needs to be replaced. For example, first, the characteristic fluctuation amount is grasped from the comparison between the measured value obtained by re-measurement of the first reference oscillation circuit and the recorded measured value, and it is determined whether or not it is within the adjustable range. .

  Here, Table 5 shows an example of recorded measurement values, re-measurement values, and their differences of the first reference oscillation circuit.

  Table 5 shows the recorded oscillation frequencies of the first reference oscillation circuits 1, 2, 3, and 4. In addition, remeasured values (oscillation frequencies) of the first reference oscillation circuits 1, 2, 3, and 4 are shown. Furthermore, the difference between the recorded oscillation frequency and the remeasured value (oscillation frequency) of the first reference oscillation circuit 1, 2, 3, 4 and the average value (0.000900) of this difference are shown. .

If this difference is within the adjustable range and it is determined that the replacement of the first reference vibrator is unnecessary, at least one of the following (1) and (2) is performed.
(1) The first reference vibrator is adjusted so as to cancel the characteristic variation (step S54).

  For example, by adjusting the control voltage Vc supplied from the tester 4 to the variable capacitance diode 33, the overall resonance frequency in which the first reference vibrator and the variable capacitance diode 33 are combined is the same as before the change. Adjust as follows. Table 6 shows numerical examples after adjustment.

  Table 6 shows the measured values (oscillation frequencies) after adjustment of the first reference oscillation circuits 1, 2, 3, and 4. As shown in Table 6, even when the characteristic of the first reference vibrator changes, the characteristic fluctuation of the first reference vibrator is canceled out by appropriately adjusting the control voltage Vc. Can be adjusted. Using the first reference vibrator adjusted in this way, it is possible to continue testing the product while maintaining the consistency of the test results.

Note that the first reference oscillation circuit may be measured while changing the control voltage Vc to obtain a control voltage value that matches the recorded measurement value. Further, a control voltage value necessary for canceling the grasped characteristic variation may be calculated using a predetermined conversion formula. In any case, the obtained control voltage value is recorded in the storage medium 44 of the tester 4 and used in the subsequent product test and the measurement of the first reference oscillation circuit.
(2) The test conditions are adjusted so as to cancel the characteristic variation of the first reference vibrator (step S55).

  For example, an offset value for the measured value is set. Most simply, for example, for an oscillation frequency under a predetermined condition, an offset value having the same absolute value as that of the fluctuation of the oscillation frequency of the first reference vibrator and an opposite sign is set. A similar offset value (an offset value with respect to a measured value is an offset value having the same absolute value and an opposite sign) may be set for a specification value for determining good / bad. Table 7 shows a numerical example (an example in which an offset value is added to a measured value).

  Table 7 shows the measured oscillation frequencies of the mass-produced products A, B, C, and D. Further, the oscillation frequencies after correction (−0.00090) of the mass-produced products A, B, C, and D are shown. Furthermore, the results of OK / NG determination (good / bad determination) for the mass-produced products A, B, C, D are shown. As shown in Table 7, the test results can also be obtained by adjusting the measurement conditions, that is, by setting an offset value that cancels the characteristic fluctuation of the first reference vibrator with respect to the measurement value. The product test can be continued while maintaining the consistency of the product.

  The set offset value is recorded in the storage medium 44 of the tester 4 and used in the subsequent product test and measurement of the first reference oscillation circuit (when the offset value is set with respect to the specification value, Applies to recorded first reference oscillator circuit measurements).

  In step S51 shown in FIG. 6, when it is determined that the fluctuation amount exceeds the predetermined range and cannot be adjusted, or when it is determined that replacement is necessary for other reasons, the process proceeds to step S52. In step S52, the first reference vibrator is replaced. Next, in step S53, the first plurality of reference oscillation circuits are measured using the replaced first reference vibrator, and the obtained measurement values and the recorded measurement values (the first before the characteristic change). From the comparison with the measured value using the reference vibrator), the characteristics of the first reference vibrator before replacement (first reference vibrator before characteristic change), and the characteristics of the replaced first reference vibrator In step S54 or step S55, the same adjustment as described above is performed so as to grasp the difference between the two and advance to step S6 shown in FIG. By performing such adjustment, the consistency of the test results can be maintained even when the first reference vibrator is replaced.

  Next, the management of the reference vibrator at a new test location will be described.

  In order to perform a product test at a new test location, after executing Steps S1 and S2 shown in FIG. 5 described above, the process of FIG. 7 described below is executed.

  FIG. 7 is a flowchart of the steps executed to perform a product test at a new test location. FIG. 8 is a flowchart of the process for preparing the second plurality of reference oscillation circuits in step S110 shown in FIG. 7, and FIG. 9 is a flowchart of the adjustment 2 process in step S130 shown in FIG.

  First, in step S110, at least a part of the first plurality of reference oscillation circuits that are mounted and stored on the test board 3 for management of the first reference resonator at the existing test location is removed. And prepared as a second reference oscillation circuit. Details will be described later.

  Next, in step S120, the second reference oscillation circuit is measured using the second reference oscillator having the same specifications as the first reference oscillator.

  In step S130, adjustment 2 is performed. In this adjustment 2, a difference in characteristics between the first reference vibrator and the second reference vibrator is grasped in comparison with the measurement result recorded for management at the existing test location. Then, the variable capacitance element is adjusted in step S131 shown in FIG. 9 or the test condition is adjusted by setting an offset value for the measured value / usage value in step S132 so as to offset this difference. .

  In the existing test point, for the continuous management of the reference oscillator, the same number of new reference oscillation circuits as the number supplied to form the second reference oscillation circuit are supplemented, and the first reference vibration is supplied. It is preferable to continue storing a sufficient number of first reference oscillation circuits for management by recording and storing the measurement values measured by the child.

  On the other hand, in order to manage the reference vibrator at a new test location, it is preferable to prepare as many second reference oscillation circuits as the number of existing test locations. For this reason, a test board used for testing a product at an existing test location is stored by mounting on the test board while it is not being used for product testing (removed from the test equipment). One of the first plurality of reference oscillation circuits is removed and prepared as a second reference oscillator (step S111 and step S112 shown in FIG. 8). Then, when a new reference oscillation circuit is mounted on this test board, and when a product is tested using this board (when this board is attached to a test device to test the product) Preferably, before the product is tested, the newly installed reference oscillation circuit is measured and the measured value is recorded. The measurement is performed by a switch provided on the test board. The first reference oscillator is connected to the newly installed reference oscillation circuit, and the necessary power supply voltage and input signal are received from the tester for testing the product. This is performed by supplying (step S113). Then, by storing this new reference oscillation circuit on the test board as one of the first plurality of reference oscillation circuits, the first reference oscillation circuit using a predetermined number of first reference oscillation circuits is continuously used. It is possible to manage one reference vibrator.

  By repeating this operation until all the first reference oscillation circuits are replaced with new reference oscillation circuits (steps S114 and S115), the existing test is performed while minimizing the influence on the test schedule at the existing test points. The second reference for performing reference oscillator management at a new test location using the same number (N in step S114) of reference transducers as those used for reference oscillator management at the location. It can be prepared as an oscillation circuit.

  Although it is possible to remove all the first reference oscillation circuits at once and replace them with new reference oscillation circuits, there is a possibility that the fluctuation of the first reference oscillator cannot be detected. Therefore, it is preferable to prepare a second reference oscillation circuit by removing at least one, preferably two or more, more preferably three or more first reference oscillation circuits.

  Here, the second reference vibrator is prepared by attaching a reference oscillation circuit to a test board for testing a product at an existing test location, and using the test board as a test apparatus for testing the product. Measurement is performed using a tester for installation and product testing, and then removed from the test board when the test board is removed from the test equipment (when not used for testing). That is, it is possible to carry out using only those used for testing at existing test locations without preparing a dedicated board or measuring instrument. In addition, the test board used for testing the product at the existing test location is the first reference oscillation circuit except for the first reference oscillator and the one removed to form the second reference oscillation circuit. Is attached and can be used for product testing. For this reason, the influence which it has with respect to the schedule of the product test in the existing test location can be suppressed to the minimum.

  In step S130 (adjustment 2) shown in FIG. 7 described above, after the difference in characteristics from the first reference vibrator is grasped and adjusted, the second test is performed in the same manner as in the management at the existing test location. A measurement value obtained by measuring the second reference oscillation circuit using the reference oscillator is recorded, and the second reference oscillation circuit is stored. Then, the product using the second reference vibrator is tested (step S140), and the second reference oscillation circuit is remeasured at an appropriate time (step S150), and the characteristics of the second reference vibrator are measured. The presence or absence of fluctuation is determined (step S160). If it is determined that the characteristic variation has occurred, adjustment 1 '(adjustment of the reference vibrator or adjustment of the test conditions) is performed (step S170). If adjustment is impossible, the reference oscillator is replaced, and the difference between the characteristics of the reference oscillator before replacement (before fluctuation) is grasped by measuring the stored reference oscillation circuit, and the same adjustment is performed. If it is determined that the characteristic of the reference vibrator has changed (step S180), the changed reference oscillation circuit is replaced and measured (step S190).

  Note that the measurement value to be recorded may be the result of measurement performed for grasping the characteristics of the second reference vibrator. In this case, the measurement value corrected based on the adjustment result may be recorded instead of the measurement value as it is. Alternatively, remeasurement may be performed after adjustment, and the measured value may be recorded.

  Here, Tables 8 and 9 show numerical examples in the adjustment of the reference vibrator for use in another test location.

  Table 8 shows the specification values of the second reference vibrator. Table 9 shows the oscillation frequencies recorded by the first reference oscillator of the second reference oscillation circuits 1, 2, 3, and 4. In addition, measurement values (initial oscillation frequencies) of the second reference oscillation circuits 1, 2, 3, and 4 at the second reference vibrator are shown. Further, measured values (adjusted oscillation frequencies) of the second reference oscillation circuits 1, 2, 3, and 4 at the second reference vibrator are shown.

  As shown in Table 9, the initial oscillation frequency measured by the second reference oscillator with the second reference oscillator is different from the recorded oscillation frequency measured with the first reference oscillator. Yes. This is because the characteristics of the second reference vibrator and the characteristics of the first reference vibrator are different. However, by appropriately setting the control voltage Vc applied to the variable capacitance diode 33, it is possible to adjust so as to cancel out the difference in characteristics between the reference vibrators. In other words, the adjusted oscillation frequency using the second reference vibrator matches the recorded oscillation frequency measured with the first reference vibrator within the range of measurement variation as shown in Table 9. To do. By using the second reference vibrator thus adjusted, it is possible to ensure consistency between the test result at the new test location and the test result at the existing test location.

It is a figure which shows the structure of the test system to which one Embodiment of the management method of the reference | standard vibrator of this invention was applied. It is a figure which shows the circuit element in the test board shown in FIG. It is a figure which shows the test board different from the test board shown in FIG. It is a figure which shows the state by which the reference | standard crystal oscillator and the variable capacitor for adjustment were connected to the oscillation circuit which is a product. It is a flowchart of the process performed in order to manage the reference | standard vibrator | oscillator in a single test location. It is a flowchart of the process of the adjustment 1 in step S5 shown in FIG. It is a flowchart of the process performed in order to perform the product test in a new test location. 8 is a flowchart of a process for preparing a second plurality of reference oscillation circuits in step S110 shown in FIG. It is a flowchart of the process of the adjustment 2 in step S130 shown in FIG.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Wafer 2 Prober 3, 3_1 Test board 4 Tester 10 Semiconductor chip (product chip)
11, 33 Variable capacitance diode 12 CMOS inverter 13 Input terminal 21 Stage 31 Reference crystal resonator 32 Reference oscillation circuit 34 Probe needle 35 Switch 36, 37 Capacitor 38 Socket 41 CPU
42 Power Supply / Signal Unit 43 Measuring Instrument 44 Storage Medium 100 Test Equipment

Claims (7)

A method for managing a reference resonator used for testing a plurality of oscillation circuits produced continuously,
Measuring a plurality of reference oscillation circuits using the first reference oscillator, recording the obtained measurement values, and storing the plurality of reference oscillation circuits;
Testing the plurality of continuously produced oscillator circuits using the first reference oscillator;
Thereafter, re-measurement of the plurality of reference oscillation circuits is performed using the first reference vibrator, the obtained measurement value is compared with the recorded measurement value, and the first reference vibrator is compared. If a characteristic variation is recognized, adjust the first reference vibrator or the test condition so as to cancel the characteristic variation,
After the adjustment for canceling the fluctuation, the test of the plurality of oscillation circuits that are continuously produced is further continued. A method for managing a reference resonator used for testing a plurality of oscillation circuits produced continuously,
Measuring a plurality of reference oscillation circuits using the first reference oscillator, recording the obtained measurement values, and storing the plurality of reference oscillation circuits;
Testing the plurality of continuously produced oscillator circuits using the first reference oscillator;
Thereafter, re-measurement of the plurality of reference oscillation circuits is performed using the first reference vibrator, the obtained measurement value is compared with the recorded measurement value, and the first reference vibrator is compared. If the need for replacement is found,
The first reference vibrator is replaced with a new first reference vibrator, the plurality of reference oscillation circuits are measured using the replaced first reference vibrator, and the obtained measurement values are obtained. Comparing the characteristic of the replaced first reference vibrator with the recorded measurement value and the difference between the characteristic of the first reference vibrator before the change before the change,
Adjusting the exchanged first reference vibrator or the test conditions so as to cancel the grasped difference,
After the adjustment for canceling out the difference, the test of the plurality of oscillation circuits that are continuously produced is further continued. When at least one characteristic variation of the plurality of reference oscillation circuits is recognized by comparing the measured value obtained by the re-measurement with the recorded measured value,
Replace the reference oscillation circuit in which the characteristic variation is recognized with a new reference oscillation circuit,
The new reference oscillation circuit is measured using the first reference oscillator, and the obtained measurement value is replaced with the measurement value of the reference oscillation circuit in which the characteristic variation is recognized, and recorded.
3. The new reference oscillation circuit is added to the plurality of reference oscillation circuits excluding the reference oscillation circuit in which the characteristic variation is recognized, and stored as a plurality of new reference oscillation circuits. The reference oscillator management method described. The plurality of reference oscillation circuits are stored together with the first reference oscillator on a test board for connecting the oscillation circuit to a test for testing the oscillation circuit and storing the plurality of reference oscillation circuits. Done by
4. The reference vibrator management method according to claim 1, wherein re-measurement of the plurality of reference oscillation circuits is performed in a state of being mounted on the test board. 5.   The adjustment is performed by supplying a control voltage to a variable capacitive element connected to the reference vibrator from a tester that tests the oscillation circuit, and changing the capacitance value of the variable capacitive element, thereby 5. The reference vibrator management method according to claim 1, wherein the management method is performed by adjusting a resonance frequency. A reference oscillator management method used for testing a plurality of continuously produced oscillation circuits at a plurality of test points,
When testing the plurality of oscillation circuits using the first reference oscillator at the first test location, a test board equipped with the first reference oscillator and the plurality of first reference oscillation circuits is tested. Attached to the
Using the switch mounted on the test board, the reference vibrator is connected to each of the plurality of first reference oscillation circuits, the measurement result is recorded,
While the test board is not attached to the test apparatus, a part of the plurality of first reference oscillation circuits is removed from the test board to prepare as a second reference oscillation circuit,
A second reference oscillator for testing the plurality of oscillation circuits at a second test location is connected to the second reference oscillation circuit and measured, and the obtained measurement value and the recorded measurement Comparing the values and grasping the difference in characteristics between the first reference vibrator and the second reference vibrator,
A method of managing a reference vibrator, characterized by adjusting a test condition at the second reference vibrator or the second test location so as to cancel the grasped difference.   A part of the plurality of first reference oscillation circuits is removed from the test board, and the same number of new reference oscillation circuits as the removed first reference oscillation circuits are part of the plurality of first reference oscillation circuits. Is mounted on the test board, the first reference vibrator is connected to the newly mounted reference oscillation circuit, the newly mounted reference oscillation circuit is measured, and the obtained measurement result is recorded. 7. The reference vibrator management method according to claim 6, wherein a predetermined number of the second reference oscillation circuits are prepared by repeating the operation a plurality of times.

Images