JPH1154575A - Charged particle line testing device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To receive a sampling-measurement trigger signal and realize sampling measurement during a measurement period, by obtaining secondary electrons by secondary electron detection means, receiving corresponding integral data, and storing the data into a predetermined address position of a waveform recorder. SOLUTION: A memory address calculator 36 receives both time difference data 35, and supplies storage address information 59b generated for memory address of a waveform recorder 80, with sampling pulse information to a signal processor 70. The signal processor 70 receives the sampling pulse information 59a and the storage address information 59b, performs a predetermined conversion calculation on integral data 69, and supplies writing data 79 with a predetermined address signal 78 to a waveform recorder 80. As it is assumed that a trigger signal 51 generated on the side of DUT driver 40 and a serial sampling pulse signal 58b operate in asynchronous relation, the storage address information 59b is generated at random, and all the sampling data can be stored into a memory.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to sampling measurement of a circuit under test such as a chip under test by a charged particle beam test apparatus.

[0002]

2. Description of the Related Art For a prior art example, FIG. 5 is a block diagram of a main part relating to sampling measurement of a charged particle beam test apparatus, FIG.
This will be described below with reference to the detection waveform of FIG. 7 and an explanatory diagram of a baseline error.

[0003] The main components related to the sampling measurement technique are as follows:
As shown in FIG. 5, the DUT driver 40, the burst sampling waveform generator 50, the lens barrel 60, and the signal processor 7
0, a waveform recording unit 80, and a control unit 90. An electron gun 62, a beam blanker 64, XY deflection means 65, a grid 66
And secondary electron detecting means 68. Note that the sampling measurement method according to this configuration is a continuous sampling measurement method of a burst-like number of N times a predetermined number of pulses with one trigger signal 51 in order to shorten the measurement time to 1 / N. .

[0004] The DUT driving section 40 includes a device under test (D
A drive signal 48 at a desired timing is applied to each terminal of the UT) to drive a circuit or a wiring pattern to be measured on a portion irradiated with the pulse of the electron beam 61 in a testable state. There are application examples. A trigger signal 51 generated for the burst sampling waveform generation unit 50 to trigger the sampling timing is a DU signal.
The T drive unit 40 repeatedly generates a trigger signal accurately synchronized with a desired timing of the drive signal 48 supplied to the DUT. Although the trigger signal generating means for generating the trigger signal 51 is generated from the DUT drive unit 40, the trigger signal generating means may be supplied from another device.

[0005] The burst sampling waveform generator 50
It has a variable delay means 52 and a burst pulse generation means 54. The operation of the burst sampling waveform generator will be described. Said precisely synchronized trigger signal 5
6A, after a predetermined delay amount Td by the variable delay means 52, the burst pulse generation means 54 outputs a burst-like continuous sampling pulse signal 58 having a predetermined pulse number N and a constant interval T as shown in FIG. Is supplied to the beam blanker 64 of the lens barrel unit 60, and at the same time, the sampling pulse information 59a and the storage address information 5
9b is supplied to the signal processing unit 70. The storage address information 59b is address information starting from the normal trigger signal 51. For example, as shown in FIG. 6B, the value of the trigger signal 51 is set to 0, and a predetermined addition value is added at regular intervals T. The generated address information 59b is output. Incidentally, the generation means of the address information 59b may be provided on the signal processing unit 70 side described later. Here, the method of giving the predetermined delay amount is as shown in the sampling pulse generation diagram of FIG.
A predetermined parameter is set and controlled by the control unit 90 so as to give a delay time obtained by adding the delay amount ΔT of a minute unit to d. Thus, the entire waveform on the time axis of the DUT is sampled. The above-mentioned predetermined number of pulses N, fixed interval T, and delay amounts Td and ΔT of the variable delay means 52 can be changed to arbitrary conditions by setting conditions from the control unit 90.

The lens barrel unit 60 supplies the burst-like sampling pulse signal 58 to the beam blanker 64 to irradiate the DUT in a burst with an electron beam 61 pulse, as shown in the detection waveform of FIG. DU of irradiation site
Secondary electrons corresponding to the T potential are sequentially captured by the secondary electron detecting means 68, and integrated data obtained by integrating the entire peak of the peak-like electric signal is converted into digital signal data 69
Is supplied to the signal processing unit 70. Note that the detection waveform in FIG. 7A is based on the assumption that the deflection of the electron beam by the X / Y deflecting means 65 is in a fixed state, and the DUT potential at the irradiation site is a steady voltage. Note that the DUT is placed on the XY moving stage immediately below and can be moved to a desired irradiation site, and has no direct relation to the subject, and further generates secondary electrons in conjunction with the pulse of the electron beam 61. Although a control signal for controlling the amount is supplied from the signal processing unit 70 to the grid 66, there is no direct connection with the subject in this case, and further, there is no direct connection with the XY deflection means 65. .

The signal processing section 70 receives the integration data 69 from the secondary electron detecting means 68, receives the sampling pulse information 59a and the storage address information 59b,
Write data 79 obtained by subjecting integral data 69 to predetermined arithmetic conversion
Together with a predetermined address signal 78 for storing the
0. The waveform recording unit 80 receives the address and the data and stores the address and the data in the corresponding address position as shown in the explanatory diagram of the address of the storage memory in FIG.
The above-described memory storage operation is performed for every 61 pulses of the electron beam.
This is repeated in bursts every N times each time the trigger signal 51 is generated. Eventually, when the measurement data is stored in all the memories of the waveform recording unit 80, the entire waveform on the time axis of the DUT is sampled, and the measurement is completed. The control unit 90 controls the entire operation described above, reads out the contents of the waveform recording unit 80, converts it into a desired display form and output form, and uses it.

[0008]

By the way, the above-mentioned conventional continuous sampling and measuring method in a burst configuration has the following difficulties. That is, since the generation of the electron beam 61 generated from the electron gun 62 of the lens barrel 60 is performed in a burst manner, there is a generation period of the electron beam 61 and a pause period. In the first several pulse trains of the electron beam 61 due to the activation of the next trigger signal 51 during this pause period, as shown in FIG. 7A, a level difference occurs in the detected waveform even if other conditions are the same. As a result, as shown in FIG. 7B, a phenomenon occurs in which a baseline error occurs with respect to the original baseline of the potential measurement of the irradiation site, and it is as if D
An erroneous measurement is made as if the circuit voltage at the measurement site of the UT fluctuates slightly. This is particularly unfavorable in a charged particle beam test apparatus that requires accurate measurement, and has practical problems. Therefore, the problem to be solved by the present invention is to constantly irradiate the electron beam 61 pulse,
An object of the present invention is to provide a charged particle beam test apparatus capable of receiving a trigger signal 51 for sampling measurement and realizing sampling measurement in a period to be measured.

[0009]

FIGS. 1 and 2 show a solution according to the present invention. First, in order to solve the above problem, in the configuration of the present invention, the trigger signal for starting the sampling measurement is synchronized with the means of the DUT drive unit 40 for driving the measurement site of the device under test to a predetermined state at a predetermined timing. In response to the generation of 51, the DUT is continuously irradiated with a pulsed charged particle beam pulse for sampling (for example, a charged particle beam such as an electron beam or an ion beam), and the secondary electrons obtained therefrom are detected. DUT
In a charged particle beam tester that samples and measures the potential of an irradiation site, a charged particle beam pulse is applied at a constant interval T
A time difference detecting unit 3 for receiving the trigger signal 51 and the continuous sampling pulse signal 58b, and for obtaining a time difference between the two, comprising a means of a continuous sampling pulse generating unit 32 for generating a continuous sampling pulse signal 58b for irradiating the UT.
And means for receiving the time difference data 35 obtained above and generating storage address information 59b corresponding to the irradiation timing of the charged particle beam pulse. The storage address information 59b is provided. Receiving means for obtaining secondary electrons from the DUT-irradiated portion by the secondary electron detecting means 68, receiving the corresponding integral data 69, and storing the integrated data 69 at a predetermined address position of the waveform recording section 80. It is. As described above, the storage address information 5 corresponding to the trigger signal 51 for starting the sampling measurement is stored.
9b can be generated, and as a result of the steady irradiation of the electron beam 61 pulses, accurate sampling measurement with a baseline error eliminated can be realized.

FIG. 3 and FIG. 4 show a solution according to the present invention. Second, in order to solve the above problem, in the configuration of the present invention, the trigger for starting the sampling measurement is synchronized with the means of the DUT driving unit 40 for driving the measurement site of the device under test to a predetermined state at a predetermined timing. In response to the generation of the signal 51, a charged particle beam test apparatus for continuously irradiating a pulsed charged particle beam pulse for sampling to the DUT and sampling and measuring the potential of the DUT irradiated portion by detecting secondary electrons obtained therefrom. , A means 53 for generating a clock signal synchronized with the generation of the trigger signal 51, receiving the trigger signal 51 and the clock signal 53, and generating a continuous pulse signal 23 synchronized with the clock signal 53
The synchronous sampling pulse generator 22 receives the continuous pulse signal 23 each time the trigger signal 51 is generated, and generates a continuous sampling delay pulse signal 58c to which a predetermined delay amount ΔT or a random delay amount is sequentially added.
And a variable delay means 52, and a trigger signal 51
Receiving the continuous pulse signal 23 from the continuous sampling delay pulse signal generating means, and obtaining the time difference data 25 between the trigger signal 51 and the continuous pulse signal 23. Receiving the time difference data 25, receiving the delay amount data 23b to be sequentially added from the continuous sampling delay pulse signal generating means, and generating the storage address information 59b corresponding to the irradiation timing of the charged particle beam pulse. Means for receiving the storage address information 59b, storing the secondary electrons from the DUT-irradiated portion at the predetermined address position of the waveform recording section 80 upon receiving the corresponding integral data 69 obtained by the secondary electron detection means 68. There is a configuration means provided with a means for performing.

The means for generating a continuous sampling pulse signal normally generates a pulse signal at a fixed time interval, and generates a pulse signal based on a random number in the vicinity of the generation of the trigger signal. The charged particle beam test apparatus may be configured to generate a pulse signal to be output by synthesizing a pulse signal at a time interval and a pulse signal based on the random number. The means for generating a continuous sampling pulse signal includes a register for storing a predetermined numerical value, a random number generator for generating a random number, a multiplexer for selectively outputting the register output signal or the random number generator signal, And a counter that counts and outputs pulses generated by the clock source based on the multiplexer output value.

[0012]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of the present invention will be described below in detail with reference to the drawings together with embodiments.

(Embodiment 1) An embodiment of the present invention will be described below with reference to a main part configuration diagram relating to sampling measurement of a charged particle beam test apparatus shown in FIG. 1, a timing diagram for detecting a time difference shown in FIG. explain. Elements corresponding to the conventional configuration are denoted by the same reference numerals.

The main configuration of the present invention is as shown in FIG.
DUT drive unit 40 and continuous sampling generation operation unit 30
, A lens barrel unit 60, a signal processing unit 70, and a waveform recording unit 80
And a control unit 90. This configuration is the same as the conventional configuration except for the continuous sampling generation operation unit 30. In the present invention, it is assumed that the circuit that generates the electron beam 61 and the trigger signal 51 that is generated by the DUT driving unit 40 operate in an asynchronous relationship.

The continuous sampling generation operation section 30 always outputs a continuous sampling pulse signal 58 at a constant interval T.
b is supplied to the beam blanker 64 to continuously and continuously irradiate the electron beam 61 pulse to the measurement site of the DUT. In response to the trigger signal 51, a time difference between the trigger point and the continuous sampling pulse signal 58b is obtained, and the storage address information 5 of the memory corresponding to the time difference is obtained.
9b as a means for arithmetically generating and supplying this to the signal processing unit 70. The storage address information 59b outputs the time difference data for the first time in order to use the address information with the starting point of the normal trigger signal 51 as a zero value. The address information 59b obtained by sequentially adding the added value to the time difference data is calculated and output.

The more specific internal configuration of the continuous sampling generation / calculation section 30 includes a continuous sampling pulse generation section 32, a time difference detection section 34, and a memory address calculation section 3.
6 The continuous sampling pulse generator 32 always outputs a continuous sampling pulse signal 5 at a constant interval T.
8b is supplied to the beam blanker 64. Thereby, the electron beam 61 pulse is always generated at a constant interval regardless of the required sampling measurement timing.
Is irradiated with a pulse on the measurement site. As a result, the advantage that the baseline error shown in FIG. 7B is eliminated can be obtained.

The time difference detecting section 34 is means for calculating a time difference between the trigger signal 51 from the DUT driving section 40 and the continuous sampling pulse signal 58b generated at a constant interval T. This detection means has a configuration example of an RS flip-flop 142, a flip-flop 144, a clock time counting means 148, a clock source 150, and a fraction time measuring means 152 shown in FIG. Here, the flip-flop 144 and the clock time counting means 148
Is for obtaining the time difference data 35a in units of clock 151 hours, and the fractional time measuring means 152 is for obtaining the time difference data 35b shorter than the clock 151 hours. This operation will be described with reference to FIG. The clock source 150 supplies, for example, a clock 151 having a 1 GHz reference time of 1 GHz to each unit. The RS flip-flop 142 outputs a time difference signal 143 of an analog amount that is set by the trigger signal 51 and then cleared by the continuous sampling pulse signal 58b at a constant interval T. The flip-flop 144 outputs a clock synchronization time difference signal 145 synchronized with the clock 151. The clock time counting means 148 outputs the clock synchronization time difference signal 1
45 counts the clock time in 1 nanosecond units of the valid period, and outputs the time difference data 35a. On the other hand, the fraction time measuring means 152 integrates the pulse width of the front fraction time difference 153a between the clock synchronization time difference signal 145 and the time difference signal 143 shown in FIG. 2A with a positive voltage, and the pulse width of the rear fraction time difference 153b. Is integrated with a negative voltage, and the time difference data 35b obtained by quantizing and converting the integration amount of the resulting positive or negative voltage into digital data in units of tens of picoseconds is output. These two time difference data 35 are supplied to the memory address operation unit 36.

The memory address calculation section 36 receives the time difference data 35 and supplies storage address information 59b calculated and generated for the memory address of the waveform recording section 80 to the signal processing section 70 together with sampling pulse information 59a. Note that this calculation may be performed only when the first trigger signal 51 is received, and the subsequent storage address information 59b may be output by adding an added value corresponding to the fixed interval T. If the storage address information 59b calculated and generated by the memory address calculation unit 36 is an unnecessary memory address value exceeding the storage area on the waveform recording unit 80 side, the storage address information 59b and the sampling pulse information 5
A determination process for not outputting 9a may be provided.

On the other hand, the signal processing section 70 performs substantially the same signal processing as that in the related art. That is, receiving the above-mentioned sampling pulse information 59a and storage address information 59b, it supplies the waveform recording unit 80 together with a predetermined address signal 78 to store the write data 79 obtained by subjecting the integrated data 69 to predetermined conversion. However, at this time, if the storage address information 59b is an unnecessary memory address value exceeding the storage area, a determination process for not performing the storage process is provided outside the sampling measurement period.

FIG. 8 shows an embodiment 32b of the continuous sampling pulse generator. The counter 321 has a clock source 3
The pulse generated at 22 is counted by the numerical value (corresponding to time T) set by REG323, and an output signal is generated from 58b. Then, every time this output timing occurs, RE
The initial value set in G323 is loaded, similarly counted, and the next output signal is generated.

FIG. 10 is a timing chart showing the operation of the continuous sampling pulse generator 32b. The continuous sampling pulse signal 58b is continuously generated at time intervals T. The continuous sampling pulse signal 58b operates in an asynchronous relationship with the trigger signal 51. Therefore, the phase between the trigger signal 51 and the continuous sampling pulse signal 58b generally differs every time, and the time difference detection unit 34 uses Tm11,
Will show measured values like Tm12, Tm13

By the way, the trigger signal 51 generated by the DUT driver 40 and the continuous sampling pulse signal 58b are assumed to operate asynchronously, so that random storage address information 59b is generated. For this reason, the same address may be sampled and measured, which increases the time, but eventually all the target sampling data is stored in the memory. The signal processing unit 70 detects this, and notifies the control unit 90 of the end of the sampling measurement based on the detection.

If desired, the signal processing unit 70 is provided with means for counting the number of times the same address is sampled and measured, and it is determined whether or not the randomness of the trigger signal 51 and the continuous sampling pulse signal 58b is biased. A randomizing control means for slightly increasing or decreasing the setting condition of the constant interval T of the continuous sampling pulse generating unit 32 via the control unit 90 to the extent that no baseline error occurs. May be additionally provided. Of course, the storage address information 5 obtained by adding this slight increase / decrease
It goes without saying that 9b is calculated and output. In this case, the trigger signal 51 and the continuous sampling pulse signal 5
8b can be reduced or eliminated. A means for unconditionally adding the randomization control means or a means for detecting a sampling measurement at the same address from the beginning of the sampling measurement may be provided, and the randomization control means may be executed each time this detection is performed. .

According to the first embodiment of the present invention, there is provided a means for constantly supplying continuous electron beam 61 pulses at a constant interval T, and a time difference between the trigger signal 51 and the continuous sampling pulse signal 58b is determined to store the storage address information. By providing the means for generating 59b, it is possible to eliminate the pause period of the electron beam 61 pulse as in the related art, and it is possible to eliminate the problem that the conventional baseline error caused by the pause period is caused. Benefits are obtained.

(Embodiment 2) An embodiment of the present invention will be described below with reference to another configuration diagram of a main part relating to sampling measurement of the charged particle beam test apparatus of FIG. 3 and an example of a circuit configuration of time difference detection of FIG. .

The main configuration of the present invention is as shown in FIG.
The configuration is such that the continuous sampling generation operation unit 30 of the first embodiment is replaced with a synchronous sampling generation operation unit 30b.
The internal configuration of the synchronous sampling generation operation unit 30b is as follows.
Synchronous sampling pulse generator 22 and variable delay means 52
, The time difference detection unit 24, the memory address calculation unit 36b,
Timing synchronous clock signal 5 from DUT driver 40
3

The timing synchronous clock signal 53 from the DUT driving section 40 is supplied to the synchronous sampling generation operation section 30b.
Supplied to This is a clock synchronized with the trigger signal 51 and is, for example, a timing clock of several tens of MHz used inside the DUT driving unit 40, and this clock frequency value is notified to the control unit 90 in advance by communication means and is known. . The generation of the trigger signal 51 generated from the side of the DUT driving unit 40 is based on the timing synchronization clock signal 5.
Although it is synchronous in FIG. 3, the occurrence time and cycle are variously different depending on the test conditions of the DUT, so that it is asynchronous with the fixed interval T.

The synchronous sampling pulse generator 22 has a D
Timing synchronous clock signal 53 from UT drive unit 40
In response to this, in synchronism with this, a continuous pulse signal 23 with a predetermined constant interval T is supplied to the time difference detection unit 24 and the variable delay unit 52. As a result, the trigger signal 51 is not completely random as in the first embodiment, but is generated and output in synchronization with the clock of the timing synchronization clock signal 53. As a result, the fractional time measuring means 152 shown in FIG.
Is obtained. Also, the DUT driving unit 40
When the trigger signal 51 is received, the delay amount data 23b, which becomes the delay amount Tx obtained by sequentially adding the predetermined delay amount ΔT, is calculated and supplied to the variable delay means 52 and the memory address calculator 36b.

The time difference detector 24 receives the timing synchronization clock signal 53 and the trigger signal 51 from the DUT driver 40, receives the synchronized continuous pulse signal 23 from the synchronization sampling pulse generator 22, and executes the operation. This is a means for obtaining the time difference between the trigger signal 51 and the continuous pulse signal 23 as in Example 1, and supplying this time difference data 25 to the memory address calculation unit 36b. However, in this case, as shown in the internal configuration example of FIG. 4, since the trigger signal 51 is synchronized with the timing synchronization clock signal 53, the fractional time measuring means 152 shown in FIG. .

The variable delay means 52 receives the continuous pulse signal 23 and the delay amount data 23b from the synchronous sampling pulse generator 22 and receives a continuous sampling delay pulse signal 58 having a delay amount corresponding to the delay amount data 23b.
Output c. As a result, there is an advantage that sampling measurement can be sequentially performed in a unit time of a predetermined delay amount ΔT within one clock period of the timing synchronization clock signal 53. Accordingly, the present invention has an advantage that the probability of sampling and measuring the same storage address is greatly reduced as in the first embodiment, and the measurement loss time is shortened. In addition, the continuous sampling delay pulse signal 58 of the delay amount Tx added sequentially.
Since this is c, the minute time is changed to the original continuous constant interval T, but it does not affect the occurrence of the baseline error shown in FIG. 7B. In the delay control to the variable delay unit 52, a predetermined delay amount ΔT is sequentially provided and output. However, the delay control may be performed such that a desired random delay amount is provided instead. . As a result of the random electron beam 61 being generated by the random delay control, even if there is a synchronous relationship with the trigger signal 51 generated by the DUT driving unit 40, it is possible to obtain sampling at all target positions. Is obtained. Therefore,
In this configuration example, the present invention can be implemented in any of a synchronous relationship and an asynchronous relationship with the trigger signal 51 generated by the DUT driving unit 40.

The memory address calculating section 36b receives the time difference data 25 from the time difference detecting section 24, receives the delay amount data 23b from the synchronous sampling pulse generating section 22, and receives a waveform recording section 80 corresponding to the time difference and the delay data.
The storage address information 59b corresponding to the memory address is calculated and generated, and supplied to the signal processing unit 70 together with the sampling pulse information 59a. If the storage address information 59b calculated and generated by the memory address calculation unit 36b is a memory address exceeding the storage area of the waveform recording unit 80, the storage address information 59b and the sampling pulse information 59a are not output if desired. May be provided.

(Embodiment 3) In general, the continuous sampling pulse generator is configured so as to be capable of performing an operation asynchronously with the trigger signal 51 as in the above embodiment. However, perfect randomness is not compensated,
For example, if the continuous sampling pulse generator and the trigger signal are fixed by chance to maintain an integral multiple, the same address may be repeatedly sampled and measured. May not be possible. The third embodiment is configured so that the operation related to the asynchronous operation can be performed more reliably. FIG. 9 shows another embodiment 32c of the continuous sampling pulse generator. The counter 321 is connected to the clock source 32
The pulse generated by 2 is counted by the output value of the multiplexer 325 to generate an output signal from 58c. The multiplexer 325 normally selects and outputs a numerical value (corresponding to time T) set by the REG 323. But,
Immediately before the generation of the trigger signal 51, that is, when the test end signal 320 indicating the end of the previous test is applied, the random number generator 3
The numerical value of 24 (corresponding to time TR) is selectively output. After once selecting and outputting the numerical value indicated by the random number generator 324, the multiplexer 325 selects and outputs the numerical value set by the REG 323 again. Then, each time the output timing is generated, the numerical value set by the multiplexer 325 is loaded and counted in the same manner to generate the next output signal.

FIG. 11 is a timing chart showing the operation of the continuous sampling pulse generator 32c according to the third embodiment. Normally, the continuous sampling pulse signal 58c according to the third embodiment is continuously generated at time intervals T. However, in the vicinity where the trigger signal 51 is generated, the signal is generated at a time interval TR based on the numerical value generated by the random number generator 324. Moreover, since this timing is based on a random number, it shows a different value every time. That is, as shown in the figure, it occurs at time intervals of TR1, TR2, and TR3.
As described above, the phase between the trigger signal 51 and the continuous sampling pulse signal 58c is different each time, and the time difference detection unit 34 detects Tm21, Tm22, and Tm as shown in the drawing.
A measurement like 23 will be shown. Since this time difference is based on a random number, randomness is compensated.

In the description of the above embodiment, the present invention is applied to a test apparatus using electron beam irradiation. However, the present invention is also applicable to a test apparatus that irradiates a DUT with a charged particle beam pulse such as an ion beam and detects secondary electrons generated therefrom. The same can be applied.

[0035]

According to the present invention, the following effects can be obtained from the above description of each embodiment. According to the present invention, the electron beam 61 pulse is constantly radiated, and a means for receiving a trigger signal for sampling measurement and generating a storage address corresponding thereto is provided. There is no pause period, and as a result, there is obtained a great advantage that accurate and stable sampling measurement with a baseline error eliminated is realized, and highly accurate sampling measurement can be realized.

[Brief description of the drawings]

FIG. 1 is a main part configuration diagram relating to sampling measurement of a charged particle beam test apparatus of the present invention.

FIG. 2 is an explanatory diagram and a circuit configuration example of a time difference detection timing according to the present invention.

FIG. 3 is a configuration diagram of another main part relating to sampling measurement of the charged particle beam test device according to the present invention.

FIG. 4 is a circuit configuration example of time difference detection according to the present invention.

FIG. 5 is a main part configuration diagram relating to sampling measurement of a conventional charged particle beam test apparatus.

FIG. 6 is a diagram illustrating a conventional burst sampling measurement method and memory storage.

FIG. 7 is an explanatory diagram of a detected waveform and a baseline error.

FIG. 8 shows Embodiment 32 of a continuous sampling pulse generator.
show b

FIG. 9 shows another embodiment 32c of the continuous sampling pulse generator.

FIG. 10 is a timing chart showing an operation of a continuous sampling pulse generator 32b.

FIG. 11 is a timing chart showing the operation of the continuous sampling pulse generation unit 32c.

[Explanation of symbols]

 Reference Signs List 22 Synchronous sampling pulse generating section 24, 34 Time difference detecting section 30 Continuous sampling generating calculating section 30b Synchronous sampling generating calculating section 32 Continuous sampling pulse generating section 36, 36b Memory address calculating section 40 DUT driving section 50 Burst sampling waveform generating section 52 Variable delay Means 54 Burst pulse generating means 60 Barrel unit 62 Electron gun 68 Secondary electron detecting means 70 Signal processing unit 80 Waveform recording unit 90 Control unit 142 RS flip-flop 144 Flip-flop 148 Clock time counting unit 150 Clock source 152 Fractional time Measuring means

Claims (4)

[Claims] 1. A measurement part of a device under test (DUT) is driven to a predetermined state, and upon receiving a trigger signal for starting a sampling measurement, a charged particle beam pulse for sampling is continuously irradiated on the DUT. A charged particle beam test apparatus for sampling and measuring the potential of a DUT irradiation site by detecting secondary electrons obtained therefrom, means for generating a continuous sampling pulse signal for irradiating the DUT with the charged particle beam pulse at a constant interval T; Means for receiving a trigger signal and the continuous sampling pulse signal to determine a time difference between the two, and receiving the time difference data determined above, generating storage address information corresponding to the irradiation timing of the charged particle beam pulse, Upon receiving the stored address information, the secondary electrons from the DUT irradiating area are obtained by the secondary electron detecting means. Receiving data, a waveform recording unit of a predetermined address and means for storing the position, or that comprises a charged particle beam testing apparatus characterized. 2. A measurement site of a device under test is driven to a predetermined state, and upon receiving a trigger signal for starting a sampling measurement, a DUT is continuously irradiated with a charged particle beam pulse for sampling. In a charged particle beam test apparatus for sampling and measuring the potential of a DUT irradiation site by detecting secondary electrons, means for generating a clock signal synchronized with generation of the trigger signal, receiving the trigger signal and the clock signal, A continuous pulse signal synchronized with the clock signal is generated. Each time the trigger signal is generated, a continuous sampling delay pulse signal is generated by receiving the continuous pulse signal and sequentially adding a predetermined delay amount or a random delay amount. Means for receiving the trigger signal and receiving a continuous pulse signal from the continuous sampling delay pulse signal generating means. Means for obtaining time difference data between the trigger signal and the continuous pulse signal; receiving the time difference data obtained above; receiving delay amount data to be sequentially added from the continuous sampling delay pulse signal generating means; Means for generating storage address information corresponding to the pulse irradiation timing; receiving the storage address information; receiving corresponding data obtained by secondary electron detection means from secondary electron irradiation sites; A charged particle beam test apparatus comprising: means for storing data at a predetermined address position; 3. The means for generating a continuous sampling pulse signal according to claim 1 normally generates a pulse signal at a fixed time interval, and generates a pulse signal based on a random number in the vicinity of the occurrence of a trigger signal. A charged particle beam test apparatus for generating a pulse signal to be synthesized by combining the pulse signal at the predetermined time interval and the pulse signal based on the random number. 4. A means for generating a continuous sampling pulse signal according to claim 3, comprising: a register for storing a predetermined numerical value; a random number generator for generating a random number; and the register output signal or the random number generator output signal. A charged particle beam test apparatus, comprising: a multiplexer for selectively outputting a signal; and a counter for counting and outputting pulses generated by a clock source based on the output value of the multiplexer.