WO2013027739A1 - Deterioration diagnostic circuit and deterioration diagnostic method - Google Patents

Deterioration diagnostic circuit and deterioration diagnostic method Download PDF

Info

Publication number
WO2013027739A1
WO2013027739A1 PCT/JP2012/071109 JP2012071109W WO2013027739A1 WO 2013027739 A1 WO2013027739 A1 WO 2013027739A1 JP 2012071109 W JP2012071109 W JP 2012071109W WO 2013027739 A1 WO2013027739 A1 WO 2013027739A1
Authority
WO
WIPO (PCT)
Prior art keywords
signal
deterioration
circuit
test block
output
Prior art date
Application number
PCT/JP2012/071109
Other languages
French (fr)
Japanese (ja)
Inventor
永典 實吉
Original Assignee
日本電気株式会社
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 日本電気株式会社 filed Critical 日本電気株式会社
Priority to US14/240,024 priority Critical patent/US20140218060A1/en
Publication of WO2013027739A1 publication Critical patent/WO2013027739A1/en

Links

Images

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01RMEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/28Testing of electronic circuits, e.g. by signal tracer
    • G01R31/2851Testing of integrated circuits [IC]
    • G01R31/2855Environmental, reliability or burn-in testing
    • G01R31/2856Internal circuit aspects, e.g. built-in test features; Test chips; Measuring material aspects, e.g. electro migration [EM]
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01RMEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/28Testing of electronic circuits, e.g. by signal tracer
    • G01R31/2851Testing of integrated circuits [IC]
    • G01R31/2855Environmental, reliability or burn-in testing
    • G01R31/2872Environmental, reliability or burn-in testing related to electrical or environmental aspects, e.g. temperature, humidity, vibration, nuclear radiation
    • G01R31/2879Environmental, reliability or burn-in testing related to electrical or environmental aspects, e.g. temperature, humidity, vibration, nuclear radiation related to electrical aspects, e.g. to voltage or current supply or stimuli or to electrical loads

Definitions

  • the present invention relates to a deterioration diagnosis circuit and a deterioration diagnosis method for a semiconductor integrated circuit.
  • the characteristics of a semiconductor integrated circuit deteriorate due to factors such as use after manufacture and environmental conditions.
  • the semiconductor integrated circuit proceeds more than a certain level, the semiconductor integrated circuit is determined to be defective.
  • it is generally difficult to determine whether the cause of the failure is due to lifetime (ie, reasonable deterioration due to aging deterioration) or accidental failure. It is. Therefore, knowing the degree of progress of deterioration of the semiconductor integrated circuit after shipment is extremely important for identifying the true cause of the failure of the semiconductor integrated circuit.
  • the semiconductor integrated circuit By knowing the progress of deterioration of the semiconductor integrated circuit, it is possible to easily feed back the progress of deterioration when an unexpected failure or unexpected deterioration is observed to the design of the semiconductor integrated circuit. Further, by acquiring and analyzing log information of the degree of progress of deterioration and predicting an appropriate replacement time of the semiconductor integrated circuit, it can be used for setting an optimal maintenance time of the semiconductor integrated circuit. It is desirable that the information on the degree of progress of deterioration of the semiconductor integrated circuit can be output to the outside of the semiconductor integrated circuit without using an external measuring instrument, and further stored.
  • Non-Patent Document 1 discloses a configuration in which the resolution of the deterioration degree in the practical range is improved and the measurement time is shortened by using two CMOS (Complementary Metal Oxide Semiconductor) ring oscillators. Specifically, Non-Patent Document 1 describes a configuration in which the ratio of the oscillation frequency is measured using two ring oscillators.
  • CMOS Complementary Metal Oxide Semiconductor
  • Patent Document 1 discloses a configuration for measuring deterioration of a MOSFET (Metal-Oxide-Semiconductor Field-Effect Transistor) using two CMOS ring oscillators.
  • the semiconductor integrated circuit described in Patent Document 1 includes two ring oscillators, and is connected so that a part of its own circuit is branched and replaced with a part of the other circuit when measuring deterioration. Then, by comparing the output phases of the own circuit and the other circuit, a change in the delay amount of the signal due to the deterioration of the circuit characteristics is detected.
  • Patent Document 2 discloses a configuration for determining whether or not a circuit has failed depending on whether the signal propagation time of the degradation diagnosis target circuit is earlier or later than the reference value.
  • Patent Document 2 performs circuit deterioration diagnosis using only the deterioration diagnosis circuit.
  • Patent Document 3 discloses a configuration for detecting aged deterioration of a signal delay path.
  • the semiconductor integrated circuit device described in Patent Document 3 writes a comparison result of delay amounts of different delay paths to a memory before shipment, performs the same measurement after shipment, and changes to a delay path if the magnitude relationship of the delay amounts differs. Judge that a failure has occurred.
  • Non-Patent Document 1 In the method described in Non-Patent Document 1 for determining the degree of deterioration of a semiconductor integrated circuit using a change in the oscillation frequency of a ring oscillator, the measured oscillation frequency varies greatly under the influence of the environment. There is a problem.
  • the influence of the environment on the oscillation frequency includes, for example, fluctuations in the chip temperature and power supply voltage of the semiconductor integrated circuit.
  • Non-Patent Document 1 alleviates the influence of the environment to some extent by measuring the ratio of the oscillation frequencies using two ring oscillators.
  • the configuration of Non-Patent Document 1 generates a signal having a difference frequency between two ring oscillators in order to measure a change in oscillation frequency.
  • Non-Patent Document 1 has a problem that the circuit configuration becomes complicated.
  • the semiconductor integrated circuit described in Patent Document 1 compares the delay using only a part of its own circuit and the other circuit. For this reason, the configuration described in Patent Document 1 has a problem that an accurate characteristic deterioration amount based on the delay deterioration amount of the entire ring oscillator cannot be obtained.
  • the deterioration diagnosis circuit described in Patent Document 2 performs deterioration diagnosis without using the signal of the actual circuit, the result of the deterioration diagnosis may not match the state of the actual circuit.
  • the deterioration diagnosis circuit described in Patent Document 2 has a problem that it requires generation of a highly accurate reference signal for deterioration diagnosis and complicated setting.
  • the semiconductor integrated circuit device described in Patent Document 3 needs to measure a delay value and record the value in a nonvolatile memory before shipment. For this reason, the semiconductor integrated circuit device described in Patent Document 3 has a problem in that the cost increases due to the use of expensive components such as a memory and an increase in manufacturing procedures.
  • An object of the present invention is to provide a technique for solving the problem of more accurately measuring the state of deterioration of a semiconductor integrated circuit with a simple configuration.
  • the deterioration diagnosis circuit of the present invention includes a test block including a first circuit to be subjected to deterioration diagnosis, a reference block including a second circuit having the same configuration as the first circuit, and a measurement mode. Deterioration of the elements constituting the test block by comparing the characteristics of the first signal output from the test block with the characteristics of the second signal output from the reference block when the signal shown in FIG. Determination means for determining the presence or absence of the control, and control means for outputting a signal indicating the measurement mode to the determination means.
  • the degradation diagnosis method of the present invention is characterized in that when a signal indicating a measurement mode is input, the characteristics of the first signal output from the test block including the first circuit to be diagnosed for degradation and the first By comparing the characteristics of the second signal output from the reference block including the second circuit having the same configuration as the circuit, the presence or absence of deterioration of the elements constituting the test block is determined.
  • the present invention makes it possible to more accurately measure the state of degradation of a semiconductor integrated circuit with a simple configuration.
  • FIG. 1 is a diagram showing a configuration of a deterioration diagnosis circuit according to the first embodiment of the present invention.
  • the deterioration diagnosis circuit 100 includes a deterioration diagnosis block 101 and a control means 105.
  • the deterioration diagnosis block 101 includes a test block 102, a reference block 103, and a determination unit 104.
  • a delay which is a kind of characteristic of the logic gate constituting the semiconductor integrated circuit may increase. .
  • the amount of stress is, for example, the intensity of stress or the length of a period during which stress is applied. Therefore, there is a difference in the degree of deterioration (deterioration degree) between the test block 102 and the reference block 103. Accordingly, the delay of the semiconductor integrated circuit in the test block 102 causes a larger delay in the logic gate than in the reference block 103. As a result, the signal output from the test block 102 is gradually delayed with respect to the signal output from the reference block 103. The determination unit 104 detects this delay and determines the deterioration state of the test block 102.
  • the determination unit 104 has two operation states, a standby mode and a measurement mode.
  • the standby mode is a state in which the deterioration diagnosis block 101 does not measure the deterioration of the test block 102.
  • the measurement mode is a state in which the deterioration diagnosis block 101 measures the deterioration of the test block 102.
  • the test block includes a first circuit to be subjected to deterioration diagnosis.
  • the reference block includes a second circuit having the same configuration as the first circuit.
  • the test block 102 and the reference block 103 output the first signal and the second signal to the determination unit 104, respectively.
  • the control unit 105 outputs a signal indicating the standby mode or the measurement mode to the determination unit 104.
  • the determination unit 104 performs the following operation when the signal input from the control unit 105 indicates the measurement mode. That is, in the measurement mode, the determination unit 104 compares the first signal input from the test block 102 and the second signal input from the reference block 103, and degrades the elements constituting the test block 102. The presence or absence of is determined. It is possible to apply stress to the test block 102 and the reference block 103 to advance the deterioration of each semiconductor integrated circuit. The stress can be applied by whether or not the power supply voltage is applied, the power supply voltage is changed, the ambient temperature is changed, the operating frequency is changed, the circuit load is changed, and the like. However, the types of stress are not limited to these.
  • any stress that affects the life of the semiconductor integrated circuit may be applied to one or both of the test block 102 and the reference block 103.
  • a plurality of types of stress may be simultaneously applied to the test block 102 and the reference block 103.
  • the type and strength of the applied stress be different between the test block 102 and the reference block 103.
  • the test block 102 is set to a state in which deterioration is advanced by applying stress, and the power is not applied by connecting the power supply terminal of the semiconductor integrated circuit of the reference block 103 to the ground potential. That is, it is good also as a state which deterioration does not advance.
  • stress may be applied to both the test block 102 and the reference block 103 even in the standby state, and a difference may be provided in the operating conditions such as temperature and operating frequency.
  • the positions of the test block 102 and the reference block 103 may be interchanged.
  • the determination unit 104 compares the timings of the two signals, the output of the test block 102 and the output of the reference block 103, and obtains the difference.
  • the test block 102 and the reference block 103 may include an oscillation circuit, and the output of the oscillation circuit may be output to the determination unit 104.
  • the determination unit 104 compares the characteristics of the signals output from the test block 102 and the reference block 103 and determines the deterioration state of the test block 102. If the test block 102 and the reference block 103 are provided with an oscillation circuit, the oscillation frequency or the signal phase may be used as the signal characteristics.
  • FIG. 2 is a diagram illustrating a configuration of a modified example of the deterioration diagnosis circuit of the first embodiment.
  • the deterioration diagnosis block 106 shown in FIG. 2 includes a characteristic adjustment unit 107 between the test block 102 and the determination unit 104.
  • the characteristic adjusting unit 107 adjusts the characteristic of the signal output from the test block 102 and inputs it to the determining unit 104.
  • the output of the determination unit 104 is input to the characteristic adjustment unit 107.
  • the characteristic adjustment unit 107 adjusts the output characteristic (for example, delay time) of the test block 102 so that the difference between the characteristic of the test block 102 and the characteristic of the reference block 102 is not detected in the output of the determination unit 104.
  • the determination unit 104 may have a function of instructing the characteristic adjustment unit 107 whether or not to adjust the characteristic of the signal input to the characteristic adjustment unit 107.
  • the characteristic adjustment amount performed by the characteristic adjustment unit 107 can be the deterioration amount of the test block 102.
  • the characteristic adjustment unit 107 may be disposed between the reference block 103 and the determination unit 104.
  • the characteristic adjustment amount of the characteristic adjustment unit 107 in a state where the difference between the characteristic of the test block 102 and the characteristic of the reference block 103 is not detected in the output of the determination unit 104 is set as the deterioration amount of the test block 102. it can.
  • the characteristic adjusting means 107 may be provided both between the test block 102 and the determination means 104 and between the reference block 103 and the determination means 104.
  • the difference between the characteristic adjustment amounts of the two characteristic adjustment means 107 in a state where the difference between the characteristic of the test block 102 and the characteristic of the reference block 103 is not detected in the output of the determination means 104 is the deterioration of the test block 102. It can be an amount.
  • the deterioration diagnosis block 106 may have a function of outputting the characteristic adjustment amount of the characteristic adjustment unit 107 to the outside.
  • the deterioration diagnosis block 106 determines that the determination unit 104 determines the test block 102. It may be determined that is not deteriorated. In this case, the determination unit 104 does not need to instruct the characteristic adjustment unit 107 to adjust the signal characteristic. Therefore, the determination unit 104 may determine that the test block 102 has not deteriorated because an instruction to adjust the characteristic of the signal to the characteristic adjustment unit 107 is unnecessary.
  • the operation states of the test block and the reference block can be arbitrarily set.
  • the test block may be operated even in the standby mode, and the reference block may be stopped.
  • the above-mentioned “characteristic” is a predetermined observable electric characteristic provided for signals output from the test block 102 and the reference block 103, and can be used for deterioration diagnosis of a semiconductor integrated circuit. Anything is acceptable. Therefore, “characteristics” include the output voltage, output current, voltage amplitude, current amplitude, and the ability to drive the load with respect to the phase, change timing, frequency, and signal magnitude aspects of the time aspect of the signal. included.
  • the determination unit 104 may include a phase comparator for detection of a signal phase, change timing, and frequency difference.
  • the determination means 104 may include a voltage comparator for detecting a difference in signal voltage.
  • the difference between the currents of the signals output from the test block 102 and the reference block 103 may be obtained, for example, from the difference between the voltages generated by the respective signals in the resistor.
  • the ability of a signal to drive a load may be obtained from, for example, the amount of change in voltage or current generated in the signal when a predetermined load is connected to the signal.
  • the first embodiment described above measures the deterioration state of a semiconductor integrated circuit with a simple configuration.
  • the same applies to the test block 102 and the reference block 103 even when receiving environmental fluctuations (for example, power supply voltage and ambient temperature) during measurement.
  • environmental fluctuations for example, power supply voltage and ambient temperature
  • BTI Bias Temperature Instability
  • the determination unit 104 determines the deterioration state based on signals output from the test block 102 and the reference block 103.
  • the test block 102 may operate even in the standby mode in which the control unit 105 does not output a signal indicating the measurement mode.
  • the reference block 103 may stop operating.
  • the deterioration diagnosis circuit 100 according to the first embodiment can continuously shift from the standby state to the measurement state while maintaining a state in which the test block 102 is stressed. For this reason, the deterioration diagnosis circuit of the first embodiment also has an effect of reducing the possibility that the deterioration amount is erroneously determined due to the influence of BTI.
  • FIG. 3 is a diagram illustrating a configuration of a deterioration diagnosis circuit according to the second embodiment.
  • the deterioration diagnosis circuit 200 shown in FIG. 3 includes a deterioration diagnosis block 220 and a control unit 211.
  • the deterioration diagnosis block 220 includes a test block 201, a reference block 202, and a determination unit 203. It should be noted that elements and explanations that are not necessary for the description of the operation are omitted as appropriate, and the configuration shown in FIG. 3 is not limited to the only configuration that can achieve the object of the present invention.
  • the test block 201 and the reference block 202 shown in FIG. 3 are composed of logic gates.
  • the test block 201 is configured as a ring oscillator in which the outputs of N NAND (not AND) gates 2011, 2012,.
  • the potentials of input terminals (not shown) of the NAND gates 2011, 2012,... 201N are fixed to the “H” level. Accordingly, each of the NAND gates 2011,... 201N operates as an inverter.
  • a NOR (not OR) gate or another type of logic gate may be used as the logic gate.
  • the number of stages of logic gates is not particularly limited as long as the oscillation conditions of the ring oscillator are satisfied. However, the smaller the number of logic gates, the smaller the area of the deterioration diagnosis circuit.
  • the reference block 202 includes a ring oscillator including N NAND gates 2021, 2022,... 202N.
  • the output of the frequency divider 204 is inverted and input to the input of the NAND gate 2021 that does not constitute a loop.
  • the potentials of input terminals (not shown) of the NAND gates 2022,... 202N are fixed to the “H” level. Therefore, NAND gates 2022,... 202N each operate as an inverter.
  • the determination unit 203 includes a frequency divider 204, a DFF (D-flop flop) group 205, a switch group 206, an EXOR (exclusive-OR) group 207, an inverter group 212, and a deterioration amount calculation unit 214.
  • the frequency divider 204 divides a signal input from any one of the branch points between the logic gates included in the test block 201 (hereinafter, referred to as “node”), and thereby the DFF group 205 and the reference block 202.
  • the switch group 206 includes N switches 2061, 2062,... 206N, and inputs the output of one node selected from the nodes of the reference block 202 to the inverter group 212.
  • the inverter group 212 includes three NAND gates 2121 to 2123 connected in series. The potentials of input terminals (not shown) of the NAND gates 2121 to 2123 are fixed to the “H” level. Accordingly, the NAND gates 2121 to 2123 each operate as an inverter. The inverter group 212 inverts the signal from the node selected by the switch group 206 every time it passes through the NAND gates 2121 to 2123. Outputs from the NAND gates 2121 to 2123 are input to the DFF group 205.
  • the DFF group 205 includes DFFs 2051 to 2053.
  • the DFFs 2051 to 2053 hold the signals respectively input from the NAND gates 2121 to 2123 at the rising edge from the frequency divider 204.
  • the EXOR group 207 includes EXOR gates 2071 and 2072.
  • the EXOR gate 2071 performs an exclusive OR operation on the outputs of the DFFs 2051 and 2052 and outputs the result as an output 208 to the deterioration amount calculation means 214.
  • the EXOR gate 2072 performs an exclusive OR operation on the outputs of the DFFs 2052 and 2053 and outputs the result to the deterioration amount calculation unit 214 as an output 209.
  • the deterioration amount calculation unit 214 calculates the deterioration amount of the test block 201 based on the outputs 208 and 209.
  • a switch 213 connects and disconnects the node of the test block 201 and the frequency divider 204.
  • the switch 213 is controlled by a control signal 210 output from the control unit 211.
  • the control signal 210 can connect or disconnect between the node of the test block 201 and the frequency divider 204.
  • a state in which the deterioration state is not measured is referred to as a standby mode
  • a state in which the deterioration state is measured is referred to as a measurement mode.
  • operations in the standby mode and the measurement mode will be described.
  • the control unit 211 outputs a control signal 210 so that the switch 213 disconnects the test block 201 and the frequency divider 204. Then, the test block 201 self-oscillates as a ring oscillator.
  • the test block 201 may be configured so that the operating condition of the ring oscillator can be controlled by an external control signal (not shown).
  • the operation of the reference block 202 is the same as that of the test block 201. Due to the difference in stress (voltage, temperature, etc.) applied to the test block 201 and the reference block 202, the deterioration status of the test block 201 under different stress conditions can be evaluated.
  • a specific node of the test block 201 and the frequency divider 204 are connected by the control signal 210. Transition from the standby mode to the measurement mode is performed by outputting a control signal 210 so that the control unit 211 closes the switch 213. That is, the only difference in the circuit configuration between the standby mode and the measurement mode is whether the switch 213 is open or closed. Therefore, the deterioration diagnosis circuit 200 can shift from the standby mode to the measurement mode without changing the circuit configuration of each block while applying stress to the test block 201 and the reference block 202.
  • the signal of the node of the test block 201 is input to the frequency divider 204.
  • the frequency divider 204 divides the signal input from the test block 201 by a predetermined frequency division ratio. As a result, the output of the frequency divider 204 alternately changes between the “H” level and the “L” level in a cycle corresponding to the frequency division ratio. For example, when the output signal level of the frequency divider 204 immediately before entering the measurement mode is “H”, the output of the frequency divider 204 subsequently changes in the order of “L” and “H”. First, the operation when the output of the frequency divider 204 changes to “L” will be described. The output of the frequency divider 204 is inverted and input to the input of the NAND gate 2021.
  • the output of the NAND gate 2021 is fixed to “H”.
  • the reference block 202 starts oscillating.
  • any one of the switches 2061 to 206N is selected. This selection procedure will be described later.
  • the output of the frequency divider 204 changes to “H”.
  • the output level of the node selected by the reference block 202 is held at the output of the DFF group 205.
  • the output level held by the DFF group 205 is logically operated by the EXOR group 207 at the subsequent stage and output as an output 208 and an output 209.
  • the oscillation operation of the reference block 202 stops when the output of the frequency divider 204 changes from “L” to “H”. Since the stop of the oscillation operation and the latching of the input data from the switch group 206 in the DFF group 205 are performed at the same time, the output signal of the DFF group 205 may be disturbed.
  • a delay circuit may be provided between the frequency divider 204 and the reference block 202 so that the DFF group 205 can correctly latch the input signal when the output of the frequency divider 204 changes.
  • a circuit that holds the signal output from the frequency divider 204 to the reference block 202 until the DFF group 205 latches the input may be provided.
  • the oscillation start and stop of the reference block 202 may be controlled separately from the output of the frequency divider 204 using the control signal 210.
  • a method for calculating the deterioration state in the deterioration amount calculating unit 214 will be described.
  • the test block 201 is deteriorated.
  • the deterioration state can be relatively calculated even when the operating environment of at least one of the test block 201 and the reference block 202 is changed.
  • the relative deterioration state can be calculated in the same manner.
  • the connection between the test block 201 and the frequency divider 204 is described based on the configuration of FIG.
  • the delay amount of the signal output from the test block is measured by the same operation.
  • the amount of deterioration of the test block can be obtained.
  • the interval at which the output of the frequency divider 204 changes when the test block 201 and the frequency divider 204 are used is expressed by the following equation (1).
  • T is a delay amount per stage of the NAND gate of the test block 201
  • ⁇ T is a delay amount that increases due to deterioration per stage of the NAND gate of the test block 201.
  • N is the number of NAND gate stages of the test block 201
  • M is the frequency division ratio of the frequency divider 204. That is, the value of Equation (1) is the total sum of delay amounts generated in the test block at intervals at which the output of the frequency divider 204 changes. Since the internal circuits of the NAND gates 2011 to 201N are the same, it is assumed that the increase in delay amount at the time of performance deterioration in the NAND gates 2011 to 201N is all equal. On the other hand, the delay amount represented by the following expression (2) is obtained in the reference block 202 as well. T ⁇ X (2) Expression (2) is a delay amount output from the reference block 202 while the output of the frequency divider 204 changes.
  • X is the total number of stages of NAND gates through which the signal in the reference block 202 has passed from the fall of the output of the frequency divider 204 to the next rise.
  • the reference block 202 also requires a counter and a large amount of DFF. Therefore, in the second embodiment, in order to simplify and reduce the area of the deterioration detection circuit, the difference between the delay amount of the test block 201 and the delay amount of the reference block 202 is obtained without directly obtaining X by a counter or the like. Measuring. Therefore, the deterioration detection circuit 220 selects a node of the reference block 202 to be input to the DFF group at the start of the measurement mode. The specific procedure will be described below.
  • the switch group 206 selects a certain node until the output of the frequency divider 204 rises. Then, a node in which the levels of the output 208 and the output 209 at the time point when the output of the frequency divider 204 falls is searched for. That is, a node where the output 208 is “L” and the output 209 is “H” or the output 208 is “H” and the output 209 is “L” is searched.
  • the opening / closing control of the switch group 206 and the investigation of the levels of the outputs 208 and 209 may be performed by the determination unit 203 or may be performed by another device under the control of the determination unit 203 from the outside.
  • FIG. 5 is a diagram illustrating signal timing in the second embodiment. In FIG.
  • the output of the frequency divider 204 is a signal obtained by dividing the output of the test block 201.
  • Signals A to C and signals A ′ to C ′ indicate inputs and outputs of the DFFs 2051 to 2053, respectively.
  • FIG. 5 shows a timing chart when one of the switches 2061 to 206N in which the output 208 is “L” and the output 209 is “H” is selected and closed at the start of the measurement mode.
  • the circuit configuration of the test block 201 and the reference block 202 as ring oscillators is the same. Therefore, when the test block 201 is not deteriorated as compared with the reference block 202, it is between the output of the frequency divider 201 and the signals (A to C) input from the reference block to the DFF group.
  • the timing does not change ((a) of FIG. 5).
  • the rise timing of the frequency divider 204 is delayed with respect to the input signals (A to C) to the DFF group.
  • the output signal B ′ of the DFF 2052 is inverted.
  • the outputs 208 and 209 are inverted, and the output 208 becomes “H” and the output 209 becomes “L” ((b) in FIG. 5).
  • the timing of the input signals A to C to the DFF group 205 is delayed by one NAND gate of the inverter group 212 in the order of A, B, and C.
  • the rise timing of the frequency divider 204 is delayed from t1 to t2 across the inversion time of the output signal B ′ of the DFF 2052, the delay amount from the fall of the frequency divider 204 to the rise is averaged.
  • the number of gate stages through which the signal in the test block 201 has passed is “2 ⁇ N ⁇ M”.
  • the deterioration amount calculation unit 214 determines the deterioration amount per gate stage in the test block 201. Can be obtained as 1 / (2 ⁇ N ⁇ M).
  • the determination unit 203 may select the node again by the switch group 206 after both the output 208 and the output 209 are inverted and the output 208 becomes “H” and the output 209 becomes “L”. For example, the determination unit 203 may shift the node selected in the reference block 202 backward by one in a state where the rising timing of the frequency divider 204 is delayed until t2 in FIG.
  • Signals A to C when the node is shifted backward by one are indicated by broken lines in FIG.
  • the relationship between the signals indicated by broken lines in FIG. 5 is that the signal B in FIG. 5 is a new signal A, and the signals obtained by sequentially inverting them are B and C.
  • the relationship between the rising timing of the frequency divider 204 and the input signals A to C to the DFF group 205 becomes the same as that at t1 in FIG.
  • the node selected by the switch group 206 is shifted backward, whereby the output 208 can be set to “L” and the output 209 can be set to “H”. .
  • the deterioration amount calculating means 214 further proceeds with the deterioration by 2 / (2 ⁇ N XM) can be calculated. Thereafter, by sequentially shifting the selected nodes backward in accordance with the delay of the rise timing of the frequency divider 204, the deterioration amount calculating unit 214 can calculate the deterioration state of the test block with a simple circuit configuration. . In selecting a node, the timing at which the output 208 becomes “L” and the output 209 becomes “H” may be not only t1 in FIG. 5 but also t3.
  • the deterioration amount calculating means 214 can measure the deterioration amount of the test block 201 in the same procedure as described above. That is, when the node is selected, the deterioration amount calculating means 214 is not limited to the test block whether the output 208 is “L” and the output 209 is “H” at t1 or t4 in FIG. The amount of degradation 201 can be calculated.
  • the deterioration amount calculating unit 214 can measure the progress of deterioration by sequentially shifting the nodes of the reference block 202 that inputs signals to the determining unit 203 backward. That is, the deterioration diagnosis circuit of the second embodiment measures the delay amount generated in the test block by comparing the output of the test block and the reference block without using a counter or a large number of DFFs. ing. As a result, the degradation diagnosis circuit of the second embodiment has an effect that the degradation state of the semiconductor integrated circuit can be measured more accurately with a simple circuit configuration. Further, the deterioration diagnosis circuit of the second embodiment can continuously shift from the standby mode to the measurement mode by the control signal 210 without changing the circuit configuration of the test block 201 and the reference block 202.
  • the deterioration diagnosis circuit of the second embodiment also has an effect of reducing the possibility that the deterioration amount is erroneously determined due to the influence of BTI.
  • the determination unit 203 may include a characteristic adjustment unit as in the modification of the first embodiment.
  • the characteristic adjusting unit adjusts the delay amount of the input signal so that the deterioration amount of the test block 201 approaches 0 in the deterioration amount calculating unit 214.
  • the characteristic adjustment means may be provided between at least one of the output of the test block 201 and the input of the frequency divider 204 and between the output of the frequency divider 204 and the input of the NAND gate 2021.
  • the deterioration amount of the test block 201 is obtained based on the delay adjustment amount of the characteristic adjustment unit when the deterioration amount calculation unit 214 does not detect the deterioration of the test block 201. it can. Furthermore, in the modification of the second embodiment, the determination unit 203 may have a function of outputting the characteristic adjustment amount of the characteristic adjustment unit to the outside.
  • the configuration of the second embodiment shown in FIG. 3 is a configuration that compares the signals A to C delayed by one stage by the inverter group 212 with the rising timing of the frequency divider 204. That is, the resolution of the delay amount of the test block 201 to be detected is fixed to one gate.
  • FIG. 4 is a diagram illustrating a configuration of a deterioration diagnosis circuit according to the third embodiment.
  • the deterioration diagnosis circuit 300 shown in FIG. 4 includes a deterioration diagnosis block 320 and a control means 311.
  • the deterioration diagnosis block 320 includes a test block 301, a reference block 302, and a determination unit 303. It should be noted that elements and descriptions unnecessary for the description of the operation are omitted as appropriate, and the configuration shown in FIG. 4 is clearly not limited to the only configuration that can achieve the object of the present invention. .
  • the test block 301 and the reference block 302 shown in FIG. 4 are composed of logic gates.
  • the test block 301 is a ring oscillator composed of NAND gates 3011 to 301N
  • the reference block 302 is a ring oscillator composed of NAND gates 3021 to 302N.
  • the determination unit 303 includes a frequency divider 304, a switch group 306, a DFF group 305, an EXOR group 307, and a deterioration amount calculation unit 314.
  • the frequency divider 304 divides a signal input from any node of the test block 301 and outputs the divided signal to the DFF group 305 and the NAND gate 3021 of the reference block 302.
  • the switch group 306 branches each node of the reference block 302 into three and connects them to DFFs 3051 to 3052 constituting the DFF group 305.
  • the output of the DFF group 305 latches the output from the three nodes selected by the switch group 306 at the rise from the frequency divider 304.
  • the EXOR group 307 performs two exclusive OR operations among the three outputs from the DFF group 305.
  • Outputs 308 and 309 are outputs of EXORs 3071 and 3072, respectively, and are output to the deterioration amount calculation means 314.
  • the deterioration amount calculation unit 314 calculates the deterioration amount of the test block 301 based on the outputs 308 and 309.
  • the switch group 306 includes three switches for each node of the reference block 302.
  • switches 3061a to 3061c, 3062a to 3062c,... 306Na to 306Nc constitute a switch group 306.
  • a switch in the switch group 306 is closed (conducts)
  • the output of the node is connected to any DFF in the DFF group 305.
  • the switch 3061a is closed, the output of the NAND gate 302N is input to the DFF 3051 through the signal line A.
  • the switch 3062b or 3062c When the switch 3062b or 3062c is closed, the output of the NAND gate 3021 is input to the DFF 3052 or 3053 through the signal line B or C, respectively.
  • the operation of each switch constituting the switch group 306 is the same for other nodes.
  • the switches 3061a and the like constituting the switch group 306 are individually controlled. With this configuration, the wiring of the signals A to C can be connected to any node of the reference block 302. That is, the delay amount between the signals A and B and the delay amount between the signals B and C can be set in units of the number of NAND gates. As a result, in the third embodiment, it is possible to change the resolution for detecting the change in the rise timing of the output of the frequency divider 204 described in FIG.
  • the determination unit 303 can be configured so that a signal that causes a delay of one gate stage is input to the DFF group 305 by closing the switch 3061 a, the switch 3062 b, and the switch 3063 c.
  • the switches 3064b and 3067c are separated by 3 stages by a NAND gate, respectively, so that the DFF group 305 has a delay of 3 stages of NAND gates. The signal that caused the error is input.
  • the delay between the change points of the signals A and B and B and C in FIG. 5 is the length of three NAND gates. Therefore, when the outputs 308 and 309 are inverted, a delay corresponding to three stages of gates is generated in the test block 301. At this time, the number of gate stages through which the signal in the test block 301 has passed is 2 ⁇ N ⁇ M, and it is considered that the delay in the three stages of gates has increased in the test block 301 due to deterioration. For this reason, the deterioration amount calculating means 314 can determine the deterioration amount per gate stage in the test block 301 as 3 / (2 ⁇ N ⁇ M).
  • the oscillation operation of the reference block 302 stops when the output of the frequency divider 304 changes from “L” to “H”. Since the stop of the oscillation operation and the latching of the input data from the switch group 306 in the DFF group 305 are performed at the same time, the output signal of the DFF group 305 may be disturbed.
  • a delay circuit may be provided between the frequency divider 304 and the reference block 302 so that the DEF group 305 can correctly latch the input signal when the output of the frequency divider 304 changes.
  • a circuit that holds the signal output from the frequency divider 304 to the reference block 302 until the DFF group 305 latches the input may be provided.
  • the oscillation start and stop of the reference block 302 may be controlled separately from the output of the frequency divider 304 using the control signal 310.
  • the degradation diagnosis circuit of the third embodiment also has the effect that the degradation state of the semiconductor integrated circuit can be measured with a simple circuit configuration, similarly to the degradation diagnosis circuit of the second embodiment.
  • the measurement unit (resolution) of the deterioration amount of the test block 301 to be measured can be changed. By increasing the measurement unit of the deterioration amount, it is possible to average the delay amount variation due to each NAND gate and the delay amount error due to noise.
  • the deterioration diagnosis circuit of the third embodiment is in a standby mode by the control signal 310 without changing the circuit configuration of the test block 301 and the reference block 302, similarly to the deterioration diagnosis circuit of the second embodiment. It is possible to make a continuous transition from measurement mode to measurement mode. For this reason, the deterioration diagnosis circuit of the third embodiment also has an effect of reducing the influence of recovery of deterioration due to BTI.
  • the determination unit 303 may include a characteristic adjustment unit as in the modification of the first embodiment. The characteristic adjustment unit adjusts the delay amount of the input signal so that the deterioration amount of the test block 301 approaches 0 in the deterioration amount calculation unit 314.
  • the characteristic adjustment means may be provided between at least one of the output of the test block 301 and the input of the frequency divider 304 and between the output of the frequency divider 304 and the input of the NAND gate 3021.
  • the deterioration amount of the test block 301 is obtained based on the delay adjustment amount of the characteristic adjustment means when the deterioration amount calculation means 314 does not detect the deterioration of the test block 301. it can.
  • the determination unit 303 may have a function of outputting the characteristic adjustment amount of the characteristic adjustment unit to the outside.
  • a test block including a first circuit to be diagnosed for deterioration;
  • a reference block including a second circuit having the same configuration as the first circuit;
  • Determination means for determining the presence or absence of the deterioration of the elements constituting the test block,
  • Control means for outputting a signal indicating the measurement mode to the determination means;
  • a deterioration diagnosis circuit comprising: (Appendix 2) The deterioration diagnosis circuit according to appendix 1, wherein the first characteristic and the second characteristic are electrical characteristics that are changed as a result of the deterioration, which are included in the first signal and the second signal, respectively.
  • the first characteristic and the second characteristic each include at least one of a phase, a change timing, a frequency, an output voltage, an output current, a voltage amplitude, a current amplitude, and an ability to drive a load.
  • (Appendix 4) 4. The deterioration diagnosis circuit according to any one of appendices 1 to 3, wherein the determination unit detects a difference between a characteristic of an element constituting the test block and a characteristic of an element constituting the reference block.
  • Appendix 5 When a signal indicating the measurement mode is input, at least one of the first signal and the second signal according to an instruction from the determination unit so that the first characteristic and the second characteristic are equal to each other.
  • the determination means determines the presence or absence of the deterioration depending on the necessity of the instruction.
  • the deterioration diagnosis circuit according to any one of appendices 1 to 3. (Appendix 6) The deterioration diagnosis circuit according to appendix 5, wherein the determination unit outputs an adjustment amount necessary for performing the adjustment as a deterioration amount indicating the degree of deterioration. (Appendix 7) 7. The deterioration diagnosis circuit according to any one of appendices 1 to 6, wherein one or both of the first circuit and the second circuit has a circulation configuration in which a signal is output at a predetermined timing. (Appendix 8) 8.
  • Appendix 9 The deterioration diagnosis circuit according to any one of appendices 1 to 8, wherein the determination unit compares a delay time of an element constituting the test block with a delay time of an element constituting the reference block.
  • the determination means includes A frequency divider that outputs a third signal obtained by dividing the first signal; A latch circuit that latches a fourth signal and a fifth signal having different delay amounts with respect to the second signal by the third signal and outputs the latched signal, respectively; An EXOR (exclusive-OR) circuit for obtaining an exclusive OR of the outputs of the latch circuit; A deterioration amount calculating means for calculating the deterioration amount based on an output of the EXOR circuit; A deterioration diagnosis circuit according to any one of appendices 1 to 9. (Appendix 11) The deterioration diagnosis circuit according to appendix 10, wherein the fourth signal and the fifth signal are generated by delaying the second signal.
  • the fourth signal and the fifth signal are the second signal output from positions corresponding to the preceding stage and the succeeding stage in the propagation path of the predetermined signal of the reference block.
  • Deterioration diagnosis circuit (Appendix 13) The deterioration diagnosis circuit according to any one of appendices 1 to 12, wherein when the control means does not output a signal indicating the measurement mode, the test block operates and the reference block stops operating. (Appendix 14) 14. The deterioration diagnosis circuit according to any one of appendices 1 to 13, wherein the test block and the reference block operate in an operating environment in which stresses applied to the test block and the reference block are different.

Abstract

In order to more accurately measure a deterioration state of a semiconductor integrated circuit by means of a simpler configuration, a deterioration diagnostic circuit of the present invention is provided with: a block to be tested, which includes a first circuit to be subjected to deterioration diagnosis; a reference block, which includes a second circuit that is provided with a configuration identical to that of the first circuit; a determining means, which determines whether elements constituting the block to be tested are deteriorated or not by comparing characteristics of first signals, which are outputted from the block to be tested in the cases where signals indicating measurement mode are inputted, with characteristics of second signals outputted from the reference block; and a control means, which outputs, to the determining means, the signals that indicate the measurement mode.

Description

劣化診断回路および劣化診断方法Deterioration diagnosis circuit and deterioration diagnosis method
 本発明は、半導体集積回路の劣化診断回路および劣化診断方法に関する。 The present invention relates to a deterioration diagnosis circuit and a deterioration diagnosis method for a semiconductor integrated circuit.
 半導体集積回路は製造後の使用や環境条件などの要因によってその特性が劣化してゆく。そして、半導体集積回路の劣化が一定以上進行した場合には、その半導体集積回路は不良と判定される。しかしながら、半導体集積回路が不良と判定された場合に、不良となった原因が寿命(すなわち経年劣化による妥当な劣化)によるものか、あるいは偶発的な故障等によるものかを見極めることは一般には困難である。
 従って、出荷後の半導体集積回路の劣化の進行度を知ることは、半導体集積回路の不良の真の原因を特定するために極めて重要である。半導体集積回路の劣化の進行度を知ることにより、想定外の故障や予想以上の劣化が観測された場合の劣化の進行度を半導体集積回路の設計へ容易にフィードバックすることができる。また、劣化の進行度のログ情報を取得して解析し半導体集積回路の適切な交換時期を予想することで、半導体集積回路の最適なメンテナンス時期の設定にも役立てることができる。
 半導体集積回路の劣化の進行度の情報は、外部測定器を用いることなく半導体集積回路の外部へ出力でき、さらには保存できることが望ましい。そのための手段として、半導体集積回路内にリング発振器を構成しその発振周波数の変化を測定することで半導体集積回路の劣化の進行度を算出する技術が知られている。
 非特許文献1は、2台のCMOS(Complementary Metal Oxide Semiconductor)リング発振器を用いることによって実用域における劣化度の分解能を向上させ、測定時間を短縮する構成を開示している。具体的には、非特許文献1は、2つのリング発振器を用いてその発振周波数の比率を測定する構成を記載している。
 特許文献1は、2台のCMOSリング発振器を用いてMOSFET(Metal−Oxide−Semiconductor Field−Effect Transistor)の劣化を測定するための構成を開示している。特許文献1に記載された半導体集積回路は、2個のリング発振器を備え、劣化の測定時には自回路の一部を分岐して他方の回路の一部と置き換えるように接続する。そして、自回路と他方の回路との出力の位相を比較することにより、回路の特性劣化による信号の遅延量の変化を検出する。
 また、特許文献2は、劣化診断対象回路の信号伝播時間が基準値に対して早いか遅いかにより回路が故障しているかどうかを判定する構成を開示している。特許文献2に記載された劣化診断回路は、劣化診断回路のみを用いて回路の劣化診断を行う。
 さらに、特許文献3は、信号遅延パスの経年劣化を検出するための構成を開示している。特許文献3に記載された半導体集積回路装置は、出荷前に異なる遅延パス同士の遅延量の比較結果をメモリに書き込み、出荷後に同様の測定を行って遅延量の大小関係が異なれば遅延パスに故障が発生していると判断する。
The characteristics of a semiconductor integrated circuit deteriorate due to factors such as use after manufacture and environmental conditions. When the deterioration of the semiconductor integrated circuit proceeds more than a certain level, the semiconductor integrated circuit is determined to be defective. However, when it is determined that the semiconductor integrated circuit is defective, it is generally difficult to determine whether the cause of the failure is due to lifetime (ie, reasonable deterioration due to aging deterioration) or accidental failure. It is.
Therefore, knowing the degree of progress of deterioration of the semiconductor integrated circuit after shipment is extremely important for identifying the true cause of the failure of the semiconductor integrated circuit. By knowing the progress of deterioration of the semiconductor integrated circuit, it is possible to easily feed back the progress of deterioration when an unexpected failure or unexpected deterioration is observed to the design of the semiconductor integrated circuit. Further, by acquiring and analyzing log information of the degree of progress of deterioration and predicting an appropriate replacement time of the semiconductor integrated circuit, it can be used for setting an optimal maintenance time of the semiconductor integrated circuit.
It is desirable that the information on the degree of progress of deterioration of the semiconductor integrated circuit can be output to the outside of the semiconductor integrated circuit without using an external measuring instrument, and further stored. As a means for that purpose, a technique is known in which a ring oscillator is configured in a semiconductor integrated circuit, and the degree of deterioration of the semiconductor integrated circuit is calculated by measuring changes in the oscillation frequency.
Non-Patent Document 1 discloses a configuration in which the resolution of the deterioration degree in the practical range is improved and the measurement time is shortened by using two CMOS (Complementary Metal Oxide Semiconductor) ring oscillators. Specifically, Non-Patent Document 1 describes a configuration in which the ratio of the oscillation frequency is measured using two ring oscillators.
Patent Document 1 discloses a configuration for measuring deterioration of a MOSFET (Metal-Oxide-Semiconductor Field-Effect Transistor) using two CMOS ring oscillators. The semiconductor integrated circuit described in Patent Document 1 includes two ring oscillators, and is connected so that a part of its own circuit is branched and replaced with a part of the other circuit when measuring deterioration. Then, by comparing the output phases of the own circuit and the other circuit, a change in the delay amount of the signal due to the deterioration of the circuit characteristics is detected.
Patent Document 2 discloses a configuration for determining whether or not a circuit has failed depending on whether the signal propagation time of the degradation diagnosis target circuit is earlier or later than the reference value. The deterioration diagnosis circuit described in Patent Document 2 performs circuit deterioration diagnosis using only the deterioration diagnosis circuit.
Further, Patent Document 3 discloses a configuration for detecting aged deterioration of a signal delay path. The semiconductor integrated circuit device described in Patent Document 3 writes a comparison result of delay amounts of different delay paths to a memory before shipment, performs the same measurement after shipment, and changes to a delay path if the magnitude relationship of the delay amounts differs. Judge that a failure has occurred.
特開2010−087275号公報JP 2010-087275 A 特開2008−147245号公報JP 2008-147245 A 特開2010−287860号公報JP 2010-287860 A
 非特許文献1に記載されている、リング発振器の発振周波数の変化を用いて半導体集積回路の劣化の進行度を判定する方法には、測定される発振周波数が環境の影響を受けて大きく変動するという問題がある。環境が発振周波数に与える影響には、例えば半導体集積回路のチップ温度や電源電圧の変動がある。非特許文献1は、2つのリング発振器を用いてその発振周波数の比率を測定することで環境による影響をある程度緩和している。しかし、非特許文献1の構成は、発振周波数の変化を測定するために2つのリング発振器の差周波の信号を生成している。このため、非特許文献1の構成には、回路構成が複雑になるという課題がある。
 また、特許文献1に記載された半導体集積回路は、自回路及び他方の回路の一部のみを用いて遅延を比較する。このため、特許文献1に記載された構成は、リング発振器全体の遅延劣化量に基づいた正確な特性劣化量を求めることができないという課題がある。
 さらに、特許文献2に記載された劣化診断回路は、実回路の信号を用いることなく劣化診断を行うため、劣化診断の結果が実回路の状態と整合しない恐れがある。加えて、特許文献2に記載された劣化診断回路には、劣化診断のための高精度な基準信号の生成や複雑な設定が必要であるという課題もある。
 さらに、特許文献3に記載された半導体集積回路装置は、出荷前に遅延値を測定してその値を不揮発性メモリに記録しておくことを必要とする。このため、特許文献3に記載された半導体集積回路装置は、メモリ等の高価な部品の使用や製造時の手順の増加によりコストが上昇するという課題がある。
 本発明の目的は、簡易な構成により半導体集積回路の劣化の状態をより正確に測定するという課題を解決するための技術を提供することにある。
In the method described in Non-Patent Document 1 for determining the degree of deterioration of a semiconductor integrated circuit using a change in the oscillation frequency of a ring oscillator, the measured oscillation frequency varies greatly under the influence of the environment. There is a problem. The influence of the environment on the oscillation frequency includes, for example, fluctuations in the chip temperature and power supply voltage of the semiconductor integrated circuit. Non-Patent Document 1 alleviates the influence of the environment to some extent by measuring the ratio of the oscillation frequencies using two ring oscillators. However, the configuration of Non-Patent Document 1 generates a signal having a difference frequency between two ring oscillators in order to measure a change in oscillation frequency. For this reason, the configuration of Non-Patent Document 1 has a problem that the circuit configuration becomes complicated.
In addition, the semiconductor integrated circuit described in Patent Document 1 compares the delay using only a part of its own circuit and the other circuit. For this reason, the configuration described in Patent Document 1 has a problem that an accurate characteristic deterioration amount based on the delay deterioration amount of the entire ring oscillator cannot be obtained.
Furthermore, since the deterioration diagnosis circuit described in Patent Document 2 performs deterioration diagnosis without using the signal of the actual circuit, the result of the deterioration diagnosis may not match the state of the actual circuit. In addition, the deterioration diagnosis circuit described in Patent Document 2 has a problem that it requires generation of a highly accurate reference signal for deterioration diagnosis and complicated setting.
Furthermore, the semiconductor integrated circuit device described in Patent Document 3 needs to measure a delay value and record the value in a nonvolatile memory before shipment. For this reason, the semiconductor integrated circuit device described in Patent Document 3 has a problem in that the cost increases due to the use of expensive components such as a memory and an increase in manufacturing procedures.
An object of the present invention is to provide a technique for solving the problem of more accurately measuring the state of deterioration of a semiconductor integrated circuit with a simple configuration.
 本発明の劣化診断回路は、劣化の診断の対象となる第1の回路を含む試験用ブロックと、第1の回路と同一の構成を備える第2の回路を含む参照用ブロックと、測定モードを示す信号が入力された場合に試験用ブロックから出力される第1の信号の特性と参照用ブロックから出力される第2の信号の特性とを比較することで試験用ブロックを構成する素子の劣化の有無を判定する判定手段と、測定モードを示す信号を判定手段に出力する制御手段と、を備える。
 本発明の劣化診断方法は、測定モードを示す信号が入力された場合に、劣化の診断の対象となる第1の回路を含む試験用ブロックから出力される第1の信号の特性と第1の回路と同一の構成を備える第2の回路を含む参照用ブロックから出力される第2の信号の特性とを比較することで試験用ブロックを構成する素子の劣化の有無を判定する。
The deterioration diagnosis circuit of the present invention includes a test block including a first circuit to be subjected to deterioration diagnosis, a reference block including a second circuit having the same configuration as the first circuit, and a measurement mode. Deterioration of the elements constituting the test block by comparing the characteristics of the first signal output from the test block with the characteristics of the second signal output from the reference block when the signal shown in FIG. Determination means for determining the presence or absence of the control, and control means for outputting a signal indicating the measurement mode to the determination means.
The degradation diagnosis method of the present invention is characterized in that when a signal indicating a measurement mode is input, the characteristics of the first signal output from the test block including the first circuit to be diagnosed for degradation and the first By comparing the characteristics of the second signal output from the reference block including the second circuit having the same configuration as the circuit, the presence or absence of deterioration of the elements constituting the test block is determined.
 本発明は、簡易な構成により半導体集積回路の劣化の状況をより正確に測定することを可能にする。 The present invention makes it possible to more accurately measure the state of degradation of a semiconductor integrated circuit with a simple configuration.
第1の実施形態の劣化診断回路の構成を示す図である。It is a figure which shows the structure of the deterioration diagnostic circuit of 1st Embodiment. 第1の実施形態の劣化診断回路の変形例の構成を示す図である。It is a figure which shows the structure of the modification of the deterioration diagnostic circuit of 1st Embodiment. 第2の実施形態の劣化診断回路の構成を示す図である。It is a figure which shows the structure of the deterioration diagnostic circuit of 2nd Embodiment. 第3の実施形態の劣化診断回路の構成を示す図である。It is a figure which shows the structure of the deterioration diagnostic circuit of 3rd Embodiment. 第2の実施形態における、DFFへの入出力信号及びEX−ORの出力信号の論理を示す図である。It is a figure which shows the logic of the input-output signal to DFF and the output signal of EX-OR in 2nd Embodiment.
 次に、本発明の実施形態について図面を参照して説明する。
(第1の実施形態)
 図1は、本発明の第1の実施形態による劣化診断回路の構成を示す図である。
 図1に示すように、劣化診断回路100は、劣化診断ブロック101と制御手段105とを備える。そして、劣化診断ブロック101は、試験用ブロック102、参照用ブロック103及び判定手段104を備える。
 試験用ブロック102にストレスが印加された結果、試験用ブロックの回路を構成する半導体集積回路に劣化が生じると、半導体集積回路を構成する論理ゲートの特性の一種である遅延が増大する場合がある。後述のように、試験用ブロック102と参照用ブロック103とでは、ストレスの量に差が設けられる。ストレスの量とは、例えば、ストレスの強度又はストレスが印加される期間の長さである。そのため、試験用ブロック102と参照用ブロック103とでは、劣化の大きさ(劣化度)に差が生じる。従って、試験用ブロック102の半導体集積回路の劣化により、論理ゲートに参照用ブロック103よりも大きな遅延が生じる。その結果、参照用ブロック103が出力する信号に対して、試験用ブロック102が出力する信号に次第に遅れが生じる。判定手段104は、この遅れを検出して試験用ブロック102の劣化状態を判定する。
 判定手段104は、待機モードと測定モードとの2つの動作状態を備える。待機モードは、劣化診断ブロック101において試験用ブロック102の劣化の測定が行われない状態である。測定モードは、劣化診断ブロック101において試験用ブロック102の劣化の測定が行われる状態である。
 試験用ブロックは、劣化診断の対象となる第1の回路を含む。参照用ブロックは、第1の回路と同一の構成を備える第2の回路を含む。試験用ブロック102及び参照用ブロック103は、それぞれ第1の信号及び第2の信号を判定手段104に出力する。制御手段105は、待機モード又は測定モードを示す信号を判定手段104に出力する。
 判定手段104は、制御手段105から入力された信号が測定モードを指示している場合には、以下の動作を行う。すなわち、測定モードにおいて、判定手段104は、試験用ブロック102から入力された第1の信号及び参照用ブロック103から入力された第2の信号を比較し、試験用ブロック102を構成する素子の劣化の有無を判定する。
 試験用ブロック102及び参照用ブロック103には、それぞれの半導体集積回路の劣化を進行させるためのストレスを印加することが可能である。ストレスの印加は、電源電圧の印加の有無、電源電圧の変更、周囲温度の変更、動作周波数の変更、回路の負荷の変更、等によって可能である。しかし、ストレスの種類はこれらに限られない。半導体集積回路の寿命に影響を与える要因であれば、これら以外のストレスを試験用ブロック102及び参照用ブロック103の一方あるいは両方に印加してもよい。
 また、複数の種類のストレスを試験用ブロック102及び参照用ブロック103に同時に印加してもよい。このとき、印加するストレスの種類及び強さを試験用ブロック102と参照用ブロック103とで異なるようにすることが好ましい。
 例えば、待機モードにおいて、試験用ブロック102にストレスを印加して劣化を進行させる状態に設定するとともに、参照用ブロック103の半導体集積回路の電源端子を接地電位に接続することによりストレスがかからない状態、すなわち劣化が進行しない状態としてもよい。あるいは、待機状態においても試験用ブロック102および参照用ブロック103の両方にストレスをかけ、その際に温度や動作周波数などの動作条件に差分を設けてもよい。なお、試験用ブロック102と参照用ブロック103の位置は、入れ替わってもよい。
 一方、測定モードにおいては、判定手段104は、試験用ブロック102の出力と参照用ブロック103の出力との2つの信号のタイミングを比較し、その差分を求める。
 具体的な構成として、試験用ブロック102および参照用ブロック103は、発振回路を備え、その発振回路の出力を判定手段104に出力してもよい。
 判定手段104は、試験用ブロック102及び参照用ブロック103から出力される信号の特性を比較し、試験用ブロック102の劣化状態を判定する。試験用ブロック102および参照用ブロック103が発振回路を備える場合であれば、信号の特性として、発振周波数や信号の位相を用いればよい。
 図2は、第1の実施形態の劣化診断回路の変形例の構成を示す図である。図2に示す劣化診断ブロック106は、試験用ブロック102と判定手段104との間に特性調整手段107を備える。特性調整手段107は、試験用ブロック102から出力される信号の特性を調整して判定手段104に入力する。判定手段104の出力は、特性調整手段107に入力される。特性調整手段107は、判定手段104の出力において試験用ブロック102の特性と参照用ブロック102の特性の差が検出されないように試験用ブロック102の出力の特性(例えば遅延時間)を調整する。なお、判定手段104は、特性調整手段107に入力される信号の特性の調整を行うか否かを特性調整手段107に指示する機能を備えていてもよい。図2に示す第1の実施形態の変形例では、特性調整手段107が実施した特性の調整量を、試験用ブロック102の劣化量とすることができる。
 なお、特性調整手段107は、参照用ブロック103と判定手段104との間に配置されていてもよい。この場合も、判定手段104の出力において試験用ブロック102の特性と参照用ブロック103の特性の差が検出されない状態における特性調整手段107の特性調整量を試験用ブロック102の劣化量とすることができる。さらに、特性調整手段107を試験用ブロック102と判定手段104との間、及び、参照用ブロック103と判定手段104との間の両方に備えていてもよい。この場合には、判定手段104の出力において試験用ブロック102の特性と参照用ブロック103の特性の差が検出されない状態における2つの特性調整手段107の特性調整量の差を試験用ブロック102の劣化量とすることができる。加えて、第1の実施形態の変形例においては、劣化診断ブロック106は、特性調整手段107の特性調整量を外部に出力する機能を備えていてもよい。
 ここで、劣化診断ブロック106は、特性調整手段107による信号の特性の調整を行うことなく試験用ブロック102の特性と参照用ブロック103の特性が等しい場合には、判定手段104は試験用ブロック102が劣化していないと判断してもよい。この場合、判定手段104は特性調整手段107に対して信号の特性の調整を指示する必要はない。従って、判定手段104は、特性調整手段107に対する信号の特性の調整の指示が不要であることによって、試験用ブロック102が劣化していないと判断してもよい。
 なお、劣化診断回路が待機モードにある場合に、試験用ブロックと参照用ブロックとの動作状態は任意に設定可能である。例えば、待機モードにおいても試験用ブロックを動作させ、参照用ブロックは動作を停止させるようにしてもよい。
 さらに、上記の「特性」は、試験用ブロック102及び参照用ブロック103から出力される信号が備える観測可能な所定の電気的特性であって、半導体集積回路の劣化診断に利用可能な特性であればどのようなものでもよい。従って、「特性」には、信号の時間的な側面に関する、位相、変化タイミング、周波数や、信号の大きさの側面に関する、出力電圧、出力電流、電圧振幅、電流振幅、負荷を駆動する能力が含まれる。
 本実施形態では、上記のような「特性」の差を検出することによって、劣化の発生を診断する。そのため、具体的な特性の差の検出方法は限定されない。例えば、信号の位相や変化タイミング、周波数の差の検出のために、判定手段104は位相比較器を備えていてもよい。信号の電圧の差の検出のために、判定手段104は電圧比較器を備えていてもよい。また、試験用ブロック102及び参照用ブロック103から出力される信号の電流の差は、例えば、それぞれの信号が抵抗器に生じさせる電圧の差から求めてもよい。さらに、信号が負荷を駆動する能力は、例えば、信号に所定の負荷を接続した場合にその信号に生じる電圧あるいは電流の変化量から求めてもよい。
 以上説明した第1の実施形態は、簡単な構成で半導体集積回路の劣化状態を測定する。そして、2つのブロックを備える劣化診断ブロックの測定を制御することで、測定時の環境変動(例えば電源電圧や周囲温度)を受けた場合にも、試験用ブロック102及び参照用ブロック103に同様の環境変動を与えた上で遅延時間の差を算出することで、環境変動の影響を相殺することが可能となる。
 また、BTI(Bias Temperature Instability)と呼ばれる、回路素子に加えられたストレスが除去されると素子の劣化状態が回復する現象が知られている。測定モードにおける回路素子に印加されるストレスが、待機モードにおける回路素子のストレスと異なっていると、BTIにより劣化状態が誤って判定される恐れがある。第1の実施形態の劣化診断回路は、測定モードにおいては判定手段104が試験用ブロック102及び参照用ブロック103から出力される信号に基づいて劣化状態を判断する。そして、上述したように、試験用ブロック102は、制御手段105が測定モードを示す信号を出力していない待機モードの場合にも動作するようにしてもよい。待機モードにおいては、参照用ブロック103は動作を停止していてもよい。このような動作により、第1の実施形態の劣化診断回路100は、試験用ブロック102にストレスを付加した状態を維持しながら、待機状態から測定状態へ連続的に移行することが可能である。このため、第1の実施形態の劣化診断回路は、BTIの影響により劣化量が誤って判定される恐れを低減できるという効果も奏する。
(第2の実施形態)
 図3は、第2の実施形態の劣化診断回路の構成を示す図である。図3に示す劣化診断回路200は、劣化診断ブロック220及び制御手段211を備える。劣化診断ブロック220は、試験用ブロック201、参照用ブロック202及び判定手段203を備える。なお、動作の説明に不要な要素や説明等は適宜省略されており、また、図3に示す構成は本発明の目的を達成しうる唯一の構成として限定されるものではない。
 図3に示す試験用ブロック201及び参照用ブロック202は、論理ゲートで構成される。第2の実施形態では、試験用ブロック201は、N個のNAND(not AND)ゲート2011、2012、・・・201Nの出力がループ状に循環するリング発振器として構成される。それぞれのNANDゲート2011、2012、・・・201Nの、図示されていない入力端子の電位は“H”レベルに固定されている。従って、NANDゲート2011、・・・201Nは、それぞれインバータとして動作する。しかし、論理ゲートとしてNOR(not OR)ゲートや他の種類の論理ゲートを使用してもよい。また論理ゲートの段数はリング発振器の発振条件を満たしていれば特に制限はない。ただし、論理ゲートの段数が少ないほど劣化診断回路を小面積化することが可能となる。
 参照用ブロック202は、試験用ブロック201と同様にN個のNANDゲート2021、2022・・・202Nからなるリング発振器を備えている。ここで、分周器204の出力は、NANDゲート2021の、ループを構成しない側の入力に反転されて入力されている。また、NANDゲート2022、・・・202Nの、図示されていない入力端子の電位は“H”レベルに固定されている。従って、NANDゲート2022、・・・202Nは、それぞれインバータとして動作する。
 判定手段203は、分周器204、DFF(D−flip flop)群205、スイッチ群206、EXOR(exclusive−OR)群207、インバータ群212及び劣化量算出手段214を備える。
 分周器204は、試験用ブロック201が備える論理ゲート間の分岐点のいずれか1つ(以下、「ノード」という。)から入力された信号を分周してDFF群205及び参照用ブロック202のNANDゲート2021に出力する。
 スイッチ群206はN個のスイッチ2061、2062・・・206Nで構成され、参照用ブロック202のノードから選択された1つのノードの出力をインバータ群212に入力する。すなわち、スイッチ2061~206Nは、同時にはいずれか1つのスイッチのみがON(短絡)となる。
 インバータ群212は、直列に接続された3個のNANDゲート2121~2123で構成される。NANDゲート2121~2123の図示されていない入力端子の電位は“H”レベルに固定されている。従って、NANDゲート2121~2123は、それぞれインバータとして動作する。インバータ群212は、スイッチ群206が選択したノードからの信号をNANDゲート2121~2123を通過する毎に反転させる。そして、NANDゲート2121~2123の出力はDFF群205に入力される。
 DFF群205は、DFF2051~2053を備える。DFF2051~2053は、NANDゲート2121~2123からそれぞれ入力された信号を分周器204からの立ち上がりで保持する。EXOR群207は、EXORゲート2071及び2072を備える。EXORゲート2071は、DFF2051及び2052の出力の排他的論理和演算を行い、出力208として劣化量算出手段214へ出力する。EXORゲート2072は、DFF2052及び2053の出力の排他的論理和演算を行い、出力209として劣化量算出手段214へ出力する。
 劣化量算出手段214は、出力208及び209に基づいて、試験用ブロック201の劣化量を算出する。劣化量算出手段214における具体的な劣化量の算出方法は後述する。
 試験用ブロック201のノードと分周器204の接続及び切断は、スイッチ213によって行われる。スイッチ213は、制御手段211が出力する制御信号210によって制御される。制御信号210により、試験用ブロック201のノードと分周器204との間を接続または分離することができる。
 第2の実施形態において、劣化状況の測定を行わない状態を待機モード、劣化状況の測定を行う状態を測定モードと呼ぶ。以下に、待機モード及び測定モードの動作について説明する。以下に説明する動作は一例であり、本発明の目的を達成しうる唯一の動作ではないことは明らかである。
 待機モードにおいては、制御手段211は、試験用ブロック201と分周器204との間をスイッチ213により切断するように制御信号210を出力する。そして、試験用ブロック201は、リング発振器として自己発振する。ここで、図示しない外部からの制御信号によりリング発振器の動作条件を制御可能なように試験用ブロック201を構成してもよい。参照用ブロック202の動作も試験用ブロック201と同様である。試験用ブロック201と参照用ブロック202とに印加されるストレス(電圧、温度等)の違いにより、ストレスの異なる条件下における試験用ブロック201の劣化状況の評価が可能となる。
 測定モードにおいては、試験用ブロック201の特定のノードと分周器204とが制御信号210によって接続される。待機モードから測定モードへの移行は制御手段211がスイッチ213を閉じるように制御信号210を出力することにより行われる。すなわち、待機モードと測定モードとの回路構成の相違点は、スイッチ213が開いているか閉じているかのみである。従って、劣化診断回路200は、試験用ブロック201及び参照用ブロック202にストレスを印加したまま、それぞれのブロックの回路構成を変更することなく待機モードから測定モードへ移行することが可能である。
 測定モードにおいては、試験用ブロック201のノードの信号が分周器204に入力される。分周器204は、試験用ブロック201から入力された信号を所定の分周比で分周する。その結果、分周器204の出力は分周比に応じた周期で“H”レベルと“L”レベルとが交互に変化する。例えば、測定モードに入る直前の分周器204の出力信号レベルが“H”であった場合には、分周器204の出力が以降“L”、“H”の順に変化する。
 まず、分周器204の出力が“L”に変化したときの動作について説明する。分周器204の出力は、NANDゲート2021の入力に反転されて入力されている。従って、分周器204の出力信号が“H”の場合は、NANDゲート2021の出力は“H”に固定されている。そして、分周器204の出力信号が“L”になると、参照用ブロック202は発振を開始する。同時にスイッチ2061~206Nのうちいずれか1個のスイッチが選択される。この選択手順については後述する。
 次に、分周比に相当する個数の信号が試験用ブロック201から入力されると、分周器204の出力は“H”へと変化する。このとき、DFF群205の出力では、参照用ブロック202で選択されたノードの出力レベルが保持される。DFF群205によって保持された出力レベルは後段のEXOR群207によって論理演算されて出力208及び出力209として出力される。
 なお、分周器204の出力が“L”から“H”に変化することより参照用ブロック202の発振動作が停止する。発振動作の停止とDFF群205におけるスイッチ群206からの入力データのラッチとが同時に行われるため、DFF群205の出力信号に乱れが生じる可能性がある。DFF群205が分周器204の出力変化時の入力信号を正しくラッチできるように、分周器204と参照用ブロック202との間に遅延回路を設けてもよい。あるいは、分周器204から参照用ブロック202へ出力される信号をDFF群205が入力をラッチするまで保持する回路を設けてもよい。あるいは制御信号210を利用して、分周器204の出力とは別に参照用ブロック202の発振開始及び停止を制御してもよい。
 以下では、劣化量算出手段214における劣化状態の算出方法について説明する。ここでは簡単のため、試験用ブロック201のみが劣化するものとして説明する。ただし、試験用ブロック201と参照用ブロック202との少なくとも一方の動作環境を変化させた場合においても劣化状態を相対的に算出できることは容易に理解される。さらに、試験用ブロック201と参照ブロック202とを入れ換えた場合などにおいても同様に相対的な劣化状態を算出できる。
 試験用ブロック201と分周器204との接続は、説明が容易な図3の構成に基づいて説明している。しかし、これ以外の構成のリング発振器あるいは参照用ブロック202との間でタイミングの比較が可能な他の回路を使用した場合も、同様な動作によって試験用ブロックが出力する信号の遅延量を測定し、試験用ブロックの劣化量を求めることが可能である。
 図3に示す劣化検出回路220において、試験用ブロック201と分周器204を用いた場合の分周器204の出力が変化する間隔は、以下の式(1)で示される。
(T+ΔT)×2×N×M  ・・・(1)
 ここで、Tは試験用ブロック201のNANDゲート一段あたりの遅延量、ΔTは試験用ブロック201のNANDゲート一段あたりの劣化により増加する遅延量である。また、Nは試験用ブロック201のNANDゲートの段数、Mは分周器204の分周比である。すなわち、式(1)の値は、分周器204の出力が変化する間隔における試験用ブロックで発生する遅延量の総和である。なお、NANDゲート2011~201Nの内部の回路は同一であるので、NANDゲート2011~201Nにおける性能劣化時の遅延量の増加は全て等しいとした。
 一方、参照用ブロック202でも同様に以下の式(2)で表される遅延量が得られる。
 T×X  ・・・(2)
 式(2)は、分周器204の出力が変化する間に参照用ブロック202から出力される遅延量である。
 ここで、Xは分周器204の出力の立ち下がりから次の立ち上がりまでの間に参照用ブロック202内の信号が通過したNANDゲートの総段数である。一般的には、Xを測定するためには、参照用ブロック202においてもカウンタや多量のDFFを必要とする。そこで、第2の実施形態では劣化検出回路を簡素化して小面積化するため、カウンタ等により直接Xを求めることなく、試験用ブロック201の遅延量と参照用ブロック202の遅延量との差を測定している。このため、劣化検出回路220は、測定モードの開始時に、DFF群に入力する参照用ブロック202のノードの選定を行う。以下に、その具体的な手順を説明する。
 まず、分周器204の出力の立ち上がり時点までにスイッチ群206はあるノードを選択しておく。そして、分周器204の出力の立ち下がり時点における出力208および出力209のレベルが一致しないノードが探される。すなわち、出力208が“L”かつ出力209が“H”、あるいは出力208が“H”かつ出力209が“L”のどちらかとなるノードが探索される。ここで、スイッチ群206の開閉制御及び出力208、209のレベルの調査は、判定手段203が行ってもよいし、あるいは判定手段203の外部からの制御により他の機器が行ってもよい。
 図5は、第2の実施形態における信号のタイミングを示す図である。図5において、分周器204の出力は、試験用ブロック201の出力を分周した信号である。信号A~C及び信号A‘~C’はそれぞれDFF2051~2053の入力及び出力を示す。図5は、測定モードの開始時に、出力208が“L”かつ出力209が“H”となるようなスイッチ2061~206Nのいずれか1つが選択されて閉じられた場合のタイミングチャートを示す。
 試験用ブロック201と参照用ブロック202とのリング発振器としての回路構成は同一である。従って、参照用ブロック202と比較して試験用ブロック201が劣化していない場合には、分周器201の出力と参照用ブロックからDFF群へ入力される信号(A~C)との間のタイミングは変化しない(図5の(a))。しかし、試験用ブロック201の劣化が進行して試験用ブロック202の出力の遅延が増大すると、分周器204の立ち上がりタイミングがDFF群への入力信号(A~C)に対して遅れるようになる。そして、試験用ブロック201の劣化が進行して分周器204の立ち上がりタイミングが図5のt1からt2まで遅れると、DFF2052の出力信号B‘が反転する。その結果、出力208、209はいずれも反転して出力208が“H”、出力209が“L”となる(図5の(b))。
 NANDゲート2121~2123は直列に接続されているので、DFF群205への入力信号A~Cのタイミングは、A、B、Cの順にインバータ群212のNANDゲート1段分ずつ遅れる。そして、分周器204の立ち上がりタイミングがDFF2052の出力信号B‘の反転時刻を挟んでt1からt2まで遅れた場合の、分周器204の立ち下がりから立ち上がりまでの間の遅延量は、平均してNANDゲート1段分である。試験用ブロック201において劣化によってゲート1段分の遅延が増加したと考える。ここで、試験用ブロック201での信号が通過したゲート段数は“2×N×M“である。このため、劣化量算出手段214は、信号A’~C‘及び出力208、209が図5の(a)から(b)に変化した場合に、試験用ブロック201におけるゲート1段あたりの劣化量を、1/(2×N×M)として求めることができる。
 なお、判定手段203は、出力208及び出力209がいずれも反転して出力208が“H”、出力209が“L”となった後、再度スイッチ群206によるノードの選択を行ってもよい。例えば、判定手段203は、分周器204の立ち上がりタイミングが図5のt2まで遅れている状態で、参照用ブロック202において選択されるノードを後方へ1つシフトさせてもよい。ノードを後方へ1つシフトさせた場合の信号A~Cを、図5の破線で示す。図5に破線で示した各信号の関係は、図5において信号Bを新たな信号Aとし、それらを順に反転させた信号を新たにB、Cとしたものとなる。その結果、ノードを後方へ1つシフトさせることにより、分周器204の立ち上がりタイミングとDFF群205への入力信号A~Cとの関係は図5のt1におけるものと同様となる。
 このように、出力208及び出力209が反転した後、スイッチ群206によって選択されるノードを後方へシフトさせることにより、出力208が“L”かつ出力209が“H”という状態とすることができる。その後、さらに試験用ブロック201の劣化が進行して出力208が“H”かつ出力209が“L”となった場合、劣化量算出手段214は、劣化がさらに進行して2/(2×N×M)となったことを算出できる。
 以降、分周器204の立ち上がりタイミングの遅延に従って選択するノードを順次後方にシフトさせることで、簡素な回路構成により試験用ブロックの劣化の状況を劣化量算出手段214において算出することが可能となる。
 なお、ノードの選択の際に、出力208が“L”、出力209が“H”となるタイミングは図5のt1だけではなく、t3である場合もある。しかし、この場合も試験用ブロック201の性能劣化の進行により分周器204の立ち上がりタイミングがt3からt4まで遅れると、出力208が“H”、出力209が“L”に反転する。そして、劣化量算出手段214は、上記と同様の手順で試験用ブロック201の劣化量を測定することができる。
 すなわち、ノードを選択する際に、出力208が“L”、出力209が“H”となるタイミングが図5のt1であってもt4であっても、劣化量算出手段214は、試験用ブロック201の劣化量を算出することができる。
 以降、判定手段203に信号を入力する参照用ブロック202のノードを順次後方へずらしていくことで劣化量算出手段214において劣化の進行度を測定することが可能となる。すなわち、第2の実施形態の劣化診断回路は、カウンタや多数のDFFを使用することなく、試験用ブロック及び参照用ブロックの出力を比較することで、試験用ブロックで発生する遅延量を測定している。その結果、第2の実施形態の劣化診断回路は、簡易な回路構成により半導体集積回路の劣化の状態をより正確に測定できるという効果を奏する。
 また、第2の実施形態の劣化診断回路は、試験用ブロック201及び参照用ブロック202の回路構成を変更することなく、制御信号210により待機モードから測定モードへ連続的に移行することが可能である。このため、第2の実施形態の劣化診断回路は、BTIの影響により劣化量が誤って判定される恐れを低減できるという効果も奏する。
 なお、第2の実施形態においてもその変形例として、第1の実施形態の変形例と同様に判定手段203は特性調整手段を備えていてもよい。特性調整手段は、劣化量算出手段214において試験用ブロック201の劣化量が0に近づくように、入力される信号の遅延量を調整する。
 特性調整手段は、試験用ブロック201の出力と分周器204の入力との間、及び、分周器204の出力とNANDゲート2021の入力との間の少なくとも一方に備えられていてもよい。第2の実施形態の変形例においては、劣化量算出手段214において試験用ブロック201の劣化が検出されない状態における特性調整手段の遅延調整量に基づいて、試験用ブロック201の劣化量を求めることができる。
 さらに、第2の実施形態の変形例において、判定手段203は、特性調整手段の特性調整量を外部に出力する機能を備えていてもよい。
(第3の実施形態)
 図3に示した第2の実施形態の構成は、インバータ群212で1段ずつ遅延させた信号A~Cと分周器204の立ち上がりタイミングとを比較する構成であった。すなわち、検出される試験用ブロック201の遅延量の分解能はゲート1段分に固定されている。
 第3の実施形態では、分周器204からの出力とタイミングを比較する信号をノード毎に自由に選択できる構成について説明する。
 図4は、第3の実施形態の劣化診断回路の構成を示す図である。図4に示す劣化診断回路300は、劣化診断ブロック320及び制御手段311を備える。劣化診断ブロック320は、試験用ブロック301、参照用ブロック302及び判定手段303を備える。なお、動作の説明に不要な要素や説明等は適宜省略されており、また、図4に示す構成は本発明の目的を達成しうる唯一の構成として限定されることがないのは明らかである。
 図4に示す試験用ブロック301及び参照用ブロック302は、論理ゲートで構成される。試験用ブロック301は、NANDゲート3011~301Nで構成されたリング発振器であり、参照用ブロック302は、NANDゲート3021~302Nで構成されたリング発振器である。
 判定手段303は、分周器304、スイッチ群306、DFF群305、EXOR群307及び劣化量算出手段314を備える。
 分周器304は、試験用ブロック301のいずれかのノードから入力された信号を分周してDFF群305及び参照用ブロック302のNANDゲート3021にそれぞれ出力する。スイッチ群306は、参照用ブロック302の各ノードを3つに分岐してそれぞれをDFF群305を構成するDFF3051~3052に接続する。
 DFF群305の出力は、スイッチ群306により選択された3個所のノードからの出力を分周器304からの立ち上がりでラッチする。EXOR群307は、DFF群305からの3つの出力のうち2つの排他的論理和演算を行う。出力308および出力309はそれぞれEXOR3071及び3072の出力であり、劣化量算出手段314へ出力される。
 劣化量算出手段314は、出力308及び309に基づいて、試験用ブロック301の劣化量を算出する。劣化量算出手段314における具体的な劣化量の算出方法は後述する。
 スイッチ群306は、参照用ブロック302のノード毎に3個のスイッチを備えている。図4では、スイッチ3061a~3061c、3062a~3062c、・・・306Na~306Ncが、スイッチ群306を構成する。スイッチ群306のあるスイッチが閉じる(導通する)と、当該ノードの出力がDFF群305のいずれかのDFFと接続される。
 例えば、スイッチ3061aが閉じると、NANDゲート302Nの出力が、信号線Aを通じてDFF3051に入力される。また、スイッチ3062bまたは3062cが閉じると、NANDゲート3021の出力が、それぞれ信号線BまたはCを通じてDFF3052または3053に入力される。スイッチ群306を構成する各スイッチの動作は、他のノードについても同様である。
 第3の実施形態では、スイッチ群306を構成するスイッチ3061a等が個別に制御される。この構成により、信号A~Cの配線を参照用ブロック302の任意のノードと接続することができる。すなわち、信号AとBとの間の遅延量及び信号BとCとの間の遅延量をそれぞれNANDゲートの個数単位で設定できる。その結果、第3の実施形態では、図5で説明した分周器204の出力の立ち上がりタイミングの変化を検出する分解能を変化させることが可能である。
 図3で説明した構成では、インバータ群212において、DFF2051~2053に入力される信号にはゲート1段ずつの遅延を生じさせていた。図4においては、例えば、スイッチ3061a、スイッチ3062b、スイッチ3063cを閉じることで、DFF群305にゲート1段ずつの遅延を生じさせた信号が入力されるように判定手段303を構成できる。
 さらに、図4に示す構成では、例えば、スイッチ3061aの他に、図示されないがそれぞれNANDゲートで3段ずつ離れたスイッチ3064b、3067cを閉じることで、DFF群305にはNANDゲート3段ずつの遅延を生じさせた信号が入力される。この場合、図5における信号AとB、BとCとの変化点間の遅延はNANDゲート3段分の長さとなる。従って、出力308及び309が反転した場合には、試験用ブロック301においてゲート3段分の遅延が発生していることになる。
 このとき、試験用ブロック301での信号が通過したゲート段数は2×N×Mであり、試験用ブロック301では劣化によってゲート3段分の遅延が増加したと考えられる。このため、劣化量算出手段314は、試験用ブロック301におけるゲート1段あたりの劣化量を、3/(2×N×M)として求めることができる。
 なお、分周器304の出力が“L”から“H”に変化することより参照用ブロック302の発振動作が停止する。発振動作の停止とDFF群305におけるスイッチ群306からの入力データのラッチとが同時に行われるため、DFF群305の出力信号に乱れが生じる可能性がある。DEF群305が分周器304の出力変化時の入力信号を正しくラッチできるように、分周器304と参照用ブロック302との間に遅延回路を設けてもよい。あるいは、分周器304から参照用ブロック302へ出力される信号をDFF群305が入力をラッチするまで保持する回路を設けてもよい。あるいは制御信号310を利用して、分周器304の出力とは別に参照用ブロック302の発振開始及び停止を制御してもよい。
 このように、第3の実施形態の劣化診断回路も、第2の実施形態の劣化診断回路と同様に、簡易な回路構成により半導体集積回路の劣化の状態を測定できるという効果を奏する。
 そして、第3の実施形態では、スイッチ群306を備えることにより、測定される試験用ブロック301の劣化量の測定単位(分解能)を変化させることができる。劣化量の測定単位を大きくすることにより、NANDゲート毎の遅延量のばらつきや雑音による遅延量の誤差を平均化できるという効果が得られる。
 なお、第3の実施形態の劣化診断回路は、第2の実施形態の劣化診断回路と同様に、試験用ブロック301及び参照用ブロック302の回路構成を変更することなく、制御信号310により待機モードから測定モードへ連続的に移行することが可能である。このため、第3の実施形態の劣化診断回路も、BTIによる劣化の回復の影響を低減できるという効果も奏する。
 なお、第3の実施形態においてもその変形例として、第1の実施形態の変形例と同様に判定手段303は特性調整手段を備えていてもよい。特性調整手段は、劣化量算出手段314において試験用ブロック301の劣化量が0に近づくように、入力される信号の遅延量を調整する。
 特性調整手段は、試験用ブロック301の出力と分周器304の入力との間、及び、分周器304の出力とNANDゲート3021の入力との間の少なくとも一方に備えられていてもよい。第3の実施形態の変形例においては、劣化量算出手段314において試験用ブロック301の劣化が検出されない状態における特性調整手段の遅延調整量に基づいて、試験用ブロック301の劣化量を求めることができる。
 さらに、第3の実施形態の変形例において、判定手段303は、特性調整手段の特性調整量を外部に出力する機能を備えていてもよい。
 以上、第1~第3の実施形態を参照して本願発明の実施形態を説明した。しかし、本願発明が適用可能な形態は上述した実施形態に限定されるものではない。本願発明の構成や詳細説明には、本願発明のスコープ内で当業者が理解し得る様々な変更をすることができる。
 この出願は、2011年8月24日に出願された日本出願特願2011−182438を基礎とする優先権を主張し、その開示の全てをここに取り込む。
 なお、本発明の実施形態は以下の付記1~12のようにも記載されうるがこれらには限定されない。
(付記1)
 劣化の診断の対象となる第1の回路を含む試験用ブロックと、
前記第1の回路と同一の構成を備える第2の回路を含む参照用ブロックと、
測定モードを示す信号が入力された場合に前記試験用ブロックから出力される第1の信号の第1の特性と前記参照用ブロックから出力される第2の信号の第2の特性とを比較することで前記試験用ブロックを構成する素子の前記劣化の有無を判定する判定手段と、
前記測定モードを示す信号を前記判定手段に出力する制御手段と、
を備える劣化診断回路。
(付記2)
 前記第1の特性及び前記第2の特性は、前記第1の信号及び前記第2の信号がそれぞれ備える、前記劣化の結果変化する電気的特性である、付記1に記載された劣化診断回路。
(付記3)
 前記第1の特性及び前記第2の特性は、位相、変化タイミング、周波数、出力電圧、出力電流、電圧振幅、電流振幅、及び負荷を駆動する能力、の少なくとも1つをそれぞれ含む、付記2に記載された劣化診断回路。
(付記4)
 前記判定手段は、前記試験用ブロックを構成する素子の特性と前記参照用ブロックを構成する素子の特性との差を検出する、付記1乃至3のいずれかに記載された劣化診断回路。
(付記5)
 前記測定モードを示す信号が入力された場合に、前記第1の特性および前記第2の特性が等しくなるように、前記判定手段からの指示に従って前記第1の信号および前記第2の信号の少なくとも一方に所定の調整を行う特性調整手段を備え、
 前記判定手段は、前記指示の要否によって前記劣化の有無を判定する、
 付記1乃至3のいずれかに記載された劣化診断回路。
(付記6)
 前記判定手段は、前記調整を行うために必要な調整量を前記劣化の程度を示す劣化量として出力する、付記5に記載された劣化診断回路。
(付記7)
 前記第1の回路と前記第2の回路との一方もしくは両方が、所定のタイミングで信号を出力する循環型の構成となっている、付記1乃至6のいずれかに記載された劣化診断回路。
(付記8)
 前記第1の回路と前記第2の回路との一方もしくは両方が、論理ゲートが循環的に接続された構成となっている、付記1乃至7のいずれかに記載された劣化診断回路。
(付記9)
 前記判定手段は、前記試験用ブロックを構成する素子の遅延時間と前記参照用ブロックを構成する素子の遅延時間とを比較する、付記1乃至8のいずれかに記載された劣化診断回路。
(付記10)
 前記判定手段は、
前記第1の信号を分周した第3の信号を出力する分周器と、
前記第2の信号に対して互いに異なる遅延量を持つ第4の信号及び第5の信号を前記第3の信号でラッチしてそれぞれ出力するラッチ回路と、
前記ラッチ回路の出力の排他的論理和を求めるEXOR(exclusive−OR)回路と、
前記EXOR回路の出力に基づいて前記劣化量を算出する劣化量算出手段と、
を備える付記1乃至9のいずれかに記載された劣化診断回路。
(付記11)
 前記第4の信号及び前記第5の信号は、前記第2の信号を遅延させることで生成される、付記10に記載された劣化診断回路。
(付記12)
 前記第4の信号及び前記第5の信号は、前記参照用ブロックの、所定の信号の伝播経路における前段および後段に相当する位置から出力される前記第2の信号である、付記10に記載された劣化診断回路。
(付記13)
 前記制御手段が前記測定モードを示す信号を出力していない場合に、前記試験用ブロックは動作し前記参照用ブロックは動作を停止する、付記1乃至12のいずれかに記載された劣化診断回路。
(付記14)
 前記試験用ブロック及び前記参照用ブロックは、前記試験用ブロック及び前記参照用ブロックに印加されるストレスが異なる動作環境で動作する、付記1乃至13のいずれかに記載された劣化診断回路。
(付記15)
 測定モードを示す信号が入力された場合に、劣化診断の対象となる第1の回路を含む試験用ブロックから出力される第1の信号の第1の特性と前記第1の回路と同一の構成を備える第2の回路を含む参照用ブロックから出力される第2の信号の第2の特性とを比較することで前記試験用ブロックを構成する素子の劣化の有無を判定する、
劣化診断方法。
Next, embodiments of the present invention will be described with reference to the drawings.
(First embodiment)
FIG. 1 is a diagram showing a configuration of a deterioration diagnosis circuit according to the first embodiment of the present invention.
As shown in FIG. 1, the deterioration diagnosis circuit 100 includes a deterioration diagnosis block 101 and a control means 105. The deterioration diagnosis block 101 includes a test block 102, a reference block 103, and a determination unit 104.
As a result of the stress applied to the test block 102, when the semiconductor integrated circuit constituting the circuit of the test block is deteriorated, a delay which is a kind of characteristic of the logic gate constituting the semiconductor integrated circuit may increase. . As will be described later, there is a difference in the amount of stress between the test block 102 and the reference block 103. The amount of stress is, for example, the intensity of stress or the length of a period during which stress is applied. Therefore, there is a difference in the degree of deterioration (deterioration degree) between the test block 102 and the reference block 103. Accordingly, the delay of the semiconductor integrated circuit in the test block 102 causes a larger delay in the logic gate than in the reference block 103. As a result, the signal output from the test block 102 is gradually delayed with respect to the signal output from the reference block 103. The determination unit 104 detects this delay and determines the deterioration state of the test block 102.
The determination unit 104 has two operation states, a standby mode and a measurement mode. The standby mode is a state in which the deterioration diagnosis block 101 does not measure the deterioration of the test block 102. The measurement mode is a state in which the deterioration diagnosis block 101 measures the deterioration of the test block 102.
The test block includes a first circuit to be subjected to deterioration diagnosis. The reference block includes a second circuit having the same configuration as the first circuit. The test block 102 and the reference block 103 output the first signal and the second signal to the determination unit 104, respectively. The control unit 105 outputs a signal indicating the standby mode or the measurement mode to the determination unit 104.
The determination unit 104 performs the following operation when the signal input from the control unit 105 indicates the measurement mode. That is, in the measurement mode, the determination unit 104 compares the first signal input from the test block 102 and the second signal input from the reference block 103, and degrades the elements constituting the test block 102. The presence or absence of is determined.
It is possible to apply stress to the test block 102 and the reference block 103 to advance the deterioration of each semiconductor integrated circuit. The stress can be applied by whether or not the power supply voltage is applied, the power supply voltage is changed, the ambient temperature is changed, the operating frequency is changed, the circuit load is changed, and the like. However, the types of stress are not limited to these. Any stress that affects the life of the semiconductor integrated circuit may be applied to one or both of the test block 102 and the reference block 103.
A plurality of types of stress may be simultaneously applied to the test block 102 and the reference block 103. At this time, it is preferable that the type and strength of the applied stress be different between the test block 102 and the reference block 103.
For example, in the standby mode, the test block 102 is set to a state in which deterioration is advanced by applying stress, and the power is not applied by connecting the power supply terminal of the semiconductor integrated circuit of the reference block 103 to the ground potential. That is, it is good also as a state which deterioration does not advance. Alternatively, stress may be applied to both the test block 102 and the reference block 103 even in the standby state, and a difference may be provided in the operating conditions such as temperature and operating frequency. Note that the positions of the test block 102 and the reference block 103 may be interchanged.
On the other hand, in the measurement mode, the determination unit 104 compares the timings of the two signals, the output of the test block 102 and the output of the reference block 103, and obtains the difference.
As a specific configuration, the test block 102 and the reference block 103 may include an oscillation circuit, and the output of the oscillation circuit may be output to the determination unit 104.
The determination unit 104 compares the characteristics of the signals output from the test block 102 and the reference block 103 and determines the deterioration state of the test block 102. If the test block 102 and the reference block 103 are provided with an oscillation circuit, the oscillation frequency or the signal phase may be used as the signal characteristics.
FIG. 2 is a diagram illustrating a configuration of a modified example of the deterioration diagnosis circuit of the first embodiment. The deterioration diagnosis block 106 shown in FIG. 2 includes a characteristic adjustment unit 107 between the test block 102 and the determination unit 104. The characteristic adjusting unit 107 adjusts the characteristic of the signal output from the test block 102 and inputs it to the determining unit 104. The output of the determination unit 104 is input to the characteristic adjustment unit 107. The characteristic adjustment unit 107 adjusts the output characteristic (for example, delay time) of the test block 102 so that the difference between the characteristic of the test block 102 and the characteristic of the reference block 102 is not detected in the output of the determination unit 104. The determination unit 104 may have a function of instructing the characteristic adjustment unit 107 whether or not to adjust the characteristic of the signal input to the characteristic adjustment unit 107. In the modification of the first embodiment shown in FIG. 2, the characteristic adjustment amount performed by the characteristic adjustment unit 107 can be the deterioration amount of the test block 102.
The characteristic adjustment unit 107 may be disposed between the reference block 103 and the determination unit 104. Also in this case, the characteristic adjustment amount of the characteristic adjustment unit 107 in a state where the difference between the characteristic of the test block 102 and the characteristic of the reference block 103 is not detected in the output of the determination unit 104 is set as the deterioration amount of the test block 102. it can. Further, the characteristic adjusting means 107 may be provided both between the test block 102 and the determination means 104 and between the reference block 103 and the determination means 104. In this case, the difference between the characteristic adjustment amounts of the two characteristic adjustment means 107 in a state where the difference between the characteristic of the test block 102 and the characteristic of the reference block 103 is not detected in the output of the determination means 104 is the deterioration of the test block 102. It can be an amount. In addition, in the modification of the first embodiment, the deterioration diagnosis block 106 may have a function of outputting the characteristic adjustment amount of the characteristic adjustment unit 107 to the outside.
Here, when the characteristic of the test block 102 is equal to the characteristic of the reference block 103 without the characteristic adjustment unit 107 adjusting the characteristic of the signal, the deterioration diagnosis block 106 determines that the determination unit 104 determines the test block 102. It may be determined that is not deteriorated. In this case, the determination unit 104 does not need to instruct the characteristic adjustment unit 107 to adjust the signal characteristic. Therefore, the determination unit 104 may determine that the test block 102 has not deteriorated because an instruction to adjust the characteristic of the signal to the characteristic adjustment unit 107 is unnecessary.
When the deterioration diagnosis circuit is in the standby mode, the operation states of the test block and the reference block can be arbitrarily set. For example, the test block may be operated even in the standby mode, and the reference block may be stopped.
Furthermore, the above-mentioned “characteristic” is a predetermined observable electric characteristic provided for signals output from the test block 102 and the reference block 103, and can be used for deterioration diagnosis of a semiconductor integrated circuit. Anything is acceptable. Therefore, “characteristics” include the output voltage, output current, voltage amplitude, current amplitude, and the ability to drive the load with respect to the phase, change timing, frequency, and signal magnitude aspects of the time aspect of the signal. included.
In the present embodiment, the occurrence of deterioration is diagnosed by detecting the difference in “characteristic” as described above. Therefore, a specific method for detecting a difference in characteristics is not limited. For example, the determination unit 104 may include a phase comparator for detection of a signal phase, change timing, and frequency difference. The determination means 104 may include a voltage comparator for detecting a difference in signal voltage. Further, the difference between the currents of the signals output from the test block 102 and the reference block 103 may be obtained, for example, from the difference between the voltages generated by the respective signals in the resistor. Furthermore, the ability of a signal to drive a load may be obtained from, for example, the amount of change in voltage or current generated in the signal when a predetermined load is connected to the signal.
The first embodiment described above measures the deterioration state of a semiconductor integrated circuit with a simple configuration. By controlling the measurement of the deterioration diagnosis block including two blocks, the same applies to the test block 102 and the reference block 103 even when receiving environmental fluctuations (for example, power supply voltage and ambient temperature) during measurement. By calculating the difference in delay time after giving environmental fluctuations, it becomes possible to offset the influence of environmental fluctuations.
In addition, a phenomenon called BTI (Bias Temperature Instability) is known in which a deterioration state of an element is recovered when stress applied to the circuit element is removed. If the stress applied to the circuit element in the measurement mode is different from the stress of the circuit element in the standby mode, the deterioration state may be erroneously determined by the BTI. In the deterioration diagnosis circuit of the first embodiment, in the measurement mode, the determination unit 104 determines the deterioration state based on signals output from the test block 102 and the reference block 103. As described above, the test block 102 may operate even in the standby mode in which the control unit 105 does not output a signal indicating the measurement mode. In the standby mode, the reference block 103 may stop operating. By such an operation, the deterioration diagnosis circuit 100 according to the first embodiment can continuously shift from the standby state to the measurement state while maintaining a state in which the test block 102 is stressed. For this reason, the deterioration diagnosis circuit of the first embodiment also has an effect of reducing the possibility that the deterioration amount is erroneously determined due to the influence of BTI.
(Second Embodiment)
FIG. 3 is a diagram illustrating a configuration of a deterioration diagnosis circuit according to the second embodiment. The deterioration diagnosis circuit 200 shown in FIG. 3 includes a deterioration diagnosis block 220 and a control unit 211. The deterioration diagnosis block 220 includes a test block 201, a reference block 202, and a determination unit 203. It should be noted that elements and explanations that are not necessary for the description of the operation are omitted as appropriate, and the configuration shown in FIG. 3 is not limited to the only configuration that can achieve the object of the present invention.
The test block 201 and the reference block 202 shown in FIG. 3 are composed of logic gates. In the second embodiment, the test block 201 is configured as a ring oscillator in which the outputs of N NAND (not AND) gates 2011, 2012,. The potentials of input terminals (not shown) of the NAND gates 2011, 2012,... 201N are fixed to the “H” level. Accordingly, each of the NAND gates 2011,... 201N operates as an inverter. However, a NOR (not OR) gate or another type of logic gate may be used as the logic gate. The number of stages of logic gates is not particularly limited as long as the oscillation conditions of the ring oscillator are satisfied. However, the smaller the number of logic gates, the smaller the area of the deterioration diagnosis circuit.
Similar to the test block 201, the reference block 202 includes a ring oscillator including N NAND gates 2021, 2022,... 202N. Here, the output of the frequency divider 204 is inverted and input to the input of the NAND gate 2021 that does not constitute a loop. In addition, the potentials of input terminals (not shown) of the NAND gates 2022,... 202N are fixed to the “H” level. Therefore, NAND gates 2022,... 202N each operate as an inverter.
The determination unit 203 includes a frequency divider 204, a DFF (D-flop flop) group 205, a switch group 206, an EXOR (exclusive-OR) group 207, an inverter group 212, and a deterioration amount calculation unit 214.
The frequency divider 204 divides a signal input from any one of the branch points between the logic gates included in the test block 201 (hereinafter, referred to as “node”), and thereby the DFF group 205 and the reference block 202. Are output to the NAND gate 2021.
The switch group 206 includes N switches 2061, 2062,... 206N, and inputs the output of one node selected from the nodes of the reference block 202 to the inverter group 212. That is, only one of the switches 2061 to 206N is turned ON (short circuit) at the same time.
The inverter group 212 includes three NAND gates 2121 to 2123 connected in series. The potentials of input terminals (not shown) of the NAND gates 2121 to 2123 are fixed to the “H” level. Accordingly, the NAND gates 2121 to 2123 each operate as an inverter. The inverter group 212 inverts the signal from the node selected by the switch group 206 every time it passes through the NAND gates 2121 to 2123. Outputs from the NAND gates 2121 to 2123 are input to the DFF group 205.
The DFF group 205 includes DFFs 2051 to 2053. The DFFs 2051 to 2053 hold the signals respectively input from the NAND gates 2121 to 2123 at the rising edge from the frequency divider 204. The EXOR group 207 includes EXOR gates 2071 and 2072. The EXOR gate 2071 performs an exclusive OR operation on the outputs of the DFFs 2051 and 2052 and outputs the result as an output 208 to the deterioration amount calculation means 214. The EXOR gate 2072 performs an exclusive OR operation on the outputs of the DFFs 2052 and 2053 and outputs the result to the deterioration amount calculation unit 214 as an output 209.
The deterioration amount calculation unit 214 calculates the deterioration amount of the test block 201 based on the outputs 208 and 209. A specific method for calculating the deterioration amount in the deterioration amount calculating unit 214 will be described later.
A switch 213 connects and disconnects the node of the test block 201 and the frequency divider 204. The switch 213 is controlled by a control signal 210 output from the control unit 211. The control signal 210 can connect or disconnect between the node of the test block 201 and the frequency divider 204.
In the second embodiment, a state in which the deterioration state is not measured is referred to as a standby mode, and a state in which the deterioration state is measured is referred to as a measurement mode. Hereinafter, operations in the standby mode and the measurement mode will be described. Obviously, the operations described below are examples and are not the only operations that can achieve the objectives of the present invention.
In the standby mode, the control unit 211 outputs a control signal 210 so that the switch 213 disconnects the test block 201 and the frequency divider 204. Then, the test block 201 self-oscillates as a ring oscillator. Here, the test block 201 may be configured so that the operating condition of the ring oscillator can be controlled by an external control signal (not shown). The operation of the reference block 202 is the same as that of the test block 201. Due to the difference in stress (voltage, temperature, etc.) applied to the test block 201 and the reference block 202, the deterioration status of the test block 201 under different stress conditions can be evaluated.
In the measurement mode, a specific node of the test block 201 and the frequency divider 204 are connected by the control signal 210. Transition from the standby mode to the measurement mode is performed by outputting a control signal 210 so that the control unit 211 closes the switch 213. That is, the only difference in the circuit configuration between the standby mode and the measurement mode is whether the switch 213 is open or closed. Therefore, the deterioration diagnosis circuit 200 can shift from the standby mode to the measurement mode without changing the circuit configuration of each block while applying stress to the test block 201 and the reference block 202.
In the measurement mode, the signal of the node of the test block 201 is input to the frequency divider 204. The frequency divider 204 divides the signal input from the test block 201 by a predetermined frequency division ratio. As a result, the output of the frequency divider 204 alternately changes between the “H” level and the “L” level in a cycle corresponding to the frequency division ratio. For example, when the output signal level of the frequency divider 204 immediately before entering the measurement mode is “H”, the output of the frequency divider 204 subsequently changes in the order of “L” and “H”.
First, the operation when the output of the frequency divider 204 changes to “L” will be described. The output of the frequency divider 204 is inverted and input to the input of the NAND gate 2021. Therefore, when the output signal of the frequency divider 204 is “H”, the output of the NAND gate 2021 is fixed to “H”. When the output signal of the frequency divider 204 becomes “L”, the reference block 202 starts oscillating. At the same time, any one of the switches 2061 to 206N is selected. This selection procedure will be described later.
Next, when the number of signals corresponding to the frequency division ratio is input from the test block 201, the output of the frequency divider 204 changes to “H”. At this time, the output level of the node selected by the reference block 202 is held at the output of the DFF group 205. The output level held by the DFF group 205 is logically operated by the EXOR group 207 at the subsequent stage and output as an output 208 and an output 209.
Note that the oscillation operation of the reference block 202 stops when the output of the frequency divider 204 changes from “L” to “H”. Since the stop of the oscillation operation and the latching of the input data from the switch group 206 in the DFF group 205 are performed at the same time, the output signal of the DFF group 205 may be disturbed. A delay circuit may be provided between the frequency divider 204 and the reference block 202 so that the DFF group 205 can correctly latch the input signal when the output of the frequency divider 204 changes. Alternatively, a circuit that holds the signal output from the frequency divider 204 to the reference block 202 until the DFF group 205 latches the input may be provided. Alternatively, the oscillation start and stop of the reference block 202 may be controlled separately from the output of the frequency divider 204 using the control signal 210.
Hereinafter, a method for calculating the deterioration state in the deterioration amount calculating unit 214 will be described. Here, for the sake of simplicity, it is assumed that only the test block 201 is deteriorated. However, it is easily understood that the deterioration state can be relatively calculated even when the operating environment of at least one of the test block 201 and the reference block 202 is changed. Further, when the test block 201 and the reference block 202 are interchanged, the relative deterioration state can be calculated in the same manner.
The connection between the test block 201 and the frequency divider 204 is described based on the configuration of FIG. However, even when a ring oscillator having another configuration or another circuit capable of timing comparison with the reference block 202 is used, the delay amount of the signal output from the test block is measured by the same operation. The amount of deterioration of the test block can be obtained.
In the deterioration detection circuit 220 shown in FIG. 3, the interval at which the output of the frequency divider 204 changes when the test block 201 and the frequency divider 204 are used is expressed by the following equation (1).
(T + ΔT) × 2 × N × M (1)
Here, T is a delay amount per stage of the NAND gate of the test block 201, and ΔT is a delay amount that increases due to deterioration per stage of the NAND gate of the test block 201. N is the number of NAND gate stages of the test block 201, and M is the frequency division ratio of the frequency divider 204. That is, the value of Equation (1) is the total sum of delay amounts generated in the test block at intervals at which the output of the frequency divider 204 changes. Since the internal circuits of the NAND gates 2011 to 201N are the same, it is assumed that the increase in delay amount at the time of performance deterioration in the NAND gates 2011 to 201N is all equal.
On the other hand, the delay amount represented by the following expression (2) is obtained in the reference block 202 as well.
T × X (2)
Expression (2) is a delay amount output from the reference block 202 while the output of the frequency divider 204 changes.
Here, X is the total number of stages of NAND gates through which the signal in the reference block 202 has passed from the fall of the output of the frequency divider 204 to the next rise. In general, in order to measure X, the reference block 202 also requires a counter and a large amount of DFF. Therefore, in the second embodiment, in order to simplify and reduce the area of the deterioration detection circuit, the difference between the delay amount of the test block 201 and the delay amount of the reference block 202 is obtained without directly obtaining X by a counter or the like. Measuring. Therefore, the deterioration detection circuit 220 selects a node of the reference block 202 to be input to the DFF group at the start of the measurement mode. The specific procedure will be described below.
First, the switch group 206 selects a certain node until the output of the frequency divider 204 rises. Then, a node in which the levels of the output 208 and the output 209 at the time point when the output of the frequency divider 204 falls is searched for. That is, a node where the output 208 is “L” and the output 209 is “H” or the output 208 is “H” and the output 209 is “L” is searched. Here, the opening / closing control of the switch group 206 and the investigation of the levels of the outputs 208 and 209 may be performed by the determination unit 203 or may be performed by another device under the control of the determination unit 203 from the outside.
FIG. 5 is a diagram illustrating signal timing in the second embodiment. In FIG. 5, the output of the frequency divider 204 is a signal obtained by dividing the output of the test block 201. Signals A to C and signals A ′ to C ′ indicate inputs and outputs of the DFFs 2051 to 2053, respectively. FIG. 5 shows a timing chart when one of the switches 2061 to 206N in which the output 208 is “L” and the output 209 is “H” is selected and closed at the start of the measurement mode.
The circuit configuration of the test block 201 and the reference block 202 as ring oscillators is the same. Therefore, when the test block 201 is not deteriorated as compared with the reference block 202, it is between the output of the frequency divider 201 and the signals (A to C) input from the reference block to the DFF group. The timing does not change ((a) of FIG. 5). However, when the deterioration of the test block 201 progresses and the delay of the output of the test block 202 increases, the rise timing of the frequency divider 204 is delayed with respect to the input signals (A to C) to the DFF group. . When the deterioration of the test block 201 progresses and the rising timing of the frequency divider 204 is delayed from t1 to t2 in FIG. 5, the output signal B ′ of the DFF 2052 is inverted. As a result, the outputs 208 and 209 are inverted, and the output 208 becomes “H” and the output 209 becomes “L” ((b) in FIG. 5).
Since the NAND gates 2121 to 2123 are connected in series, the timing of the input signals A to C to the DFF group 205 is delayed by one NAND gate of the inverter group 212 in the order of A, B, and C. When the rise timing of the frequency divider 204 is delayed from t1 to t2 across the inversion time of the output signal B ′ of the DFF 2052, the delay amount from the fall of the frequency divider 204 to the rise is averaged. One NAND gate. In the test block 201, it is considered that the delay for one gate stage has increased due to deterioration. Here, the number of gate stages through which the signal in the test block 201 has passed is “2 × N × M”. Therefore, when the signals A ′ to C ′ and the outputs 208 and 209 change from (a) to (b) in FIG. 5, the deterioration amount calculation unit 214 determines the deterioration amount per gate stage in the test block 201. Can be obtained as 1 / (2 × N × M).
Note that the determination unit 203 may select the node again by the switch group 206 after both the output 208 and the output 209 are inverted and the output 208 becomes “H” and the output 209 becomes “L”. For example, the determination unit 203 may shift the node selected in the reference block 202 backward by one in a state where the rising timing of the frequency divider 204 is delayed until t2 in FIG. Signals A to C when the node is shifted backward by one are indicated by broken lines in FIG. The relationship between the signals indicated by broken lines in FIG. 5 is that the signal B in FIG. 5 is a new signal A, and the signals obtained by sequentially inverting them are B and C. As a result, by shifting the node backward by one, the relationship between the rising timing of the frequency divider 204 and the input signals A to C to the DFF group 205 becomes the same as that at t1 in FIG.
As described above, after the output 208 and the output 209 are inverted, the node selected by the switch group 206 is shifted backward, whereby the output 208 can be set to “L” and the output 209 can be set to “H”. . Thereafter, when the deterioration of the test block 201 further proceeds and the output 208 becomes “H” and the output 209 becomes “L”, the deterioration amount calculating means 214 further proceeds with the deterioration by 2 / (2 × N XM) can be calculated.
Thereafter, by sequentially shifting the selected nodes backward in accordance with the delay of the rise timing of the frequency divider 204, the deterioration amount calculating unit 214 can calculate the deterioration state of the test block with a simple circuit configuration. .
In selecting a node, the timing at which the output 208 becomes “L” and the output 209 becomes “H” may be not only t1 in FIG. 5 but also t3. However, also in this case, when the rise timing of the frequency divider 204 is delayed from t3 to t4 due to the progress of performance deterioration of the test block 201, the output 208 is inverted to “H” and the output 209 is inverted to “L”. Then, the deterioration amount calculating means 214 can measure the deterioration amount of the test block 201 in the same procedure as described above.
That is, when the node is selected, the deterioration amount calculating means 214 is not limited to the test block whether the output 208 is “L” and the output 209 is “H” at t1 or t4 in FIG. The amount of degradation 201 can be calculated.
Thereafter, the deterioration amount calculating unit 214 can measure the progress of deterioration by sequentially shifting the nodes of the reference block 202 that inputs signals to the determining unit 203 backward. That is, the deterioration diagnosis circuit of the second embodiment measures the delay amount generated in the test block by comparing the output of the test block and the reference block without using a counter or a large number of DFFs. ing. As a result, the degradation diagnosis circuit of the second embodiment has an effect that the degradation state of the semiconductor integrated circuit can be measured more accurately with a simple circuit configuration.
Further, the deterioration diagnosis circuit of the second embodiment can continuously shift from the standby mode to the measurement mode by the control signal 210 without changing the circuit configuration of the test block 201 and the reference block 202. is there. For this reason, the deterioration diagnosis circuit of the second embodiment also has an effect of reducing the possibility that the deterioration amount is erroneously determined due to the influence of BTI.
In the second embodiment, as a modification, the determination unit 203 may include a characteristic adjustment unit as in the modification of the first embodiment. The characteristic adjusting unit adjusts the delay amount of the input signal so that the deterioration amount of the test block 201 approaches 0 in the deterioration amount calculating unit 214.
The characteristic adjustment means may be provided between at least one of the output of the test block 201 and the input of the frequency divider 204 and between the output of the frequency divider 204 and the input of the NAND gate 2021. In the modification of the second embodiment, the deterioration amount of the test block 201 is obtained based on the delay adjustment amount of the characteristic adjustment unit when the deterioration amount calculation unit 214 does not detect the deterioration of the test block 201. it can.
Furthermore, in the modification of the second embodiment, the determination unit 203 may have a function of outputting the characteristic adjustment amount of the characteristic adjustment unit to the outside.
(Third embodiment)
The configuration of the second embodiment shown in FIG. 3 is a configuration that compares the signals A to C delayed by one stage by the inverter group 212 with the rising timing of the frequency divider 204. That is, the resolution of the delay amount of the test block 201 to be detected is fixed to one gate.
In the third embodiment, a configuration in which a signal whose timing is compared with the output from the frequency divider 204 can be freely selected for each node will be described.
FIG. 4 is a diagram illustrating a configuration of a deterioration diagnosis circuit according to the third embodiment. The deterioration diagnosis circuit 300 shown in FIG. 4 includes a deterioration diagnosis block 320 and a control means 311. The deterioration diagnosis block 320 includes a test block 301, a reference block 302, and a determination unit 303. It should be noted that elements and descriptions unnecessary for the description of the operation are omitted as appropriate, and the configuration shown in FIG. 4 is clearly not limited to the only configuration that can achieve the object of the present invention. .
The test block 301 and the reference block 302 shown in FIG. 4 are composed of logic gates. The test block 301 is a ring oscillator composed of NAND gates 3011 to 301N, and the reference block 302 is a ring oscillator composed of NAND gates 3021 to 302N.
The determination unit 303 includes a frequency divider 304, a switch group 306, a DFF group 305, an EXOR group 307, and a deterioration amount calculation unit 314.
The frequency divider 304 divides a signal input from any node of the test block 301 and outputs the divided signal to the DFF group 305 and the NAND gate 3021 of the reference block 302. The switch group 306 branches each node of the reference block 302 into three and connects them to DFFs 3051 to 3052 constituting the DFF group 305.
The output of the DFF group 305 latches the output from the three nodes selected by the switch group 306 at the rise from the frequency divider 304. The EXOR group 307 performs two exclusive OR operations among the three outputs from the DFF group 305. Outputs 308 and 309 are outputs of EXORs 3071 and 3072, respectively, and are output to the deterioration amount calculation means 314.
The deterioration amount calculation unit 314 calculates the deterioration amount of the test block 301 based on the outputs 308 and 309. A specific method for calculating the deterioration amount in the deterioration amount calculating means 314 will be described later.
The switch group 306 includes three switches for each node of the reference block 302. In FIG. 4, switches 3061a to 3061c, 3062a to 3062c,... 306Na to 306Nc constitute a switch group 306. When a switch in the switch group 306 is closed (conducts), the output of the node is connected to any DFF in the DFF group 305.
For example, when the switch 3061a is closed, the output of the NAND gate 302N is input to the DFF 3051 through the signal line A. When the switch 3062b or 3062c is closed, the output of the NAND gate 3021 is input to the DFF 3052 or 3053 through the signal line B or C, respectively. The operation of each switch constituting the switch group 306 is the same for other nodes.
In the third embodiment, the switches 3061a and the like constituting the switch group 306 are individually controlled. With this configuration, the wiring of the signals A to C can be connected to any node of the reference block 302. That is, the delay amount between the signals A and B and the delay amount between the signals B and C can be set in units of the number of NAND gates. As a result, in the third embodiment, it is possible to change the resolution for detecting the change in the rise timing of the output of the frequency divider 204 described in FIG.
In the configuration described with reference to FIG. 3, in the inverter group 212, a signal input to the DFFs 2051 to 2053 is delayed by one gate. In FIG. 4, for example, the determination unit 303 can be configured so that a signal that causes a delay of one gate stage is input to the DFF group 305 by closing the switch 3061 a, the switch 3062 b, and the switch 3063 c.
Further, in the configuration shown in FIG. 4, for example, in addition to the switch 3061a, the switches 3064b and 3067c, which are not shown in the figure, are separated by 3 stages by a NAND gate, respectively, so that the DFF group 305 has a delay of 3 stages of NAND gates. The signal that caused the error is input. In this case, the delay between the change points of the signals A and B and B and C in FIG. 5 is the length of three NAND gates. Therefore, when the outputs 308 and 309 are inverted, a delay corresponding to three stages of gates is generated in the test block 301.
At this time, the number of gate stages through which the signal in the test block 301 has passed is 2 × N × M, and it is considered that the delay in the three stages of gates has increased in the test block 301 due to deterioration. For this reason, the deterioration amount calculating means 314 can determine the deterioration amount per gate stage in the test block 301 as 3 / (2 × N × M).
Note that the oscillation operation of the reference block 302 stops when the output of the frequency divider 304 changes from “L” to “H”. Since the stop of the oscillation operation and the latching of the input data from the switch group 306 in the DFF group 305 are performed at the same time, the output signal of the DFF group 305 may be disturbed. A delay circuit may be provided between the frequency divider 304 and the reference block 302 so that the DEF group 305 can correctly latch the input signal when the output of the frequency divider 304 changes. Alternatively, a circuit that holds the signal output from the frequency divider 304 to the reference block 302 until the DFF group 305 latches the input may be provided. Alternatively, the oscillation start and stop of the reference block 302 may be controlled separately from the output of the frequency divider 304 using the control signal 310.
As described above, the degradation diagnosis circuit of the third embodiment also has the effect that the degradation state of the semiconductor integrated circuit can be measured with a simple circuit configuration, similarly to the degradation diagnosis circuit of the second embodiment.
In the third embodiment, by providing the switch group 306, the measurement unit (resolution) of the deterioration amount of the test block 301 to be measured can be changed. By increasing the measurement unit of the deterioration amount, it is possible to average the delay amount variation due to each NAND gate and the delay amount error due to noise.
Note that the deterioration diagnosis circuit of the third embodiment is in a standby mode by the control signal 310 without changing the circuit configuration of the test block 301 and the reference block 302, similarly to the deterioration diagnosis circuit of the second embodiment. It is possible to make a continuous transition from measurement mode to measurement mode. For this reason, the deterioration diagnosis circuit of the third embodiment also has an effect of reducing the influence of recovery of deterioration due to BTI.
In the third embodiment, as a modification, the determination unit 303 may include a characteristic adjustment unit as in the modification of the first embodiment. The characteristic adjustment unit adjusts the delay amount of the input signal so that the deterioration amount of the test block 301 approaches 0 in the deterioration amount calculation unit 314.
The characteristic adjustment means may be provided between at least one of the output of the test block 301 and the input of the frequency divider 304 and between the output of the frequency divider 304 and the input of the NAND gate 3021. In the modification of the third embodiment, the deterioration amount of the test block 301 is obtained based on the delay adjustment amount of the characteristic adjustment means when the deterioration amount calculation means 314 does not detect the deterioration of the test block 301. it can.
Furthermore, in the modification of the third embodiment, the determination unit 303 may have a function of outputting the characteristic adjustment amount of the characteristic adjustment unit to the outside.
The embodiment of the present invention has been described above with reference to the first to third embodiments. However, the form to which the present invention can be applied is not limited to the above-described embodiment. Various changes that can be understood by those skilled in the art can be made to the configuration and detailed description of the present invention within the scope of the present invention.
This application claims the priority on the basis of Japanese application Japanese Patent Application No. 2011-182438 for which it applied on August 24, 2011, and takes in those the indications of all here.
The embodiment of the present invention can be described as in the following supplementary notes 1 to 12, but is not limited thereto.
(Appendix 1)
A test block including a first circuit to be diagnosed for deterioration;
A reference block including a second circuit having the same configuration as the first circuit;
When the signal indicating the measurement mode is input, the first characteristic of the first signal output from the test block is compared with the second characteristic of the second signal output from the reference block. Determination means for determining the presence or absence of the deterioration of the elements constituting the test block,
Control means for outputting a signal indicating the measurement mode to the determination means;
A deterioration diagnosis circuit comprising:
(Appendix 2)
The deterioration diagnosis circuit according to appendix 1, wherein the first characteristic and the second characteristic are electrical characteristics that are changed as a result of the deterioration, which are included in the first signal and the second signal, respectively.
(Appendix 3)
The first characteristic and the second characteristic each include at least one of a phase, a change timing, a frequency, an output voltage, an output current, a voltage amplitude, a current amplitude, and an ability to drive a load. Degradation diagnostic circuit described.
(Appendix 4)
4. The deterioration diagnosis circuit according to any one of appendices 1 to 3, wherein the determination unit detects a difference between a characteristic of an element constituting the test block and a characteristic of an element constituting the reference block.
(Appendix 5)
When a signal indicating the measurement mode is input, at least one of the first signal and the second signal according to an instruction from the determination unit so that the first characteristic and the second characteristic are equal to each other. Provided with characteristic adjustment means for performing predetermined adjustment on one side
The determination means determines the presence or absence of the deterioration depending on the necessity of the instruction.
The deterioration diagnosis circuit according to any one of appendices 1 to 3.
(Appendix 6)
The deterioration diagnosis circuit according to appendix 5, wherein the determination unit outputs an adjustment amount necessary for performing the adjustment as a deterioration amount indicating the degree of deterioration.
(Appendix 7)
7. The deterioration diagnosis circuit according to any one of appendices 1 to 6, wherein one or both of the first circuit and the second circuit has a circulation configuration in which a signal is output at a predetermined timing.
(Appendix 8)
8. The deterioration diagnosis circuit according to any one of appendices 1 to 7, wherein one or both of the first circuit and the second circuit has a configuration in which logic gates are cyclically connected.
(Appendix 9)
The deterioration diagnosis circuit according to any one of appendices 1 to 8, wherein the determination unit compares a delay time of an element constituting the test block with a delay time of an element constituting the reference block.
(Appendix 10)
The determination means includes
A frequency divider that outputs a third signal obtained by dividing the first signal;
A latch circuit that latches a fourth signal and a fifth signal having different delay amounts with respect to the second signal by the third signal and outputs the latched signal, respectively;
An EXOR (exclusive-OR) circuit for obtaining an exclusive OR of the outputs of the latch circuit;
A deterioration amount calculating means for calculating the deterioration amount based on an output of the EXOR circuit;
A deterioration diagnosis circuit according to any one of appendices 1 to 9.
(Appendix 11)
The deterioration diagnosis circuit according to appendix 10, wherein the fourth signal and the fifth signal are generated by delaying the second signal.
(Appendix 12)
The fourth signal and the fifth signal are the second signal output from positions corresponding to the preceding stage and the succeeding stage in the propagation path of the predetermined signal of the reference block. Deterioration diagnosis circuit.
(Appendix 13)
The deterioration diagnosis circuit according to any one of appendices 1 to 12, wherein when the control means does not output a signal indicating the measurement mode, the test block operates and the reference block stops operating.
(Appendix 14)
14. The deterioration diagnosis circuit according to any one of appendices 1 to 13, wherein the test block and the reference block operate in an operating environment in which stresses applied to the test block and the reference block are different.
(Appendix 15)
The same configuration as the first circuit and the first characteristic of the first signal output from the test block including the first circuit to be subjected to deterioration diagnosis when a signal indicating the measurement mode is input The presence or absence of deterioration of the elements constituting the test block is determined by comparing the second characteristic of the second signal output from the reference block including the second circuit including:
Deterioration diagnosis method.
 100、200、300 劣化診断回路
 101 劣化診断ブロック
 102、201、301 試験用ブロック
 103、202、302 参照用ブロック
 104、203、303 判定手段
 105、211、311 制御手段
 106 劣化診断ブロック
 107 特性調整手段
 204、304 分周器
 205、305 DFF群
 206、306 スイッチ群
 207、307 EXOR群
 208、308 出力
 209、309 出力
 210、310 制御信号
 213、313 スイッチ
 214、314 劣化量算出手段
 2011~201N、2021~202N、2121~2123 NANDゲート
 2051~2053、3051~3053 Dフリップフロップ
 2061~206N スイッチ
 3061a~306Na、3061b~306Nb、3061c~306Nc スイッチ
 2071、2072、3071、3072 EXORゲート
100, 200, 300 Deterioration diagnosis circuit 101 Degradation diagnosis block 102, 201, 301 Test block 103, 202, 302 Reference block 104, 203, 303 Judgment means 105, 211, 311 Control means 106 Degradation diagnosis block 107 Characteristic adjustment means 204, 304 Frequency divider 205, 305 DFF group 206, 306 Switch group 207, 307 EXOR group 208, 308 Output 209, 309 Output 210, 310 Control signal 213, 313 Switch 214, 314 Degradation amount calculation means 2011-201N, 2021 ~ 202N, 2121 ~ 2123 NAND gate 2051 ~ 2053, 3051 ~ 3053 D flip-flop 2061 ~ 206N Switch 3061a ~ 306Na, 3061b ~ 306Nb, 3061c ~ 306Nc Switch 2071,2072,3071,3072 EXOR gate

Claims (10)

  1. 劣化の診断の対象となる第1の回路を含む試験用ブロックと、
    前記第1の回路と同一の構成を備える第2の回路を含む参照用ブロックと、
    測定モードを示す信号が入力された場合に前記試験用ブロックから出力される第1の信号の第1の特性と前記参照用ブロックから出力される第2の信号の第2の特性とを比較することで前記試験用ブロックを構成する素子の前記劣化の有無を判定する判定手段と、
    前記測定モードを示す信号を前記判定手段に出力する制御手段と、
    を備える劣化診断回路。
    A test block including a first circuit to be diagnosed for deterioration;
    A reference block including a second circuit having the same configuration as the first circuit;
    When the signal indicating the measurement mode is input, the first characteristic of the first signal output from the test block is compared with the second characteristic of the second signal output from the reference block. Determination means for determining the presence or absence of the deterioration of the elements constituting the test block,
    Control means for outputting a signal indicating the measurement mode to the determination means;
    A deterioration diagnosis circuit comprising:
  2. 前記第1の特性及び前記第2の特性は、前記第1の信号及び前記第2の信号がそれぞれ備える、前記劣化の結果変化する電気的特性である、請求項1に記載された劣化診断回路。 2. The deterioration diagnosis circuit according to claim 1, wherein the first characteristic and the second characteristic are electrical characteristics that change as a result of the deterioration, which are included in the first signal and the second signal, respectively. .
  3. 前記測定モードを示す信号が入力された場合に、前記第1の特性および前記第2の特性が等しくなるように、前記判定手段からの指示に従って前記第1の信号および前記第2の信号の少なくとも一方に所定の調整を行う特性調整手段を備え、
    前記判定手段は、前記指示の要否によって前記劣化の有無を判定する、
    請求項1又は2に記載された劣化診断回路。
    When a signal indicating the measurement mode is input, at least one of the first signal and the second signal according to an instruction from the determination unit so that the first characteristic and the second characteristic are equal to each other. Provided with characteristic adjustment means for performing predetermined adjustment on one side
    The determination means determines the presence or absence of the deterioration depending on the necessity of the instruction.
    The deterioration diagnosis circuit according to claim 1 or 2.
  4. 前記判定手段は、前記調整を行うために必要な調整量を前記劣化の程度を示す劣化量として出力する、請求項3に記載された劣化診断回路。 The deterioration diagnosis circuit according to claim 3, wherein the determination unit outputs an adjustment amount necessary for performing the adjustment as a deterioration amount indicating the degree of deterioration.
  5. 前記判定手段は、
    前記第1の信号を分周した第3の信号を出力する分周器と、
    前記第2の信号に対して互いに異なる遅延量を持つ第4の信号及び第5の信号を前記第3の信号でラッチしてそれぞれ出力するラッチ回路と、
    前記ラッチ回路の出力の排他的論理和を求めるEXOR(exclusive−OR)回路と、
    前記EXOR回路の出力に基づいて前記劣化量を算出する劣化量算出手段と、
    を備える請求項1乃至4のいずれかに記載された劣化診断回路。
    The determination means includes
    A frequency divider that outputs a third signal obtained by dividing the first signal;
    A latch circuit that latches a fourth signal and a fifth signal having different delay amounts with respect to the second signal by the third signal and outputs the latched signal, respectively;
    An EXOR (exclusive-OR) circuit for obtaining an exclusive OR of the outputs of the latch circuit;
    A deterioration amount calculating means for calculating the deterioration amount based on an output of the EXOR circuit;
    A deterioration diagnosis circuit according to claim 1, comprising:
  6. 前記第4の信号及び前記第5の信号は、前記第2の信号を遅延させることで生成される、請求項5に記載された劣化診断回路。 The deterioration diagnosis circuit according to claim 5, wherein the fourth signal and the fifth signal are generated by delaying the second signal.
  7. 前記第4の信号及び前記第5の信号は、前記参照用ブロックの、所定の信号の伝播経路における前段および後段に相当する位置から出力される前記第2の信号である、請求項5に記載された劣化診断回路。 The said 4th signal and the said 5th signal are the said 2nd signal output from the position corresponded to the front | former stage and back | latter stage in the propagation path of a predetermined | prescribed signal of the said reference block. Deterioration diagnosis circuit.
  8. 前記制御手段が前記測定モードを示す信号を出力していない場合に、前記試験用ブロックは動作し前記参照用ブロックは動作を停止する、請求項1乃至7のいずれかに記載された劣化診断回路。 8. The deterioration diagnosis circuit according to claim 1, wherein when the control means does not output a signal indicating the measurement mode, the test block operates and the reference block stops operating. .
  9. 前記試験用ブロック及び前記参照用ブロックは、前記試験用ブロック及び前記参照用ブロックに印加されるストレスが異なる動作環境で動作する、請求項1乃至8のいずれかに記載された劣化診断回路。 The deterioration diagnosis circuit according to claim 1, wherein the test block and the reference block operate in an operating environment in which stresses applied to the test block and the reference block are different.
  10. 測定モードを示す信号が入力された場合に、劣化の診断の対象となる第1の回路を含む試験用ブロックから出力される第1の信号の第1の特性と前記第1の回路と同一の構成を備える第2の回路を含む参照用ブロックから出力される第2の信号の第2の特性とを比較することで前記試験用ブロックを構成する素子の劣化の有無を判定する、劣化診断方法。 When a signal indicating a measurement mode is input, the first characteristic of the first signal output from the test block including the first circuit to be diagnosed for deterioration is the same as the first circuit. Deterioration diagnosis method for determining presence / absence of deterioration of elements constituting said test block by comparing with second characteristic of second signal output from reference block including second circuit having configuration .
PCT/JP2012/071109 2011-08-24 2012-08-15 Deterioration diagnostic circuit and deterioration diagnostic method WO2013027739A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US14/240,024 US20140218060A1 (en) 2011-08-24 2012-08-15 Degradation diagnosing circuit and degradation diagnosing method

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2011182438 2011-08-24
JP2011-182438 2011-08-24

Publications (1)

Publication Number Publication Date
WO2013027739A1 true WO2013027739A1 (en) 2013-02-28

Family

ID=47746478

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2012/071109 WO2013027739A1 (en) 2011-08-24 2012-08-15 Deterioration diagnostic circuit and deterioration diagnostic method

Country Status (2)

Country Link
JP (1) JPWO2013027739A1 (en)
WO (1) WO2013027739A1 (en)

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20190068493A (en) * 2016-04-19 2019-06-18 아날로그 디바이시즈 글로벌 언리미티드 컴퍼니 Wear monitor device
JP2020036095A (en) * 2018-08-27 2020-03-05 株式会社デンソー Pulse signal abnormality detection device
US10794950B2 (en) 2016-04-19 2020-10-06 Analog Devices Global Wear-out monitor device
JP2021503091A (en) * 2017-11-15 2021-02-04 プロテアンテクス リミテッド Integrated circuit margin measurement and failure prediction device
US11391771B2 (en) 2017-11-23 2022-07-19 Proteantecs Ltd. Integrated circuit pad failure detection
US11740281B2 (en) 2018-01-08 2023-08-29 Proteantecs Ltd. Integrated circuit degradation estimation and time-of-failure prediction using workload and margin sensing
US11762789B2 (en) 2018-12-30 2023-09-19 Proteantecs Ltd. Integrated circuit I/O integrity and degradation monitoring
US11762013B2 (en) 2018-04-16 2023-09-19 Proteantecs Ltd. Integrated circuit profiling and anomaly detection
US11929131B2 (en) 2019-12-04 2024-03-12 Proteantecs Ltd. Memory device degradation monitoring

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH03170818A (en) * 1989-11-29 1991-07-24 Shimadzu Corp Automatic null balancing measuring apparatus
JPH11351883A (en) * 1998-06-10 1999-12-24 Japan Aviation Electronics Ind Ltd Closed-loop optical interference angular velocity meter
JP2004253891A (en) * 2003-02-18 2004-09-09 Makoto Ogawa Semiconductor circuit
JP2004266243A (en) * 2002-09-13 2004-09-24 Chartered Semiconductor Mfg Ltd Test structure for on-chip real-time reliability testing
WO2011027553A1 (en) * 2009-09-07 2011-03-10 日本電気株式会社 Aging diagnostic device, aging diagnostic method

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH03170818A (en) * 1989-11-29 1991-07-24 Shimadzu Corp Automatic null balancing measuring apparatus
JPH11351883A (en) * 1998-06-10 1999-12-24 Japan Aviation Electronics Ind Ltd Closed-loop optical interference angular velocity meter
JP2004266243A (en) * 2002-09-13 2004-09-24 Chartered Semiconductor Mfg Ltd Test structure for on-chip real-time reliability testing
JP2004253891A (en) * 2003-02-18 2004-09-09 Makoto Ogawa Semiconductor circuit
WO2011027553A1 (en) * 2009-09-07 2011-03-10 日本電気株式会社 Aging diagnostic device, aging diagnostic method

Cited By (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11269006B2 (en) 2016-04-19 2022-03-08 Analog Devices International Unlimited Company Exposure monitor device
JP2019518326A (en) * 2016-04-19 2019-06-27 アナログ・デヴァイシズ・グローバル Wear monitoring device
US11686763B2 (en) 2016-04-19 2023-06-27 Analog Devices International Unlimited Company Exposure monitor device
US10794950B2 (en) 2016-04-19 2020-10-06 Analog Devices Global Wear-out monitor device
KR102172350B1 (en) 2016-04-19 2020-11-02 아날로그 디바이시즈 글로벌 언리미티드 컴퍼니 Wear monitor device
KR20190068493A (en) * 2016-04-19 2019-06-18 아날로그 디바이시즈 글로벌 언리미티드 컴퍼니 Wear monitor device
US11385282B2 (en) 2017-11-15 2022-07-12 Proteantecs Ltd. Integrated circuit margin measurement and failure prediction device
JP2021503091A (en) * 2017-11-15 2021-02-04 プロテアンテクス リミテッド Integrated circuit margin measurement and failure prediction device
JP7130757B2 (en) 2017-11-15 2022-09-05 プロテアンテクス リミテッド Integrated circuit margin measurement and failure prediction equipment
US11841395B2 (en) 2017-11-15 2023-12-12 Proteantecs Ltd. Integrated circuit margin measurement and failure prediction device
US11391771B2 (en) 2017-11-23 2022-07-19 Proteantecs Ltd. Integrated circuit pad failure detection
US11740281B2 (en) 2018-01-08 2023-08-29 Proteantecs Ltd. Integrated circuit degradation estimation and time-of-failure prediction using workload and margin sensing
US11762013B2 (en) 2018-04-16 2023-09-19 Proteantecs Ltd. Integrated circuit profiling and anomaly detection
JP7095491B2 (en) 2018-08-27 2022-07-05 株式会社デンソー Pulse signal abnormality detector
JP2020036095A (en) * 2018-08-27 2020-03-05 株式会社デンソー Pulse signal abnormality detection device
US11762789B2 (en) 2018-12-30 2023-09-19 Proteantecs Ltd. Integrated circuit I/O integrity and degradation monitoring
US11929131B2 (en) 2019-12-04 2024-03-12 Proteantecs Ltd. Memory device degradation monitoring

Also Published As

Publication number Publication date
JPWO2013027739A1 (en) 2015-03-19

Similar Documents

Publication Publication Date Title
WO2013027739A1 (en) Deterioration diagnostic circuit and deterioration diagnostic method
TWI548886B (en) Aging detection circuit and method thereof
US8674774B2 (en) Aging diagnostic device, aging diagnostic method
US8072253B2 (en) Clock adjusting circuit and semiconductor integrated circuit device
JP2005098981A (en) Semiconductor integrated circuit device, measurement result managing system, and management server
US20140218060A1 (en) Degradation diagnosing circuit and degradation diagnosing method
US20130002274A1 (en) Aging degradation diagnosis circuit and aging degradation diagnosis method for semiconductor integrated circuit
JP5173216B2 (en) Semiconductor integrated circuit system, semiconductor integrated circuit, operating system, and control method of semiconductor integrated circuit
JP4644205B2 (en) Test apparatus, test method, and electronic device
US8560993B2 (en) Semiconductor device and method of testing the same
JP2017103629A (en) Delay circuit, dll circuit, and fault relief method for delay circuit
CN109752636B (en) Test circuit for monitoring temperature instability
EP3292420B1 (en) Ring oscillator test circuit
US20140125419A1 (en) Method and system for testing oscillator circuit
JP4040393B2 (en) Jitter test circuit, semiconductor device equipped with jitter test circuit, and jitter test method
US7168020B2 (en) Circuit and method for testing embedded phase-locked loop circuit
Simevski et al. Low-complexity integrated circuit aging monitor
JP6242183B2 (en) Semiconductor integrated circuit, method for testing semiconductor integrated circuit, and method for suppressing rush current in semiconductor integrated circuit
JP2005140759A (en) Semiconductor integrated circuit and failure detection method for semiconductor integrated circuit
JP5731618B2 (en) Critical path circuit for performance monitoring
US8618823B2 (en) Semiconductor device
JP5741817B2 (en) Semiconductor integrated circuit
KR101917718B1 (en) Semiconductor integrated circuit
Lee et al. Rigorous test flow for PLL to identify weak devices
JP2001296334A (en) Integrated circuit and failure detection method

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 12825094

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2013530029

Country of ref document: JP

Kind code of ref document: A

WWE Wipo information: entry into national phase

Ref document number: 14240024

Country of ref document: US

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 12825094

Country of ref document: EP

Kind code of ref document: A1