Measurement substrate, measurement device, and measurement method
The measurement substrate and device facilitate efficient and flexible testing of multiple semiconductor memories by allowing simultaneous irradiation and calculation of soft error rates within a limited area, addressing the inefficiencies of existing methods by enabling high-density memory arrangement and arbitrary memory selection and configuration.
Patent Information
- Application Number
- PCT/JP2023/018406
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2023-05-17
- Publication Date
- 2024-11-21
AI Technical Summary
Existing methods for measuring the soft error rate (SER) of semiconductor memory in accelerated testing are limited by restrictions on radiation irradiation area and inefficiencies when testing multiple memories simultaneously, as they require a large board area and do not allow for selection or configuration of individual memory circuits.
A measurement substrate and device that allows simultaneous irradiation and measurement of multiple arbitrarily selected semiconductor memories using a configuration with overlapping irradiation substrates and a control board to calculate the soft error occurrence rate, enabling efficient testing within a limited radiation area and allowing for selection and configuration of memory circuits.
Enables high-density arrangement and testing of multiple semiconductor memories, maintaining test efficiency even if one memory fails, and allowing for arbitrary selection and configuration of memory circuits, thereby improving the overall testing process.
Smart Images

Figure JP2023018406_21112024_PF_FP_ABST
Abstract
Description
Measurement substrate, measurement device, and measurement method
[0001] The present invention relates to a measurement substrate, a measurement device, and a measurement method.
[0002] As a method for measuring the soft error rate (hereinafter referred to as "SER") of a semiconductor memory (hereinafter abbreviated as "memory"), Non-Patent Document 1 discloses an accelerated test using a particle accelerator radiation source.
[0003] A soft error is an event in which a single particle such as a neutron, proton, or heavy particle enters an LSI such as a memory, causing the data (bit) stored in the LSI to be inverted due to the charge generated by the nuclear reaction. SER refers to the soft error that occurs per unit time or per unit bit.
[0004] Hidenori Iwashita, Gentaro Funatsu, HirotakaSato, Takashi Kamiyama, Michihiro Furusaka, StephenA. Wender, Eric Pitcher, YoshiakiKiyanagi, “Energy-Resolved Soft-Error Rate Measurements for 1?800 MeV Neutrons by the Time-of-Flight Technique at LANSCE,” IEEE Transactions on Nuclear Science, vol. 67, no. 11, pp. 2363-2369, November2020.
[0005] In accelerated testing, the memory under test is irradiated with high-intensity radiation from a particle accelerator radiation source, allowing for rapid measurement of natural SER.
[0006] Furthermore, in order to improve the efficiency of the accelerated test, multiple memories (e.g., FPGAs (Field Programmable Gate Arrays)) are mounted on one board, and the multiple memories are simultaneously irradiated with radiation to measure the SER.
[0007] However, in such accelerated testing, there is a limit to the area that can be irradiated with radiation, and furthermore, the area of the board on which the memory to be inspected is mounted is large, making it difficult to test many memories simultaneously.
[0008] Furthermore, if multiple memories are mounted on a single board in order to test many memories simultaneously, if one of the memories fails, test results cannot be obtained, resulting in a problem of poor test efficiency.Furthermore, since the circuit configuration of each memory mounted on a single board cannot be changed, there is a drawback in that it is not possible to select and test any memory.
[0009] The present disclosure has been made in consideration of the above circumstances, and its purpose is to provide a measurement substrate, a measurement device, and a measurement method that are capable of simultaneously irradiating radiation on a plurality of arbitrarily selected semiconductor memories and measuring the soft error occurrence rate.
[0010] A measurement substrate according to one embodiment of the present disclosure is a measurement substrate used to measure the soft error rate, and includes a plurality of irradiation substrates each having a first region on which a semiconductor device is mounted and a second region on which a connecting member for connecting the semiconductor device is mounted, and at least two of the irradiation substrates are arranged so as to overlap in the normal direction such that the first regions of the irradiation substrates do not overlap in the normal direction.
[0011] A measuring apparatus according to one aspect of the present disclosure is a measuring apparatus that measures the soft error rate by irradiating a semiconductor device with radiation, and includes a plurality of irradiation substrates each having a first region on which the semiconductor device is mounted and a second region on which a connecting member for connecting the semiconductor device is mounted, the measurement substrate being arranged so that the first region of each irradiation substrate is located within a predetermined irradiation region onto which the radiation is irradiated, and a control substrate connected to each irradiation substrate, which acquires error signals detected in each semiconductor device when the radiation is irradiated, and calculates the soft error rate based on the error signals.
[0012] A measurement method according to one aspect of the present disclosure is a method for measuring a soft error rate by irradiating radiation onto semiconductor devices, the method comprising: irradiating radiation onto a plurality of measurement substrates each having a first region on which the semiconductor devices are mounted and a second region on which connecting members for connecting the semiconductor devices are mounted; irradiating the measurement substrates with radiation such that the first region of each measurement substrate is positioned within a predetermined irradiation region onto which the radiation is irradiated; acquiring error signals detected in each of the semiconductor devices; and calculating the soft error rate based on the error signals.
[0013] According to the present disclosure, it is possible to simultaneously irradiate a plurality of arbitrarily selected semiconductor memories with radiation and measure the soft error occurrence rate.
[0014] Fig. 1 is a perspective view of a measurement substrate according to an embodiment. Fig. 2 is a plan view of a measurement substrate according to an embodiment. Fig. 3 is a plan view of an irradiation substrate according to an embodiment. Fig. 4 is a perspective view of a measurement substrate according to a modified example. Fig. 5 is an explanatory diagram showing the configuration of a measurement device according to an embodiment. Fig. 6 is a block diagram showing the electrical configuration of a measurement device according to an embodiment. Fig. 7 is a schematic diagram of an acceleration test device using a particle accelerator radiation source.
[0015] Hereinafter, an embodiment of the present disclosure will be described with reference to the drawings. Fig. 1 is a perspective view showing the configuration of a measurement substrate 100 according to the embodiment, Fig. 2 is a plan view, and Fig. 3 is a plan view showing the configuration of an irradiation substrate P mounted on the measurement substrate 100.
[0016] 3, the irradiation substrate P has a long rectangular shape in a plan view. The irradiation substrate P includes a memory 11 (semiconductor memory), a wiring pattern 12, and an HDMI (High-Definition Multimedia Interface) terminal 14.
[0017] For example, an FPGA, which allows the internal logic circuit configuration to be arbitrarily set by a user through setting operations, can be used as the memory 11. The memory 11 has a rectangular shape in a plan view. The memory 11 is provided on one end side of the irradiation substrate P in the long side direction.
[0018] The irradiation substrate P can be equipped with multiple types of memories 11. That is, the memories 11 equipped on the irradiation substrate P differ in size, number of pins, and connection conditions. For this reason, the irradiation substrate P is designed with wiring that corresponds to each memory 11.
[0019] The HDMI terminal 14 is provided on the other end side in the long side direction of the irradiation substrate P. The memory 11 and the HDMI terminal 14 are electrically connected by a plurality of wiring patterns 12. By connecting a dedicated cable (W1 shown in FIG. 5 ) to the HDMI terminal 14, it is possible to electrically connect it to a control substrate 23, which will be described later.
[0020] In order to arrange many memories 11 within a determined area (within the irradiation area R1 shown in FIG. 2 ), the irradiation substrate P minimizes the number of devices mounted on the irradiation substrate P. Specifically, the irradiation substrate P has an area (first area) on which semiconductor devices such as memories 11 are mounted, and an area (second area) on which connecting members (HDMI terminals 14 and wiring patterns 12) that connect the semiconductor devices are mounted.
[0021] The irradiation substrate P is mounted with the semiconductor devices and connecting members described above. Furthermore, as necessary, capacitors, inductors, and flash memories are mounted to satisfy the drive conditions of the semiconductor devices to be irradiated. No other electronic components are mounted on the irradiation substrate P. In other words, the irradiation substrate P only has a first region for mounting the semiconductor devices, a second region for mounting the connecting members that connect the semiconductor devices, and a region for mounting other necessary components. This allows the area of the irradiation substrate P to be reduced.
[0022] 3, two long holes 13A and 13B are drilled along each of the two long sides of the irradiation substrate P. As will be described later, the long holes 13A and 13B are used as holes through which bolts 16 (see FIGS. 1 and 2) are inserted when multiple irradiation substrates P are arranged in an overlapping manner.
[0023] When a soft error is detected in the memory 11 during the accelerated test, the irradiation substrate P transmits an error signal and data on the type of error to the control substrate 23 (details of which will be described later) shown in FIGS.
[0024] 1 and 2, the measurement substrate 100 includes a plurality of (six in the figures) irradiation substrates P1 to P6 and a fixing plate 51. Hereinafter, when each of the irradiation substrates P1 to P6 shown in FIGS. 1 and 2 is specifically indicated, it will be described with a suffix such as "irradiation substrate P1," and when it is not specifically indicated or when it is referred to collectively, it will be described without a suffix such as "irradiation substrate P." Note that, in this embodiment, an example will be described in which six irradiation substrates P are mounted on the measurement substrate 100, but the number of irradiation substrates P is not limited to six.
[0025] The fixed plate 51 has a rectangular shape, and on one side thereof, three irradiation substrates P2, P4, and P6 are arranged in parallel at positions spaced a predetermined distance apart in the normal direction of the fixed plate 51. On the front side of each of the irradiation substrates P2, P4, and P6, three irradiation substrates P1, P3, and P4 are arranged in parallel at positions spaced a predetermined distance apart in the normal direction of the irradiation substrates P2, P4, and P6.
[0026] Specifically, bolts 16 are inserted through the elongated holes 13A of the three front-side irradiation substrates P1, P3, and P5 and the elongated holes 13B of the three rear-side irradiation substrates P2, P4, and P6, and are then fixed in place by installing spacers 15. Therefore, as shown in Figure 2, the three front-side irradiation substrates P1, P3, and P5 and the three rear-side irradiation substrates P2, P4, and P6 are arranged with a distance L1 offset in the longitudinal direction.
[0027] By installing the spacer 15, a predetermined distance of space is secured between the fixed plate 51 and the rear side irradiation substrates P2, P4, P6, and between the rear side irradiation substrates P2, P4, P6 and the front side irradiation substrates P1, P3, P5.
[0028] By setting the offset L1, the memories 11 of the irradiation substrates P1, P3, and P5 and the memories 11 of the irradiation substrates P2, P4, and P6 do not overlap in the direction perpendicular to the fixing plate 51 (the normal direction). That is, when radiation is irradiated from the normal direction of the measurement substrate 100, the radiation is irradiated onto the memories 11 mounted on each irradiation substrate P without being blocked. Furthermore, by using the above-described arrangement, it is possible to concentrate the memories 11 of each irradiation substrate P within a limited range. Specifically, the memories 11 of each irradiation substrate P can be arranged within the irradiation region R1 shown in FIG. 2. That is, the memory 11 (first region) of each irradiation substrate P is arranged so as to be located within the predetermined irradiation region R1 irradiated with radiation.
[0029] Figure 4 is a perspective view showing a modified example of the measurement board 100 shown in Figures 1 and 2. The measurement board 100A shown in Figure 4 includes six irradiation boards P1 to P6. The measurement board 100A is placed on the fixing plate 51 so that the short sides of the two irradiation boards P on the memory 11 side are adjacent to each other.
[0030] Specifically, the irradiation substrate P1 is arranged so that the short side on the memory 11 side is adjacent to the short side on the memory 11 side of the irradiation substrate P2. Similarly, the irradiation substrates P3 and P4, and P5 and P6 are arranged so that the short sides on the memory 11 side are adjacent to each other. By arranging the six irradiation substrates P as shown in FIG. 4, it is possible to concentrate the memories 11 of each irradiation substrate P within a limited area, similar to the measurement substrate 100 shown in FIGS. 1 and 2. That is, the memories 11 (first region) of each irradiation substrate P are arranged so as to be located within a predetermined irradiation region irradiated with radiation. Although not shown in FIG. 4, spacers are installed between the fixing plate 51 shown in FIG. 4 and each irradiation substrate P, ensuring a predetermined distance of space between the fixing plate 51 and each irradiation substrate P.
[0031] Next, a description will be given of a measurement device for measuring the SER (soft error rate) using the above-described measurement substrates 100 and 100A. Fig. 5 is a configuration diagram of a measurement device 200 according to this embodiment, and Fig. 6 is a block diagram showing the electrical configuration of the measurement device 200.
[0032] As shown in Figures 5 and 6, the measuring device 200 includes a measuring board 100A, a personal computer 21, a power supply 22, and a control board 23. In the following, an example will be shown in which the measuring board 100A shown in Figure 4 is used in the measuring device 200. The measuring board 100 shown in Figures 1 and 2 may be used instead of the measuring board 100A.
[0033] 5 and 6 outputs a power supply signal to a voltage conversion unit 23B (details of which will be described later) of the control board 23. The power supply signal is a signal including a plurality of voltages required to drive the memory 11 (e.g., FPGA).
[0034] 6, the control board 23 includes a control unit 23 A and a voltage conversion unit 23 B. It is desirable to install the control board 23 in a location that is less susceptible to the effects of radiation when an accelerated test is performed.
[0035] The control board 23 has a connection portion CN shown in Fig. 5, and is connected to each irradiation board P mounted on the measurement board 100A by a cable W1. The control board 23 performs data communication through the cable W1.
[0036] The control board 23 performs data communication with the personal computer 21. The control board 23 and the personal computer 21 can be connected and disconnected, and the user can connect the personal computer 21 to the control board 23 as needed to obtain various data acquired by the control board 23. Furthermore, by connecting the personal computer 21 to the control board 23, setting data can be written to the memory 11 mounted on the irradiation board P.
[0037] The voltage conversion unit 23B shown in Figure 6 includes an irradiation unit power supply 25A and a control unit power supply 25B. Each of the power supplies 25A and 25B includes a voltage conversion circuit such as a switching regulator, and converts the power signal supplied from the power supply 22 into a desired voltage. That is, the irradiation unit power supply 25A generates a voltage for driving the irradiation substrate P and outputs this voltage to the measurement substrate 100A in response to an on / off signal output from the control unit 23A. The control unit power supply 25B generates a voltage for driving the control unit 23A and outputs it to the control unit 23A.
[0038] The cable W1 connecting the control unit 23A and the measurement board 100A includes the following signal lines S1 to S7. Specifically, a reconfiguration signal line S1 that transmits a reconfiguration signal to the measurement board 100A and a JTAG signal line S2 that transmits a JTAG signal are provided between the control unit 23A and the measurement board 100A. The reconfiguration signal is a signal for reconfiguring the memory 11 (for example, FPGA) when an error occurs in the memory 11. The JTAG signal is a signal that includes design data for the irradiation board P.
[0039] Between the control unit 23A and the measurement board 100A, there are provided a serial signal line S3 for transmitting and receiving serial signals, a clock signal line S4 for transmitting clock signals, an error signal line S5 for transmitting error signals, and an arbitrary signal line S6 for transmitting arbitrary signals. The serial signal includes a signal indicating the type of error detected on the irradiation board P. The clock signal is a clock for operating the memory 11. The error signal is a signal indicating the count value of errors detected on the irradiation board P.
[0040] A power supply signal line S7 for transmitting a power supply signal is provided between the voltage conversion section 23B and the measurement board 100A.
[0041] Furthermore, if the memory 11 mounted on the irradiation substrate P to be irradiated with radiation is an SRAM (Static Random Access Memory), an FPGA for controlling the SRAM is mounted on the irradiation substrate P, and the SRAM is controlled via this FPGA. Therefore, the signal lines included in the cable W1 connecting the measurement substrate 100A and the control substrate 23 are the same as the signal lines S1 to S7 described above.
[0042] The personal computer 21 transmits a JTAG signal to the control unit 23A of the control board 23. The JTAG signal is a signal that includes design data for the irradiation substrate P. The JTAG signal transmitted from the personal computer 21 is transmitted to the irradiation substrate P via the control unit 23A of the control board 23. The design data is written to the memory 11 mounted on each irradiation substrate P by the JTAG signal.
[0043] The personal computer 21 receives the error signal transmitted from the control unit 23A, and during the accelerated test, monitors the occurrence state of the error signal in the memory 11 of each irradiation substrate P. The personal computer 21 displays the SER (soft error rate) of each memory 11 transmitted from the control board 23 on a display (not shown) or the like.
[0044] As mentioned above, the personal computer 21 can be connected and disconnected from the control board 23, and can be connected and used when writing design data to each memory 11 and when monitoring the status of the memory 11 during an accelerated test.
[0045] Fig. 7 is a schematic diagram of an acceleration test apparatus using a particle accelerator radiation source. As shown in Fig. 7, the acceleration test apparatus MA includes a radiation source 31 and a shielding body 32. The shielding body 32 is made of a material that blocks radiation, such as concrete.
[0046] The shield 32 has two opposing side surfaces, and an opening 34 is formed through the shield 32, penetrating from one side surface 32a to the other side surface 32b. That is, a rectangular space is formed from the one side surface 32a to the other side surface 32b.
[0047] 1 and 2 is placed on the other side surface 32b. More specifically, the measurement substrate 100 or 100A is placed so that the irradiation surface 33, which is the area irradiated with radiation emitted from the radiation source 31 and passed through the opening 34, coincides with the irradiation area R1 (see FIG. 2) of the irradiation substrate P. Therefore, the radiation emitted from the radiation source 31 and passed through the opening 34 is irradiated onto the irradiation area R1 in which the six markings 11 are provided.
[0048] Next, a description will be given of the operation of the measuring device 200 according to this embodiment. As an initial setting, the personal computer 21 shown in Figures 5 and 6 is connected to the control board 23, and design data is written to the memory 11 mounted on each irradiation board P.
[0049] 2 is set on the irradiation surface 33 of the acceleration test apparatus MA shown in Fig. 7. That is, the memories 11 (six memories 11) of each irradiation substrate P mounted on the measurement substrate 100 or 100A are set so as to coincide with the irradiation surface 33.
[0050] 7 is activated to emit radiation. The radiation emitted from the radiation source 31 passes through the openings 34 of the shielding body 32 and is irradiated onto each memory 11. That is, radiation is irradiated onto the memories 11 of a plurality of irradiation substrates P, each of which has an area (first area) on which the memories 11 (semiconductor devices) are mounted, and an area (second area) on which the HDMI 14 (connecting member) connecting the memories 11 and the wiring pattern 12 (connecting member) are mounted.
[0051] When each irradiation substrate P detects a soft error in the memory 11, it transmits an error signal to the control unit 23A of the control substrate 23 shown in FIG.
[0052] The control unit 23A receives the error signals transmitted from the measurement boards 100 and 100A and counts the error signals. The control unit 23A calculates the SER (soft error rate) based on the count value of the error signals. The control unit 23A transmits the calculated SER to the personal computer 21.
[0053] The personal computer 21 receives the SER transmitted from the control unit 23A and displays the SER on a display (not shown) or the like in response to a user request. The user can operate the personal computer 21 to recognize the SER of each irradiation substrate P. In this way, it becomes possible to measure the SER (soft error rate) using the measurement substrate 100A on which a plurality of irradiation substrates P, each equipped with a memory 11, are mounted.
[0054] In this way, the measuring apparatus 200 according to this embodiment is a measuring apparatus 200 that measures the soft error rate by irradiating radiation onto a memory 11 (semiconductor device), and is equipped with a plurality of irradiation substrates P each having a first region on which the memory 11 is mounted and a second region on which connecting members (wiring pattern 12, HDMI terminal 14) that connect the memory 11 are mounted, and is equipped with measurement substrates 100, 100A in which at least two irradiation substrates P are arranged to overlap in the normal direction so that the first regions of each irradiation substrate P do not overlap in the normal direction, and a control substrate 23 that is connected to each irradiation substrate P and calculates the soft error rate (SER) based on error signals detected in each memory 11.
[0055] In this embodiment, the memory-mounted board that has been conventionally used is divided into an irradiation board P and a control board 23, and further, the memories 11 of the multiple irradiation boards P1 to P6 are arranged so that they do not overlap in the normal direction. Therefore, since the irradiation boards P1 to P6 can be arranged in a concentrated manner in the irradiation region R1 shown in Figure 2, even if the radiation irradiation surface 33 is limited as shown in Figure 7, the memories 11 can be arranged with high density on this irradiation surface 33. This makes it possible to perform accelerated testing on a large number of memories 11.
[0056] Since the design data of each irradiation board P can be changed by operating the personal computer 21, even if one of the memories 11 mounted on each irradiation board P fails, as in the conventional case, the accelerated test can be performed using the other normal memories 11. This solves the problem of reduced test efficiency. Furthermore, since the setting data of each memory 11 can be changed as desired, it becomes possible to appropriately select the memory to be tested and perform the accelerated test.
[0057] Furthermore, by using an FPGA as the memory 11 (semiconductor memory), it becomes possible to easily set design data for the memory 11 .
[0058] 1, spacers 15 are installed between the fixed plate 51 and the rear-side irradiation substrates P2, P4, and P6, and between the rear-side irradiation substrates P2, P4, and P6 and the front-side irradiation substrates P1, P3, and P5. This ensures a space between the irradiation substrates P without causing contact between the irradiation substrates P, making it possible to stably fix the irradiation substrates P to the fixed plate 51.
[0059] The present disclosure is not limited to the above-described embodiments, and various modifications are possible within the scope of the gist of the present disclosure.
[0060] REFERENCE SIGNS LIST 11 Memory 12 Wiring pattern 13A, 13B Slot 14 HDMI terminal 15 Spacer 21 Personal computer 22 Power supply 23 Control board 23A Control unit 23B Voltage conversion unit 25A Irradiation unit power supply 25B Control unit power supply 31 Radiation source 32 Shielding body 33 Irradiation surface 34 Opening 51 Fixing plate 100, 100A Measurement board 200 Measurement device P (P1 to P6) Irradiation board R1 Irradiation area
Claims
1. A measurement board used to measure the soft error occurrence rate, comprising a plurality of irradiation boards each having a first region for mounting a semiconductor device and a second region for mounting a connection member for connecting the semiconductor device, wherein at least two of the irradiation boards are arranged so that the first regions of the irradiation boards do not overlap in the normal direction.
2. The measurement substrate according to claim 1, wherein a spacer is provided between the overlapping irradiation substrates to ensure a distance in the normal direction between the respective irradiation substrates.
3. The measuring board according to claim 1 or 2, wherein the semiconductor device includes an FPGA (Field Programmable Gate Array).
4. A measuring device for measuring a soft error occurrence rate by irradiating a semiconductor device with radiation, comprising: a measuring board having a plurality of irradiation boards each having a first region for mounting the semiconductor device and a second region for mounting a connecting member for connecting the semiconductor device, the measuring board being arranged so that the first region of each irradiation board is located within a predetermined irradiation region irradiated with the radiation; and a control board connected to each irradiation board, for acquiring an error signal detected in each of the semiconductor devices when irradiated with the radiation, and for calculating the soft error occurrence rate based on the error signal.
5. The measuring device according to claim 4, wherein at least two of the irradiation substrates are arranged to overlap in the normal direction such that the first regions of the respective irradiation substrates do not overlap in the normal direction.
6. The measuring device according to claim 5, further comprising a spacer disposed between the overlapping irradiation substrates to ensure a distance in the normal direction between the respective irradiation substrates.
7. The measuring apparatus according to any one of claims 4 to 6, wherein the semiconductor device is an FPGA (Field Programmable Gate Array).
8. A measurement method for measuring a soft error occurrence rate by irradiating a semiconductor device with radiation, comprising: a plurality of irradiation substrates each having a first region on which the semiconductor devices are mounted and a second region on which connecting members for connecting the semiconductor devices are mounted, the measurement substrate being arranged so that the first region of each irradiation substrate is located within a predetermined irradiation region irradiated with the radiation; an error signal detected in each of the semiconductor devices is obtained; and the soft error occurrence rate is calculated based on the error signal.
Citation Information
Patent Citations
Irradiation test method for semiconductor device
JP2010091334A
Soft error inspection method, soft error inspection device, and soft error inspection system
WO2018225136A1
Laminated substrate for soft error detection and soft error detection device
WO2022249220A1
