Apparatus and method for detecting failed bits of flash memory
Patent Information
- Application Number
- CN202510562585.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2025-04-21
- Filing Date
- 2025-04-30
- Publication Date
- 2026-10-23
AI Technical Summary
[0007]基于上述,本发明诸实施例所述转换电路基于不同数据总线的当前逻辑值向电容性导线汲取放电电流。响应于当前逻辑值的失效位元越少,则转换电路向电容性导线汲取越小的放电电流,使得电容性导线的节点电压的下降速度越慢(下降时间长度越长)。反之,响应于当前逻辑值的失效位元越多,则转换电路向电容性导线汲取越大的放电电流,使得电容性导线的节点电压的下降速度越快(下降时间长度越短)。下降时间长度越短意味着,失效位元检测所花费的时间越少。闪存的编程操作包括多个编程循环(programmingloop),以进行步进脉冲编程(increment-step-pulse-programming,ISPP)操作。一般而言,编程操作的最初几个编程循环的当前逻辑值具有较多的失效位元。本发明诸实施例所述失效位元检测装置可以在编程操作的最初几个编程循环中有效缩短失效位元检测所花费的时间。
Smart Images

Figure CN122967736A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to a memory circuit, and more particularly to a device and method for detecting faulty bits in flash memory. Background Technology
[0002] Integrated circuits incorporating high-capacity and high-performance 3D NAND flash memory are continuously evolving. To overcome the challenges of size reduction, the NAND flash memory industry has transitioned from planar architectures to 3D architectures. The storage capacity of 3D vertical NAND flash memory has increased steadily over the past decade to meet the needs of various industries, including mobile and data centers. The increase in the number of stacked word-line (WL) layers has played a key role in achieving higher storage capacity or bit density. This paradigm shift allows for increased storage density through the vertical stacking of storage cells, rather than solely relying on cell miniaturization. Charge-trap technology uses continuous silicon nitride layers as storage nodes (SNs). Compared to floating gate technology, charge-trap technology is widely used in 3D NAND flash memory arrays due to its simpler process and higher vertical scalability.
[0003] With technological advancements, integrated circuits are stacking more and more layers, increasing bit density and enabling NAND flash memory products to have larger storage capacities. As capacity increases, fail bit detectors will play a crucial role in chips. Summary of the Invention
[0004] The present invention provides a flash memory fault bit detection device and method to determine whether the current logic value of the data bus of the flash memory page buffer is faulty.
[0005] In one embodiment of the present invention, the aforementioned failure bit detection device includes a capacitive wire, a pre-charge circuit, a conversion circuit, and a determination circuit. The pre-charge circuit is coupled to the capacitive wire. During pre-charge, the pre-charge circuit pre-charges the capacitive wire. Multiple input terminals of the conversion circuit are coupled to different data buses of the page buffer of the flash memory. During a determination period following the pre-charge period, the conversion circuit converts the current logic values of the different data buses into discharge currents. A current terminal of the conversion circuit is coupled to the capacitive wire. During the determination period, the conversion circuit draws discharge current from the capacitive wire. The determination circuit is coupled to the capacitive wire. During the determination period, the determination circuit checks the node voltage of the capacitive wire. The determination circuit determines a fall time length based on the fall rate of the node voltage during the determination period. The determination circuit determines whether the current logic value of the different data buses is failure based on the fall time length.
[0006] In one embodiment of the present invention, the above-described fault bit detection method includes: precharging the capacitive wires of the flash memory by a precharge circuit during a precharge period; converting the current logic values of different data buses of the flash memory's page buffer into discharge currents by a conversion circuit after the precharge period; drawing discharge currents from the capacitive wires by the conversion circuit during the determination period; checking the node voltages of the capacitive wires by the determination circuit during the determination period; determining the fall time length based on the fall rate of the node voltages during the determination period; and determining whether the current logic values of the different data buses are faulty based on the fall time length.
[0007] Based on the above, the conversion circuit described in the embodiments of the present invention draws discharge current from the capacitive wires based on the current logic values of different data buses. The fewer faulty bits in response to the current logic value, the smaller the discharge current drawn by the conversion circuit from the capacitive wires, resulting in a slower drop in the node voltage of the capacitive wires (a longer drop time). Conversely, the more faulty bits in response to the current logic value, the larger the discharge current drawn by the conversion circuit from the capacitive wires, resulting in a faster drop in the node voltage of the capacitive wires (a shorter drop time). A shorter drop time means less time is spent detecting faulty bits. Flash memory programming operations include multiple programming loops for increment-step-pulse-programming (ISPP) operations. Generally, the current logic value of the first few programming loops of the programming operation has a large number of faulty bits. The faulty bit detection device described in the embodiments of the present invention can effectively shorten the time spent detecting faulty bits in the first few programming loops of the programming operation.
[0008] To make the above features and advantages of the present invention more apparent and understandable, specific embodiments are described below in conjunction with the accompanying drawings for detailed explanation. Attached Figure Description
[0009] Figure 1 This is a schematic diagram of a circuit block of a non-volatile memory integrated circuit according to an embodiment of the present invention.
[0010] Figure 2 This is a circuit block diagram of a non-volatile memory integrated circuit according to an embodiment of the present invention.
[0011] Figure 3 This is a circuit block diagram of a fault bit detection device according to an embodiment of the present invention.
[0012] Figure 4 This is a schematic flowchart of a flash memory fault bit detection method according to an embodiment of the present invention.
[0013] Figure 5 This is a circuit block diagram illustrating a pre-charging circuit, a judgment circuit, and a conversion circuit, according to one embodiment.
[0014] Figure 6 This is a schematic diagram of signals during the pre-charging period and the determination period, drawn according to an embodiment of the present invention.
[0015] Figure 7 This is a circuit block diagram of a determination circuit and a conversion circuit, drawn according to another embodiment.
[0016] Figure 8 This is a circuit block diagram of the determination circuit and the conversion circuit, drawn according to yet another embodiment.
[0017] Figure 9 This is a circuit block diagram of the determination circuit 320 and the conversion circuit 340, drawn according to a further embodiment.
[0018] Explanation of reference numerals in the attached figures:
[0019] 11: External Chip
[0020] 100: Non-volatile memory integrated circuits
[0021] 101, 102, 103: String of storage cells
[0022] 110: Memory cell array
[0023] 120: Control Circuit
[0024] 121: Interface Circuit
[0025] 122: Command Decoder
[0026] 123: Address decoder
[0027] 124: Word Line Operation Logic
[0028] 125: Bitline Operation Logic
[0029] 126: Word Line Decoding Circuit
[0030] 127: Bit-line decoding circuit
[0031] 128: Page Buffer
[0032] 300: Failure Bit Detection Device
[0033] 310: Pre-charge circuit
[0034] 320: Judgment Circuit
[0035] 321: Reference Voltage Generation Circuit
[0036] 322, 325: Comparators
[0037] 323: Clock Generation Circuit
[0038] 324: Counter
[0039] 330: Capacitive wire
[0040] 340: Conversion circuit
[0041] 350: Bias voltage generation circuit
[0042] 601, 611: Dashed lines
[0043] 602, 612: Solid lines
[0044] 620: Reference time length
[0045] 621, 622: Length of descent
[0046] 710, 810, 910: Temperature compensation circuit
[0047] BL1, BLm, BLn: Bit lines
[0048] CK5: Clock signal
[0049] CK5': Compensated clock signal
[0050] CNT5: Count value
[0051] CNT5': Compensated count value
[0052] CSL: Common Source Line
[0053] DBUS_0 to DBUS_n: Data Bus
[0054] EN5: Enable signal
[0055] FBIT: Node Voltage
[0056] GSL: Grounding Selection Line
[0057] Ibias5: Current Source
[0058] Ifail: Discharge current
[0059] Isink_1, Isink_n: Current components
[0060] M, M': Storage unit
[0061] MN51_1, MN51_n, MN52: Transistors
[0062] MP51: Precharged transistor
[0063] P61: During pre-charging
[0064] P62: Judgment Period
[0065] PASS3: Result Signal
[0066] Pre5: Precharge control signal
[0067] REF5: Reference value
[0068] S410, S420, S430, S440, S450, S460: Steps
[0069] SSL0, SSL1, SSL2, SSL3: Serial select lines
[0070] Sub0, Sub1, Sub2, Sub3: Sub-blocks
[0071] SW5_1, SW5_n: Switches
[0072] Vbias: Bias voltage
[0073] VDD: Power supply voltage source
[0074] Vref: Reference voltage
[0075] Vref': Compensated reference voltage
[0076] VSS: Reference Voltage Source
[0077] WL1, WLi, WLi+1, WLn-1, WLn, WLm: Word lines
[0078] X, Y, Z: Direction Detailed Implementation
[0079] The term "coupled (or connected)" as used throughout this specification (including the claims) may refer to any direct or indirect connection means. For example, if the text describes a first device coupled (or connected) to a second device, it should be interpreted as the first device being directly connected to the second device, or the first device being indirectly connected to the second device through other devices or some connection means. The terms "first," "second," etc., used throughout this specification (including the claims) are used to name elements or distinguish different embodiments or scopes, and are not intended to limit the upper or lower limit of the number of elements, nor to limit the order of elements. Furthermore, wherever possible, elements / components / steps using the same reference numerals in the drawings and embodiments represent the same or similar parts. Elements / components / steps using the same reference numerals or the same terms in different embodiments can be referred to mutually in the relevant descriptions.
[0080] Figure 1 This is a circuit block diagram of a non-volatile memory integrated circuit 100 according to an embodiment of the present invention. The non-volatile memory integrated circuit 100 includes a memory cell array 110 and a control circuit 120. The control circuit 120 is coupled to the memory cell array 110. Figure 1 The storage cell array 110 shown can be used as Figure 3 This is one of many implementation examples of the storage cell array 110 shown, and Figure 3 The control circuit 120 shown can be used as Figure 1 This is one of many embodiments of the control circuit 120 shown. Depending on the design, in some embodiments, the control circuit 120 may be implemented as hardware circuitry. In other embodiments, the control circuit 120 may be implemented as a combination of hardware, firmware, and software (i.e., programs).
[0081] In hardware terms, the control circuit 120 described above can be implemented as logic circuitry on an integrated circuit. For example, the functions of the control circuit 120 can be implemented in various logic blocks, modules, and circuits within one or more hardware controllers, microcontrollers, hardware processors, microprocessors, application-specific integrated circuits (ASICs), digital signal processors (DSPs), field-programmable gate arrays (FPGAs), central processing units (CPUs), and / or other processing units. The functions of the control circuit 120 can be implemented as hardware circuitry, such as various logic blocks, modules, and circuits within an integrated circuit, using hardware description languages (such as Verilog HDL or VHDL) or other suitable programming languages.
[0082] In software and / or firmware form, the functions of the control circuit 120 can be implemented as programming codes. For example, the control circuit 120 can be implemented using general programming languages (such as C, C++, or assembly languages) or other suitable programming languages. The programming codes can be recorded / stored in a non-transitory machine-readable storage medium. In some embodiments, the non-transitory machine-readable storage medium includes, for example, semiconductor memory and / or a storage device. An electronic device (e.g., a CPU, hardware controller, microcontroller, hardware processor, or microprocessor) can read and execute the programming codes from the non-transitory machine-readable storage medium to implement the functions of the control circuit 120.
[0083] Figure 1 The equivalent circuit of a three-dimensional (3D) NAND flash memory integrated circuit is shown in a three-dimensional (3D) manner. The memory cell array 110 has multiple memory cells (e.g., memory cells M'). Figure 1The memory cells in the illustrated memory cell array 110 are arranged in a 3D manner in the XYZ coordinate system. However, this does not mean that the circuitry of the memory cell array 110 is limited to a 3D arrangement. In one example, the memory cell array 110 can be divided into four sub-blocks: Sub0, Sub1, Sub2, and Sub3. The control circuitry 120 can independently control the operation of each sub-block (e.g., sub-block Sub0).
[0084] Figure 1 Each storage cell string shown (e.g., storage cell string 101, 102, or 103) includes multiple storage cells (e.g., storage cell M') connected in series along the Z direction. Storage cell array 110 includes multiple word lines (WL), for example... Figure 1 The word lines shown are WL1, ..., WLi, WLi+1, ..., WLn-1, WLn, ..., WLm. A word line (e.g., word line WLn) can correspond to a layer in the XY plane. Each of word lines WL1 to WLm is coupled to multiple memory cells in a corresponding row of these memory cells. Multiple memory cells in each memory cell string (e.g., memory cell string 101) correspond to a word line (e.g., word line WLi+1). Memory cell M can be configured as a string select transistor coupled to a string select line (SSL, e.g., string select lines SSL0, SSL1, SSL2, and SSL3), and another memory cell can be configured as a ground select transistor coupled to a ground select line (GSL). The string select transistor and the ground select transistor are located on opposite sides of each memory cell string (e.g., memory cell string 101). In this example, multiple memory cell strings (e.g., memory cell strings 101, 102, etc.) coupled to the same string selection line (e.g., string selection line SSL0) on the same plane (e.g., the plane defined by the X and Z directions) can be defined as a subblock (e.g., subblock Sub0).
[0085] Each of the multiple bit lines (BLs) is coupled to multiple memory cells in a corresponding column of these memory cells. That is, each string of memory cells (e.g., string 101, 102, or 103) is connected to the corresponding bit line (e.g., bit line BL1, BLn, or BLm) via a corresponding string select transistor on a string select line (e.g., string select line SSL0). Strings of memory cells in the same column (e.g., string 101) in different sub-blocks (e.g., sub-blocks Sub0, Sub1, etc.) along the Y direction are connected to the corresponding bit line (e.g., bit line BL1). The string select line (e.g., string select line SSL0) can be a conductive line or conductive layer formed above the topmost word line layer (e.g., word line WL1). Each string of memory cells (e.g., string 101, 102, or 103) can be connected to the same common source line (CSL) via a corresponding ground select transistor on the ground select line GSL. The ground select line GSL can be a conductive line or conductive layer formed below the bottommost word line layer (e.g., word line WLm). The common source line CSL can be a conductive layer formed above the substrate of the non-volatile memory integrated circuit 100. The serial select lines (e.g., serial select lines SSL0~SSL3) in the memory cell array 110 can be on the same conductive layer, but divided into independent lines. Each independent line (serial select line) can independently control the operation of the corresponding sub-block (e.g., sub-blocks Sub0, Sub1, etc.) in the memory cell array 110.
[0086] In one example, in a single-level cell (SLC) configuration, multiple memory cells (including memory cell M') coupled to the same word line or word line layer (e.g., word line WLn) within a sub-region (e.g., sub-region Sub0) can be defined as a page. In another example, in a triple-level cell (TLC) configuration, multiple memory cells (including memory cell M') coupled to the same word line or word line layer (e.g., word line WLn) within a sub-region (e.g., sub-region Sub0) can be defined as three pages. In the TLC configuration, these three pages include a high page, a middle page, and a low page. Memory cells M' located on the same word line (e.g., word line WLn) are subjected to the same voltage. Each word line (e.g., word line WLn) can be connected to a driver circuit, such as an X-decoder (or scan driver).
[0087] In one example, one or more dummy lines or layers (not shown) may be provided between the string select lines (e.g., string select lines SSL0~SSL3) and the word line layer (e.g., word line WL1), and between the ground select line GSL and the bottommost word line layer (e.g., word line WLm). In another example, one or more dummy lines or layers (not shown) may be provided in the middle portion of each memory cell string (e.g., memory cell string 101, 102, or 103). The non-volatile memory integrated circuit 100 also includes control circuitry 120 for implementing corresponding operations on the memory cell array 110.
[0088] Figure 2 This is a circuit block diagram of a non-volatile memory integrated circuit 100 according to an embodiment of the present invention. Figure 2 The non-volatile memory integrated circuit 100 shown can be referenced. Figure 1 Related explanations. External chip 11 sends commands to non-volatile memory integrated circuit 100 and transmits and receives data from non-volatile memory integrated circuit 100. Based on the control of external chip 11 (e.g., memory controller), Figure 2 The non-volatile memory integrated circuit 100 shown provides data access functionality to the external chip 11. Figure 2 In the embodiment shown, the control circuit 120 includes an interface circuit 121, a command decoder 122, an address decoder 123, word line operation logic 124, bit line operation logic 125, word line decoding circuit 126, bit line decoding circuit 127, and a page buffer 128.
[0089] Command decoder 122 receives commands (such as programming commands, read mode commands, write mode commands, etc.) from external chip 11 through interface circuit 121. Upon receiving a command, command decoder 122 decodes the command to perform the corresponding operation. For example, command decoder 122 transmits the array address to address decoder 123 based on the command decoding result. Based on the control of command decoder 122, word line operation logic 124 informs word line decoding circuit 126 of the word line to be selected (WL) and the word line to be deselected. Address decoder 123 decodes the array address and transmits the address decoding result to word line decoding circuit 126 and bit line decoding circuit 127. Based on the control of address decoder 123 and word line operation logic 124, word line decoding circuit 126 selects the corresponding word line of memory cell array 110.
[0090] Based on the control of command decoder 122, bit line operation logic 125 notifies bit line decoding circuit 127 of the bit line to be selected (BL) and the bit line to be deselected. Based on the control of address decoder 123 and bit line operation logic 125, bit line decoding circuit 127 sends and receives data to and from external chip 11 through interface circuit 121. Bit line decoding circuit 127 selects the corresponding bit line of memory cell array 110 based on the control of bit line operation logic 125. Page buffer 128 includes a sense amplifier to sense the signal of the bit line to generate corresponding data.
[0091] Figure 3 This is a circuit block diagram of a fault bit detection device 300 according to an embodiment of the present invention. Figure 3 The memory cell array 110, page buffer 128, and bit line decoding circuit 127 shown can be referenced. Figure 2 The relevant explanations are omitted here. Based on practical applications, the failure bit detection device 300 can be arranged in any functional circuit of the non-volatile memory integrated circuit 100. Figure 3 The failure bit detection device 300 shown includes a pre-charging circuit 310, a judgment circuit 320, a capacitive wire 330, and a conversion circuit 340. The capacitive wire 330 can be made of any conductive material (e.g., metal). Because the capacitive wire 330 is wound very long, its equivalent capacitance (parasitic capacitance) is very large. Therefore, the capacitive wire 330 can store charge. The capacitive wire 330 is coupled to the pre-charging circuit 310, the judgment circuit 320, and the conversion circuit 340.
[0092] Page buffer 128 has multiple data buses DBUS_0 to DBUS_n. The number n of data buses DBUS_1 to DBUS_n can be any integer determined according to the actual design and application. Multiple inputs of conversion circuit 340 are coupled to different data buses DBUS_0 to DBUS_n of page buffer 128. Depending on the application, conversion circuit 340 can be arranged in any functional circuit of non-volatile memory integrated circuit 100. For example, conversion circuit 340 can be arranged in the page buffer region.
[0093] Figure 4 This is a flowchart illustrating a method for detecting faulty bits in flash memory according to an embodiment of the present invention. Please refer to... Figure 3 and Figure 4In step S410, the pre-charge circuit 310 pre-charges the capacitive wire 330 during the pre-charge period. The capacitive wire includes a wire-wound capacitor (parasitic capacitance, not shown) to store the pre-charge charge output by the pre-charge circuit 310 during the pre-charge period. The pre-charge circuit 310 stops pre-charging the capacitive wire 330 during a determination period after the pre-charge period.
[0094] During the determination period after the pre-charge period, the conversion circuit 340 converts the current logic values of the different data buses DBUS_1 to DBUS_n of the page buffer 128 into the discharge current Ifail (step S420). The programming operation of the memory cell array 110 (flash memory) includes multiple programming loops to perform increment-step-pulse-programming (ISPP) operations on the current target memory cell in the memory cell array 110. Generally, the current logic values of the first few programming loops of the programming operation have more invalid bits, that is, the current logic values of the different data buses DBUS_1 to DBUS_n of the page buffer 128 have more invalid bits. The current logic value of the target data bus in the data buses DBUS_1 to DBUS_n is a first logic value (e.g., logic value "1"), indicating that the programming result of the corresponding word line in the memory cell array 110 is invalid.
[0095] In response to the current logic value of the target data bus indicating a programming failure, the conversion circuit 340 converts the current logic value of the target data bus into a corresponding current component of the discharge current Ifail. The current logic value of the target data bus is a second logic value (e.g., logic value "0"), indicating that the programming result for the corresponding word line in the memory cell array 110 is successful. In response to the current logic value of the target data bus indicating a successful programming result, the conversion circuit 340 abandons the conversion of the current logic value of the target data bus into a corresponding current component of the discharge current. Therefore, the more failure bits in the current logic value of the data buses DBUS_1 to DBUS_n, the larger the discharge current Ifail. Conversely, the fewer failure bits in the current logic value of the data buses DBUS_1 to DBUS_n, the smaller the discharge current Ifail.
[0096] The current terminal of the switching circuit 340 is coupled to the capacitive wire 330. During the decision period, the switching circuit 340 draws a discharge current Ifail from the capacitive wire 330 (step S430), causing the node voltage FBIT of the capacitive wire 330 to continuously decrease during the decision period. The more data buses DBUS_1 to DBUS_n that indicate a programming failure (i.e., the more failure bits of the current logic value of more data buses DBUS_1 to DBUS_n), the larger the discharge current Ifail, resulting in a faster decrease in the node voltage FBIT.
[0097] During the determination period, the determination circuit 320 checks the node voltage FBIT of the capacitive wire 330 (step S440). The determination circuit 320 determines the fall time length based on the rate of fall of the node voltage FBIT during the determination period (step S450). The smaller the discharge current Ifail drawn by the switching circuit 340 to the capacitive wire 330, the slower the rate of fall of the node voltage FBIT of the capacitive wire 330 (the longer the fall time). Conversely, the larger the discharge current Ifail drawn by the switching circuit 340 to the capacitive wire 330, the faster the rate of fall of the node voltage FBIT of the capacitive wire 330 (the shorter the fall time). In step S460, the determination circuit 320 determines whether the current logic values of the different data buses DBUS_1 to DBUS_n of the page buffer 128 are invalid based on the fall time length.
[0098] For example (but not limited to), the decision circuit 320 compares the node voltage FBIT with a reference voltage during the decision period. The level of the reference voltage can be determined according to the actual design and application. The decision circuit 320 determines the fall time length based on the comparison result of the node voltage FBIT and the reference voltage. The decision circuit 320 compares the fall time length with a reference time length. The reference time length can be determined according to the actual design and application. The decision circuit 320 determines whether the current logic value of different data buses DBUS_1 to DBUS_n is invalid based on the comparison result of the fall time length and the reference time length.
[0099] The determination circuit 320 notifies the next-level circuit (not shown) via the result signal PASS3 whether the current logic value of the different data buses DBUS_1 to DBUS_n of the page buffer 128 is invalid. For example (but not limited to), a high logic level for the result signal PASS3 indicates that the current logic value check has passed, while a low logic level for the result signal PASS3 indicates that the current logic value check has failed. If the result signal PASS3 indicates that the current logic value check has failed, the non-volatile memory integrated circuit 100 will perform the next programming cycle on the current target memory cell in the memory cell array 110 until the result signal PASS3 indicates that the current logic value check has passed.
[0100] In summary, the conversion circuit 340 draws a discharge current Ifail from the capacitive wire 330 based on the current logic values of the different data buses DBUS_1 to DBUS_n of the page buffer 128. The fewer faulty bits corresponding to the current logic value, the smaller the discharge current Ifail drawn by the conversion circuit 340 from the capacitive wire 330, resulting in a slower drop rate (longer fall time) of the node voltage FBIT of the capacitive wire 330. Conversely, the more faulty bits corresponding to the current logic value, the larger the discharge current Ifail drawn by the conversion circuit 340 from the capacitive wire 330, resulting in a faster drop rate (shorter fall time) of the node voltage FBIT of the capacitive wire 330. A shorter fall time means less time is spent detecting faulty bits. Generally, the current logic value in the first few programming cycles of a programming operation has more faulty bits. The faulty bit detection device 300 can effectively shorten the time spent detecting faulty bits in the first few programming cycles of a programming operation.
[0101] Figure 5 This is a circuit block diagram of a pre-charging circuit 310, a judgment circuit 320, and a conversion circuit 340, drawn according to an embodiment. Figure 5 The pre-charging circuit 310, the judgment circuit 320, the capacitive wire 330, and the conversion circuit 340 shown can be referenced. Figure 3 Related explanations. Figure 5 The pre-charging circuit 310, the judgment circuit 320, and the conversion circuit 340 shown can be used as... Figure 3 This is one of many implementation examples of the pre-charging circuit 310, the judgment circuit 320, and the conversion circuit 340 shown.
[0102] exist Figure 5In the illustrated embodiment, the pre-charge circuit 310 includes a pre-charge transistor MP51. A first terminal (e.g., source) of the pre-charge transistor MP51 is coupled to a power supply voltage source VDD. A second terminal (e.g., drain) of the pre-charge transistor MP51 is coupled to a capacitive wire 330. A control terminal (e.g., gate) of the pre-charge transistor MP51 is controlled by a pre-charge control signal Pre5. During the pre-charge period, the pre-charge transistor MP51 is turned on to pre-charge the capacitive wire 330. During a subsequent determination period after the pre-charge period, the pre-charge transistor MP51 is turned off to stop pre-charging the capacitive wire 330.
[0103] exist Figure 5 In the illustrated embodiment, the conversion circuit 340 includes multiple switches, such as switches SW5_1, ..., SW5_n. The first terminal of each switch SW5_1 to SW5_n is coupled to a capacitive wire 330. The control terminal of each switch SW5_1 to SW5_n is coupled to its corresponding data bus DBUS_1 to DBUS_n of the page buffer 128. Figure 5 In the illustrated embodiment, the conversion circuit 340 further includes a plurality of transistors, such as transistors MN51_1, ..., MN51_n. A first terminal (e.g., drain) of each transistor MN51_1 to MN51_n is coupled to a second terminal of a corresponding switch SW5_1 to SW5_n. For example, the first terminal of transistor MN51_1 is coupled to the second terminal of switch SW5_1, and the first terminal of transistor MN51_n is coupled to the second terminal of switch SW5_n. A second terminal (e.g., source) of each transistor MN51_1 to MN51_n is coupled to a reference voltage source VSS. The control terminal (e.g., gate) of each transistor MN51_1 to MN51_n is controlled by a bias voltage Vbias.
[0104] For example, if the current logic values of data buses DBUS_1 to DBUS_n are all logic values "0" (indicating that the programming result is successful), then switches SW5_1 to SW5_n are all off, so the discharge current Ifail is very small (even zero). If the current logic value of data bus DBUS_1 is all logic values "1" (indicating that the programming result of a certain bit is unsuccessful), then switch SW5_1 is on, so the conversion circuit 340 converts the current logic value of data bus DBUS_1 into a corresponding current component Isink_1 in the discharge current Ifail. Similarly, if the current logic value of data bus DBUS_n is all logic values "1", then switch SW5_n is on, so the conversion circuit 340 converts the current logic value of data bus DBUS_n into a corresponding current component Isink_n in the discharge current Ifail.
[0105] exist Figure 5 In the illustrated embodiment, the failure bit detection device 300 further includes a bias voltage generation circuit 350. The bias voltage generation circuit 350 is coupled to the conversion circuit 340 to provide a bias voltage Vbias. This embodiment does not limit the specific implementation of the bias voltage generation circuit 350. For example, the bias voltage generation circuit 350 includes a current source Ibias5 and a transistor MN52. A first terminal (e.g., drain) of transistor MN52 is coupled to the current source Ibias5. A second terminal (e.g., source) of transistor MN52 is coupled to a reference voltage source VSS. A control terminal (e.g., gate) of transistor MN52 is coupled to the first terminal of transistor MN52 and the control terminal of each transistor MN51_1 to MN51_n to provide the bias voltage Vbias.
[0106] Figure 6 This is a schematic diagram illustrating the signals during the pre-charging period P61 and the determination period P62, according to an embodiment of the present invention. Please refer to... Figure 5 and Figure 6 Based on the control of the precharge control signal Pre5, the precharge transistor MP51 is turned on during the precharge period at P61 to precharge the capacitive wire 330. Therefore, the node voltage FBIT of the capacitive wire 330 is pulled up to the precharge level (e.g., the voltage level of the power supply voltage source VDD or another voltage level). The precharge transistor MP51 is turned off during the judgment period P62 after P61 during the precharge period to stop precharging the capacitive wire 330. The more invalid bits of the current logic value of the data bus DBUS_1 to DBUS_n, the larger the discharge current Ifail, resulting in a faster drop in the node voltage FBIT (e.g., ...). Figure 6 (As shown by the dashed line 601). Conversely, the fewer failed bits in the current logic values of data buses DBUS_1 to DBUS_n, the smaller the discharge current Ifail, resulting in a slower drop in the node voltage FBIT (as shown by the dashed line 601). Figure 6 As shown by the solid line 602).
[0107] exist Figure 5In the illustrated embodiment, the determination circuit 320 includes a reference voltage generation circuit 321, a comparator 322, a clock generation circuit 323, a counter 324, and a comparator 325. The reference voltage generation circuit 321 generates the original reference voltage Vref. The comparator 322 is coupled to the reference voltage generation circuit 321 to receive the original reference voltage Vref. The comparator 322 is coupled to the capacitive wire 330 to receive the node voltage FBIT. During the determination period, the comparator 322 compares the node voltage FBIT with the original reference voltage Vref. The comparator 322 generates an enable signal EN5 based on the comparison result between the node voltage FBIT and the original reference voltage Vref. The more invalid bits the current logic value of the data bus DBUS_1 to DBUS_n has, the faster the node voltage FBIT decreases, causing the enable signal EN5 to transition to the disabled state earlier (e.g., ...). Figure 6 (As shown by the dashed line 611). Conversely, the fewer invalid bits in the current logic values of data buses DBUS_1 to DBUS_n, the slower the node voltage FBIT decreases, causing the enable signal EN5 to transition to the disabled state later (as shown by the dashed line 611). Figure 6 (As shown by the solid line 612).
[0108] Clock generation circuit 323 generates the original clock signal CK5. The enable terminal of counter 324 is coupled to the output terminal of comparator 322 to receive enable signal EN5. Enable signal EN5 controls the counting time of counter 324 as the falling time. Based on the control of enable signal EN5, counter 324 counts the original clock signal CK5 within the counting time to generate the count value CNT5. The more invalid bits of the current logic value of data bus DBUS_1 to DBUS_n, the earlier enable signal EN5 transitions to the disabled state, resulting in a shorter counting time of counter 324 (e.g., ...). Figure 6 The fall time length is shown in 621). Conversely, the fewer invalid bits in the current logic values of data bus DBUS_1 to DBUS_n, the later the enable signal EN5 transitions to the disable state, resulting in a longer counting time length for counter 324 (as shown in 621). Figure 6 The descent time length is shown in 622).
[0109] Comparator 325 is coupled to the output of counter 324 to receive the count value CNT5. During the judgment period, comparator 325 compares the count value CNT5 with the reference value REF5 (equivalent to judgment circuit 320 comparing the fall time length with the reference time length 620). Based on the comparison result of the count value CNT5 and the reference value REF5, comparator 325 determines whether the current logic value of data bus DBUS_1 to DBUS_n is invalid. The more invalid bits the current logic value of data bus DBUS_1 to DBUS_n has, the shorter the fall time length (e.g., fall time length 621). Because the fall time length 621 is shorter than the reference time length 620 (i.e., the count value CNT5 is less than the reference value REF5), the result signal PASS3 is low (indicating that the current logic value check result is invalid). Conversely, the fewer invalid bits the current logic value of data bus DBUS_1 to DBUS_n has, the longer the counting time length of counter 324 (e.g., fall time length 622). Once the count value CNT5 is greater than the reference value REF5 (that is, the fall time length 622 is longer than the reference time length 620), the result signal PASS3 is immediately pulled up (indicating that the current logic value check result is passed).
[0110] Figure 7 This is a circuit block diagram of the determination circuit 320 and the conversion circuit 340, drawn according to another embodiment. Figure 7 The judgment circuit 320 and the conversion circuit 340 shown can be used as Figure 3 This is one of many implementation examples of the judgment circuit 320 and the conversion circuit 340 shown. Figure 7 The pre-charging circuit 310, judgment circuit 320, capacitive wire 330, conversion circuit 340, and bias voltage generation circuit 350 shown can be referenced. Figure 5 The relevant explanations are omitted here.
[0111] exist Figure 7 In the embodiment shown, the determination circuit 320 includes a reference voltage generation circuit 321, a temperature compensation circuit 710, a comparator 322, a clock generation circuit 323, a counter 324, and a comparator 325. Figure 7 The reference voltage generation circuit 321, comparator 322, clock generation circuit 323, counter 324, and comparator 325 shown can be referenced. Figure 5 and Figure 6 The relevant explanations are omitted here. Figure 7The determination circuit 320 shown also includes a temperature compensation circuit 710. The input of the temperature compensation circuit 710 is coupled to the output of the reference voltage generation circuit 321 to receive the original reference voltage Vref. The temperature compensation circuit 710 compensates the original reference voltage Vref based on the current temperature to generate a compensated reference voltage Vref' as a reference voltage for comparator 322.
[0112] In response to the current temperature rise, the current of current source Ibias5 will decrease slightly, causing the discharge time of capacitive wire 330 to increase. Figure 7 The determination circuit 320 shown uses a temperature compensation circuit 710 to appropriately adjust the original reference voltage Vref to generate a compensated reference voltage Vref' that varies with temperature. When the temperature rises, the compensated reference voltage Vref' will rise accordingly, thus allowing the counter 324 to stop counting earlier.
[0113] Figure 8 This is a circuit block diagram of the determination circuit 320 and the conversion circuit 340, drawn according to yet another embodiment. Figure 8 The judgment circuit 320 and the conversion circuit 340 shown can be used as Figure 3 This is one of many implementation examples of the judgment circuit 320 and the conversion circuit 340 shown. Figure 8 The pre-charging circuit 310, judgment circuit 320, capacitive wire 330, conversion circuit 340, and bias voltage generation circuit 350 shown can be referenced. Figure 5 The relevant explanations are omitted here.
[0114] exist Figure 8 In the embodiment shown, the determination circuit 320 includes a reference voltage generation circuit 321, a comparator 322, a clock generation circuit 323, a temperature compensation circuit 810, a counter 324, and a comparator 325. Figure 8 The reference voltage generation circuit 321, comparator 322, clock generation circuit 323, counter 324, and comparator 325 shown can be referenced. Figure 5 and Figure 6 The relevant explanations are omitted here. Figure 8 The judgment circuit 320 shown also includes a temperature compensation circuit 810. The input of the temperature compensation circuit 810 is coupled to the output of the clock generation circuit 323 to receive the original clock signal CK5. The temperature compensation circuit 810 compensates the original clock signal CK5 based on the current temperature to generate a compensated clock signal CK5' as the clock signal for the counter 324.
[0115] In response to the current temperature rise, the current of current source Ibias5 will decrease slightly, causing the discharge time of capacitive wire 330 to increase. Figure 8The judgment circuit 320 shown uses a temperature compensation circuit 810 to appropriately adjust the original clock signal CK5 to generate a compensated clock signal CK5' that changes with temperature. When the temperature rises, the compensated clock signal CK5' will slow down accordingly, thereby extending the descent time.
[0116] Figure 9 This is a circuit block diagram of the determination circuit 320 and the conversion circuit 340, drawn according to a further embodiment. Figure 9 The judgment circuit 320 and the conversion circuit 340 shown can be used as Figure 3 This is one of many implementation examples of the judgment circuit 320 and the conversion circuit 340 shown. Figure 9 The pre-charging circuit 310, judgment circuit 320, capacitive wire 330, conversion circuit 340, and bias voltage generation circuit 350 shown can be referenced. Figure 5 The relevant explanations are omitted here.
[0117] exist Figure 9 In the embodiment shown, the determination circuit 320 includes a reference voltage generation circuit 321, a comparator 322, a clock generation circuit 323, a counter 324, a temperature compensation circuit 910, and a comparator 325. Figure 9 The reference voltage generation circuit 321, comparator 322, clock generation circuit 323, counter 324, and comparator 325 shown can be referenced. Figure 5 and Figure 6 The relevant explanations are omitted here. Figure 9 The determination circuit 320 shown also includes a temperature compensation circuit 910. The input of the temperature compensation circuit 910 is coupled to the output of the counter 324 to receive the count value CNT5. The temperature compensation circuit 910 compensates the count value CNT5 based on the current temperature to generate a compensated count value CNT5'.
[0118] Comparator 325 is coupled to the output of temperature compensation circuit 910 to receive the compensated count value CNT5'. During the determination period, comparator 325 compares the compensated count value CNT5' with the reference value REF5. Based on the comparison result of the compensated count value CNT5' and the reference value REF5, comparator 325 determines whether the current logic value of data bus DBUS_1 to DBUS_n is invalid. In response to the rise in the current temperature, the current of current source Ibias5 will decrease slightly, causing the discharge time of capacitive wire 330 to increase. Figure 9 The determination circuit 320 shown uses a temperature compensation circuit 910 to appropriately adjust the count value CNT5 to generate a compensated count value CNT5' that changes with temperature. When the temperature rises, the compensated count value CNT5' will decrease accordingly, thereby compensating for the increased discharge time of the capacitive wire 330.
[0119] Although the present invention has been disclosed above with reference to embodiments, it is not intended to limit the present invention. Any person skilled in the art may make some modifications and refinements without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention shall be determined by the appended claims.
Claims
1. A flash memory fault bit detection device, comprising: A capacitive wire; A pre-charge circuit is coupled to the capacitive wire, wherein the pre-charge circuit pre-charges the capacitive wire during a pre-charge period; A conversion circuit, wherein a plurality of input terminals of the conversion circuit are coupled to different data buses of a page buffer of the flash memory, the conversion circuit converts the current logic value of the different data buses into a discharge current during a decision period after the precharge period, a current terminal of the conversion circuit is coupled to the capacitive wire, and the conversion circuit draws the discharge current from the capacitive wire during the decision period. as well as A judgment circuit is coupled to the capacitive wire, wherein the judgment circuit checks a node voltage of the capacitive wire during the judgment period, the judgment circuit determines a fall time length based on a fall rate of the node voltage during the judgment period, and the judgment circuit determines whether the current logic value of the different data buses is invalid based on the fall time length.
2. The failure bit detection device according to claim 1, wherein the capacitive wire includes a wire-wound capacitor for storing the pre-charge charge output by the pre-charge circuit during the pre-charge period.
3. The failure bit detection device according to claim 1, wherein, The current logic value of a target data bus in the different data buses is a first logic value, indicating that the programming result of a corresponding word line in the flash memory is invalid; The current logic value of the target data bus is a second logic value, indicating that the programming result of a corresponding word line in the flash memory is successful; In response to the current logic value of the target data bus indicating a programming failure, the conversion circuit converts the current logic value of the target data bus into a corresponding current component in the discharge current; as well as In response to the current logic value of the target data bus indicating a successful programming result, the conversion circuit abandons the conversion of the current logic value of the target data bus into the corresponding current component in the discharge current.
4. The failure bit detection device according to claim 1, wherein, The pre-charging circuit stops pre-charging the capacitive wire during the determination period following the pre-charging period. The switching circuit draws the discharge current from the capacitive wire during the judgment period, causing the node voltage to continuously decrease during the judgment period. The more data buses in the different data buses that indicate a programming failure, the larger the discharge current, and the faster the node voltage decreases. During the decision-making process, the circuit compares the node voltage to a reference voltage. The decision circuit determines the fall time length based on the comparison between the node voltage and the reference voltage. The decision circuit compares the descent time length with a reference time length. as well as The judgment circuit determines whether the current logic value of the different data buses is invalid based on the comparison result between the fall time length and the reference time length.
5. The failure bit detection device according to claim 1, wherein the pre-charging circuit comprises: A precharge transistor, wherein a first terminal of the precharge transistor is coupled to a power supply voltage source, a second terminal of the precharge transistor is coupled to a capacitive wire, and a control terminal of the precharge transistor is controlled by a precharge control signal.
6. The failure bit detection device according to claim 1, wherein the conversion circuit comprises: Multiple switches, wherein a first terminal of each of these switches is coupled to the capacitive wire, and a control terminal of each of these switches is coupled to a corresponding one of the different data buses of the page buffer; as well as A plurality of first transistors, wherein a first terminal of each of the first transistors is coupled to a second terminal of a corresponding of the switches, a second terminal of each of the first transistors is coupled to a first reference voltage source, and a control terminal of each of the first transistors is controlled by a bias voltage.
7. The failure bit detection device according to claim 6 further includes: A bias voltage generating circuit is coupled to the conversion circuit to provide the bias voltage.
8. The failure bit detection device according to claim 7, wherein the bias voltage generating circuit comprises: A current source; as well as A second transistor, wherein a first terminal of the second transistor is coupled to the current source, a second terminal of the second transistor is coupled to a second reference voltage source, and a control terminal of the second transistor is coupled to the first terminal of the second transistor and the control terminal of each of the first transistors to provide the bias voltage.
9. The failure bit detection device according to claim 1, wherein the judgment circuit comprises: A first comparator, coupled to the capacitive wire, receives the node voltage, wherein the first comparator compares the node voltage to a reference voltage during the determination period; A counter having a constant power input coupled to an output of a first comparator, wherein the first comparator controls a counting time length of the counter as the fall time length based on a comparison result of the node voltage and the reference voltage, and the counter counts a clock signal during the counting time length to generate a count value; and A second comparator, coupled to an output of the counter to receive the count value, wherein the second comparator compares the count value to a reference value during the determination period, and the second comparator determines whether the current logic value of the different data buses is invalid based on the comparison result of the count value and the reference value.
10. The faulty bit detection device according to claim 9, wherein the judgment circuit further comprises: A reference voltage generation circuit generates an original reference voltage; as well as A temperature compensation circuit has an input terminal coupled to an output terminal of a reference voltage generation circuit to receive the original reference voltage, wherein the temperature compensation circuit compensates the original reference voltage based on a current temperature to generate a compensated reference voltage as the reference voltage for the first comparator.
11. The faulty bit detection device according to claim 9, wherein the judgment circuit further comprises: A clock generation circuit generates a primary clock signal; as well as A temperature compensation circuit has an input terminal coupled to an output terminal of a clock generation circuit to receive the original clock signal, wherein the temperature compensation circuit compensates the original clock signal based on a current temperature to generate a compensated clock signal as the clock signal for the counter.
12. The failure bit detection device according to claim 1, wherein the judgment circuit comprises: A first comparator, coupled to the capacitive wire, receives the node voltage, wherein the first comparator compares the node voltage to a reference voltage during the determination period; A counter having a consistent power input coupled to an output of a first comparator, wherein the first comparator controls a counting time length of the counter as the fall time length based on a comparison result of the node voltage and the reference voltage, and the counter counts a clock signal during the counting time length to generate a count value; A temperature compensation circuit has an input terminal coupled to an output terminal of the counter to receive the count value, wherein the temperature compensation circuit compensates the count value based on a current temperature to generate a compensated count value. as well as A second comparator, coupled to an output of the temperature compensation circuit, receives the compensated count value, wherein the second comparator compares the compensated count value to a reference value during the determination period, and the second comparator determines whether the current logic value of the different data buses is invalid based on the comparison result between the compensated count value and the reference value.
13. A method for detecting faulty bits in flash memory, comprising: A precharge circuit of the flash memory precharges a capacitive wire of the flash memory during a precharge period, wherein the precharge circuit is coupled to the capacitive wire; A conversion circuit of the flash memory converts the current logic value of different data buses of a page buffer of the flash memory into a discharge current during a determination period after the pre-charge period, wherein multiple inputs of the conversion circuit are coupled to the different data buses; The switching circuit draws the discharge current from the capacitive wire during the determination period, wherein a current terminal of the switching circuit is coupled to the capacitive wire; During the determination period, a determination circuit of the flash memory checks the node voltage of the capacitive wire, wherein the determination circuit is coupled to the capacitive wire; The judgment circuit determines the length of a drop time based on the rate of drop of the node voltage during the judgment period; as well as The judgment circuit determines whether the current logic value of the different data buses is invalid based on the fall time length.
14. The failure bit detection method according to claim 13, wherein the capacitive wire includes a wire-wound capacitor for storing the precharge charge output by the precharge circuit during the precharge period.
15. The failure bit detection method according to claim 13, wherein the current logic value of a target data bus in the different data buses is a first logic value indicating that the programming result of a corresponding word line in the flash memory is failure, and the current logic value of the target data bus is a second logic value indicating that the programming result of a corresponding word line in the flash memory is success, and the failure bit detection method further includes: In response to the current logic value of the target data bus indicating a programming failure, the conversion circuit converts the current logic value of the target data bus into a corresponding current component in the discharge current. as well as In response to the current logic value of the target data bus indicating that the programming result is passed, the conversion circuit abandons the conversion of the current logic value of the target data bus into the corresponding current component in the discharge current.
16. The method for detecting faulty bits according to claim 13, further comprising: The pre-charging circuit stops pre-charging the capacitive wire during the determination period following the pre-charging period. The switching circuit draws the discharge current from the capacitive wire during the judgment period, causing the node voltage to continuously decrease during the judgment period. The more data buses in the different data buses that indicate a programming failure, the larger the discharge current, and the faster the node voltage decreases. The decision circuit compares the node voltage to a reference voltage during the decision period. The judgment circuit determines the fall time length based on the comparison between the node voltage and the reference voltage. The judgment circuit compares the descent time length with a reference time length; as well as The judgment circuit determines whether the current logic value of the different data buses is invalid based on the comparison result between the fall time length and the reference time length.
