Sandisk Corporation

Sandisk Technologies Inc.

Sunrise Memory Corporation

Siu Lung Chan

Yan Li

Raul-Adrian Cernea

Seungpil Lee

Eli Harari

Robert D Norman

Christopher John Petti

Scott Brad Herner

Khandker N Quader

Sayeef Salahuddin

Frank Sai-Keung Lee

Mehrdad Mofidi

Man Lung Mui

Yupin K Fong

Teruhiko Kamei

Shahzad Khalid

Kwang Ho Kim

Jeffrey W Lutze

Kwang-Ho Kim

Min She

Chang Hua Siau

Xiaowei Jiang

Yupin Kawing Fong

An electronic device with embedded access to a high-bandwidth, high-capacity fast-access memory includes (a) a memory circuit fabricated on a first semiconductor die, wherein the memory circuit includes numerous modular memory units, each modular memory unit having (i) a three-dimensional array of storage transistors, and (ii) a group of conductors exposed to a surface of the first semiconductor die, the group of conductors being configured for communicating control, address and data signals associated the memory unit; and (b) a logic circuit fabricated on a second semiconductor die, wherein the logic circuit also includes conductors each exposed at a surface of the second semiconductor die, wherein the first and second semiconductor dies are wafer-bonded, such that the conductors exposed at the surface of the first semiconductor die are each electrically connected to a corresponding one of the conductors exposed to the surface of the second semiconductor die. The three-dimensional array of storage transistors may be formed by NOR memory strings.

Device with embedded high-bandwidth, high-capacity memory using wafer bonding

Memory device including modular memory units and modular circuit units for concurrent memory operations

Disclosed herein are techniques for sensing multiple reference levels in non-volatile storage elements without changing the voltage on the selected word line. One aspect includes determining a first condition of a selected non-volatile storage element with respect to a first reference level based on whether a sensing transistor conducts in response to a sense voltage on a sense node. Then, a voltage on the source terminal of the sensing transistor is modified after determining the first condition with respect to the first reference level. A second condition of the selected non-volatile storage element is then determined with respect to a second reference level based on whether the sensing transistor conducts in response to the sense voltage on the sense node. This allows two different reference levels to be efficiently sensed. Dynamic power is saved due low capacitance of the sensing transistor relative to the sense node.

Sensing multiple reference levels in non-volatile storage elements

Methods for compensating for variations in bit line resistance during sensing of memory cells are described. The variations in bit line resistance may occur die-to-die or plane-to-plane on the same die. In some embodiments, for each die or memory plane on a die, a plurality of bit line read voltages associated with a plurality of zones may be determined based on sensing criteria. The sensing criteria may comprise a number of fail bits. Each zone of the plurality of zones may be associated with a memory array region within a die or memory plane. Prior to performing a read or verify operation on a group of memory cells, a bit line read voltage used during sensing of the group of memory cells may be determined based on the plurality of bit line read voltages and a zone associated with the group of memory cells.

Bit line resistance compensation

Methods for compensating for variations in bit line resistance during sensing of memory cells are described. The variations in bit line resistance may occur die-to-die or plane-to-plane on the same die. In some embodiments, for each die or memory plane on a die, a plurality of bit line read voltages associated with a plurality of zones may be determined based on sensing criteria such as a number of fail bits. Each zone of the plurality of zones may be associated with a memory array region within a memory plane. Within each zone, different bit line read voltages may be applied to different bit line groupings in order to compensate for systematic variations in bit line resistance between neighboring bit lines due to the use of multiple patterning lithography techniques such as spacer-based double patterning.

Methods for operating a non-volatile storage system utilizing a shared-bit-line NAND architecture are described. A shared-bit-line NAND architecture includes one or more pairs of NAND strings, wherein each pair of the one or more pairs of NAND strings shares a common bit line. In some embodiments, a pair of NAND strings includes an odd NAND string adjacent to an even NAND string. Prior to programming a memory cell associated with the even NAND string, an odd channel associated with the odd NAND string (i.e., the NAND string of the pair that is not selected for programming) is precharged to a bit line inhibit voltage, floated, and then boosted to a second voltage greater than the bit line inhibit voltage as an even channel associated with the even NAND string is precharged. Subsequently, the odd channel may be boosted (e.g., via self-boosting) prior to programming the memory cell.

Shared-bit-line bit line setup scheme

A non-volatile memory device capable of reading and writing a large number of memory cells with multiple read/write circuits in parallel has an architecture that reduces redundancy in the multiple read/write circuits to a minimum. The multiple read/write circuits are organized into a bank of similar stacks of components. Redundant circuits such as a processor for processing data among stacks each associated with multiple memory cells are factored out. The processor is implemented with an input logic, a latch and an output logic. The input logic can transform the data received from either the sense amplifier or the data latches. The output logic further processes the transformed data to send to either the sense amplifier or the data latches or to a controller. This provides an infrastructure with maximum versatility and a minimum of components for sophisticated processing of the data sensed and the data to be input or output.

Non-volatile memory and method with shared processing for an aggregate of read/write circuits

A compact and versatile sense amp is presented. Among its other features this sense amp arrangement provides a way to pre-charge bit lines while doing data scanning. Another feature is that the sense amp circuit can provide a way to set three different bit line levels used in the quick pass write (QPW) technique using dynamic latch, where quick pass write is a technique where cells along a given word line selected for programming can be enabled, inhibited, or partially inhibited for programming. Also, it can provide a convenient way to measure the cell current.

Compact sense amplifier for non-volatile memory suitable for quick pass write

Compact sense amplifier for non-volatile memory

A non-volatile memory device capable of reading and writing a large number of memory cells with multiple read/write circuits in parallel has features to reduce power consumption during read, and program/verify operations. A read or program verify operation includes one or more sensing cycles relative to one or more demarcation threshold voltages to determine a memory state. In one aspect, selective memory cells among the group being sensed in parallel have their conduction currents turned off when they are determined to be in a state not relevant to the current sensing cycle. In another aspect, a power-consuming period is minimized by preemptively starting any operations that would prolong the period. In a program/verify operation cells not to be programmed have their bit lines charged up in the program phase. Power is saved when a set of these bit lines avoids re-charging at every passing of a program phase.

Non-volatile memory and method with power-saving read and program-verify operations

A group of non-volatile memory cells are programmed in a programming pass by a series of incremental programming pulses where each pulse is followed by a program-verify and possibly program-inhibition step. Performance is improved during the programming pass by delayed starting and prematurely terminating the various verify levels that demarcate the multiple memory states. This amounts to skipping the verifying and inhibiting steps of the fastest and slowest programming (fringe) cells of the group. A reference pulse is established when the fastest cells have all been program-verified relative to a first verify level. The starting of what verify level at what pulse will then be delayed relative to the reference pulse. Verifying stops for a given verify level when only a predetermined number of cells remain unverified relative to that given level. Any errors arising from over- or under-programming of the fringe cells are corrected by an error correction code.

Nonvolatile memory and method with reduced program verify by ignoring fastest and/or slowest programming bits

Growing community of inventors

San Jose, CA, United States of America

Siu Lung Chan