Azalea Microelectronics Corporation

Winbond Electronics Corporation

Integrated Silicon Solution Incorporated

Nexflash Technologies, Inc.

Other

Ovonyx Inc.

Windbond Electronics Corp.

Frontenac Ventures

Eungjoon Park

Ali Pourkeramati

Joo Weon Park

Robin John Jigour

Poongyeub Lee

Hsi-Hsien Hung

Jongjun Kim

Kwangho Kim

Tyler A Lowrey

Masaru Yano

Hui W Chen

Tien-Ler Lin

Kyung Joon Han

Jong Seuk Lee

Jooweon Park

A flash memory device may operate from two supply voltages, one being provided externally, and the other being generated within the flash memory device from the external supply voltage. The flash memory device may be provided with a selectable-level buffer for interfacing with either low supply voltage or high supply voltage integrated circuits. To provide even greater flexibility, the flash memory device may be provided with the capability of receiving a second supply voltage from an external source, which may take precedence over the internally-generated second supply voltage or may be combined with the internally-generated second supply voltage.

Flash memory having dual supply operation

A flash memory for code and data storage includes a code memory array having fast read access and suitability for execute in place, a data memory array having the characteristics of low bit cost and high density storage, and a suitable interface to provide access to both the code and data. The code memory array may be a NOR array or a performance-enhanced NAND array. The memory may be implemented in a single chip package or multi-chip package solution.

Flash memory for code and data storage

In accordance with an embodiment of the present invention, a semiconductor wafer has a plurality of dies each having a circuit and a plurality of contact pads. The plurality of contact pads include a first contact pad to receive a power supply voltage, a second contact pad to receive a ground voltage, and a third contact pad to receive a test control signal. The third contact pad is connected to a programmable self-test engine (PSTE) embedded on the corresponding die so that the test control signal activates the PSTE to initiate a self-test. A probe card has a plurality of sets of probe pins, each set of probe pins having three probe pins for contacting the first, second, and third contact pads of one of a corresponding number of the plurality of dies. During wafer test, the plurality of sets of probe pins come in contact with a corresponding number of dies so that the self-test is carried out simultaneously in the corresponding number of dies.

Structure and method for parallel testing of dies on a semiconductor wafer

A serial flash memory is provided with multiple configurable pins, at least one of which is selectively configurable for use in either single-bit serial data transfers or multiple-bit serial data transfers. In single-bit serial mode, data transfer is bit-by-bit through a pin. In multiple-bit serial mode, a number of sequential bits are transferred at a time through respective pins. The serial flash memory may have 16 or fewer pins, and even 8 or fewer pins, so that low pin count packaging such as the 8-pin or 16-pin SOIC package and the 8-contact MLP/QFN/SON package may be used. The availability of the single-bit serial type protocol enables compatibility with a number of existing systems, while the availability of the multiple-bit serial type protocol enables the serial flash memory to provide data transfer rates, in systems that can support them, that are significantly faster than available with standard serial flash memories.

Serial flash semiconductor memory

Method and apparatus for memory device testing at a higher clock rate than the clock rate provided by a memory tester. The method includes providing a memory tester capable of generating a first clock signal characterized by a first clock frequency, and applying the first clock signal to the memory device. The method also includes receiving a command for activating a high-clock-frequency test mode. The method generates a second clock signal in the memory device in response to the first clock signal. The second clock signal is characterized by a second clock frequency which is higher than the first clock frequency. The method then tests the memory device at the second clock frequency. In a specific embodiment, the method is applied to a serial flash memory device. The invention can also be applied to testing and operating other memory devices or systems that include synchronized circuits.

Clock frequency doubler method and apparatus for serial flash testing

To control the problem of program and erase disturb in flash memory arrays having multiple sectors of cells grouped in each isolation wells of the flash memory array, a refresh procedure is used that involves two readings of each of the cells in a 'refresh area' of a group under different read timing conditions, with other read conditions being constant or varied as desired. Cells that yield the same result in both reads are not excessively disturbed and need not be reprogrammed. However, cells that read differently may be excessively disturbed and should be reprogrammed. The refresh procedure is particularly suitable for memory arrays with small sector size and many sectors per group. The memory arrays preferably incorporate memory cells that use hot electron programming and Fowler-Nordheim erase.

Small sector floating gate flash memory

The problem of bit disturb is reduced by discharging the floating bit lines of a nonvolatile memory array during programming. An illustrative virtual ground memory array uses single transistor floating gate type memory cells that are programmed using Fowler-Nordheim ('FN') tunneling, highly conductive and lengthy bit lines, buried and relatively short sub-bit lines and a programming discharge circuit for controlling spurious voltages on the bit lines that can arise when some of the bit lines are left floating during programming. Discharge control transistor respectively coupled to the bit lines direct current into a discharge section. A discharge section may be provided for each bit line, or shared by all bit lines. The discharge section may be a fixed circuit section for use through the programming process or may be selected from multiple discharge options.

Nonvolatile memory having bit line discharge, and method of operation thereof

A memory uses multiple threshold levels in a memory cell that are not a power of two, and further uses a cell mapping technique wherein the read mapping is only a partial function The domain of read states for a single three-level memory cell, for example, has three states, but only two of them can be uniquely mapped to a bit. The domain of read states for two three-level memory cell, for example, has nine states, but only eight of them can be uniquely mapped to three bits. Although the read mapping is only partial, the voltage margin for the three-level memory cells is larger that the voltage margin available in the commonly used four-level memory cells. This increased voltage margin facilitates memory cell threshold voltage sensing, thereby increasing the reliability of the memory. Memory reliability may be further improved by increasing the voltage margin between the memory cellstate and thestate relative to the voltage margin between thestate and thestate, which more effectively accommodates charge loss from thestate through electron leakage. Asymmetrical read and program mapping may also be used to improve read reliability in the presence of ground noise or Vnoise.

Nonvolatile semiconductor memory having three-level memory cells and program and read mapping circuits therefor

In a NOR-type flash memory of either the ETOX or virtual ground type that is programmed using electron injection and erased using FN tunneling, and that has row redundancy, the typical sequence of operations used for an embedded sector erase, namely the Preprogram, Preprogram Verify, Erase, Erase Verify, Post-Program Verify, and Post-Program operations, need not be performed for the data cells on bad or shorted rows or in unused redundant rows. Instead, the bad or shorted rows or the unused redundant rows are suitably biased so that the threshold voltages of the data cells in these rows tend to converge to a threshold voltage near the UV erased threshold.

Method and apparatus for providing row redundancy in nonvolatile semiconductor memory

In an array of nonvolatile memory cells, as many memory cells as desired and indeed even the entire array of memory cells may be placed in a single region of the bulk, illustratively a p-well. Peripheral circuitry is used to in effect section the memory array into blocks and groups of blocks, and to establish suitable biasing and counter-biasing within those blocks and groups during page or block erase operations to limit erase disturb. Each group is provided with its own set of voltage switches, which furnishes the bias voltages for the various modes of operation, including erase. Each of the voltage switches furnish either a large positive voltage when its group is selected, or a large negative voltage when its group is unselected. The size of the group is established as a compromise between degree of erase disturb and substrate area required for the voltage switches.

Nonvolatile memory and method of operation thereof to control erase disturb

In accordance with one embodiment of the present invention, a non-volatile integrated circuit memory includes a flash EPROM array having a first plurality of memory cells, and an EEPROM array having a second plurality of memory cells arranged along rows and columns. Each of the first and second plurality of memory cells has a drain region spaced apart from a source region to form a channel region therebetween. The drain region has a greater depth than the source region. Each memory cell further has a floating gate and a select gate. The EEPROM array further includes a plurality of data lines each being coupled to the drain regions of a plurality of cells along at least a portion of a column of cells, and a plurality of source lines each being coupled to the source regions of a plurality of cells along at least a portion of a row of cells.

Integrated circuit having an EEPROM and flash EPROM using a memory cell with source-side programming

In accordance with an embodiment of the present invention, an array of non-volatile memory cells are arranged along rows and columns. Each memory cell has a drain region spaced apart from a source region to form a channel region therebetween. The drain region has a greater depth than the source region. Each memory cell further has a stack of floating gate and select gate extending over the channel region, the select gate of the cells along each row being connected together to form a wordline. Each of a plurality of data lines is coupled to the drain region of at least a portion of a column of cells. Each of a plurality of source lines is coupled to a source region of a plurality of cells along at least a portion of a row of cells, wherein injection of hot electrons from a portion of the channel region near the source region to the floating gate is induced in a selected memory cell in the array by applying a first voltage to a selected data line to which the drain of the selected memory cell is coupled, a second positive voltage to a word line to which the selected gate of the selected memory cell is coupled, and a third positive voltage to a source line to which the source of the selected memory cell is coupled, wherein the injection of hot electrons increases a threshold voltage of the selected cell.

Growing community of inventors

Fremont, CA, United States of America

Eungjoon Park