摘要:
In a non-volatile memory a group of memory cells is programmed respectively to their target states in parallel using a multiple-pass index programming method which reduces the number of verify steps. For each cell a program index is maintained storing the last programming voltage applied to the cell. Each cell is indexed during a first programming pass with the application of a series of incrementing programming pulses. The first programming pass is followed by verification and one or more subsequent programming passes to trim any short-falls to the respective target states. If a cell fails to verify to its target state, its program index is incremented and allows the cell to be programmed by the next pulse from the last received pulse. The verify and programming pass are repeated until all the cells in the group are verified to their respective target states. No verify operations between pulses are necessary.
摘要:
A ramp-down programming voltage is used to program a group of nonvolatile memory cells in parallel, step by step from a highest step to a lowest step. Overall programming time is improved when a conventional setup for program inhibit together with a verify after each program step are avoided. A program voltage estimate is provided for each cell indicating the programming voltage expected to program the cell to its target. Initially, all but those cells having estimates at or above the current program voltage step will be program-inhibited. Thereafter, with each descending program voltage step, additional cells will be un-inhibited. Once un-inhibited, a cell need not be re-inhibited even if programmed to its target. This is because subsequent program steps are at lower voltages and ineffective in programming the cell beyond its target. The un-inhibit operation in one implementation amounts to simply pulling the associated bit lines to ground.
摘要:
In a nonvolatile memory having an array of memory cells, wherein the memory cells are individually programmable to one of a range of threshold voltage levels, there is provided a predictive programming mode in which a predetermined function predicts what programming voltage level needs to be applied in order to program a given memory cell to a given target threshold voltage level. In this way, no verify operation needs to be performed, thereby greatly improving the performance of the programming operation. In a preferred embodiment, the predetermined function is linear and is calibrated for each memory cell under programming by one or more checkpoints. The checkpoint is an actual programming voltage that programs the memory cell in question to a verified designated threshold voltage level.
摘要:
In a non-volatile memory a group of memory cells is programmed respectively to their target states in parallel using a multiple-pass index programming method which reduces the number of verify steps. For each cell a program index is maintained storing the last programming voltage applied to the cell. Each cell is indexed during a first programming pass with the application of a series of incrementing programming pulses. The first programming pass is followed by verification and one or more subsequent programming passes to trim any short-falls to the respective target states. If a cell fails to verify to its target state, its program index is incremented and allows the cell to be programmed by the next pulse from the last received pulse. The verify and programming pass are repeated until all the cells in the group are verified to their respective target states. No verify operations between pulses are necessary.
摘要:
In a non-volatile memory a group of memory cells is programmed respectively to their target states in parallel using a multiple-pass index programming method which reduces the number of verify steps. For each cell a program index is maintained storing the last programming voltage applied to the cell. Each cell is indexed during a first programming pass with the application of a series of incrementing programming pulses. The first programming pass is followed by verification and one or more subsequent programming passes to trim any short-falls to the respective target states. If a cell fails to verify to its target state, its program index is incremented and allows the cell to be programmed by the next pulse from the last received pulse. The verify and programming pass are repeated until all the cells in the group are verified to their respective target states. No verify operations between pulses are necessary.
摘要:
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 among each stack are factored out. In one aspect, a serial bus allows communication between components in each stack, thereby reducing the number of connections in a stack to a minimum. A bus controller sends control and timing signals to control the operation of the components and their interactions through the serial bus. In a preferred embodiment, the bus transactions of corresponding components in all the similar stacks are controlled simultaneously.
摘要:
A memory has defective locations in its user portion replaceable by redundant locations in a redundant portion. Data latches in the user and redundant portions allow data sensed from or to be written to a memory to be exchanged with a data bus. A defective location latching redundancy scheme assumes the column circuits including data latches for defective columns to be still useable. The data latches for the defective columns are used to buffer corresponding redundant data that are normally accessible from their data latches in the redundant portion. In this way both the user and redundant data are available from the user data latches, and streaming data into or out of the data bus is simplified and performance improved.
摘要:
When reading data from a non-volatile storage element that is part of a group of connected non-volatile storage elements, resistance information is measured for the group. One or more read parameters are set based on the measured resistance information. The read process is then performed using the one or more parameters.
摘要:
A memory has defective locations in its user portion replaceable by redundant locations in a redundant portion. Data latches in column circuits of user and redundant portions allow data sensed from or to be written to a memory to be exchanged with a data bus. A remote redundancy scheme has the redundant data available from a central buffer accessible by any number of column circuits. Redundant data buffer circuits enable bus exchange with data from the user data latches except for defective locations when data are taken from the central buffer. In this way only addressing for the user portion is used for bus exchange. Also, accessibility to the redundant data will not be restricted by the locations of the column circuits relative to the redundant data latches and the buffered redundant data can be accessed at a finer granularity than that imposed by the column circuits.
摘要:
A non-volatile memory and method for its operation that can reduce the amount of disturb in non-selected cells during an erase process are presented. For a set of storage elements formed over a common well structure, all word-lines are initially charged with the same high voltage erase signal that charges the well to insure there is no net voltage difference between the well and word-lines. The selected word-lines are then discharged to ground while the non-selected word-lines and the well are maintained at the high voltage. This can be accomplished without increasing any pitch area circuit or adding new wires in the memory array, and at minimal additional peripheral area. Advantages include less potential erase disturb in the non-selected storage elements and a tighter erase distribution for the selected elements.