Abstract: In a nonvolatile memory array, blocks that contain only obsolete data are modified by adding charge to their cells, increasing the charge level from the programmed charge levels that represented obsolete data to elevated charge levels. The increase in overall charge in such blocks lessens the tendency of such blocks to impact data retention in neighboring blocks.

Abstract: A non-volatile memory system that has two or more sub-blocks in a block performs a check before accessing memory cells to see if the condition of a sub-block that is not being accessed could affect the memory cells being accessed. If such a sub-block is found then parameters used to access the cells may be modified according to a predetermined scheme.

Abstract: Charge leakage from a floating gate in a NAND flash memory die is reduced by applying a data retention bias to a word line extending over the floating gates. The data retention bias is applied to one or more selected word lines when the memory die is in idle mode, when no read, write, erase, or other commands are being executed in the memory die.

Abstract: A non-volatile memory chip having SLC blocks acting as a write cache for MLC blocks for high density storage requires constant copying or folding of SLC blocks into MLC blocks. To avoid the time-consuming toggling out and in of the pages of the entire SLC block for ECC checking by a controller chip, only a small sample is checked. An optimal read point for reading the memory cells in the sample of the SLC block is dynamically determined by trying different read points so that the data is read within an error budget. Once the optimal read point is determined, it is used to read the entire SLC block without further error checking. Then the SLC block can be copied (blind folded) to the MLC block with the confidence of being within the error budget.

Abstract: A NAND flash memory chip includes a first partition that has smaller memory cells, with smaller charge storage elements, and a second partition that has larger memory cells, with larger charge storage elements, in the same memory array. Data is selected for storage in the first or second partition according to characteristics, or expected characteristics, of the data.

Abstract: Memory cells that are indicated as being erased but are suspected of being partially programmed may be subject to a verification scheme that first performs a conventional read and then, if the conventional read does not indicate partial programming, performs a second read using lower read-pass voltage on at least one neighboring word line.

Abstract: Lower page data that may be endangered by programming upper page data in the same memory cells is protected during upper programming using protective upper page programming schemes. High overall programming speeds are maintained by selectively using protective upper programming schemes only where endangered data is committed and may not be recoverable from another location.

Abstract: In a multi-level cell memory array, a flag that indicates that a logical page is unwritten is subject to a two-step verification. In a first verification step, the logical page is read, and ECC decoding is applied. If the first verification step indicates that the logical page is unwritten, then a second verification step counts the number of cells that are not in an unwritten condition.

Abstract: In a non-volatile memory device, the parameters used in write and erase operation are varied based upon device age. For example, in a programming operation using a staircase waveform, the amplitude of the initial pulse can be adjusted based upon the number of erase-program cycles (hot count) of the block containing the selected physical page for the write. This arrangement can preserve performance for relatively fresh devices, while extending life as a devices ages by using gentler waveforms as the device ages.

Abstract: In a nonvolatile memory array, blocks that contain only obsolete data are modified by adding charge to their cells, increasing the charge level from the programmed charge levels that represented obsolete data to elevated charge levels. The increase in overall charge in such blocks lessens the tendency of such blocks to impact data retention in neighboring blocks.

Abstract: In a nonvolatile memory that stores data in two or more different data storage formats, such as binary and MLC, a separation scheme is used to distribute blocks containing data in one data storage format (e.g. binary) so that they are separated by at least some minimum number of blocks using another data storage format (e.g. MLC).

Abstract: A memory device cooperating with a memory controller scrambles each unit of data using a selected scrambling key before storing it in an array of nonvolatile memory cells. This helps to reduce program disturbs, user read disturbs, and floating gate to floating gate coupling that result from repeated and long term storage of specific data patterns. For a given page of data having a logical address and for storing at a physical address, the key is selected from a finite sequence thereof as a function of both the logical address and the physical address. In a block management scheme the memory array is organized into erase blocks, the physical address is the relative page number in each block. When logical address are grouped into logical groups and manipulated as a group and each group is storable into a sub-block, the physical address is the relative page number in the sub-block.

Abstract: A non-volatile memory chip having SLC blocks acting as a write cache for MLC blocks for high density storage requires constant copying or folding of SLC blocks into MLC blocks. To avoid the time-consuming toggling out and in of the pages of the entire SLC block for ECC checking by a controller chip, only a small sample is checked. An optimal read point for reading the memory cells in the sample of the SLC block is dynamically determined by trying different read points so that the data is read within an error budget. Once the optimal read point is determined, it is used to read the entire SLC block without further error checking. Then the SLC block can be copied (blind folded) to the MLC block with the confidence of being within the error budget.

Abstract: In a nonvolatile memory array, data is stored in multi-level cells (MLC) as upper-page data and lower-page data. Safe copies of both upper-page and lower-page data are stored in on-chip cache during programming. If a write fail occurs, data is recovered from on-chip cache. The controller does not have to maintain safe copies of data.

Abstract: In a nonvolatile memory system, data received from a host by a memory controller is transferred to an on-chip cache, and new data from the host displaces the previous data before it is written to the nonvolatile memory array. A safe copy is maintained in on-chip cache so that if a program failure occurs, the data can be recovered and written to an alternative location in the nonvolatile memory array.

Abstract: In a nonvolatile memory array, data is stored in multi-level cells (MLC) as upper-page data and lower-page data. Safe copies of both upper-page and lower-page data are stored in on-chip cache during programming. If a write fail occurs, data is recovered from on-chip cache. The controller does not have to maintain safe copies of data.

Abstract: In a nonvolatile memory system, data received from a host by a memory controller is transferred to an on-chip cache, and new data from the host displaces the previous data before it is written to the nonvolatile memory array. A safe copy is maintained in on-chip cache so that if a program failure occurs, the data can be recovered and written to an alternative location in the nonvolatile memory array.