Method for non-volatile memory with reduced erase/write cycling during trimming of initial programming voltage
High performance non-volatile memory devices have the programming voltages trimmed for individual types of memory pages and word lines. A group of word lines within each erasable block of memory are tested in successive program loops to minimize the problem of incurring excessive number of erase/program cycles. An optimum programming voltage for a given type of memory pages is derived from statistical results of a sample of similar of memory pages.
1. In a non-volatile memory having an array of memory cells that is organized into erasable blocks, each erasable block containing a block of word lines for accessing memory cells that are erasable together, and each word line containing at least one page of memory cells that are programmable together, a method of determining a starting programming voltage for a given page, comprising:
(a) selecting a sample of pages representative of the given page within a block;
(b) providing an associated programming voltage for programming each page of the sample, the associated programming voltage having a staircase waveform with an associated initial value and a predetermined number of steps;
(c) erasing the block containing the sample of pages;
(d) determining for every page in the sample if the page is programmable to a target pattern using the associated programming voltage with the associated initial value; and if programmable, excluding the page from further processing after accumulating the associated initial value as part of a gathered statistics, otherwise, incrementing the associated initial value by a predetermined step;
(e) repeating (c) to (d) until all pages in the sample has either been determined to be programmable or the associated initial value has been incremented to a predetermined maximum voltage;
(f) computing an average starting programming voltage for the sample from the gathered statistics to derive a starting programming voltage for the page.
2. The method as in claim 1 , wherein:
the sample is part of larger sample taken from a plurality of blocks in the memory array; and
the gathered statistics includes accumulated initial values associated with programmable pages from the plurality of blocks.
3. The method as in claim 1 , wherein
the predetermined step increment of the associated initial value is commensurate with a step of a similar programming voltage applied to the memory during a normal program operation.
4. The method as in claim 2 , wherein
the predetermined step increment of the associated initial value is commensurate with a step of a similar programming voltage applied to the memory during a normal program operation.
5. The method as in claim 1 , wherein the predetermined number of steps of the staircase waveform of the associated programming voltage is fifteen or less.
6. The method as in claim 2 , wherein the predetermined number of steps of the staircase waveform of the associated programming voltage is fifteen or less.
7. The method as in claim 1 , wherein the predetermined maximum voltage is the initial value plus an integral number of steps of the staircase waveform voltages.
8. The method as in claim 2 , wherein the predetermined maximum voltage is the initial value plus an integral number of steps of the staircase waveform voltages.
9. The method as in claim 7 , wherein the integral number of steps is more than forty.
10. The method as in claim 8 , wherein the integral number of steps is more than forty.
11. The method as in claim 1 , wherein accumulating the associated initial value as part of a gathered statistics includes:
accumulating the number of steps the initial value has incremented to when the page is program-verified; and
said computing an average starting programming voltage for the sample is by a product of an average of the number of incremented steps for the sample and the predetermined step of voltage.
12. The method as in claim 2 , wherein accumulating the associated initial value as part of a gathered statistics includes:
accumulating the number of steps the initial value has incremented to when the page is program-verified; and
said computing an average starting programming voltage for the sample is by a product of an average of the number of incremented steps for the sample and the predetermined step of voltage.
13. The method as in claim 1 , further including:
(g) performing an initial programming on the page with a staircase waveform programming voltage having a starting voltage;
(h) verifying if a target pattern has been programmed to the page in between each of the steps of the staircase waveform;
(i) obtaining a final voltage of the staircase waveform when the page is program-verified; and
(j) estimating the associated starting programming voltage by linear scaling of the final voltages; and
wherein (g)-(j) are performed prior to (a)-(f).
14. The method as in claim 2 , further including:
(g) performing an initial programming on the page with a staircase waveform programming voltage having a starting voltage;
(h) verifying if a target pattern has been programmed to the page in between each of the steps of the staircase waveform;
(i) obtaining a final voltage of the staircase waveform when the page is program-verified; and
(j) estimating the associated starting programming voltage by linear scaling of the final voltages; and
wherein (g)-(j) are performed prior to (a)-(f).
15. The method as in claim 2 , further comprising:
forming a set of samples by selecting a page from each block, the page being located in a geometrically similar location of each block;
obtaining a statistical estimation of a programming voltage from each sample of the set; and
wherein the step of computing an average starting programming voltage for the sample is by selecting a minimum statistical estimation among the set.
16. The method as in any one of claims 1 - 15 , wherein individual memory cells each stores one bit of data.
17. The method as in any one of claims 1 - 15 , wherein individual memory cells each stores more than one bit of data.