Nand multiplane operation problem

Hi,

I've been supplied by a different team in my company, with the low level routines to program my flash

and they have taken it upon themselves to make use of the multiplane operation (which is entirely new to me).

However, when I try and use it to erase my flash block-by-block every now and again one (sometimes both) of the planes fails to erase.

Each device fails in a different place, but when multiple attempts are made with the same device it always fails in the same place(s) (in spite of the fact that almost all of the flash is now already erased).

If I rewrite my function to erase the failed blocks using the "normal" erase operation it erases cleanly

Anyone any ideas what might be going on here?

TIA

tim

Reply to
tim.....
Loading thread data ...

Carefully document exactly where those failures exist and under what pattern condition(s)..sort of a 3D map: X,Y location in a plane which i assume is one chip, and the corresponding Z location which i assume is which chip. Do this for at least three different flash of the SAME brand, SAME type, and SAME batch of chips used to see if there are differences. I would bet different brands, types,etc would have different failures,with maybe one brand/type being failure free (maybe even over time = = diffusion batch). Wild guess: the impulse current for an erase creates a voltage spike that can inhibit erasure. Device sensitivity to supply spikes can vary across a chip, and the more layers (chips) there are, the worse it can be.

Reply to
Robert Baer

Your mention of current spikes makes me think -- if the flash is being used on in-house boards, check that the power supply decoupling is adequate. It could be that the chips are fine as long as they've got proper decoupling caps, but end up sensitive to spikes if they're not decoupled properly.

(By "properly" I mean enough caps, of low enough ESR, in the right spots, with the right traces. If the chip manufacturer is good at their job they'll have an ap note. If not, they may have a demo board to use as a guide to decoupling. Or, you may have some crusty old guy who's seen every possible mistake to critique your decoupling situation.)

--

Tim Wescott 
Wescott Design Services 
http://www.wescottdesign.com
Reply to
Tim Wescott

all my available devices are the same model/brand. Can't speak for batch

they all exhibit different failure locations

tim

Reply to
tim.....

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.