FM25C160B Puzzling Problem

Tip / Sign in to post questions, reply, level up, and achieve exciting badges. Know more

cross mob
Anonymous
Not applicable

Has anyone seen an entire block of FRAM memory (I'm using the FM25C160B) suddenly contain the value 2420 when looking at 4 byte words?  

   

Currently I am only using the first 30 bytes on this device to store counter information.  As the counters increment (counters are 4 byte unsigned integers, therefore when a counter needs to increment the current value is read from FRAM, the value incremented, and the new value written back to FRAM), at random times the entire FRAM device memory will change value to 2420 (0x0000096A assuming 4 byte values) from address 0 to address 256 (sixty-four 4 byte words).  The memory above that will have a smattering of 9's and a few 6's here and there and then above around address 400 everything is zero.   There are other devices on the SPI bus but I've shut those off to make sure they weren't causing the problem.  Are there Vdd or other issues that could cause this behavior?  It's very suspicious that 256 bytes flip from their previous value to all containing the bytes 0x6A 0x09 0x00 0x00 (starting at address 0, going to 256, repeating these four bytes).

   

I've obtained the same behavior on multiple boards.

   

I surely hope this is a firmware bug that I can find and fix and not a problem with the FM25C160B.

   

Thank you

0 Likes
3 Replies
Anonymous
Not applicable

Hello Steve,

   

Can you please attach your firmware so that we can review your code.

   

Thanks,

   

Krishna.

0 Likes
Anonymous
Not applicable

Krishna,

   

I am now fairly certain the bug is my firmware.  On an oscilloscope, the chip select line has a different waveform when the glitch occurs.  I will let you know when I have it fixed so this thread can be closed.

   

 

   

Thank you,

   

Steve

0 Likes
Anonymous
Not applicable

Problem solved.  Thank you for entertaining my problems.

0 Likes