- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
hi everyone,
I'm working on a custom board. There is a Max2 on it and I store the serial number of our device in the UFM. Once, one of our customer power down the device and power up again and there was no serial number anymore. After some test, I found that the UFM was erased i-e there were only '1' in the whole memory. I'm trying to understand what happened. I control the UFM with the SPI protocol. I'm sure that the serial number was already written in the UFM. I'm sure that the erase sequence was not send to the UFM. Max2 is powered with a regulated 3.3v. The Max2 VHDL isn't erased and the device starts properly. Has anyone an idea about what could happened on my board? What can cause the UFM to be reseted? thank youLink Copied
5 Replies
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
The UFM in Max II have very less write cycles (less than 100 right out of my head) see datasheet.
we were realy unlucky with the UFM and accessing it via spi was very unreliable. no matter if spi clock was 4MHz nor a few kHz. the spi signals had no over or undershots so we used 1 UFM as a mirrow of the other UFM as a fall back memory. but it happend so often that after power up both UFM delivered garbage via spi or the whole memory content was corrupted. thats why we decided not to use this UFM memory anymore. Our EPM570F100I5N talks with a EP2C70 via SPI, the 3,3V power supply is impedance controlled and stable external reset works the hdl inside max ii works but ufm not always. Same situation no spi command sent to max ii, just a power up.- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
I've used UFM for storage of calibration data in a measurement module design, in other words, the data is accessed internally and only rarely rewritten. I normal operation, write and erased are disabled by an external calibration jumper. As far as I know, no memory failure has been observed with a device yet.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
same in our application where the ufm had the configuration and operating parameters.
normaly ufm was written and only read whenever needed. did you used the spi if to access the ufm ?- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
No, I'm using the altufm_parallel interface.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
I use parallel Mode, and I meet the same question, the UFM is erased when our board power down and power up again even if I always force the erase signal to VCC.
![](/skins/images/318B71AC6E18CAB8E2CA944B2397E849/responsive_peak/images/icon_anonymous_message.png)
Reply
Topic Options
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page