- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Morning folks,
I have just created my first FPGA (EP3C10E144C8N)board with a config memory (AT25DF321A-SH-B). Whilst I seem to be able to program the device with the jic and then verify it, the FPGA does not configure. If i make a change to the program and then compile etc and verify it spots the difference and fails verification, which suggest to me that this device is at least capable of communicating during programming. After reset and the FPGA starts to look for configuration I examine the nCONFIG, nSTATUS, CONF_DONE signals and find that the nSTATUS pin is toggling suggesting to me that (according to p 9-9 of the cyclone II handbook) that it is failing configuration and attempting to start itself again, and doing so repeatedly! CONF_DONE is low (which stands to reason if it hasnt configured) and nCONFIG is high (which again is what i would expect to see). I have initially told Quartus and the jic conversion process that the config device is an EPCS16. I can only assume that the device is not working as I expect even though I was able to program and verify the jic file to the memory device. Is this true, that whilst it may program and verify ok it still might not configure the FPGA?? Would i be better of with M25P16 device??? Any comments would be most welcome. https://www.alteraforum.com/forum/attachment.php?attachmentid=8202 https://www.alteraforum.com/forum/attachment.php?attachmentid=8203 deBoogle PS couple of waveforms showing nSTATUS on CH1, and the CLK on dig CH0, SO on dig CH1 and SI on dig CH2Link Copied
4 Replies
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi:
One issue is the size is wrong. Altera doesn't have an EPCS32. Only 16 and 64 http://www.altera.com/literature/hb/cfg/cyc_c51014.pdf It's probably getting to the device verification and failing. (My guess). Because the size is incorrect. I have had good luck with the M25P family. I have used other manufactures, but I validated them on a board before releasing them for production. Not all serial flashes work even if they are the correct size. I know we had issues with Atmel parts. Lately we are using external CPU's to do the programming more often, so I haven't looked at this issue recently. Looking at the Altera site, it looks like they are using quad SPI going forward as the standard. We haven't used any of these yet.. Pete- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
At power on the Cyclone device issues a "Read Silicon ID" (opcode ABh) - Read the EPCS data sheet.
The AT25DF321A data sheet refers to this opcode as "Resume from Deep Power-Down". Try disabling the Silicon ID check (its in the advanced options when you create the EPCS programming image). Cheers, Dave- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Good Catch Dave!
I knew the Cyclone devices die a "Read Silicon ID" but didn't know you could disable it. Pete- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Many thanks Guys,
In the end I went to the 'known' compatible devices from ST and it works a treat. I will however be trying the the AT device again but with the ID switched off. Again, many thanks. D
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