2003-11-20 12:25 PM
uPSD3233BV-24U1 Flash programming problems
2011-05-17 02:56 AM
Hi,
I have a customer and he is having problems with a uPSD3233BV-24U1 uPSD. They are as follows - The main problem is that it won't program the flash The part will erase The part will read back the empty contents The part will even program the PSD section but it refuses to program the flash memory. He has tried to get the part to start programming from block 1 instead of block zero to no avail. The error it is giving is that it times out while attempting to program He has programmed the device approximately 20 times previous to these problems Any suggestions? Do you think it is simply a bad chip? The customer is a bit cautious about production issues if these arises again! Thanks a lot Manz2011-05-17 02:56 AM
Have you tried to use the same PSDsoft project files to program a different uPSD part? If do not have another part to use, maybe you can attach your zipped PSDsoft project files here, I can try it out on my DK3200 board. Also, have you tried to program the Flash through the MCU bus and see if that work for you?
2011-05-17 02:56 AM
Hello all:
I'm the customer Manz referred to. Since his original post, I have followed this problem up. Here's what I have found: 1. The uPSD part correctly programs and verifies most of the time. 2. If I ''Verify All'', the verify fails either in the flash or the PSD approximately 10% of the time. When it fails, it reports a single bit error at a particular location, e.g. ''expect 0xB1, read 0xB3'', although the errors reported are not repeatable. 3. If I ''Program/Verify All'', the operation fails perhaps approximately 10% of the time. Running ''Verify All'' immediately after such a failure shows that the part is indeed correctly programmed, i.e. the failure is in the verification step. 4. Infrequently, running ''Program/Verify All'' results in the problem described in the original post: the flash fails to program at all. This has happened three times in approximately 200 programming attempts. Replacing the part fixes the problem . 5. I am using 4-wire JTAG programming. Circuitry is as per DK3200 schematics, but with no series 100ohm resistors in TCK and TMS lines. The JTAG signals are all well defined with clean edges and fast rise times. The minimum TCK period is approximately 2us. My FlashLINK programmer is a Rev F PCB. Target is running at 3.3V. Does anyone have any insight into this problem? Any help would be greatly appreciated. Andrew Ward2011-05-17 02:56 AM
The problem has been resolved. Turns out the problems we were having were due to our FlashLINK programmer, model FL-101, which is 3 or 4 years old. The programmer works without error on a 733MHz Win98 box. The errors described above are generated when the programming is attempted from a 2.5GHz WinXP box.
We purchased a new FlashLINK programmer (model FL-101B) and this successfully programs and verifies our board from WinXP. Andrew