09-02-2015 12:32 PM
I am using the Xilinx Platform cable USB (version 1), with the IMPACT software.
I bought a huge batch of these XC95144XL chips for a project, and I can only get a fraction of them to work. They all recognise perfectly in IMPACT as the correct chips, but when I go to erase them the program seems to hang, and I have to unplug the USB cable to get the program to respond again. For the chips that DO erase, most of them give errors during programming, or stall again in the pre-programming erase. Out of the 100 I have tested so far I've gotten 6 to actually program to completion.
I have tried every setting I can find, and I can't get these to work. But I have another small batch that I ordered a few months ago (different supplier) and those all work on the first try. Is it that the chips could be all faulty, or is there another setting or program I can try? I have done the Overwrite Erase protection in the erase options as well, to no avail.
Any help would be greatly appreciated!!
09-02-2015 12:37 PM
I would contact your authorized Xilinx distributor you purchased them from.
If you did not buy them from a Xilinx authorized distributor, then we cannot help you: there are folks out there who gather scrap parts and re-sell them as good parts (clean, re-mark) -- obviously they are not good parts at all.
09-02-2015 12:43 PM
09-02-2015 12:59 PM
Given the older batch programs, and the new one does not, that makes it pretty clear there is something wrong.
Either the programmer hardware is now different, or the devices are.
Did you say that you are able to program the older batch devices now just fine?
Have you checked the revisions, errata, product change notices?
I am not finding anything....
09-02-2015 01:16 PM
The revisions appear to be identical. The only thing I can see is that the non-working ones have a TM symbol next to XC95144XL, and the working chips do not. And all reference pictures I can see on Google have the same TM.
But the revision numbers are the same otherwise.
09-02-2015 01:19 PM - edited 09-02-2015 01:23 PM
TM is Trade Mark. If TM is not there, I would question where they came from....
Sure these came through the proper channels? Something sounds strange here.
What about the date codes?
09-02-2015 09:38 PM
09-03-2015 12:15 AM
Please share complete part marking snap of problematic parts
Did the board and design is same for problematic parts & good parts?
Make sure all voltages are proper as per data sheet especially during erasing operation
09-03-2015 08:41 PM
I upgraded to the Xilinx Platform Cable USB 2 programmer and now the chips are working properly! Still odd that the old batch worked just fine, but at least now I know it must have been something with the JTAG programmer.
09-04-2015 03:28 AM
@getafixx Good. I dont see any programming algo,errata or any other changes notification for 144XL.