anderray

The JTAG HS3 is directly connected to the J3 JTAG header. The Xilinx DLC9G connects to J3 through a ribbon cable. Correct, there are no series resistors added to the JTAG lines, only the pull-ups. I have also tested with 10K pull-ups instead of the 4.7K, but there was no change.

TDI being stuck high may be a red herring. I've now seen it program successfully with the Digilent adapter and still have TDI high at the end. After a successful programming, sometimes it remains high and sometimes not. The increased TMS activity only happens when programming fails. Also, interestingly, with the Digilent JTAG adapter programming will always fail if the Saleae logic analyzer is connected to TCK. It's connection to the rest of the signals does not effect the frequency of success. There maybe be some electrical issue that the Xilinx adapter is more tolerant of. Changing the JTAG speed makes no difference. Even at 125Khz it will still fail.

The forum doesn't seem to easily show the filenames. To clarify, The first logic image is "Kintex Ultrascale Xilinx JTAG Programming Successful" And the second image is "Kintex Ultrascale Digilent JTAG Programming Failed"

The schematic is pretty simple, I've attached shots of the relevant parts. When using the Xilinx JTAG adapter I do not see TDI go high like with the Digilent adapter. At least not for multiple seconds at the end of the programming attempt. I've attached a screenshot of the capture showing the whole programming sequence for both. I have the logic captures saved from a Saleae analyzer. I could provide them if useful, but they are to large to directly attach here.

I'm having trouble programming a XCKU060 with the JTAG-HS3. Occasionally it succeeds and programs correctly, maybe 1 in 10 tries, but mostly programming fails with "ERROR: [Labtools 27-3165] End of startup status: LOW" Using the Xilinx DLC9G JTAG adapter on the same board always succeeds in programming. Looking at a logic analyzer capture of the JTAG data there is a point where it looks like the TDI gets stuck high when the programming fails. It seems like there is some state or timing issue, but I'm not sure how to figure out what it is. I've tried different JTAG speeds, even down to 125Khz and it's always the same. Any help would be apricated. Thanks