This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

CCS/PGA900: The target failed to see a correctly formatted SWD header.

Part Number: PGA900

Tool/software: Code Composer Studio

I designed my own PCB with a PGA900 and have 50 pcs. I have it professionally manufactured. I also have the PGA900EVM and the XDS200 debugger.

I use the latest version of CCS 9.0.1.

I can easily debug a PGA900 with the XDS200 with the PGA900EVM.
But with exactly the same project it doesn't work (anymore) with my own print . A few months ago it worked with CCS V8.0.2, but only after I shortened the debug cable and lowered the frequency to 50KHz.
Now I have the new CCS version and I just can't get it to work anymore. And I can't get back to the old setup, because it was on an older pc.
When I test the debug connection I get the error:

[Start: Texas Instruments XDS2xx USB Debug Probe_0]

 

Execute the command:

 

%ccs_base%/common/uscif/dbgjtag -f %boarddatafile% -rv -o -S integrity

 

[Result]

 

 

-----[Print the board config pathname(s)]------------------------------------

 

C:\Users\CH1011~1\AppData\Local\TEXASI~1\

   CCS\ccs901\0\0\BrdDat\testBoard.dat

 

-----[Print the reset-command software log-file]-----------------------------

 

This utility has selected a 560/2xx-class product.

This utility will load the program 'xds2xxu.out'.

The library build date was 'Apr 30 2019'.

The library build time was '12:14:56'.

The library package version is '8.1.0.00012'.

The library component version is '35.35.0.0'.

The controller does not use a programmable FPGA.

 

An error occurred while hard opening the controller.

 

-----[An error has occurred and this utility has aborted]--------------------

 

This error is generated by TI's USCIF driver or utilities.

 

The value is '-615' (0xfffffd99).

The title is 'SC_ERR_SWD_PROTOCOL'.

 

The explanation is:

The target failed to see a correctly formatted SWD header. The

connection to the target may be unreliable. Try lowering the

TCLK setting before trying again.

 

[End: Texas Instruments XDS2xx USB Debug Probe_0]

I have so far taken the following measures to determine the error:
- Aligning the pin levels with the jumpers on the EVM -> EVM still works and own board doesn't
- Desolder PGA900 IC from own PCB and flash in EVM socket -> works
- Check the connection lines GND, SWDCLK, SWDIO, AVDD -> The connection is implemented exactly as in the reference scheme of the EVM and everything is connected correctly.
- Several prints tested and ESD taken into account -> Same error behavior everywhere
- Various (linear) power supplies and supply voltages (5V-10V) used
- Various USB slots used for XD200
- Pull-down 10k attached to SWDIO.

However, the same error message was displayed again and again. It struck me that there are very contradictory data regarding Pull-Up, Pull-Down at SWD, which is generally very badly documented. For example, there is no schema for the XDS200 debuggers connectors (CTI20-Pin to ARM-10Pin).

I am open for further suggestions to find and eliminate the bug and am also willing to release my print design and software project if necessary.

  • Hi,

    Given you already took several measures in hardware to try to isolate this issue between the device and the Debug Probe, including validating this environment in a known good hardware such as the EVM, I can only imagine this may be a system-level problem such as ground loops, excessive noise or other hardware design that is present only on the custom board. 

    I strongly suggest you to try to use an oscilloscope to monitor the ground and Vcc lines while physically plugging the XDS200 to the board, as well as during a test connection, searching for possible noise, inadvertent DC levels, etc. Some waveforms can be seen at the post below. 

    https://e2e.ti.com/support/tools/ccs/f/81/p/661170/2442314#2442314 

    Unfortunately we don't do schematic reviews on this forum, but if you followed the PGA900 EVM information and kept the connections short and solidly connected, our experience shows the factors above become predominant in preventing a successful connection. 

    Hope this helps,

    Rafael