Questions regarding design choices of CW313 board

Embedded Security ChipWhisperer Hardware
1

Hi,

I recently received a Husky and the much more compact CW313 board for targets, which is nice.

But I had a few questions regarding some design choices about it:

  • The Husky and the CW313 are supposed to be designed to work together and I was expecting the SMAs to be respectively connected to SHUNTL and SHUNTH signal to leverage the diff input that the Husky has. But both SMAs are equivalent and wired to SHUNTL.
  • We can provide our own power rails using either the test probes or a 0.1" header, which is amazing but I think you forgot to add the GND signal to the header :slight_smile:
  • Regarding the PCB to convert CW308T targets, I’m a bit surprised that the “top” of the target is facing the CW313 edge connector because quite a few targets, e.g. MCP ones, have a big DB9 connector there, that will get on the way.

Otherwise, thanks a lot for Husky + CW312! That’s such a nice improvement overall. More compact and easier targets to design, better ADC, etc.
Overflow detection allows to programmatically find the best gain to avoid clipping.

2

The 2x SMAs was mostly for using glitch + SCA together. So it was a bit of a trade-off on deciding to route them to SHUNTL( which was more commonly used). In particular with the CW312 boards we never really found much difference in our testing on the differential input with the CW312/CW313 board due to the filtering present. Of course on a “real” board you’d have more common-mode noise.

Originally the header was all test points, then got changed to a header, then the GND test point was left… maybe that is a good idea to update for our next run!

Ah yeah - those adapter boards have outlived their plans! I meant to spin them as a 4-layer board that had an open area in the middle as well, and maybe to arrange the other way around. But for now they’ve stayed the same…

Thanks for feedback! As always lots we’d like to do still but time is the biggest problem for us all :wink: