Jake,
I apologize for this delay in replying to your post.
I am trying to get my hands on a Hercules Launchpad and a ADS7961 board, but in the meantime here are some questions based on your post.
You mentioned that you are expect to “read the channels from the chip sequentially, responding with that data as well as the previous channel”. A lot of people do get confused with this particular family of parts, but you actually get the response for the channel selected two frames ago. I hope this is not the source of the issue you are seeing. Please see a detailed explanation for this behaviour in this post: http://e2e.ti.com/support/data_converters/precision_data_converters/f/73/p/307340/1073979.aspx#1073979
Looking at the code, you seem to have constructed the packet correctly. You are reading from all 16-channel sequentially and you have not used the flipPacket function in the code. I am assuming that you are using the standard spi.h, with no modifications.
Could you clarify what you mean you data received seems to be garbage. Are they random or is there a repeatability across 16-captures that we can discern. Which of the channels are connected and are a few unconnected/grounded? It would help if you could share the following:
- Schematics of the board
- A screenshot of your logic analyser capture, preferably starting at the Ch0 request
- The source code including the .h files
Thanks.
Regards,
Sandeep