Does anyone have any experience programming the SPI interface on the AD9434 (or similar) eval board? I am a relative newbie at this, so I very well may have missed something obvious.
I am trying to talk to the eval board directly as part of a prototyping project before we commit to a full board spin. I am using a single board computer running Linux to do the interfacing. Nothing that I do either attempting to read from or write to the SPI interface seems to have any effect. The chip seems to be deaf. I can change the settings using the HSC-ADC-EVALCZ board and the pre-canned SPI software, so I know the board is alive.
I can view the various signals CS, SCLK, and SDIO at the test points on the eval board, so I know the signals are getting to the board and appear to be at the right voltage levels. After struggling to reconfigure the Linux driver, I got the various signals to look the way I expected. The only anomaly I see there is that there is a gap of several clock cycles between the chip select and the first clocking of the instruction information. I also see a similar (although a bit longer gap) after the clocking of the last data bits. My understanding of the SPI protocol is that these gaps don't matter since SCLK is not changing. I've also worked at different clock speeds, but am currently working at 1 MHz to be sure data hold times, etc. are ok. (I'm using jumper wires to interface between the microcontroller board and the header on the eval module, so the connections are not great, although on the scope they appear ok.)
A couple of additional items: Jumper P400 is in the 1.8 V position (pins 2/3 jumpered). P200 (PDWN) oddly enough doesn't seem to make a difference, but I mostly try talking to the board with it in the off position.
Am I missing a signal that the HSC-ADC-EVALCZ board provides that makes it able to talk to the 9434 board, or am I missing something else? Any help would be appreciated.