- Tue Nov 09, 2004 9:45 am
#2178
Hello,
I was making some general libraries for the RF-24G unit in MSP-430 C (The IAR compiler), and noticed that the phase of the SPI clock seems to be different than what Nordic and Laipaic cite. Currently, my code uses phase wherein data is shifted out on the rising edge of the clock, and polled on the falling edge. This reliably works with the RF-24G units. However, if you look at the Nordic/Laipac timing diagrams, they show opposite phase (poll on rising edge, shift on falling edge), for Shockburst TX and Configuration modes. If you aren't using an SPI module, you probably won't notice this (the Sparkfun example code works because it shifts data between edges). Changing to the reccomended phase during receive mode produces courrupt data, which is expected if you poll during a shift, so I'm fairly certain this is the case.
Has anyone else found this to be the case?
A.
I was making some general libraries for the RF-24G unit in MSP-430 C (The IAR compiler), and noticed that the phase of the SPI clock seems to be different than what Nordic and Laipaic cite. Currently, my code uses phase wherein data is shifted out on the rising edge of the clock, and polled on the falling edge. This reliably works with the RF-24G units. However, if you look at the Nordic/Laipac timing diagrams, they show opposite phase (poll on rising edge, shift on falling edge), for Shockburst TX and Configuration modes. If you aren't using an SPI module, you probably won't notice this (the Sparkfun example code works because it shifts data between edges). Changing to the reccomended phase during receive mode produces courrupt data, which is expected if you poll during a shift, so I'm fairly certain this is the case.
Has anyone else found this to be the case?
A.