On Friday, February 06, 2015 01:10:36 TJF wrote: > Am Freitag, 6. Februar 2015 00:06:45 UTC+1 schrieb H: > > Using the PRU is a round about way to get regular sampling. The ADC can be > > configured to sample at a regular rate and place samples in a FIFO and > > interrupt as the FIFO fills. The PRU isn't needed. See the TRM. > > That's not entirely correct. Yes, you can configure the ADCSS to measure in > continous mode, but you don't have control over the timing. To get an exact > sampling rate it needs a trigger, either one of the TIMER-[4-7] or a > software trigger (ie. from PRU code). When using the PRU for trigger, it > can also do FIFO clearance and low level data manipulation, like scaling > the samples to 16 bit (ie. to use them for audio data). The key benefits of > the PRU solution are being more flexible and coding single source on a > real-time unit.
You can control the timing on the ADC with the open/close times on each step. I have the ADC by itself working well enough for an 8KHz sampling with noise filtering using the oversampling/averaging for audio purposes. The driver convert the samples as 16bit (simple bit shift) as it unloads the FIFO. The bigger problem is noise on the ADC itself. Using the PRU would slam the internal bus twice eatting up bandwidth. It is no big deal if you don't need the bandwidth... -- Hunyue Yau http://www.hy-research.com/ -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
