Re: AIC23 _ High Pass Filter_BIG_FUNDAMENTAL_PROBLEM

Akash_DSP wrote:
Hi,

I am working on C6416 DSK which has AIC23 (codec). I am facing a bizarre
problem.

I programmed the C6416 using CCS v3.3 to just collect the analog input
through the A/D, send it to the D/A and see the waveform on the
oscilloscope. But there is a big problem. The phase delay between the input
signal (Sampling Frequency =44.1 KHz) and the output signal as seen on the
oscilloscope varies periodically which is more then one sampling period.

"Varies periodically" means that the delay between input and output changes on a periodic (in the simplest case, sinusoidally) way. Is that what you mean?

If the tone.c program given as example is modified to generate a square
wave of 17 Hz (approx) then you can actually see the differentiation of the
signal. This means that there is a problem with the High Pass filter in the
aic23. Even after disabling it through the registers the output remains the
same i.e. the HPF still functions.

The anti-alias filter before the A-to-D converter causes what you see. That is a low-pass filter, not high pass. Id you disable it, you will suffer aliasing, which is worse.

Can anyone suggest me some ways to get out of this problem and to disable
the HPF. Please reply ASAP and let me know if you need any additional
information.

Re-familiarize yourself with how sampling works and what its limitations are.

Jerry
--
Engineering is the art of making what you want from things you can get.
¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯
.