UPsampling
To Capslock and Tirroth
)
Moreover, the ad1852/53 chars go worse at higher Fs - see the datasheet: at Fs 48kHs rippple with BW20kHz is about 0.0004dB !!!, at Fs96kHs - about 0.2dB /again BW=20kHz !!!/ and at 192 kHz - some -1dB@20kHz and up to -3 dB@90kHz! So what are these 192kHz Fs for?? I'm (ALMOST) sure the sigma-delta is operating at the same clock, only the digital filter is 8x/4x/2x for Fs 48/96/192kHz The true 192kHz DAC will use something like the PCM 1704 and 8x digital filter, I think. With such design high specs can be reached)
Also, the specs in the AD's datasheet assume Fs48k/20bit, no THD or other data for Fs192kHz, am I wright??
regards, Kaloyan
To Capslock and Tirroth
)Moreover, the ad1852/53 chars go worse at higher Fs - see the datasheet: at Fs 48kHs rippple with BW20kHz is about 0.0004dB !!!, at Fs96kHs - about 0.2dB /again BW=20kHz !!!/ and at 192 kHz - some -1dB@20kHz and up to -3 dB@90kHz! So what are these 192kHz Fs for?? I'm (ALMOST) sure the sigma-delta is operating at the same clock, only the digital filter is 8x/4x/2x for Fs 48/96/192kHz The true 192kHz DAC will use something like the PCM 1704 and 8x digital filter, I think. With such design high specs can be reached
)Also, the specs in the AD's datasheet assume Fs48k/20bit, no THD or other data for Fs192kHz, am I wright??
regards, Kaloyan
Machinow,
Take a closer look at the "ripple" of the AD1853. As has been pointed out before, this is not an equiripple design and the consequences of passband ripple are not the same. In fact, most of the "ripple" at 192kHz is caused by a slow rolloff at the higher frequencies. Specifically, see "88kHz passband response 2x mode, 192kHz sample rate" on page 11.
As far as distortion at 192kHz, we see for a 25kHz input spurious responses are dominated by some fringes about -97dB. At 75kHz input there appears to be a response at 37.5kHz and -90dB. These measurements seem pretty in line for the quoted THD at 1kHz.
Take a closer look at the "ripple" of the AD1853. As has been pointed out before, this is not an equiripple design and the consequences of passband ripple are not the same. In fact, most of the "ripple" at 192kHz is caused by a slow rolloff at the higher frequencies. Specifically, see "88kHz passband response 2x mode, 192kHz sample rate" on page 11.
As far as distortion at 192kHz, we see for a 25kHz input spurious responses are dominated by some fringes about -97dB. At 75kHz input there appears to be a response at 37.5kHz and -90dB. These measurements seem pretty in line for the quoted THD at 1kHz.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.