hi
I have a commercial DAC based on the WM8741. it is configured in software mode. the DAC has a stand-by feature which leaves much of the board powered but (among other things) switches the WM8741 to low power mode.
when in normal operation, there is a fair amount of noise which goes away when the DAC chip is switched to its internal low-power mode. not huge, but enough to be noticeable in a quiet room. until recently I was using the single ended outputs. half signal level, so the noise was halved too, hence I never noticed it. looks like switching to balanced outs doubled the noise and it exceeded my acceptability threshold.
it is obviously not because of the output analog stage, which remains powered at all times.
the power pins of the WM are decoupled as per datasheet, electrolytics paralleled with 100n polyester caps, through hole type. 3 pairs, one for each power pin. the caps being not SMD, they are placed a bit far (maybe 6-10 mm) from the WM chip, due to sizing. could this be the problem?
it is powered by a LT3208 regulator, which claims "a low 20μVRMS over a 10Hz to 100kHz bandwidth".
there is an old (unanswered) post here, asking about WM8741 noise issues in software mode.
any advice welcome. I'll return with some pics.
I have a commercial DAC based on the WM8741. it is configured in software mode. the DAC has a stand-by feature which leaves much of the board powered but (among other things) switches the WM8741 to low power mode.
when in normal operation, there is a fair amount of noise which goes away when the DAC chip is switched to its internal low-power mode. not huge, but enough to be noticeable in a quiet room. until recently I was using the single ended outputs. half signal level, so the noise was halved too, hence I never noticed it. looks like switching to balanced outs doubled the noise and it exceeded my acceptability threshold.
it is obviously not because of the output analog stage, which remains powered at all times.
the power pins of the WM are decoupled as per datasheet, electrolytics paralleled with 100n polyester caps, through hole type. 3 pairs, one for each power pin. the caps being not SMD, they are placed a bit far (maybe 6-10 mm) from the WM chip, due to sizing. could this be the problem?
it is powered by a LT3208 regulator, which claims "a low 20μVRMS over a 10Hz to 100kHz bandwidth".
there is an old (unanswered) post here, asking about WM8741 noise issues in software mode.
any advice welcome. I'll return with some pics.
bump.
one thing that has just occurred to me is that the output is unfiltered and the buffer stage used has a bandwidth > 100 MHz (!). that would not explain the in-band noise, especially with no signal. but still...
what do you think about this setup (unfiltered WM8741+very high bandwidth buffer)?
one thing that has just occurred to me is that the output is unfiltered and the buffer stage used has a bandwidth > 100 MHz (!). that would not explain the in-band noise, especially with no signal. but still...
what do you think about this setup (unfiltered WM8741+very high bandwidth buffer)?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.