My ignorance knows no bounds

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
In my CS43122 DAC I'm picking up a lot of noise via the BNC connector. If I remove the cable altogether, there is a huge amount of audible crap coming from the analog outputs. Presumably this is because the connector, the digital circuit, and the analog circuit all share the same ground plane, but I really have no idea. What should I try?
 
I cut the ground plane down on a line down the center of the pulse transformer, completely severing the input from the rest of the circuit. I connected this separate ground with its own wire back to the star ground at the power supply. This cut the noise significantly, and now there is no difference between the noise with the cable plugged or unplugged.

I'm starting to believe that my Revision B with three ground planes and four power entry points will be significantly better.
 
halojoy said:
Okay. We will have to wait for
our friends with greater experiences in this thing.
Like Jocko and Elso.
But I would also think that the ground is involved in some way.

Very high performance DAC's are like wild beasts. The pcb layout is essential (I mentioned it earlier). You can get real scared if you read the datasheets or the application notes. The evaluation board from Crystal is a 4-layer pcb.

I did also fail but my failure was not to get approved for CE marking which is much harder than just get descent sound out. I think also that the CS43122 maybe can be even harder to connect proberly.

jwb, I think your main trouble is ground, supply voltage and location of the parts. MuRata-filters are good in order to filter out RF-noise.

Good luck with your revision B and keep us posted about your progress.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.