SAA7210 to PCM56, a Dutch-American connection

Status
Not open for further replies.
Jump to Latest
#61
Hi Bernard
the different frequency response of the GIC filter, I have to clone it and try in my DAC.
Click to expand...

I am busy with a tube stage, and maybe try a filter. My preference goes to a 18 dB/okt. Bessel. What also can be done is calculating the values.
http://www-users.cs.york.ac.uk/~fisher/mkfilter/trad.html

Hi Rfbrw
They are more than a little difficult to understand. I can't make out what you are trying to do. I can definitely see no trace of AN207 or any other working schematic I can think of.
Click to expand...

Have been looking at AN207's I2S vs PCM waveform patterns of CLK DATA WS and LE, then its not that difficult anymore. Looked at CS8412/14's datasheet too.
I have altered the I2S schematic again, and will not post it before its tried and tested. As we say in Holland : There are more way's to Rome.
The schematic is allmost like AN207, but clock is untouched. Chips used: 3 74/164 and 174/00 with less its impossible.

I listen critical to cd, sacd and dvd. What i know now is this simple non os pcm56 can give a fine but mostly a natural sound. If a oversampled or upsampled dac is performing better, will listen to that and stop putting effort in this pcm56 experiment.
 
#62
tubee said:


I maybe try a filter.

http://www-users.cs.york.ac.uk/~fisher/mkfilter/trad.html


I listen critical to cd, sacd and dvd. What i know now is this simple non os pcm56 can give a fine but mostly a natural sound. If a oversampled or upsampled dac is performing better, will listen to that and stop putting effort in this pcm56 experiment.
Click to expand...

Try a filter.

Thanks for the link.
The problem with my analog filter is that it is not properly implemented in my DAC.
I found that fiddling with the input resistor almost removes the ripple, perhaps the output impedance needs to be corrected too.

There is an interview in a german hifi mag with the guy who programs the digital filters for Marantz SACD players, one of the new filters does not remove all HF...
 
#63
Hi Bernard

Its imo better to use a filter wich has best impulse response. Even not removing all rubbisch, the end outcome will be better listenable i think.
I redesigned my speaker filters to 12dB series Bessel, lot of work, but great result. 6dB would also be possible but my units were not suitable for 6dB.

Ecdesigns has showed in his 1541 thread the difference between Butterwoth, Bessel and another filtertype. See if i can find it.

Oh here it is: nice step comparing of different filters on his drawing below.

http://www.diyaudio.com/forums/attachment.php?s=&postid=916328&stamp=1147687163
 
#65
AN207 problem.

Yesterday finished i2s to parallel converter according to AN207.
Used chips 74ACT00 and 74ACT74. Its kinda working🙂 but with awful distortions.
Source was PCM2707 (tested, 100% working) , DAC was AD1851 (new , not tested).
When I move finger near 74ACT74, distortion level rises also cant set volume on PC, its dead silent or loud.
 
#67
I use ground planes on both sides of PCB. One power supply for logic and second for AD1851 digital part.
Also tried simulate using pspice9, there was also problem with latch signal, it was unpredictable (2 red lines, not logic 0 or 1).
 
#74
I have "borrowed" this from unknown japanese website...
don't know if it helps?

Arne K
 

Attachments

  • image23.jpg
    image23.jpg
    91.1 KB · Views: 925
#77
tubee said:
Hi Arne K

In my last I2S - PCM design i shifted Data for one channel also 16 bit, in your "borrow" done also but then 20 bit for PCM63.
Click to expand...

All this 20 bit stuff is a red herring, a mere distraction. You know where you are and you know where you need to be. Take it step by step and it will be clear what you need to do.
 
Status
Not open for further replies.
Top Bottom