There are various ways to avoid a coupling cap, but if you use a good quality cap of the correct size you won't hear any distortion from it. Also, if you had used tubes, for sure you would have added tube distortion and coupling caps too.
Excellent results can be obtained using circuits recommended by ESS, there is no reason to ignore them: http://www.esstech.com/files/4514/4095/4306/Application_Note_Component_Selection_and_PCB_Layout.pdf
http://www.esstech.com/files/5114/4095/4310/Sabre_8_2Channel_64PIN_V3_SCH.pdf
Excellent results can be obtained using circuits recommended by ESS, there is no reason to ignore them: http://www.esstech.com/files/4514/4095/4306/Application_Note_Component_Selection_and_PCB_Layout.pdf
http://www.esstech.com/files/5114/4095/4310/Sabre_8_2Channel_64PIN_V3_SCH.pdf
A useful device, perhaps, for helping to better visualize your DAC's microcontroller interaction. Turns out some DACs (and many other ICs, for example, SRC4392, are controlled via I2S bus or sometimes maybe SPI bus. Found a ridiculously cheap little analyzer that can capture low-ish speed bus traffic for analysis. Some freeware and or other software helps simplify things a lot. The thing and some glowing reviews can be found here: USB Logic Analyzer Device Set USB Cable 24MHz 8CH 24M Hz 8 Channel for ARM FPGA M100 SCM: Amazon.com: Industrial & Scientific Just tried one myself and am surprised at how cool and effective this thing is for DIY DAC project applications.
Can anybody explain, why a mono-polar electrolytic cap should be a good option for an AC application here? I guess any (big enough) film cap (bi-polar) should do a better Job here, no?
mono poles just work fine in "one direction" in such application to my understanding..
I used a variation of IV stages followed by a differential stage.
May I suggest if you are considering modifying one of these DACs, it may be helpful read the following post and the earlier posts it references: http://www.diyaudio.com/forums/digital-line-level/314935-es9038q2m-board-55.html#post5395485
It turns out there are some other things besides caps and the ESS publically accessible downloads to be aware of for best sound quality results. If some care is take it is possible to produce a DAC better than most or possibly all on the market under $1k, IMHO. But it is fair amount of work to make it happen.
Also, it can help a lot to monitor your DAC progress if you have a very accurate headphone amp and quality headphones, or another accurate reproduction system. I could provide some information on my findings to date in that area and offer some suggestions if you would like.
I can report on using a R (4R) on the raw output of the DAC. This is preliminary finding only, our THD+N went down -3db. We are doing more tests but it seems that this mod helps with a lower quality board .
Hmmm, if a little resistance is affecting distortion compensation it might suggest different resistor values might be optimal for each of the two outputs for each channel. That is, a small adjustment to something like gain or offset of one of the two outputs might better balance distortion cancellation when the outputs are differentially summed.