Experience with this DIY DAC ?

it's not a problem that you don't have spdif transformer on board,one situated at the transport side it is enough .
I see that the vertical upsampler board is missing (i think you can order it separatelly and place it instead of those 5 jumpers)
but i see on board some nice silver mica caps (6 of them - 2 near spdif,4 near opamps),also some handy feritte beads near the dac for power supply noise rejection...i also see variable adjusting for opamp power supply,so i think it can be adjusted to 15V instead of 12V giving more headroom
Although i don't reccomend to use it on USB input because the PC's ground it's not galvanically isolated through this connection bringing you to the dac all the HF junk (and anyway USB protocol is far far away from being a quality "audiophille" music source - better hang with optical output which is galvanically isolating
the PC from the external DAC thus not affecting it's performance)
Beside all these,a lot of modifications are waiting for you on the board if you want to make it sound pretty good compared to existing (expensive) Hi-Fi CD-Player implementations
 
Funny, my experience tels me the silver mica are the worst possible capacitors in analog filter stage, and the isolation transformer at both ends are crucial for half-decent sound with s/pdif. In fact, good quality isolation transformer followed by coax-to-TTL section makes those receivers sing! I learned this from THE man himself....

The s/pdif interface & input receiver is a section on this DAC that needs a lot of attention.

Boky
 
pls google silver mica propreties and why are they so expensive and good in decoupling
My experience tells me that 40% of sound is given by a good transport with separate powered low-jitter masterclock,reclocked spdif signal (derived from the same masterclock) and finally an spdif buffer for the signal not to be so affected and dependent on the cable quality
10% is given by the interconect cables between transport and DAC (< 20cm,good quality BNC interconnects)
20% is given by the SPDIF receiver chip implementation
and only the last 30% is given by the DAC itself (considering already a good implementation and low noise power supplies)
 
Franz Gysi said:
Hello

Thanks for the hints. I will check the USB port and I am aware of the restrictions about USB for audio.

And the upsampling board is included, but was not attached on my foto.


As soon as you will try the upsampling board, could you describe the sound this dac has with it and if you consider the sound better with the board in or without this board.

There's mixed opinions about upsampling or not

When I look at this new board, i see 3 rectified sections instead of two as there's in the older version.

I suppose that the upsampling board or the USB board has a separate supply.
 
luxury54 said:
regarding the rectifier and the old version of the DAC,it's worth adding some 22nF-47nF film capacitors in the empty places beside the diodes in parallel with each one of them to surpress their switching noise...

I made a comparaison between one dac with paralleled diodes with caps and another one without.

I didn't heard a great difference between the two dacs.

The diodes in these dacs are BYV types which I think have low trr value and are soft recovery types. At least, they are good diodes.
 
As soon as you will try the upsampling board, could you describe the sound this dac has with it and if you consider the sound better with the board in or without this board.

Yes, I will do it again.

My first try yesterday with the upsampling board made me immediatly thinking, there is some error. Timing error or lost data?

The second try was different, I did not recognise an error.

I will try a little more for and back this weekend.

Franz
 
It seems, pin 20 from the CS8412 is on 3.3V level, so this modification is not needed on this version.

Nice!

After removing the first four coupling caps, I switched two times for and back, changing the configuration with or without upsampling board.

I prefer without, the sound has some more punch.

Franz
 
Franz Gysi said:
It seems, pin 20 from the CS8412 is on 3.3V level, so this modification is not needed on this version.

Nice!

After removing the first four coupling caps, I switched two times for and back, changing the configuration with or without upsampling board.

I prefer without, the sound has some more punch.

Franz

Do they have the right PLL R anc C values on the CS8416 ?
 
After some for and back tests, I strongly feel, that it is not easy to decide what is better, upsampling or not.

As already mentionned here, a A/B comparison should be possible, but with just one device it is nearby impossible to compare seriously.

At the moment I attached my (slightly modified) Zhaolu D3 DAC to the TG-Link Sndscape Cinema! card and it is still the unbeaten solution in my home.

Maybe because TG-Link is so much better than SPDIF? Or maybe the analog stage from the Zhalou is better (based on OPA627 and in my case OPA2134 as line driver).

I surely will not use the analog part of the chinese DAC, I will attach a tube stage, possibly connected with transformers.

Franz
 
I have the idea, to make a testsetup with this parts:

An externally hosted image should be here but it was not working when we last tested it.


6111 double triode and Studer input trannies.

But first I will store this project for some days/weeks and will work on the showed Studer amplifier.

Kind regards
Franz

P.S.
Luxury54: two weeks ago, I was in Bucuresti, setting up some systems for the new studios for Radio Kiss FM, Magic FM and One FM. Next time, should I have a chance to come again, I will contact you! Or are you far away from Bucuresti?
 
Franz Gysi said:


Sorry, wich R and C and what are the right values?

I just looked through the thread but could not find anything.

Thanks
Franz

/Edit: just found the values in the datasheet. Just searching on the board now

/second Edit: Yes, correct values: Rflt 1k; Cflt 220nF; Crip 10nF


These are the old values

The correct ones are

EFLT=3K
CFLT = 22 nF
CRIP = 1 nF
 
Franz Gysi said:
I have the idea, to make a testsetup with this parts:

An externally hosted image should be here but it was not working when we last tested it.


6111 double triode and Studer input trannies.

But first I will store this project for some days/weeks and will work on the showed Studer amplifier.

Kind regards
Franz

P.S.
Luxury54: two weeks ago, I was in Bucuresti, setting up some systems for the new studios for Radio Kiss FM, Magic FM and One FM. Next time, should I have a chance to come again, I will contact you! Or are you far away from Bucuresti?


Are they 1 : 1 transformers ?

I would try them without tube. My dac with Lundahl 1 : 1 transformers has enough gain.