A NOS 192/24 DAC with the PCM1794 (and WaveIO USB input)

Until recently I used an old red board 2 deck DDDAC in my workshop-no Tent shunts, no bias module, and simple single cap output to my preamp then 845 monoblocks. The sound was stunning and I never once thought about upgrading it.
Of course, my pimped lounge model with Tent shunts, the bias module, separate digital and analogue power supplies and tube output stage is better. But there's a lot of hours and pounds in between them.😊
 
  • Like
Reactions: 1 user
Sometimes it is fun to do your best pimp out effort even if the improvement could hardly justify the effort. And sometimes it's also just great to say this thing likely could be better but it is just fine the way it is. I think that is half the fun of DIY systems. You can invest in incremental upgrades if and only if you feel like it. I can't imagine the frustration of constantly chasing perfection by buying the latest bling.
 
  • Like
Reactions: 1 user
Kartick, I think this is for two reasons (no good answers)
1. It is not used a lot
2. there is no industry standard for it

It might be the ian canad works with others, but hey no one knows for sue? just try
I did this using an Ian Canada board. The i2s pin connections matched those used by Gustard and others. SQ was slightly worse than USB from Innuos Zen Mini Mk3 pimped with Phoenix USB box so I pulled out from this experiment. Best SQ so far is direct i2s from a built in Pi 4 with Ian Canada fifipi. I'm just waiting for my Q7 board to arrive which I believe is a further improvement. As I've also got some very sensitive horns now I'm also trying Sowter transformers to reduce the noise floor
 
  • Like
Reactions: 1 user
I have been using a KALI FIFO/Reclocker with my WaveIO and DDDAC PCM1794 for about 6 months using custom interface PCB's.
My DAC sounds more analog than ever. The Kali has been 100% stable with this installation.

I first referred to posts #6891 and 6911 for wire connections from Waveio->Kali->DDDAC and created a rat’s nest of wiring which worked.
Next I designed circuit boards for each side of the Kali to neaten up and simplify the connections.
It took about 300 hours of music play time for everything to fully break in.
 

Attachments

  • Kali PCBs.jpg
    Kali PCBs.jpg
    144.1 KB · Views: 166
  • Kali View from DDDAC.jpg
    Kali View from DDDAC.jpg
    200.5 KB · Views: 166
  • Kali Top.jpg
    Kali Top.jpg
    212 KB · Views: 170
Disabled Account
Joined 2002
Hello,
Of course this can be true but why can all those witty guys not make something that will work with a cable length that is a bit more comfortable for the majority of us?
A few months ago i kind of stopped this digital project because i ended up with to many boards interconnected with these pesky cables.
Some decades ago people were happy saying bye bye to their record players where everything has to be perfect to achieve good sound ( according to them)and now we ended up with something even more demanding.
Just found a new secondhand shop where they have some nice Decca/Telefunken/ Deutsche grammophon records for 75 eurocents but did not have my reading glasses but they looked pretty interesting.
greetings, eduard
 
I'm trying to run Q7 with Roopie and a DDDAC. What I'm getting is soft music against a background of static.
Is this with the old red DDDAC board? This has passive I2S delay, which is a bit picky about the impedances seen at the I2S input. I had the same symptom, and it went away once I moved to the newer blue board with the active I2S circuitry.
 
So having tested my previous DDDAC against the Mola Mola Tambaqui, dCS Bartok, MSB Discrete and Premier, Rockna Wavedream Signature DACs and finding that non could best it, I decided to build my own signature DDDAC. https://adaudio.wordpress.com/2022/...-mola-tambaqui-and-msb-discrete-and-prestige/

As I have recently moved to 100db sensitive Klipsch Cornwall speakers I decided to use the Sowter transformers as output devices. So the spec of my latest DAC is:
-Sowter output transformers

-Ian Canada FiFOPi Q7

-Ian Canada Supercapcitor Pi psu

-RPI4

-5 separate power supplies using New Class D regulators

-Choke input psu for the DAC analogue side

-Choke psu for the DAC digital side

-Organic polymer caps on the digital side both for decoupling and small bypass caps

-New Class D regs for the DAC boards, main digital regulator and the two relating to the shift register and Tent module

-Elna Silmic caps for the analogue side

And how does all this sound? I can honestly say the DAC has exceed my wildest expectations. There is absolutely no digital artefact and reverb and bass response are right up there with the finest analogue devices. I’m particularly surprised by the SPDIF input. This has always seemed the poor relation of my other builds but with these latest mods it’s sound is well up there with the Pi streamer sound.



As well as my DAC shoot out above, I listened to many expensive streamers-and tried a few via HDMI i2s but none has equalled the Pi4 with the Q7 and decent power supplies. This DAC can now compete or beat the very best up to many thousands of pounds in price. The MSB Prestige above is £29.5k alone.
 
  • Like
Reactions: 4 users
With the capacitors (Jantzen Silver Z 4,7uF x 4 pieces) it sounded somehow calmer, less dynamic compared to the transformers. And my custom transformers are not in the range of Sowters. And so the transformers remained inside, and the capacitors went into the speakers.
 
  • Like
Reactions: 1 user
Are you guys talking about caps in full balanced mode? If you are talking about caps in single ended output from the DAC chips, I can easily see why you prefer a transformer from balanced out. Howwver, its a totally different story with caps in fully balanced setup, see discussion a few posts back.

If you have noise problems that go away when you Insert the transformers in place of the caps, you have a problem elsewhere. I'd look for ground loops.
 
I had no problems with noise either with the capacitors or the transformer. Only the transformer sounds better to me, I have both SE and BAL coming out of that transformer and that's where I'm done with it. A ground loop cannot appear there either, it is completely galvanically separated from DDDAC.

I don't have a balanced system by the way, so I haven't tried a balanced connection. In the meantime, I made a small amplifier with balanced inputs, as soon as I make a suitable cable I will try to connect the DDDAC direct to that amplifier in both SE and BAL variants, I am really interested to see if there is a difference. Of course I have to get the same cable for that test first (VDH D102 MK3), it wouldn't be right with SE and BAL cable to be different.
 
Last edited: