Initial Experiments with the PCM1794 and asynchronous USB

The middle step on the journey to the DDDAC1794 NOS DAC

Of course I did not want to jump in just like that.... It was important for me to understand how the sound print would be from a PCM1794 based DAC, getting its data from an asynchronous USB receiver. Therefore I wanted to keep things simple and I ordered two ready available boards to experiment with. For the USB receiver, I picked the HiFace module, as it was easy to get and not so expensive. For similar reasons I choose for the small PCM1794 DAC board from Twisted Pear, as it has the opportunity to easily do passive I/V conversion and the opportunity to do some selections on the operation modes for the PCM1794. After all my experiments with this, the passive I/V is an absolute MUST HAVE (there are for sure valid reasons to use active I/V, but it never sounded as good as a WELL implemented ( !!! ) passive I/V. I started listening and experimented around in the Lab at home. Using LifePo Batteries to feed analog sections for example showed still the importance of clean power supply and after listening in I was kind of happy with the results, specially on HD music.. Just then, an excellent opportunity offered itself. We would do another audio weekend at the GOTO Horn lovers' sites ;-) We had a larger group of visitors coming this time, including Jean Hiraga, with whom we had sessions before, so this would be great to do some initial listening tests.

I prepared everything so we could do demos and listening tests. The most important tool was the Music, so it was great that LINN Music offers superb audiophile recordings in all three formats.... So I downloaded the album " The Super Audio Collection Volume 5 Sampler" in all three formats (44.1/16, 96/24 and 192/24).... This would allow us to really compare the difference between the formats and the performance of the Hardware. We were ready to rock!
Here is the link where you can download this Album :-)

How first tests look like :-) At the top left you see the HiFace USB Module... In the middle / left, the twisted Pear DAC Board, with my Favorite Mundorf Silver Gold Oil Couple Capacitors. At the right a simple power supply to feed the digital section. In the plastics bags (for safety ;-) the A123 Batteries for the analog sections. A small disadvantage of this setup was the fact that the maxim output voltage was only 0,5Volt. Definitively not what we look for as final solution. nevertheless, amazing how a bunch of boards and stuff can already sound so good :-) PS: small peek under the table shows the discrete regulated power supply design from Bernd. This was made for the DDDAC1543MK2, but was used by me as basic design principle for the later 1794 power supply designs...

Still looking VERY (typical?) DIY ;-) The good old way of putting things on a board and wire it up, still works for me; easy to change things and experiment around (!!)

Listening impressions It was great to have so many experienced Ears together :-) The fact that we could easily compare between formats made things clear very quickly. In the end of the PCM1794 session (we had many other things to experience...), we agreed it is awesome to listen to HD music.... It is breathtaking to hear the true texture of instruments and human voices at 192kHz compared to 44.1 material. Pianos are great, violins and bass start to live... BUT, what we also experienced is that this was not so strong at 96kHz and at 44.1 it sounded like a very normal CD player. This could not be caused by sample frequency only. And comparing 44.1/16 soundtracks with the DDDAC1543mk2, it became apparent, that it HAD TO DO with the digital filter, from which I already found out in the past, that the negative effects are most prominent if the Fs is close to the audio range (so 44.1kHz.....) I already knew what to do ;-)

Theory discussions: At Bernd's place we had a great session as well. Here we (Bernd, me, Jean Hiraga and Andre) are discussing the principle of paralleling DAC chips

First conclusions

Our first conclusions were VERY straight forward:

• 192kHz soundtracks are awesome! The low level detail is stunning
• There is no real need to use 96kHz or 88kHz if you can have 192kHz
• Red book music (44.1kHz) is pointless to play on a DAC with digital filtering (NOS beats everything)
• Power supplies must be clean
• Asynchronous USB Interface is the way to go
• Not directly mentioned, but Foobar with WASAPI drivers is a MUST (what a driver can do....unbelievable....)
• Paralleling chips works.... (other session)
• ERGO: A 1794 DAC with no digital filtering to be able to play ALL formats at highest possible level was the challenge
• Last but not least, we need higher output voltage if possible

Keep reading ...... the story continuous with designing the circuit ( Design Phase )