My X-DAC 3.0 and X-DAC 3.0 Signature used the CS4328. These are out of production as Crystal's CS4328 is long gone. Too bad as it was quite the tasty part with goodies like +-5v analog supplies for the analog stage which could drive 600 ohm pro loads directly. CAL, Entec, and Timbre Technologies also had very nice DACs using CS4328. These date from the mid-1990s when the CS4328 was SOTA.
ACG's X-DAC 3.0 (historical info only, this product is NOT in production)
An externally hosted image should be here but it was not working when we last tested it.
ACG's X-DAC 3.0 (historical info only, this product is NOT in production)
California Audio Labs Alpha (original version, not 24 bit/96 kHz) and Meridian 263 are commercial production DACs using CS4328.
I have an Alpha that needed some power control repair, so I got to look over the PCB. CS4328 is buffered by a discrete op amp, first stage is single diff pair using VN0610 SS MOSFETs with simple bipolar current sink on sources and resistive load, second stage is current source loaded common emitter using MPSA05/55, output stage is 12AX7 with both halves in parallel as cathode follower with current sink load using bipolar IC op amp and MPSA05, DC servo using LM741 controls output DC voltage. IMHO the tube usage in this case is silly since the upstream SS dominates the sound, and the currents in the first stage wasn't balanced due to resistor values, tube rolling is probably of placebo value. If you have one of these and its original condition isn't sacred to you, I suggest replacing first stage with bipolar transistor pair with emitter resistors, and replace the tube with a suitable FET or bipolar darlington, remember that the tube uses +140 so connect drain or collector to +15. Also the follower current sink set resistor should be reduced, stock value gives about 2mA, I suggest 6 - 8mA. Sounds pretty clean with a DVD as digital source after mod, less grit and clearer soundstage but brighter/sharper treble.
If you insist on keeping the tube I suggest 12AV7/5965 with increased sink current, but more plate and filament power is needed, a resistor in series with filaments can be reduced to compensate but plate, filament regulator and power transformer run hotter and you may need to remove cover or drill holes in it. Sound is similar to transistor follower but a little less bright.
For digital in, Meridian 263 supposedly has PLL clock recovery, the Alpha is basic CS8412 with input buffer.
I have an Alpha that needed some power control repair, so I got to look over the PCB. CS4328 is buffered by a discrete op amp, first stage is single diff pair using VN0610 SS MOSFETs with simple bipolar current sink on sources and resistive load, second stage is current source loaded common emitter using MPSA05/55, output stage is 12AX7 with both halves in parallel as cathode follower with current sink load using bipolar IC op amp and MPSA05, DC servo using LM741 controls output DC voltage. IMHO the tube usage in this case is silly since the upstream SS dominates the sound, and the currents in the first stage wasn't balanced due to resistor values, tube rolling is probably of placebo value. If you have one of these and its original condition isn't sacred to you, I suggest replacing first stage with bipolar transistor pair with emitter resistors, and replace the tube with a suitable FET or bipolar darlington, remember that the tube uses +140 so connect drain or collector to +15. Also the follower current sink set resistor should be reduced, stock value gives about 2mA, I suggest 6 - 8mA. Sounds pretty clean with a DVD as digital source after mod, less grit and clearer soundstage but brighter/sharper treble.
If you insist on keeping the tube I suggest 12AV7/5965 with increased sink current, but more plate and filament power is needed, a resistor in series with filaments can be reduced to compensate but plate, filament regulator and power transformer run hotter and you may need to remove cover or drill holes in it. Sound is similar to transistor follower but a little less bright.
For digital in, Meridian 263 supposedly has PLL clock recovery, the Alpha is basic CS8412 with input buffer.
More CS4328
The old CS4328 DAC chip is an interesting one indeed. Would like to play with one some day to compare with my Cambridge Audio CD-3; but they are rare and tweaking time is at a premium.
To add to your list:
Meridian 206.18 CD Player
Rotel RDP-980 DAC
McCormack DA-1 DAC
These all use the CS4328 as far as I know. From what I understand, with a well executed power supply (this is crucial) results with this DAC chip can be outstanding.
Hope this helps
The old CS4328 DAC chip is an interesting one indeed. Would like to play with one some day to compare with my Cambridge Audio CD-3; but they are rare and tweaking time is at a premium.
To add to your list:
Meridian 206.18 CD Player
Rotel RDP-980 DAC
McCormack DA-1 DAC
These all use the CS4328 as far as I know. From what I understand, with a well executed power supply (this is crucial) results with this DAC chip can be outstanding.
Hope this helps
I have two prototypes with CS4328/CS8412 complete with a nice case for sale. Those two are to 99% ready. The third one is in duty with good results.
http://home.swipnet.se/~w-50719/hifi/dac
http://home.swipnet.se/~w-50719/hifi/dac
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.