[denjo]

This DAC seems to be getting quite good reviews lately. Has anyone tried this DAC yet?

http://www.mood-lab.com/dice.htm

I like the shunt regulated PS! (Paul Hynes, are you reading this?)
And the Black Gates ...
And the computer USB interface ...

[Paul Hynes]

Hi Denjo,

Looks like a similar implementation to my own non-commercial 1543 DAC with shunt regulators on everything. Must check my office for bugs.

Seriously though, the performance of power supply regulators, and the way the power supply is implemented, can affect the audio performance of equipment radically, whether it uses digital or analogue circuitry.

The following is not a criticism of the DAC you mentioned, just a general look at shunt regulation. What I am clarifying here is that the term “shunt regulator”, used on it’s own in technical descriptions, means very little, even though it’s become an underground buzz statement.

There are just as many ways of designing circuitry for shunt regulators as there are for series regulators and, for that matter, any other electronic item. This accounts for the vast and bewildering array of product available on the market and the different levels of performance of these products. I have a file in my R&D cabinet that contains a large number of possible topologies for shunt regulator trials dating back to the early eighties. Some are good, some are not so good, some are not worth considering and one or two are exceptional.

A shunt regulator with current source drive is likely to have better supply line rejection than typical three terminal regulators, which is worthwhile, as it would reduce supply line interaction. However, under load, it’s output impedance, transient response and bandwidth may be no better than typical three terminal regulators, unless it was designed specifically with this in mind.

Regards
Paul

[denjo]

Hi Paul

Thanks for chiming in! Indeed, design and implementation are critical factors determining how good a shunt regulator is. Your non-commercial 1543 DAC with shunt regulators on everything sounds very promising. Would you like to share this DAC with other AudioCirclers? Have you compared it with the Altmann?

Best Regards
Dennis

[Paul Hynes]

Hi Dennis,

There is not a lot to share really. The DAC is a work in progress and still very basic in implementation. There is no original thinking in the DAC architecture. I’m just following the path trodden by many other DIY enthusiasts. It’s not strictly speaking the same as the Moodlab Dice application as I haven’t yet added USB capability, although it does use the same digital receiver chip and dac chip. Of course, I have high performance shunt regulators on every chip power feed. They are the same regulator circuit used in the power supplies I supplied to you, for your Squeezebox and Altman DAC. They are resized to directly replace standard three terminal regulators easily in low current applications like local regulation.

I’ve kept the whole thing as simple as possible. Currently, it’s basically a re-clocked TDA1543 with a digital receiver so I can use it with the digital sources that I have at present. I will be investigating stacking the TDA1543 chips next and I also have some TDA1541A chips to play with when I get the time. I will be implementing the USB to I2S function when I have finished the construction of a silent computer with a large hard drive array, which is currently in the design phase. This will allow me to bypass the CS8412 digital receiver. This project is not for any of my consultancy clients, just for my own system.

I haven’t compared the DAC with any other makes at this point in time so I can’t tell you how it compares. However, I did compare regulators on the TDA1543 so I could truthfully tell people what to expect from upgrading the regulators on this chip. I fitted a 7808 three terminal regulator first and was less than impressed even after plenty of burn-in time. When I applied an 8 volt high performance shunt regulator, there were some quite radical improvements, even before burn-in, in all the musically important areas like tempo, timbre, dynamic scale, low-level information portrayal, image stability and there was much less clutter in the sound field. Burn-in has improved things further. Clearly, you can’t expect the full performance from electronic systems unless you deal with the power supplies first.

I am hopeful of further improvements with chip stacking but I will have to fit a larger, heavier duty regulator before implementing this as each TDA 1543 can draw up to 60ma on an eight volt supply. I may just set the TDA1541A up first to see how that behaves. Once I have finalised the design I will have a Teflon circuit board made to "guild the lily" so to speak.

Out of interest, the Altman DAC is specified to run from a 12 volt master supply. There should be one or more regulator chips on the board to drop the 12 volts to the voltage required by the various digital chips. Unfortunately I don’t have an Altman DAC to verify this. If there are standard three terminal industrial regulator chips on board, they are ripe for upgrade. It would be very interesting if anyone who owns one could verify this.

Regards
Paul