Neil McCauley
27-10-2015, 18:13
Hi Scott. First things first ¬ are you a DAC chip maker and if so are your chips made to your specification i.e. proprietary – or do you buy third-party off the shelf so to speak?
I use third-party / off the shelf New Old Stock DAC Chips. The Computer Audio Design 1543 DAC uses, well TDA1543/N2 DAC Chips – 16 of them actually.
Is microphony, (currently a big issue in the domestic audio world) a fact, or merely conjecture – or perhaps a bit of both?
I’m a big believer in “microphonics” reducing sound quality. In fact it’s the main reason for the case material of the 1543 DAC. The entire case is made from laser cut 10mm thick acrylic. During prototype work I made a similar case from aluminium. I put the exact same parts in both cases and the difference in sound quality was amazing. Acrylic has a resonant frequency different from aluminium and I find that the high frequency dampening effects of acrylic are superior to more traditional materials. There is also the fact that acrylic is completely non-magnetic – but that’s a different issue.
The 1534 DAC uses four hockey puck sized feet. These feet are made from a very soft polymer that forms a honeycomb design. I get a lot of comments about the feet. I tried a lot of different materials and designs…
To my ears these polymer feet were vastly better at damping out vibration than anything else I could find.
The PCB that contains the TDA5143 DAC chips is also isolated. I found that the DAC chips were very susceptible to vibration, or “microphonics”.
On the assumption that the primary objective of your designs are that it should neither add nor detract from the original signal, how can you determine that this goal is being achieved at the analogue stage? Or put differently, assuming that any analogue circuit must surely add it’s own sonic signature, however small, how is the extent of that signature determined?
I wish I had a tool to tell me that a modification improved or decreased sound quality. How easy that would that be! I think most audio measurements tell us very little. Some of the best sounding audio gear I have ever heard doesn’t measure all that well. Take a look at the specifications of the Nelson Pass First Watt SIT-1 mono block amplifiers. The distortion measurements are extremely poor! It’s only 10 watts! But if you ever get the opportunity to hear these amps with a good high efficiency speaker – wow!
The only way I have found to really improve my designs is by listening.
Back to the question, I try to put as few parts in the digital, analogue and power path as necessary. Many designers use operational amplifiers, transistors, transformers or tubes in the output stage of their DACs. These technologies all have their advantages and disadvantages. Over the years I have tried them all.
I like a current output DAC chip because I can have an output stage that doesn’t use any active parts. If someone wants a “tube” sound then use a tube preamp or amplifier. To my ears I find that I get a much more natural realistic sound using extremely high quality passive components for the output stage.
Currently, what is the limiting factor on the performance of a DAC? Is it the chip, the power supply or should I mind my own d**n business?
Without a doubt power supply. Everybody talks about DAC chips. The DAC chip is crucial for sure, and I’m personally a fan of R-2R non oversampling architecture. But the power supplies are at least as important as the DAC chip choice itself.
Please click http://www.hifianswers.com/2015/10/howard-popeck-interviews-scott-berry-of-cad-computer-audio-design-part-1-of-3/ to continue reading
I use third-party / off the shelf New Old Stock DAC Chips. The Computer Audio Design 1543 DAC uses, well TDA1543/N2 DAC Chips – 16 of them actually.
Is microphony, (currently a big issue in the domestic audio world) a fact, or merely conjecture – or perhaps a bit of both?
I’m a big believer in “microphonics” reducing sound quality. In fact it’s the main reason for the case material of the 1543 DAC. The entire case is made from laser cut 10mm thick acrylic. During prototype work I made a similar case from aluminium. I put the exact same parts in both cases and the difference in sound quality was amazing. Acrylic has a resonant frequency different from aluminium and I find that the high frequency dampening effects of acrylic are superior to more traditional materials. There is also the fact that acrylic is completely non-magnetic – but that’s a different issue.
The 1534 DAC uses four hockey puck sized feet. These feet are made from a very soft polymer that forms a honeycomb design. I get a lot of comments about the feet. I tried a lot of different materials and designs…
To my ears these polymer feet were vastly better at damping out vibration than anything else I could find.
The PCB that contains the TDA5143 DAC chips is also isolated. I found that the DAC chips were very susceptible to vibration, or “microphonics”.
On the assumption that the primary objective of your designs are that it should neither add nor detract from the original signal, how can you determine that this goal is being achieved at the analogue stage? Or put differently, assuming that any analogue circuit must surely add it’s own sonic signature, however small, how is the extent of that signature determined?
I wish I had a tool to tell me that a modification improved or decreased sound quality. How easy that would that be! I think most audio measurements tell us very little. Some of the best sounding audio gear I have ever heard doesn’t measure all that well. Take a look at the specifications of the Nelson Pass First Watt SIT-1 mono block amplifiers. The distortion measurements are extremely poor! It’s only 10 watts! But if you ever get the opportunity to hear these amps with a good high efficiency speaker – wow!
The only way I have found to really improve my designs is by listening.
Back to the question, I try to put as few parts in the digital, analogue and power path as necessary. Many designers use operational amplifiers, transistors, transformers or tubes in the output stage of their DACs. These technologies all have their advantages and disadvantages. Over the years I have tried them all.
I like a current output DAC chip because I can have an output stage that doesn’t use any active parts. If someone wants a “tube” sound then use a tube preamp or amplifier. To my ears I find that I get a much more natural realistic sound using extremely high quality passive components for the output stage.
Currently, what is the limiting factor on the performance of a DAC? Is it the chip, the power supply or should I mind my own d**n business?
Without a doubt power supply. Everybody talks about DAC chips. The DAC chip is crucial for sure, and I’m personally a fan of R-2R non oversampling architecture. But the power supplies are at least as important as the DAC chip choice itself.
Please click http://www.hifianswers.com/2015/10/howard-popeck-interviews-scott-berry-of-cad-computer-audio-design-part-1-of-3/ to continue reading