Hurtig said:I see that a lot of users are intersted in the schematic. Currently I am still working on these.
But I don't really see why they are so interesting
As I have mentioned several times, it is not a matter of using the right schematic or the best components. It's "simply" a matter of "matching" the schematic, components and real-life implemnetation.
Change one transistor to a different type, and you may end up with a crappy DAC. Change the PCB layout slightly, and you may see the same result...
You should consider the art of building audio gear, as building a violin I have not yet seen the cookbook of building the perfect Stradivarius. Simply because it is pure craftmanship.
I am still waiting for you to share the actual PCB layout or analog stage or low noise voltage regulator design..... Your craftsmanship suggestions regarding Stradivarius are appreciated....
Now, are you going actually to show us something and help us better understand your brilliance, or you are still doing free advertising and waiting for the number of views and replys to keep increasing?
Boky
analog_sa said:
Not a Bryston fan by a long shot but your comparison is unfair. Bryston use some proven antijitter techniques which make the marriage of a mediocre receiver and dac tolerable. Do you know any of the "people swapping $5000 DAC's in preference of the BDA-1"?
Actually you are talking about one of the most important matters in digital design.
We discussed jitter reduction and upsampling a lot on our way through our project. But very soon we found out that state of the art powersupplies will give you a lot more performance pr invested $ and hour.
Som years ago i saw an Accuphase SACD player, I think it was called DP77 or so. It had several BB DACS working in parallel, it had completely separated digital fron analog by the means of high speed optocouplers and it actually seemed very nice. That until I saw the analog stage and the supplies for the DAC chips.
Supplies were just standard 3 pin regulators as lazy as they get, and the analog stage was composed og 2 pcs. of 4580DD JRC and 4pcs of 2114D JRC and that was that.
A lot of effort went in to make this machine a state of the art digital product, and then all the subtle results of refining the digitals has to run through 6 cheap opamps with noisy supplies.
We never understood this way of engineering, but we think it must have its roots in something different than sound itself.
Actually it is a very prestigeous job to design complicatet digital circuits, and from a technical point of view any op-amp has better specs than any discrete NFB designs can manage, so soundquality has to be considered to do otherwise than Accuphase did.
If you do consider sound quality, you pretty fast realize that jittereduction, upsampling, High end cloks and so on, are not major issues in chasing natural sound. You first of all simply need a whole through into the digitals to be able to evaluate them, this whole goes inevitably through the analog design, and that´s the viewpoint we chose to look at a complete DAC from.
That could have resulted in close down of the project, as there was an actual risc that digital sound simply was not good enough for that kind of aproach. We consider ourselves lucky that it did not turn out that way.
.Thise empirical information leads us to another surprising conclusion, namely that "digital sound" actually does not derive from the digital components, it´s in the analogs.
The "digital sound" often referred to mostly has its origin in those standard opamp based output stges and filters - funny ah!
analog_sa said:
Not a Bryston fan by a long shot but your comparison is unfair. Bryston use some proven antijitter techniques which make the marriage of a mediocre receiver and dac tolerable. Do you know any of the "people swapping $5000 DAC's in preference of the BDA-1"?
Be aware of storytelling in this business: I do not know the Bryston, but I have heard about sophisticated and patented anti jitter solutions before. Like the Bencmark DAC1.... Guess they call it "Ultra-Lock". Analog Devices call it AD1896
Still... People are ONLY focusing on the technical parts... We have a different approach. Sound is in first place. And if we win by not using som High Tech solution, we will happy knowing it.
I mean: Yeah, as an engineer you may think that anti jitter circuits will gain performance. But if in real life it degrades.... What should you then choose?? Again, I would go for the sound.
Designers going for the tech solution, would normally also tell you that a low THD op-amp will be better than an non feedback discrete analog stage. But I honestly don't believe.
Hurtig said:
Be aware of storytelling in this business: I do not know the Bryston, but I have heard about sophisticated and patented anti jitter solutions before. Like the Bencmark DAC1.... Guess they call it "Ultra-Lock". Analog Devices call it AD1896
Still... People are ONLY focusing on the technical parts... We have a different approach. Sound is in first place. And if we win by not using som High Tech solution, we will happy knowing it.
I mean: Yeah, as an engineer you may think that anti jitter circuits will gain performance. But if in real life it degrades.... What should you then choose?? Again, I would go for the sound.
Designers going for the tech solution, would normally also tell you that a low THD op-amp will be better than an non feedback discrete analog stage. But I honestly don't believe.
Hehe the best dac I have ever heard was with the 1896 .
Hurtig said:
Be aware of storytelling in this business: I do not know the Bryston, but I have heard about sophisticated and patented anti jitter solutions before. Like the Bencmark DAC1.... Guess they call it "Ultra-Lock". Analog Devices call it AD1896
Yeah, as an engineer you may think that anti jitter circuits will gain performance. But if in real life it degrades.... What should you then choose?? Again, I would go for the sound.
Wow, you must have some amazing insider info.
Do you seriously think you are addressing a congregation of morons? Anti jitter circuits degrade performance? I/V convertors are bad for sound? There are bad examples of everything but to make such sweeping statements is ridiculous.
analog_sa said:
Well... If you open a DAC1, it doesn't take an rocket scientist to see how the anti jitter circuit in a DAC1 is made. It's simply a AD1896 and a crystal-oscillator. But I agree... "Ultra Lock" does sound more expensive
I'm not sure ASRC's are that bad. I just havn't heard a succesfull implementation yet.
Hurtig said:
Well... If you open a DAC1, it doesn't take an rocket scientist to see how the anti jitter circuit in a DAC1 is made. It's simply a AD1896 and a crystal-oscillator. But I agree... "Ultra Lock" does sound more expensive
Apparently my sarcasm was wasted.
Most participants in this thread are well aware of what is in the DAC1. Why do you keep mentioning it? It is a nice mid-fi dac but certainly built to a price and compromised in many ways. Using an ASRC is a slightly contentious way to suppress jitter and certainly not the only one.
Per Nielsen said:
We both do understand that the scematics are of interest to some of the readers in this thread, but as already mentioned, they do not yet exist. Further more Hurtig is now working on a PCB for the transformers, which also is not finished yet.
We do have some old scematics containing the original design, but they do employ completely different powersupplies and in addition the sound was very different as well.
That exactly is the reason why that design - which wasn´t the first at all - was abandoned.
What to do next with the complete DAC design looks like a paradox.
We could give away the PCB layout files and a parts list, and then everybody could make their own copy of the board and mount it up with preferred components - some might even chase improvements in digital domain.
That we do know will not work the way we want it at all, both because different components will make different sound, and the chance of having it producing the desired output in the end is very small.
It is a complicated and compact design wich possibly would produce massive amounts of failure, thus giving it a bad reputation.
To underline this matter I can state a few facts from devellopment experience.
Hurtig is in fact a very good engineer, everything works the way it is supposed to do in every single matter. But in one case we burned a DAC chip, just because it was connected to a ring core transformer. We never found any logical explanation to that incident, but a ring core transformer was enough to make the whole thing fail.
Both earlier and since then, we decided to use the transformers giving the best isolation, which is definately not ring core design, but at the time it happened, we just did not have the transformer required as U/I core.
So this is what the business is about. We can´t even safely leave the choise of transformers to anyone else, without them getting into trouble as we did.
Therefore we are considering the possibilities for some premounted PCB´s to DIYérs instead.
No promises given yet though!
Be patient please! We are trying to work it out!
If somebody should think this is just "big business" they ought to try this out
.A few years of considerations and experiments and then your preferred component reaches EOL
Good luck if you try
Relax...
As I see it the whole DAC concept is not worth the trouble trying to squeeze out the schematic from these people.
If they do not want to, leave them.
There are many other DACs that you can built.
And there are lots of people around who believe they were genius, they are just one of them.
As I see it the whole DAC concept is not worth the trouble trying to squeeze out the schematic from these people.
If they do not want to, leave them.
There are many other DACs that you can built.
And there are lots of people around who believe they were genius, they are just one of them.
Clearly, the strength of this project, if any, is not the design itself. A strong design should be robust and not so easy to make fail. It also seems insulting to the intelligence of the forum members to suggest that we could so easily "botch" the project. There are many people here who have years of experience with difficult projects, and surely those folks can decide for themselves from looking at a schematic or pcb design whether they want to spend their time and effort on such a project.
Bernhard said:Relax...
As I see it the whole DAC concept is not worth the trouble trying to squeeze out the schematic from these people.
If they do not want to, leave them.
There are many other DACs that you can built.
And there are lots of people around who believe they were genius, they are just one of them.
ezkcdude said:
My sentiments exactly.
1. don't care much 'bout schematics - but other members do and have asked for the same; it would be nice to share as this is what the forums are about, in particular if you joined us all in Feb 2009
2. The strengths of this project? Ground plane, regulation, parts optimisation.... MAYBE ... but can we be certain? NO, because there's no sharing... just marketing bla bla bla...
3. Remarks that many here would not understand ground topology and their intentions and brilliance are straight down offensive. If they spent 5 years developing low noise "back-bone" for their design does not mean others went thru the same LENGTHY process to achieve the same.
4. Nothing to learn, nothing to help with or suggest ....
Boky out.
ezkcdude said:
I just want to correct a bit...
The final DAC is NOT likely to fail! It is extreme robust.
In the first prototypes we did burn a couple of DAC-chips, but we do know why. Simply because we started out using absolutely no filtering. Therefor the DAC-chip was connected directly to the analog stage input transistor. And when we tried a toroid for only the analog stage, the start up time of this was to fast. This caused the input transistor to sink current from the DAC-chip before it was powered up. But no real problem, since we do not want to use toroids, since it will generally lower performance, due to the bad isolation compared to UI-cores.
Today we use a little filtering, and even with a toroid, there is no risk of burning the DAC-chip. Actually we have done hundreds of modifications on the DAC, and has seen zero fails! Due to the use of the 2 step regulators in all supply voltages, we have a very robust design (Even though this is not the reason why we use this topology).
Also please note, that I am used to designing electronics that MUST be fail safe!
I used work as a developer at Bang&Olufsen, which is probably the company with the higest production quality in this business (Like the sound performance or not
).Since then, I have been designing industrial electronics (mainly analog and mixed signal), for equipment running 24/7 and 365 days/year.
Therefor I believe to be able to design a fail safe audio-product
Extreme_Boky said:
I guess you really do not understand the idea off this design (an audio in general).
Ask an engineer to describe why an Stradivarius is better than other violin's?? He will be totally lost!!
The same goes on this DAC. We did not spend 5 years developing a low noise regulator or ground plane.
We did spend the 5 years, listening to hundreds of small changes to the design.
As an engineer I could design a "High End" DAC in less than a week, ready for production. But it will sound exactly as boring as most other High End DAC's, and you will ot get the feeling of music. Just like listening to a ML product. Nice looks, nice measurement, but you will fall a sleep listening to it.
The special thing about this DAC is, that we really did spend 5 years of listening to small changes. And this leads to a DAC, that will actually play music!
It's not a DAC for the tech-freaks, looking for those 0,00000000001% distortion op-amps. But when you do actually listen to it, you will see the light.
So why did we post this, if we do not share schematics?? Well... we might. But you will most likely not be able to get the performance. Change a single componen or the layout, and you will have a completly different DAC. Just like changing a single bit in a Stradivarius... Then it's ruined. Try the same in a cheap violin, and you will not even hear the difference. Just like using an op-amp. Then it all sounds the same.
We have thouhgt about making a kit of this DAC, with the most tricky parts already mounted. Maybe even a complete DAC PCB, so you only have to worry about the transformers, connectors and chassis. This way we can ensure that people will make it into smoke-machine because of components mounted badly.
- Status
- Not open for further replies.