ezkcdude said:
What will be the idea of distributing the schematics??
Anyway, you seem more like the op-amp-kind of guy...
Have you ever tried a non feedback discrete design??
It's one of the nice things about this forum, that people share their designs. Otherwise, as someone asked earlier, what is the point of this thread? We are not learning anything and you are not getting any useful feedback (no pun intended).Hurtig said:
ezkcdude said:
Fully agree. My praise for to the builders they have something that looks great, well made, and they are enjoying it. Beyond that there is nothing new here. I asked in the first reply about the topology used for the output buffer and got nothing. I asked what worked and what did not. Does it have a constant current source, source followers, bjt or jfet? I did not ask for schematics, just the design approach used. nothing.
Careful parts matching and part selection is a given. Why bother posting a build in a DIY website and not be prepared for an open discussion.
Very nice looking build and my best thoughts for the builders' future success.
-David
...let us judge using our ears...?
Hi folks ;-))
....enough of those words...enough of suspiciousness...
...let us "judge" using our ears...because other ways seem not to work...
Dear Hurtig and Kurt von Kubik,
would like to have one example of that DAC ...for a reasonnable "forum"-price...--will say at least for "low"--...that will show if you really want to get reasonnable feedback...my profile for that approach to your design: experienced classical musician and...with a lot of experience in recording live performances...
so long
suchtgutenklang
Hi folks ;-))
....enough of those words...enough of suspiciousness...
...let us "judge" using our ears...because other ways seem not to work...
Dear Hurtig and Kurt von Kubik,
would like to have one example of that DAC ...for a reasonnable "forum"-price...--will say at least for "low"--...that will show if you really want to get reasonnable feedback...my profile for that approach to your design: experienced classical musician and...with a lot of experience in recording live performances...
so long
suchtgutenklang
Dear designers,great work you did here!
I also like the sound of CS4398 very much but with a slightly different analog stage approach.
I have bought a very cheap ready available dac kit from HK using the same CS8416 + CS4398
The original sound of it was very bad and different to what i obtained after some tweaking. First of all the chinese guy implemented cs8416 pretty wrong so i had to change the PLL filter
to the new values from updated datasheet,and i also connected pin 20 to Vcc with a 47k resistor for a small innerband jitter as for with the use with delta-sigma DAC's.
After trying also Oscons,all digital decoupling caps were changed to 68 uF Tantalums bypassed with 100n Wima Film as i found that Tantalum small values work best on digital decoupling.
The kit comes with 6 regulators for each required individual chip voltage,all resistors are Dale metal foil, electrolytics are Nichicon all bypassed by a 100n Wima film cap and on the coax input has a pulse transformer of 1:1 winding to ensure complete galvanic isolation.
After that i eliminated all electrolythic coupling caps after the Dac,and changed the Op amp to a high slew rate one 350V/Us - the AD826.
in this topology this opamp is used for summing the 2 differential outputs from CS4398 ,amplifying only the differences between the + and - inputs,thus all the common noise or DC offset that exist is cancelled at the outputs.
This way i obtained 0mV DC offset at the opamp output eliminating all coupling capacitors in the signal way which was the key to a more precise sound with improved detail and focus.
But the following TUBE analog output stage which i encourage you to try redefined the sound in such a manner that i couldn't believe my ears that this DAC chip has such a great potential!
The increased dynamics,the natural warm and sweet midrange so specific for tubes and the great detail and soundstage due to the great slew rate of tubes and their voltage headroom was impressive!
All this in a very simplified analog stage with the theory in mind - the less components in sound path,the best!
So i used just one double triode for stereo,the 6N6P,that is to be found on ebay at 2-3 euros one piece,the anode was fed from a 110v toroidal transformer after rectification and the filament with 6,3V 800mAH DC (beware that this tube calls for a higher 800 mAH fillament current compared to anly 350mAH that other tubes do need)
The schematics i used was anode follower configuration and since this design inverts the phase of the signal, I took the signal just from the L- and R- of the CS4398
Compared to some pretty expensive cd players and even fully tweaked NOS Marantz cd players using the famous TDA1541's with super low jitter Kwak Clock and DEM Reclocking also tubed it outperformed each one of them.
I liked it's sound so much that i decided to house it in an aluminium external CD-ROM Enclosure that will have the Tube on it's top for good ventilation.
Here is the schematic used and some pictures,encouraging everyone that has a CS4398 DAC to try it out and post the positive results of the comparison to their old analog stage
I have also made one small adapter board to be able to directly change/swap it with it's brother,the CS4397,since i heard this IC was made with the same DEM technology and in colaboration with the old Philips team that designed the famous TDA1541A. Comparison between CS4398 and CS4397 will follow.
"Currently, the Cirrus Logic (Crystal Semiconductor) super DAC known as the CS4397 is the main DAC used in new Marantz Super Audio CD/CD players. The CS4397 DAC features a built-in digital filter that enables input of not only Super Audio CD DSD streams but also PCM data up to 192 kHz/24 bits. It also contains a delta sigma modulator and DEM (dynamic element matching) circuitry inherited from the gloriously high fidelity 16-bit DAC TDA1541A S2. A switched capacitor DAC circuit is used in the final stage, and each stage has anti-jitter performance exceeding that of previous 1-bit(Bitstream) DACs. Each CS4397 has a dual channel structure.
Apparently, the same skilled Philips engineers involved in the development of the TDA1541 also participated in the CS4397 DEM circuit. The CS4397 is therefor a super DAC that inherits the tradition of Philips technology. (source: All about Marantz - stereosound)
Source: http://www.marantzphilips.nl/the_evolution_of_dac_the_digital_filter/
I also like the sound of CS4398 very much but with a slightly different analog stage approach.
I have bought a very cheap ready available dac kit from HK using the same CS8416 + CS4398
The original sound of it was very bad and different to what i obtained after some tweaking. First of all the chinese guy implemented cs8416 pretty wrong so i had to change the PLL filter
to the new values from updated datasheet,and i also connected pin 20 to Vcc with a 47k resistor for a small innerband jitter as for with the use with delta-sigma DAC's.
After trying also Oscons,all digital decoupling caps were changed to 68 uF Tantalums bypassed with 100n Wima Film as i found that Tantalum small values work best on digital decoupling.
The kit comes with 6 regulators for each required individual chip voltage,all resistors are Dale metal foil, electrolytics are Nichicon all bypassed by a 100n Wima film cap and on the coax input has a pulse transformer of 1:1 winding to ensure complete galvanic isolation.
An externally hosted image should be here but it was not working when we last tested it.
After that i eliminated all electrolythic coupling caps after the Dac,and changed the Op amp to a high slew rate one 350V/Us - the AD826.
in this topology this opamp is used for summing the 2 differential outputs from CS4398 ,amplifying only the differences between the + and - inputs,thus all the common noise or DC offset that exist is cancelled at the outputs.
This way i obtained 0mV DC offset at the opamp output eliminating all coupling capacitors in the signal way which was the key to a more precise sound with improved detail and focus.
But the following TUBE analog output stage which i encourage you to try redefined the sound in such a manner that i couldn't believe my ears that this DAC chip has such a great potential!
The increased dynamics,the natural warm and sweet midrange so specific for tubes and the great detail and soundstage due to the great slew rate of tubes and their voltage headroom was impressive!
All this in a very simplified analog stage with the theory in mind - the less components in sound path,the best!
So i used just one double triode for stereo,the 6N6P,that is to be found on ebay at 2-3 euros one piece,the anode was fed from a 110v toroidal transformer after rectification and the filament with 6,3V 800mAH DC (beware that this tube calls for a higher 800 mAH fillament current compared to anly 350mAH that other tubes do need)
The schematics i used was anode follower configuration and since this design inverts the phase of the signal, I took the signal just from the L- and R- of the CS4398
An externally hosted image should be here but it was not working when we last tested it.
Compared to some pretty expensive cd players and even fully tweaked NOS Marantz cd players using the famous TDA1541's with super low jitter Kwak Clock and DEM Reclocking also tubed it outperformed each one of them.
I liked it's sound so much that i decided to house it in an aluminium external CD-ROM Enclosure that will have the Tube on it's top for good ventilation.
Here is the schematic used and some pictures,encouraging everyone that has a CS4398 DAC to try it out and post the positive results of the comparison to their old analog stage
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
I have also made one small adapter board to be able to directly change/swap it with it's brother,the CS4397,since i heard this IC was made with the same DEM technology and in colaboration with the old Philips team that designed the famous TDA1541A. Comparison between CS4398 and CS4397 will follow.
"Currently, the Cirrus Logic (Crystal Semiconductor) super DAC known as the CS4397 is the main DAC used in new Marantz Super Audio CD/CD players. The CS4397 DAC features a built-in digital filter that enables input of not only Super Audio CD DSD streams but also PCM data up to 192 kHz/24 bits. It also contains a delta sigma modulator and DEM (dynamic element matching) circuitry inherited from the gloriously high fidelity 16-bit DAC TDA1541A S2. A switched capacitor DAC circuit is used in the final stage, and each stage has anti-jitter performance exceeding that of previous 1-bit(Bitstream) DACs. Each CS4397 has a dual channel structure.
Apparently, the same skilled Philips engineers involved in the development of the TDA1541 also participated in the CS4397 DEM circuit. The CS4397 is therefor a super DAC that inherits the tradition of Philips technology. (source: All about Marantz - stereosound)
Source: http://www.marantzphilips.nl/the_evolution_of_dac_the_digital_filter/
hifimaker said:
The design uses only BJT's... FET's may be just as unusefull in audio, as operational amplifiers.
The topology is very simpel... We use a differential pair in the input. All signal conditioning is made "single ended" in the way, that it is not symetrical. The output is symetrical push-pull, and running at around 10mA .
And naturally, the complete analog stage is DC-coupled.
Re: ...let us judge using our ears...?
No problem.... As soon as I have PCB ready, you are velcome to have one for a fair price.
I will make the PCB order in the end of this week, so within a couple of weeks I should be ready.
suchtgutenklang said:
No problem.... As soon as I have PCB ready, you are velcome to have one for a fair price.
I will make the PCB order in the end of this week, so within a couple of weeks I should be ready.
Hurtig said:
NO! You don't understand the idea of this thread - in general.
Again!!! haven't I already mentioned that remarks about your briliance and our incapability to understand it or implement it are ofending??? I don't care what you did in 5 years and how brilliant that may sound. If you are going to use this space for free advertising, you will have me here constantly busting your chops - make no mistake about it!"
Selling a kit without explaining anything about PCB design and schematics IS MARKETING.
You are using our DIY space for your personal benefit and advertising!!!!
SHOW US SCHEMATICS AND PCB LAYOUT and then HELP US implement your design so that we all achieve the level of brilliance only you posses at the moment.
Otherwise, go to the Traiding Post section and sell your product!!!!
This is now the thread I go to for laughs.
Wish Jocko and Elso were still here, that would realy make my day.
Wish Jocko and Elso were still here, that would realy make my day.
judging after my tastes i might say that CS4397 is much more analog-like than CS4398...just like many people said that AKM4396 is much better sonically than it's ultimate 32 bit brother AK4397
so i've done a little adapter board being able to swap anytime between CS4398 and CS4397 as a matter of taste
and also strange that CS4397 is a little bit more expensive than CS4398 even on farnell's site
so i've done a little adapter board being able to swap anytime between CS4398 and CS4397 as a matter of taste
and also strange that CS4397 is a little bit more expensive than CS4398 even on farnell's site
An externally hosted image should be here but it was not working when we last tested it.
Extreme_Boky said:
That is just what I was going to post !
🙂
Now let me do it a little different:
I spent 25 years perfecting a DAC that now sounds absolutely musical, realistic and holographic using PCM63K chips but it is so good that I don't even need to post schematic for it, I don't even need to show you pictures, because it is so good that you will never be able to duplicate even 40% of its sound.Sorry.
Now how's that for a thread starter ?
P.S.Guys, the point of showing something like your projects to one of the most popular audio and electronics forums on the planet is to either share some new knowledge as people have shared a million ideas with others or at least give a price to this magnificent, ultra fantastic piece of equipment. I can show you a picture of a random board with audio chips and capacitors and call it the Ultimate, again, better than yours.What is the point?
P.S.S. And why do always the two of you post immediately after each other? Do you work in the same place and just wait for someone to post? Or are you just one and the same person?
I apologize if I am a little offensive but this thread got a little too ridiculous.
I remember a few years ago when ecdesigns started the thread about his "ultimate" NOS DAC, and people asked right away for schematics. I thought he wouldn't do it, but he did. And that thread is now over 100 pages. Maybe these guys can learn something from that.
I've been following this thread but I will sign off as I feel the thread has completely derailed. It was not much to start with anyway, it was as useful as saying "we made something great" full stop.
And now you've also started contradicting yourself because in post #73 you wrote:
Which IMO contradicts thew following from post #80:
Why didn't you or Kurt mention this in the first place (I take it you mean the same with "ring core" as "toroid")?
And now you've also started contradicting yourself because in post #73 you wrote:
Kurt von Kubik said:
Which IMO contradicts thew following from post #80:
Hurtig said:
Why didn't you or Kurt mention this in the first place (I take it you mean the same with "ring core" as "toroid")?
luxury54 said:judging after my tastes i might say that CS4397 is much more analog-like than CS4398...just like many people said that AKM4396 is much better sonically than it's ultimate 32 bit brother AK4397
so i've done a little adapter board being able to swap anytime between CS4398 and CS4397 as a matter of taste
and also strange that CS4397 is a little bit more expensive than CS4398 even on farnell's site
The cs4397 is obsolete, perhaps that's why the cost is going up. I have listened extensively to Cirrus's cs4398 eval board and I'm familiar with it's basic sound.
Has anyone else listened and compared both the cs4397 and cs4398?
-David
- Status
- Not open for further replies.