German language is more precise indeed, that's why I like it more with the french for philosophical works and often for technical.
For sure if you said you want the best and say in the same time it will be balanced, with a simple logical shortcut many can understand balanced is the best. I believe that's what I and Miro1360 either was trying to check with our questions and further advices.
Cause of course balanced is not the best or the worst...all is about the chain... and final listening which is a whole and not a focused technical choice somewhere. Trade offs, always...
For sure if you said you want the best and say in the same time it will be balanced, with a simple logical shortcut many can understand balanced is the best. I believe that's what I and Miro1360 either was trying to check with our questions and further advices.
Cause of course balanced is not the best or the worst...all is about the chain... and final listening which is a whole and not a focused technical choice somewhere. Trade offs, always...
Last edited:
What's wrong with Andrea's design? It has gain and is balanced?
https://www.diyaudio.com/forums/swa...-stage-line-stage-dac-output.html#post6433106
As I have got a pair of the rochester ad1862 dacs and I have got a pcb from diy Iggy last year, l will start building this nice little dac somehow soon within the next two month, or so.
I think about a CEN i/v from Patrick or a transformer from Ivan.
I don't need a balanced output, but an appropriate transformer allowes balanced out as well which can be followed by a nice buffer or anything else.
I think about a CEN i/v from Patrick or a transformer from Ivan.
I don't need a balanced output, but an appropriate transformer allowes balanced out as well which can be followed by a nice buffer or anything else.
Last edited:
Balanced DAC
This is one example of how to do a balanced DAC.
NP D1 DAC clone with enhancement
One needs 4x DAC IC's (e.g. AD1862), namely R Ch +ve, R Ch -ve, L Ch +ve, L Ch -ve.
The two DACs with positive phase are fed as normal SE DACs.
The -ve phase will need the S-Data inverted.
To align the time delay for all 4 DACs and invert the Data at the same time, one can, as Nelson Pass did, use 2x 74HC86.
To generate a high output voltage, you need a higher value for R_iv, plus an IV converter capable of that output voltage level.
For example, for 10Vrms output per phase (20Vrms differential), you will need to us 14k for Riv, for each DAC.
If it has to be an opamp IV, you will need to use something that can take 20V or better 24V rails.
But you can also build discrete and goes up to +/-30V rails instead.
https://www.diyaudio.com/archive/bl...-out-production-pedja-rogic-ddnf-iv-stage.pdf
Balanced, or fully differential, audio equipment is NOT about long cables.
It is about lower distortion, unless you want 2nd harmonics by choice.
Forgive my ignorance
And a pair of bridged SE circuit is not what I would call fully differential.
Forgive my ignorance
Cheers,
Patrick
This is one example of how to do a balanced DAC.
NP D1 DAC clone with enhancement
One needs 4x DAC IC's (e.g. AD1862), namely R Ch +ve, R Ch -ve, L Ch +ve, L Ch -ve.
The two DACs with positive phase are fed as normal SE DACs.
The -ve phase will need the S-Data inverted.
To align the time delay for all 4 DACs and invert the Data at the same time, one can, as Nelson Pass did, use 2x 74HC86.
To generate a high output voltage, you need a higher value for R_iv, plus an IV converter capable of that output voltage level.
For example, for 10Vrms output per phase (20Vrms differential), you will need to us 14k for Riv, for each DAC.
If it has to be an opamp IV, you will need to use something that can take 20V or better 24V rails.
But you can also build discrete and goes up to +/-30V rails instead.
https://www.diyaudio.com/archive/bl...-out-production-pedja-rogic-ddnf-iv-stage.pdf
Balanced, or fully differential, audio equipment is NOT about long cables.
It is about lower distortion, unless you want 2nd harmonics by choice.
Forgive my ignorance
And a pair of bridged SE circuit is not what I would call fully differential.
Forgive my ignorance
Cheers,
Patrick
Last edited:
You might want to note what Ivan had to say;
https://www.diyaudio.com/forums/vendor-s-bazaar/327794-output-transformers-dacs-51.html#post6493031
I will be curious to have a POC about the sounding difference of a 1 meter interlink between a SE and a balanced, with say same cable quality and termination quality. I'm sure there will be a difference but not in this precise case one to be superior to the other one... At the opposite, change a capacitor in the oap choice, I'm sure I can rank it, uh !
The whole discussion is off topic imo and only useful if your next input gives you the choice to choose between both, assuming both inputt stage are of the same quality by design vis à vis of the rest of the amp... it's almost nether really equal... context, trade offs... as always. But at saying we talk about theory and focus on the only harmonic aspect of the measurement... which is not enough for me to claim here the design is the best...if the final judge is only the whole interaction that is the music you listen to at the end of all this chain. Just to precise what I thought.
End of the balanced discussion for me .... , everybody knows what his devices need or we would not diy.
The whole discussion is off topic imo and only useful if your next input gives you the choice to choose between both, assuming both inputt stage are of the same quality by design vis à vis of the rest of the amp... it's almost nether really equal... context, trade offs... as always
. But at saying we talk about theory and focus on the only harmonic aspect of the measurement... which is not enough for me to claim here the design is the best...if the final judge is only the whole interaction that is the music you listen to at the end of all this chain. Just to precise what I thought.End of the balanced discussion for me
.... , everybody knows what his devices need or we would not diy.
Last edited:
I have not said one word about what is good or bad.
It is subjective and purely one's personal choice.
There are enough people who prefer single ended tube gear with 0.1% H2.
They also prefer vinyl to DACs BTW.
I merely tried to explain why people would bother with the extra complication.
My entire audio chain is balanced, from source to power amp.
And that includes phono pre. Again, my personal choice.
One can also look at commercial designs from Curl, Hansen (Ayre), Pass Labs, etc.
More often than not, you see fully differential circuitry, again for good reasons.
Each to his own,
Patrick
It is subjective and purely one's personal choice.
There are enough people who prefer single ended tube gear with 0.1% H2.
They also prefer vinyl to DACs BTW.
I merely tried to explain why people would bother with the extra complication.
My entire audio chain is balanced, from source to power amp.
And that includes phono pre. Again, my personal choice.
One can also look at commercial designs from Curl, Hansen (Ayre), Pass Labs, etc.
More often than not, you see fully differential circuitry, again for good reasons.
Each to his own,
Patrick
Regardless of the "balanced" or "unbalanced" question: I found another old report from Lampizator that uses the AD1862 with resistor I/U for tube output. Maybe this is interesting for one or the other:
AD1862 - Denon DCD 2560
AD1862 - TEAC VRDS-X25
AD1862 - Denon DCD 2560
AD1862 - TEAC VRDS-X25
Agreed, didn't think about this as currently it is not overly important for me. It is just an option out of many.
Thanks. I know this thread. I do have a pair of transformers for my TDA1541 DAC which I am going to try here in the first place in order to get an idea about how it might sound. Beside the mentioned CEN and the trafos I'll try the I/V on the pcb as well. I like to keep it simple.
If your small pcbs turn out working properly, are you thinking of offering them for the public? Gerber or for sale? I would be happy about that.
BR.
Almost forgotten, did they not mention you have to use one of these :
https://www.diyaudio.com/forums/digital-source/140538-lossless-sd-card-player-8.html#post2031898
or at least
Building the ultimate NOS DAC using TDA1541A
Building the ultimate NOS DAC using TDA1541A
Cheers,
Patrick
.
https://www.diyaudio.com/forums/digital-source/140538-lossless-sd-card-player-8.html#post2031898
or at least
Building the ultimate NOS DAC using TDA1541A
Building the ultimate NOS DAC using TDA1541A
Cheers,
Patrick
.
Attachments
Last edited:
The value seems high to me. He had probably tried a transistor before ...(?) What would you do concretely at this point? As a start for measurements?Tube guys are brave people.
A passive Riv of 570R with a DAC Zout of 2.1k.
I wonder how it measures.
(But of course who cares about measurements, right ?)
Patrick
You first need a setup which allows you to get the DAC to output say a 1kHz sine wave.
This was what I used while matching PCM1704s.
https://www.diyaudio.com/forums/digital-line-level/325249-project-pcm1704-2.html#post5810069
Once that is working, you can now insert different IV circuit and measure them individually.
And you should probably measure at 0dB, -6dB and -20dB digital output.
If you use passive IV, you need to add the distortion of the following amplifier stage for fair comparison.
And just to demonstrate the purpose of using balanced circuitry.
This is the distortion spectrum of our discrete AD844 IV (see attached).
https://www.diyaudio.com/forums/dig...mirror-iv-converter-la-ad844.html#post6341451
The Pedja Rogic discrete IV has a similar distortion spectrum.
https://www.diyaudio.com/archive/bl...-out-production-pedja-rogic-ddnf-iv-stage.pdf
You can see that H2 is ~10x of H3.
But with good matching, you can cancel H2 in balanced mode by 20~30dB.
And then we are getting close to -100dB THD at full swing.
Opamp will give you better measurements.
Patrick
.
This was what I used while matching PCM1704s.
https://www.diyaudio.com/forums/digital-line-level/325249-project-pcm1704-2.html#post5810069
Once that is working, you can now insert different IV circuit and measure them individually.
And you should probably measure at 0dB, -6dB and -20dB digital output.
If you use passive IV, you need to add the distortion of the following amplifier stage for fair comparison.
And just to demonstrate the purpose of using balanced circuitry.
This is the distortion spectrum of our discrete AD844 IV (see attached).
https://www.diyaudio.com/forums/dig...mirror-iv-converter-la-ad844.html#post6341451
The Pedja Rogic discrete IV has a similar distortion spectrum.
https://www.diyaudio.com/archive/bl...-out-production-pedja-rogic-ddnf-iv-stage.pdf
You can see that H2 is ~10x of H3.
But with good matching, you can cancel H2 in balanced mode by 20~30dB.
And then we are getting close to -100dB THD at full swing.
Opamp will give you better measurements.
Patrick
.
