Hello everyone.
Since this thread is very long and I haven't had the opportunity to read it completely, I'll just ask directly.
Is there a way to implement balanced outputs with this DAC?
I.e. Iout + and Iout - to be generated by the DAC and not with a conversion from unbalanced to balanced after the IV conversion.
If I interpret the AD1862 data sheet correctly, the chip only has an unbalanced output. For example, is it for possible to operate 2 chips in parallel and have one of them output an inverted signal?
Or is there a brother of the AD1862 with balanced outputs?
Other equivalent alternatives?
My entire system is balanced, so I would find it a shame if the playback chain started with an unbalanced source.
Thank you for your help in advance.
Cheers Niels
Since this thread is very long and I haven't had the opportunity to read it completely, I'll just ask directly.
Is there a way to implement balanced outputs with this DAC?
I.e. Iout + and Iout - to be generated by the DAC and not with a conversion from unbalanced to balanced after the IV conversion.
If I interpret the AD1862 data sheet correctly, the chip only has an unbalanced output. For example, is it for possible to operate 2 chips in parallel and have one of them output an inverted signal?
Or is there a brother of the AD1862 with balanced outputs?
Other equivalent alternatives?
My entire system is balanced, so I would find it a shame if the playback chain started with an unbalanced source.
Thank you for your help in advance.
Cheers Niels
Output transformer, along with its effect on the sound. It will also be a filter for HF noise.
I will try that too, although I don't have a balanced system.
It is not just "put" transformer at the end of given OP stage 
For transformer driving with OP only - not good solution. Trefore the tests will be irirelevant.
Unfortunately it is common praxis in the forum that members "putting"
all kind of unknown transformers direct at the OP outputs.
.
Consult classic pro players output for the concept for Single ended OP+BJT buffer transformer driver
for the example EMT riaa end version with OPs, LHH2000 Philips, some Revox models.
It is classic good way. To achieve BAL with transformer, driven with OP+Buffer
.
For balancing SE and after that drive with buffer balancing line transformer
https://www.ranecommercial.com/kb_article.php?article=2112
https://www.edn.com/design-of-high-performance-balanced-audio-interfaces-part-7/
...
A lot of good references and datas
https://sound-au.com/articles/balanced-2.htm
...
For transformer driving with OP only - not good solution. Trefore the tests will be irirelevant.
Unfortunately it is common praxis in the forum that members "putting"
all kind of unknown transformers direct at the OP outputs.
.
Consult classic pro players output for the concept for Single ended OP+BJT buffer transformer driver
for the example EMT riaa end version with OPs, LHH2000 Philips, some Revox models.
It is classic good way. To achieve BAL with transformer, driven with OP+Buffer
.
For balancing SE and after that drive with buffer balancing line transformer
https://www.ranecommercial.com/kb_article.php?article=2112
https://www.edn.com/design-of-high-performance-balanced-audio-interfaces-part-7/
...
A lot of good references and datas
https://sound-au.com/articles/balanced-2.htm
...
I do not agree. Typical OPA has a very low output impedance, why add a buffer?
I don't have a balanced system, and I don't need a transformer at the output. So I'm just going to try it out of curiosity. If it turns out well, I'll also put balanced outputs on the DAC if I ever need it.
I have custom Permalloy core wound transformers 1:1 for that test. One has a very wide range of a couple of MHz, and the other limits the range to 80kHz quite sharply. Only the winding is different, the cores are the same. The one that limits the range sounds better to me in DDDAC. And I have a few other 1:1 transformers to try. It will be interesting to compare and measure it a bit. If I am not satisfied, I will not put it in DAC, very simple.
I don't have a balanced system, and I don't need a transformer at the output. So I'm just going to try it out of curiosity. If it turns out well, I'll also put balanced outputs on the DAC if I ever need it.
I have custom Permalloy core wound transformers 1:1 for that test. One has a very wide range of a couple of MHz, and the other limits the range to 80kHz quite sharply. Only the winding is different, the cores are the same. The one that limits the range sounds better to me in DDDAC. And I have a few other 1:1 transformers to try. It will be interesting to compare and measure it a bit. If I am not satisfied, I will not put it in DAC, very simple.
Last edited:
transformer work in two way reflect the load on the other side ...buffer is there to isolate the impedence
Yes and to give more current drive to the complex reactive load. Not to forget that is no classic series buffer but part of the feedback loop.
In the older design excellent line transformers used. Much better than contmporary. Core material, wire purity, isolation. But the designers find that only OP not suitable to drive.
I simply could not understand why some people constantly fighting achieved, scientifically verified concepts in audio praxis?
Without challenging the old, there is no new. Discovery is all about challenging what is, it's how new inventions come by.
That will seems low but You will have ringing just with these values. Also all this will transfer to the side of OP amp driver as addition to already present reactive load. Reactive include not only R and C but also L component.
I am pretty sure that type of winding is not the only diference if the information of HF side is as you said 80K few MHz. Please make some 5 minute passive measurements only: L primary, Rdc Primary, Mutual inductance (google) and capacitance.
You didn't note what was the termination value and type for HF range measurements.
Yes it is BUT the data line input to one AD chip should be inverted. Every other digital input lines remains the same.
In this case You will have doubled digital module and depending on how good is power supply and pub maybe You will get more digital noise? That is not canceling like balanced analog line...
Thanks to the accidental death of my finally good sounding Soekris 1121 I felt the urge to finish the AD1862 dac, which I did successfully some days ago. Power supplies are some leftovers from former projects, nothing fancy, and for opamps I only had to contenders with the Opa177 and LM6171, of which I like the Opa117 more.
I have to say, that I didn't expect this little dac to sound that good.
I had a glimpse of it when Ernst visited me some months ago with his one, and the very good impression I had back then is veryfied by now.
Very organic and life like sound, I can listen to whole albums without skipping a track, which says something.
Out of frustration I bought a smsl D300 before, which I can sell now I guess, although it does sound pretty good for a Delta Sigma dac.
I did like the Soekris a lot as well, compared to the Miro dac it was more spacious and airy, but also less focused and more laid back.
Considering the very simple output stage it sounds absolutely remarkable, so of course I participated in the group by next door for Patrick's boards.
Surprising is the very big sensitivity for parts, it seems to matter a lot which part to put where.
Thanks everybody for making this possible.....
I have to say, that I didn't expect this little dac to sound that good.
I had a glimpse of it when Ernst visited me some months ago with his one, and the very good impression I had back then is veryfied by now.
Very organic and life like sound, I can listen to whole albums without skipping a track, which says something.
Out of frustration I bought a smsl D300 before, which I can sell now I guess, although it does sound pretty good for a Delta Sigma dac.
I did like the Soekris a lot as well, compared to the Miro dac it was more spacious and airy, but also less focused and more laid back.
Considering the very simple output stage it sounds absolutely remarkable, so of course I participated in the group by next door for Patrick's boards.
Surprising is the very big sensitivity for parts, it seems to matter a lot which part to put where.
Thanks everybody for making this possible.....
A picture of the usual breadboard...
with an old Quanghao first gen Salas Shunt, not so good actually. Battery powered LT3045 for WaveIo, and a supply for a long gone Curryman dac as preregulator for a China LT3045 12v reg.
I will build a second one, this time with the PSU2 for a neat and tidy build.
I will build a second one, this time with the PSU2 for a neat and tidy build.
