As mentioned earlier, we intend to perform our own distortion measurements on a collection of IV converters we have on hand, for our own education.
https://www.diyaudio.com/community/...st-tht-i2s-input-nos-r-2r.354078/post-7594471
Here are the measurement results.
All thanks are due to Fran for all the measurements.
Patrick
.
https://www.diyaudio.com/community/...st-tht-i2s-input-nos-r-2r.354078/post-7594471
Here are the measurement results.
All thanks are due to Fran for all the measurements.
Patrick
.
Thank you
Would you please post results for first SEN using only K170 without cascode
SMMS I/V on dip8 socket from post 1 Is very nice
Would you please post results for first SEN using only K170 without cascode
SMMS I/V on dip8 socket from post 1 Is very nice
Last edited:
Nice comparison. I assume the sonic differences are not related to THD below a certain level.
Here is my measurement of the plain NE5534N at 440 Hz full scale dithered sine, D/A is a TDA1541A:
https://www.diyaudio.com/community/threads/tda1541a-reducing-dnl.414159/page-3#post-7729807
Here is my measurement of the plain NE5534N at 440 Hz full scale dithered sine, D/A is a TDA1541A:
https://www.diyaudio.com/community/threads/tda1541a-reducing-dnl.414159/page-3#post-7729807
Listening impressions are subjective.
Fran has published them a while ago at the Miro thread.
Patrick
Fran has published them a while ago at the Miro thread.
Patrick
Hi thanks. Interresting.
So one must conclude as Woodturner-fran like best the noisy DHT, the worse the measurement the better the sound... H2 effect ?
Should be more complicate than just the THD performance, especially in I/V yask where there are a lot of more thingsto look at.
Note that if listening impression are subjectives, a graphic, even more if an isolated one, makes no sound. It just gives you a general idea of the circuit performance designed with that good numbers in mind. Which is okay but not suffice as demonstrate the no corelation between subjective listening impression (the goal of the design) and electrical numbers as objective benchmark technic to compare two circuitrie on the measured paramater.
So THD aloe is not enough,, so questions remains, what are the others parameters to measure and how to rank them and balanced them in a design - trade offs as some could be oposed - to acheive an excellent I/V stage (according the particular dac ic constraint).
I keep in mind for instance, just at seeing the layout that a huge through hole design as with the DHT vs the tinny FC CEN smd pcb both acheive good subjective listening result but exhibiting again different thd numbers.
So one must conclude as Woodturner-fran like best the noisy DHT, the worse the measurement the better the sound... H2 effect ?
Should be more complicate than just the THD performance, especially in I/V yask where there are a lot of more thingsto look at.
Note that if listening impression are subjectives, a graphic, even more if an isolated one, makes no sound. It just gives you a general idea of the circuit performance designed with that good numbers in mind. Which is okay but not suffice as demonstrate the no corelation between subjective listening impression (the goal of the design) and electrical numbers as objective benchmark technic to compare two circuitrie on the measured paramater.
So THD aloe is not enough,, so questions remains, what are the others parameters to measure and how to rank them and balanced them in a design - trade offs as some could be oposed - to acheive an excellent I/V stage (according the particular dac ic constraint).
I keep in mind for instance, just at seeing the layout that a huge through hole design as with the DHT vs the tinny FC CEN smd pcb both acheive good subjective listening result but exhibiting again different thd numbers.
It has been very interesting carrying out the measurements.
As is pointed out in the PDF, there are limitations in the measurements, particularly in that it also includes the whole chain from SD card player through DAC along with the IV stage. But at least everything else is held constant in these tests, so that only IV stage and its power supply changes.
THD is not the same as subjective impression for sure. Several friends here have loved that 3a5 stage - I have built 3 of them at this point and they are in daily use in their homes. I have one or two others yet to try/measure so there might be another update in a few weeks, time allowing.
Fran
As is pointed out in the PDF, there are limitations in the measurements, particularly in that it also includes the whole chain from SD card player through DAC along with the IV stage. But at least everything else is held constant in these tests, so that only IV stage and its power supply changes.
THD is not the same as subjective impression for sure. Several friends here have loved that 3a5 stage - I have built 3 of them at this point and they are in daily use in their homes. I have one or two others yet to try/measure so there might be another update in a few weeks, time allowing.
Fran
Many thanks for the measurements.
No experience with 3A5 but the distortion appears excessive and for me, completely unacceptable in a source component. My 10Y line stage has probably two orders of magnitude better distortion, so very likely the culprit is the topology. Dhts are great provided they work completely unloaded, i.e. active load +buffering. Or maybe the high Riv also has a contribution...
No experience with 3A5 but the distortion appears excessive and for me, completely unacceptable in a source component. My 10Y line stage has probably two orders of magnitude better distortion, so very likely the culprit is the topology. Dhts are great provided they work completely unloaded, i.e. active load +buffering. Or maybe the high Riv also has a contribution...
I'm interested.
A few months ago I did extensive measurements on some SEN variants I'm working on.
Although I can get truly excellent measurements under some conditions, these conditions are not where always where I expected/hoped to find them. Most new high-performance DACs that I have looked at are specified their best (in terms of THD+N) at full output volume (no digital attenuation, 0dB) and are typically designed to deliver 4Vrms at this point.
In my test set-up (Iancanada's ES9038Q2M DAC board) I get best results with 10-12dB digital attenuation and I/V resistor designed for 1Vrms at this level. I always suspected that I needed to up the SEN bias to fully exploit the high current delivered by ESS chip (some 8mA), but there does not seem to be a strict correlation between bias current and output sweet spot.
As I cannot exclude that the DAC I'm using is actually the culprit, I stopped the experiments deciding that an ultra low noise and distortion current source would be needed to make real progress. Never got down to making this....
A few months ago I did extensive measurements on some SEN variants I'm working on.
Although I can get truly excellent measurements under some conditions, these conditions are not where always where I expected/hoped to find them. Most new high-performance DACs that I have looked at are specified their best (in terms of THD+N) at full output volume (no digital attenuation, 0dB) and are typically designed to deliver 4Vrms at this point.
In my test set-up (Iancanada's ES9038Q2M DAC board) I get best results with 10-12dB digital attenuation and I/V resistor designed for 1Vrms at this level. I always suspected that I needed to up the SEN bias to fully exploit the high current delivered by ESS chip (some 8mA), but there does not seem to be a strict correlation between bias current and output sweet spot.
As I cannot exclude that the DAC I'm using is actually the culprit, I stopped the experiments deciding that an ultra low noise and distortion current source would be needed to make real progress. Never got down to making this....
Nic,
You need to understand the distortion mechanism in the floating SEN IV first of all.
It is not just about bias.
For DAC with a lot of output current, it is also not necessarily the best iv circuit.
Patrick
You need to understand the distortion mechanism in the floating SEN IV first of all.
It is not just about bias.
For DAC with a lot of output current, it is also not necessarily the best iv circuit.
Patrick
From your article I understand that also input capacitance and leakage current are important factors to take into consideration. I believe to have these under control too.
I have been listening to ESS DAC's with SEN I/V for years and everything else I have tried has lost in comparison.
Therefore I will continue to fiddle with this circuit to squeeze the last drop of juice out of it 🙂
I have been listening to ESS DAC's with SEN I/V for years and everything else I have tried has lost in comparison.
Therefore I will continue to fiddle with this circuit to squeeze the last drop of juice out of it 🙂
Final warning to those concerned. Play nicely and without the personal insults or you will find yourselves unable to post further in this thread.
Several years ago ESS held a seminar on their dac modulator. They pointed out things they thought audiophiles were hearing which were not necessarily shown in standard measurements. What ESS said has been discussed in this forum enough by now that pretty much everyone should know what they are, and what ESS most likely meant by their choice of words.
Anyone still not familiar with that?
Moving along to the more recent past, there was some discussion about how I/V circuits must respond to to some of what must be coming out of DSD modulators. That discussion was in a thread about DSD dac design, specifically MarcelvdG's RTZ DSD Dac.
Anyone still not familiar with that?
Moving along to the more recent past, there was some discussion about how I/V circuits must respond to to some of what must be coming out of DSD modulators. That discussion was in a thread about DSD dac design, specifically MarcelvdG's RTZ DSD Dac.
Last edited:
Do you know what the internal impedance of the ES9038Q2M is? I may be wrong but the older chips were in the order of 200ohms, right? Is non linear capacitance much of a distortion factor at such a low impedance?
Have you tried Owen's IV?
Impedances are given by @Iancanda at: https://www.diyaudio.com/community/...dac-hats-for-raspberry-pi.302687/post-5442603
Following below are some explanations offered by ESS about how their dacs sound from their modulator seminar several years ago. IME not many people seem to "get" some of what ESS is trying to say, at least not at first. Probably more on the new side to folks have haven't taken a deep dive into how sigma-delta modulators work.
From the ESS Hyperstream Modulator presentation, there is a slide which initiates discussion about audiophiles with the words, "Understanding what audiophiles are hearing."
"The surprising reality is that sigma-delta DACs can be audibly distinguished from a conventional DAC despite measuring very much better than that DAC."
"...an important point: The human ear detects signals well below the noise level of the DAC."
"The ear is exquisitely sensitive to "unusual" noise sources. Your ancestors camped out by a waterfall (white noise) and yet their 'ears pricked up" when they heard a hint of a predator moving in the undergrowth. (The equivalent visual phenomenon is "seeing something out of the corner of your eye). Noise, to a large degree, can be accommodated by the ear and is not troubling, but the tiniest "anomalous" noise is raised to the conscious level."
"Sigma-delta modulators create non-periodic steady state noise (non-PSS) artifacts..."
"Periodic Steady State analysis is common in RF circuits. It means that the system is forced to repeat a pattern of behavior over and over again with a certain time period. Any artifact is presumed to also repeat in this time period."
"Audio measurements such as THD and DNR are done in the Periodic Steady State. Therefore, they will not activate non-PSS noise. You will not find non-PSS noise by looking at THD, DNR, and SNR."
"As the audio signal moves, the noise does not remain the same."
"Non-PSS noise is the biggest issue, but experiments suggest there are more problems. For example: Audiophiles rate as inferior systems that have variable excess phase noise."
"We find that an unconditionally stable loop sounds better in listening tests."
-------------------------------------------------------
Link to Yumpu Presentation:
https://www.yumpu.com/en/document/read/23182504/noise-shaping-sigma-delta-dacs-ess-technology-inc
"WHY AUDIOPHILES KNOW BY LISTENING IF ITS A SIGMA-DELTA DAC" starts on page 28 of the yompu presentation.
At ASR:
https://www.audiosciencereview.com/...-‘hump’-investigation.5752/page-5#post-134135
Additional info at 6 Moons:
Where ESS talks about doing MANY LISTENIG TESTS.
https://6moons.com/audioreviews2/resonessence/2.html
More at ASR about ESS and TI paper on modulator idle tones:
https://www.audiosciencereview.com/...ess-thd-‘hump’-investigation.5752/post-150331
From the ESS Hyperstream Modulator presentation, there is a slide which initiates discussion about audiophiles with the words, "Understanding what audiophiles are hearing."
"The surprising reality is that sigma-delta DACs can be audibly distinguished from a conventional DAC despite measuring very much better than that DAC."
"...an important point: The human ear detects signals well below the noise level of the DAC."
"The ear is exquisitely sensitive to "unusual" noise sources. Your ancestors camped out by a waterfall (white noise) and yet their 'ears pricked up" when they heard a hint of a predator moving in the undergrowth. (The equivalent visual phenomenon is "seeing something out of the corner of your eye). Noise, to a large degree, can be accommodated by the ear and is not troubling, but the tiniest "anomalous" noise is raised to the conscious level."
"Sigma-delta modulators create non-periodic steady state noise (non-PSS) artifacts..."
"Periodic Steady State analysis is common in RF circuits. It means that the system is forced to repeat a pattern of behavior over and over again with a certain time period. Any artifact is presumed to also repeat in this time period."
"Audio measurements such as THD and DNR are done in the Periodic Steady State. Therefore, they will not activate non-PSS noise. You will not find non-PSS noise by looking at THD, DNR, and SNR."
"As the audio signal moves, the noise does not remain the same."
"Non-PSS noise is the biggest issue, but experiments suggest there are more problems. For example: Audiophiles rate as inferior systems that have variable excess phase noise."
"We find that an unconditionally stable loop sounds better in listening tests."
-------------------------------------------------------
Link to Yumpu Presentation:
https://www.yumpu.com/en/document/read/23182504/noise-shaping-sigma-delta-dacs-ess-technology-inc
"WHY AUDIOPHILES KNOW BY LISTENING IF ITS A SIGMA-DELTA DAC" starts on page 28 of the yompu presentation.
At ASR:
https://www.audiosciencereview.com/...-‘hump’-investigation.5752/page-5#post-134135
Additional info at 6 Moons:
Where ESS talks about doing MANY LISTENIG TESTS.
https://6moons.com/audioreviews2/resonessence/2.html
More at ASR about ESS and TI paper on modulator idle tones:
https://www.audiosciencereview.com/...ess-thd-‘hump’-investigation.5752/post-150331
Do you mean https://www.diyaudio.com/community/threads/return-to-zero-shift-register-firdac.379406/post-7636229 and
https://www.diyaudio.com/community/threads/return-to-zero-shift-register-firdac.379406/post-7686002 ? Or rather the discussion about partly passive filtering?
We have a very-low-level frequency-modulated tone added to the audio signal with some modulator types. As its carrier frequency is very close to zero, it looks like harmonic distortion on a DFT plot until you zoom in. It is in fact due to intermodulation between quite-high-level frequency-modulated tones around half the DSD rate.
Last edited:
There is a lot of good stuff in that thread. However, part of what I was thinking about is the case where the slew rate coming out of the dac exceeds what the I/V amp can keep up with so some of the RF from the dac feeds forward around I/V amp through the integrator cap to the I/V amp output, were it is presumably absorbed by the open loop output impedance of the IV amp's output stage.
There is also the subject of how measured noise isn't necessarily noise in the conventional sense, and how some of the so-called noise can sound more like a distortion is also of possible interest.
Even something as old hat as current noise in resistors starts to demand some new respect. At least if ESS is to be believed: Noise correlated with the audio signal has more effect on perception than noise which is unchanging.
There is also the subject of how measured noise isn't necessarily noise in the conventional sense, and how some of the so-called noise can sound more like a distortion is also of possible interest.
Even something as old hat as current noise in resistors starts to demand some new respect. At least if ESS is to be believed: Noise correlated with the audio signal has more effect on perception than noise which is unchanging.
Last edited:
A further addition to the collection -- the xen JFET passive IV :
https://www.diyaudio.com/community/...st-tht-i2s-input-nos-r-2r.354078/post-7160371
Tube lovers must give it a try.
Essentially second harmonics dominant.
Patrick
.
https://www.diyaudio.com/community/...st-tht-i2s-input-nos-r-2r.354078/post-7160371
Tube lovers must give it a try.
Essentially second harmonics dominant.
Patrick
.
