Some power line pollution. I'll try later if it gets lower if I supply the DAM with a regulated supply.
Because of that I was delaying my publication. My measurements are too poor, so I managed to improve my measurements, using best cables and battery in DAM (to avoid capacitive coupling the line/main source, now getting the DAM connected only to the PC). So I'll publish some results in the coming posts, starting with the "noise floor":
Attachments
The figures has title explaining, but I have to add: the decoupling in "DIYBras" and "only caps" tests are with 470µF x 16V Nichicon FP for each rail. The original are "raw", without cap or mods, of course. The 4V are BIG cap coupled directly to audio board (with R-Zener to protect against surges).
Note some low level 60Hz, like the 50Hz with zfe graphs; is very difficult to totally eliminate it.
The signal applied in DAm was 1kHz sinewave with 192kHz sample rate input.
Note some low level 60Hz, like the 50Hz with zfe graphs; is very difficult to totally eliminate it.
The signal applied in DAm was 1kHz sinewave with 192kHz sample rate input.
Attachments
Last edited:
A warning concerning my sound card measurements: My Fireface 800 seems to have a "jitter problem". Signals ~120 dB below the main peak can result from this, and are not necessarily from the source. So this affects mainly audio-out measurements of full level test tones.
Details: I got not as clean graphs as Paul after modification and looked for the reason.
Here e.g. a 24kHz -1 dB test signal
The symmetric pattern around the min peak is typical for jitter. Following Julian Dunn [1] this would result from about 50ps jitter modulated by a 2kHz (next to square) wave.
This jitter could either be caused by the DAC or the ADC. The DAM should not have that much jitter (and I tested with an other DAC where I get a similar picture), so it is from the ADC.
-----
[1] Julian Dunn: Audio Precision Application Note 5, Measurement Techniques for Digital Audio page 21-22
http://www.ap.com/download/file/17
Details: I got not as clean graphs as Paul after modification and looked for the reason.
Here e.g. a 24kHz -1 dB test signal
The symmetric pattern around the min peak is typical for jitter. Following Julian Dunn [1] this would result from about 50ps jitter modulated by a 2kHz (next to square) wave.
This jitter could either be caused by the DAC or the ADC. The DAM should not have that much jitter (and I tested with an other DAC where I get a similar picture), so it is from the ADC.
-----
[1] Julian Dunn: Audio Precision Application Note 5, Measurement Techniques for Digital Audio page 21-22
http://www.ap.com/download/file/17
Well,
With all of this wonderful dithering on what is best for the shift register PS and the fact that I removed the components from my board and I am waiting for someone to come up with THE WAY I have returned to LPs.
After a thorough reworking of my turntable's motor system and cleaning up solder joints in my phono amp I can hear that I thought I was getting close to good sound with the digital setup but I was nowhere close. So I guess I am appreciative for this, taking my DAC out of commission and having to depend, again, on the turntable.
I look forward to getting the DAC back in working condition and hope springs eternal that digital will triumph in the end. I love the convenience of the computer. I had been using a yet to be released LINUX OS/Player which is the best I had ever heard and it is not a finished project yet. Not my work! When it is available I will enthusiastically let it be known.
In comparison to WINDOWS based players it is a revelation and yet there is still so far to go, damnit.
There is value to using both formats - especially when one is not working!
After all is said and done I suspect SOEKRIS's recommendations will be found to be best. From what I can see measurement methodology and the supply supplying the shift register supply are confusing us as to what is best. I am completely incapable of taking measurements so please none of you intrepid folks whose efforts I greatly admire take this the wrong way.
I suspect whoever suggested using battery based supplies for the SOEKRIS for measurement taking is on the right track. I have assembled a "hybrid" supply using batteries as giant capacitors and hope this will isolate the SOEKRIS from the noise of the power line but was wanting to get the board mods done (when these became available) before installing - I have been putting this off for months for various reasons. blah, blah, blah ...
With all of this wonderful dithering on what is best for the shift register PS and the fact that I removed the components from my board and I am waiting for someone to come up with THE WAY I have returned to LPs.
After a thorough reworking of my turntable's motor system and cleaning up solder joints in my phono amp I can hear that I thought I was getting close to good sound with the digital setup but I was nowhere close. So I guess I am appreciative for this, taking my DAC out of commission and having to depend, again, on the turntable.
I look forward to getting the DAC back in working condition and hope springs eternal that digital will triumph in the end. I love the convenience of the computer. I had been using a yet to be released LINUX OS/Player which is the best I had ever heard and it is not a finished project yet. Not my work! When it is available I will enthusiastically let it be known.
In comparison to WINDOWS based players it is a revelation and yet there is still so far to go, damnit.
There is value to using both formats - especially when one is not working!
After all is said and done I suspect SOEKRIS's recommendations will be found to be best. From what I can see measurement methodology and the supply supplying the shift register supply are confusing us as to what is best. I am completely incapable of taking measurements so please none of you intrepid folks whose efforts I greatly admire take this the wrong way.
I suspect whoever suggested using battery based supplies for the SOEKRIS for measurement taking is on the right track. I have assembled a "hybrid" supply using batteries as giant capacitors and hope this will isolate the SOEKRIS from the noise of the power line but was wanting to get the board mods done (when these became available) before installing - I have been putting this off for months for various reasons. blah, blah, blah ...
Now I only use the DAM with batteries. I always thought about doing this with a DAC, but only the DAM inspired me to actually do this. My old DACs used many shunt supplyes, class A output stages (DAM not need it), so consumed too much, and did not support all modern PCM sampling.
rickmcinnis, I'm afraid your dream to match the digital with the vinyl is something quite complicated to happen fully. Vinyl have colorations and always have measurable distortion; is a fact. The digital has some colorations, in different ways. The subjective result differs. Many prefer the vinyl, and the best sound system I've ever heard was playing a vinyl.
I consider the recordings and electronic reproduction system something flawed (like some folks in Beyond the Ariel thread, and the late Harvey "Gizmo" Rosemberg), so it is no sin prefer something theoretically "lower" technologycally in theory (vinyl, in the opinion of measurements fans). Also, different technologies of speakers sound different, and in this case also we can choose what "less hurt our ears." Or "pick your poison".
I use digital as in Brazil is very difficult to get decent vinyl. In digital, we purchase and download ...
Only my 2 cents...
But all of this reality don't stop me to trying to evolve the digital.
rickmcinnis, I'm afraid your dream to match the digital with the vinyl is something quite complicated to happen fully. Vinyl have colorations and always have measurable distortion; is a fact. The digital has some colorations, in different ways. The subjective result differs. Many prefer the vinyl, and the best sound system I've ever heard was playing a vinyl.
I consider the recordings and electronic reproduction system something flawed (like some folks in Beyond the Ariel thread, and the late Harvey "Gizmo" Rosemberg), so it is no sin prefer something theoretically "lower" technologycally in theory (vinyl, in the opinion of measurements fans). Also, different technologies of speakers sound different, and in this case also we can choose what "less hurt our ears." Or "pick your poison".
I use digital as in Brazil is very difficult to get decent vinyl. In digital, we purchase and download ...
Only my 2 cents...
But all of this reality don't stop me to trying to evolve the digital.
DIYBras,
We are not in disagreement!
No question "analogue" is full of noise, distortion, coloration - I would be the last to say it is perfect. I want digital to surpass IN ALL WAYS analogue.
Interesting how each format can teach us about the other. I find they each have their strong points. The winner will be based on which does the most better (and then to decide the relative importance of each category). Something that I would never bother to actually try to do. I will leave the verdict to instinct.
I look forward to getting the DAC working again - just waiting a little while to make sure the output resistor value is settled since I have to take the board to a friend to get the component replaced.
My comment was that I thought I was getting close to surpassing analogue on most fronts. Going back to the 'table and doing some long overdue maintenance and improving showed me that there is lots to do t get digital where we want it. Analogue is still noisy, colored and distorted just less annoying than my digital setup!
Dynamics - no comparison, I mean the leading edge of notes, along with bass response that is much deeper and defined than digital. One would think this would not even be a contest in the bass.
I thought I was going back out of necessity and found I might want to stay for awhile.
Most of my vinyl is old. I will but new stuff on the odd occasion but it is really expensive. I find the new pressings to be better initially but I swear I think they start showing wear quickly. Most recent example the initially exquisite BEATLES monos - but they are sounding a little funny much too soon and more pops than there should be. Extremely annoying ...
We are not in disagreement!
No question "analogue" is full of noise, distortion, coloration - I would be the last to say it is perfect. I want digital to surpass IN ALL WAYS analogue.
Interesting how each format can teach us about the other. I find they each have their strong points. The winner will be based on which does the most better (and then to decide the relative importance of each category). Something that I would never bother to actually try to do. I will leave the verdict to instinct.
I look forward to getting the DAC working again - just waiting a little while to make sure the output resistor value is settled since I have to take the board to a friend to get the component replaced.
My comment was that I thought I was getting close to surpassing analogue on most fronts. Going back to the 'table and doing some long overdue maintenance and improving showed me that there is lots to do t get digital where we want it. Analogue is still noisy, colored and distorted just less annoying than my digital setup!
Dynamics - no comparison, I mean the leading edge of notes, along with bass response that is much deeper and defined than digital. One would think this would not even be a contest in the bass.
I thought I was going back out of necessity and found I might want to stay for awhile.
Most of my vinyl is old. I will but new stuff on the odd occasion but it is really expensive. I find the new pressings to be better initially but I swear I think they start showing wear quickly. Most recent example the initially exquisite BEATLES monos - but they are sounding a little funny much too soon and more pops than there should be. Extremely annoying ...
That's it, rickmcinnis! Thanks for sharing your experience.
To add, one time I had the opportunity to witness one pre from Audiopax (I don't remember the model, but is close to the Model 88 amplifier), where the user has the opportunity to adjust the distortion of it. I could not set the same, the actual owner did not permits, but this sounded "magical" in some recordings. So I did some experiments with DIY pres and amplifiers with appreciable/adjustable THD. I was also inspired by certain sound good reports coming from systems with digital+SET. The results were mixed, but taught a lot! Vinyl and SET really are not equal but have some things in common (if vinyl teaches something, SET too). So I do not discarded the possibility, in some cases, using a deliberately distorted system, which gives an effect not far from the vinyl... and better sometimes!
In my experience we have to start from a clean DAC without "digititis" even in this case. The digital problems cannot be hidden by faults in the rest of the system.
In fact before that I have a DIY 300B "cheap" SET, for fun, and in some cases the magic reminds a little bit of vinyl. And I also have a low THD amplifier for recordings that sound better with such system.
THD, DF and noise alone explains some things but not all - so "more experimentation is needed".
To add, one time I had the opportunity to witness one pre from Audiopax (I don't remember the model, but is close to the Model 88 amplifier), where the user has the opportunity to adjust the distortion of it. I could not set the same, the actual owner did not permits, but this sounded "magical" in some recordings. So I did some experiments with DIY pres and amplifiers with appreciable/adjustable THD. I was also inspired by certain sound good reports coming from systems with digital+SET. The results were mixed, but taught a lot! Vinyl and SET really are not equal but have some things in common (if vinyl teaches something, SET too). So I do not discarded the possibility, in some cases, using a deliberately distorted system, which gives an effect not far from the vinyl... and better sometimes!
In my experience we have to start from a clean DAC without "digititis" even in this case. The digital problems cannot be hidden by faults in the rest of the system.
In fact before that I have a DIY 300B "cheap" SET, for fun, and in some cases the magic reminds a little bit of vinyl. And I also have a low THD amplifier for recordings that sound better with such system.
THD, DF and noise alone explains some things but not all - so "more experimentation is needed".
After some major hesitation I removed the 0.1 Ohm resistors again (whose installing costed me some gray hair)
Again white noise playing.
a) stock +2700uF (red) over factory mod +1000uF (green)
Now the bump of the "stock" version is gone and it is more or less always better.
b) increasing the additional capacity gives only slight improvements
stock +6800uF (red) over stock +2700uF (green).
The caps are electrolytic 16V Panasonic FM. I did not see much difference in the plots for different type electrolytics of the same capacity. The polymers have not yet arrived.
Again white noise playing.
a) stock +2700uF (red) over factory mod +1000uF (green)
Now the bump of the "stock" version is gone and it is more or less always better.
b) increasing the additional capacity gives only slight improvements
stock +6800uF (red) over stock +2700uF (green).
The caps are electrolytic 16V Panasonic FM. I did not see much difference in the plots for different type electrolytics of the same capacity. The polymers have not yet arrived.
I had around 2000uF total of basic electrolytics (random trashcan content) first, then i upgraded to 16x 680uF polymers and there is a significant upgrade to sound quality. I have not moded the opamp feedback yet.
The Dam got very very close to Schiit Yggdrasil that i have here for testing.
Sry i failed at uploading pictures. Some pictures can be found here:Soren's R-2R Dac