What there may turn out to be is a need for a better dac. Trying to cover up or obscure distortion with more distortion of a different type is generally not a very good way to go.
Mark,
I listen to a lot of youtube talk and podcasts with my diy dac and I have found that I get a very desirable "non fatiguing quality" from circuits that make second harmonic. I am also using an R-core power transformer as coupling transformer and that too improves things. I found a source, cheap, and they turned out an excellent 1:1 coupling trafo using the two windings that are intended for mains. Better than any E-I I tried so far (for coupling). I also like to tinker and the newer dacs don´t really allow this to the extent I like.
-Alex
I listen to a lot of youtube talk and podcasts with my diy dac and I have found that I get a very desirable "non fatiguing quality" from circuits that make second harmonic. I am also using an R-core power transformer as coupling transformer and that too improves things. I found a source, cheap, and they turned out an excellent 1:1 coupling trafo using the two windings that are intended for mains. Better than any E-I I tried so far (for coupling). I also like to tinker and the newer dacs don´t really allow this to the extent I like.
-Alex
Even though I started with the popular non-oversampling dac projects, I am no longer into that, I find 88.2 better. And I use some first order low passes, as well as the filter function of the trafo.
-Alex
-Alex
Alexandre,
What kind of DIY dac is it? You may be aware that we have a DIY ES9038Q2M dac modding thread here and it has just the thing for you: Adjustable 2nd Harmonic Distortion! Set it to as much or as little as you like...
EDIT: However, I still think you are trying to compensate for some remaining dac distortion with more distortion from a transformer. People try to do that all the time when first learning how to record and mix records. They think 'warmth' is a function of finding just the right transformer or tube distortion, or something similar. It never works right of course, and they are wrong about where so-called warmth comes from. It is a property of the complete lack of distortion, or as low as one can get it (which isn't easy).
What kind of DIY dac is it? You may be aware that we have a DIY ES9038Q2M dac modding thread here and it has just the thing for you: Adjustable 2nd Harmonic Distortion! Set it to as much or as little as you like...
EDIT: However, I still think you are trying to compensate for some remaining dac distortion with more distortion from a transformer. People try to do that all the time when first learning how to record and mix records. They think 'warmth' is a function of finding just the right transformer or tube distortion, or something similar. It never works right of course, and they are wrong about where so-called warmth comes from. It is a property of the complete lack of distortion, or as low as one can get it (which isn't easy).
Last edited:
I made a front end which can be programmed to feed different dac chips. It uses a xilinx xc95108 cpld. I did it this way to be able to try different chips. I have spdif optical interfaces (input and output, the output clocks a RME soundcard). There is a separate reclocker board with a Tent Labs oscillator, soon to be replaced with a selected NDK from Jocko.
So you can see I am into making the whole dac, and I have different chips available here to try, with different analog stages. I have already tried the philips chips as well as the AD1865. I can tweak almost any of these to give me excellent sound, truly excellent in the sense that I would have to pay a lot if I were to buy a commercial product this good.
I have little experience with the newer dacs though.
-Alex
So you can see I am into making the whole dac, and I have different chips available here to try, with different analog stages. I have already tried the philips chips as well as the AD1865. I can tweak almost any of these to give me excellent sound, truly excellent in the sense that I would have to pay a lot if I were to buy a commercial product this good.
I have little experience with the newer dacs though.
-Alex
You should try AK4137 upsampled and converted to 11.2MHz DSD into an ESS dac clocked with a top of the line Crystek clock on it's own dedicated voltage regulator. Some internal adjustment to the ESS registers for DPLL bandwidth and HD compensation, a superbly regulated pair of AVCC supplies, and a proper I/V output stage is quite good. Well worth trying. May seem odd, ESS dacs just seem to sound better with DSD, if it is done properly in hardware to minimize jitter.
Thanks for the suggestion. Before I do that I have to compare my prototype (when it is ready) with a good commercial dac, I think.
There is one more thing. I bought most of my parts back when our currency wasn´t devalued. So, for now, I prefer to work with what I have.
Another important thing: I need my system to sound very good with pink noise. Very useful for me sometimes, to mask outside noises and aid relaxation or even concentration.
Thanks
Alex
Another important thing: I need my system to sound very good with pink noise. Very useful for me sometimes, to mask outside noises and aid relaxation or even concentration.
Thanks
Alex
+1.
Are you still having the THAT thing downstream ?
And why not use a pair of 2SK372 instead of LSK489 ?
Patrick
Hi Mark,
In other words, lay people on average, cannot be trusted to specify a system that works properly - using DACs as your example did. No doubt many will have a problem with my statement, but a cold hard look at their systems will show that in the majority of cases, this is in fact true. A look at what is available as alternative audio accessories should bear this out as well. Examples: power cords, wire lifts, green markers, CD stabilizing rings and (my favorite), the belt driven CD player. We could include bi-wiring and similar things like wire composition. Oxygen-free wire (not for long) is pretty funny, as are any claims to having directional wire, or wire that has been broken in. If any of these things had any chance of being valid, NASA and physics labs everywhere would all have instruments available to condition and determine wire direction. Then there is another humdinger. Mr. Bybee and his fanciful objects d'art.
I could go on, but ...
-Chris
Well, actually the first condition is a noise floor above that of the measuring system. What that means is that for there to be something heard, it must appear above the noise floor of the system you are measuring. I have very often had people describing things they hear that I can't hear, and later on they describe something completely different that may even conflict with their previous observations. This is a trap easily set by manipulating their state of mind. Confirmation bias. The other thing I see is someone advancing an idea. If that is by the "lead person" (their hearing is supposed to be better), other people fall in line and agree with the opinion that has more weight. This is how rumours are spread. Very popular in the high end audio circus.
I have. A distortion measuring set and an oscilloscope set up to look for glitches or notches in the output waveform. Only visible in systems that do not use a reconstruction filter (or any other term you want to give it). Also, abnormal outputs from op amps that can't slew quickly enough. Again, a good filter really reduces that problem.
If you can hear it (and I mean if you really can hear something), it can be detected and quantified. It really helps to understand the system you are examining as that will suggest areas to look in.
You have to get creative in finding possible things to test for. As I said, if you can really hear it, it is there to be seen.
Standard tests for a non-standard situation? Some standard tests will show departures from perfection, but if you are digging around in the grass (sic), you have to seriously wonder if these things are audible at all. Take the NOS crowd for example. For reasons known only to the gods of distortion and misuse of technology, they appear to like hearing the effects of their system attempting to reproduce a rather strong 44.1 KHz carrier. Given that the proper filter for that sampling frequency is a vicious 7th order L-C affair they refuse to use ... Each sampling rate has an acceptable filter order, and the entire reason we have sampling at 176.4 KHz and 352.8 KHz is so a lower order filter can be used to reduce the HF components that would otherwise be running through our systems and ruining our collective days.
In other words, lay people on average, cannot be trusted to specify a system that works properly - using DACs as your example did. No doubt many will have a problem with my statement, but a cold hard look at their systems will show that in the majority of cases, this is in fact true. A look at what is available as alternative audio accessories should bear this out as well. Examples: power cords, wire lifts, green markers, CD stabilizing rings and (my favorite), the belt driven CD player. We could include bi-wiring and similar things like wire composition. Oxygen-free wire (not for long) is pretty funny, as are any claims to having directional wire, or wire that has been broken in. If any of these things had any chance of being valid, NASA and physics labs everywhere would all have instruments available to condition and determine wire direction. Then there is another humdinger. Mr. Bybee and his fanciful objects d'art.
I could go on, but ...
-Chris
Not from what I have seen so far. Maybe the biggest and baddest is better, I dont know.
That's OK... everything is OK.
THx-RM
Last edited:
Hi Chris,
Not at all. Humans can hear somewhat below the noise floor. People might argue about how far below the noise floor, but that's about it.
No time now to address your other extensive and thoughtful comments, but I will try when I have time.
-Mark
Hello Robert.
I get it that you are able to reliably discern all these individual sounds, it just takes practice, ie millions of A/B's, including OTF A/B's.
I have taken a look at your videos Meet Mr Lounge Audio.
I note that you use different resistor types throughout the circuit (ditto capacitors) including carbon composition.
What steered you to particular resistor types in particular locations ?.
I don't agree entirely, when a sonic change is solid and learned/imprinted then it's identifiable under good ABX conditions.
It seems that Robert has done enough ABing to be fully confident in his ability to discern the kinds of changes in signature/character that his component/materials choices bring....the key is knowing by experience what to listen for.
Same with the 40cm RCA cables that I say are directional, once the changes are pointed out to the listener, 'blind freddie' would pick them reliably in ABX testing.
Magazine flowery review reports I ignore, owner flowery review reports I ignore, sensible descriptions detailing the subjective natures of changes I take notice.
When multiple developers independently report same subjective findings that's pretty good proof in my books, proof of subjective effect but not necessarily correct explanation of cause.
'Unexplained' or 'Impossible' effects are some that I have explored....the key here is to understand that there are consequential downstream effects like changes in nature of system excess noise that are moderated by for example cable type and cable direction.
Changes at source end can be very small indeed and still cause system output to change identifiably.
In the case of the RG59 RCA cables, I am well satisfied that the cables are as identical and as end for end symmetrical as is reasonably possible....ok they wouldn't pass RF application (is there an RCA that does ?) but pretty close and certainly 'good enough' for audio interconnect application.
So I reasonably trust/consider these cables to be suitable for purpose and identical/interchangeable.
In application I consistently find that there are subtle and identifiable changes caused according to installed direction.
Removing the black sheath changed the overall sound of these cables but did not remove the directional characteristic.....actually it is now more apparent.
So, now is the question, what is the cause of this behaviour ?.
Could it be the copper braid or the copper center wire ?.
If so how so...is it the copper forming/drawing process that is responsible ?.
Or, is it the dielectric material that is responsible in some way ?
Or, is it the assembly/manufacturing/extruding process including the machinery involved ?.
Or, is it that all of the above plus time gets embedded into the final assembly ?
The next question is how/why does this directional characteristic cause changes in system excess noise ?
Perhaps Mr Bybee has some idea
.Dan.
Last edited:
Scott,
Looking at your preamp schematic I see what I think should be corrected. But just to be mean I'll see if you noticed it.
RNM,
I can't imagine who is left at Crown who could talk knowledgeably about their products.
Of general interest I have been looking at relays and their topology for use in a volume control. Now as far as I can tell the two obvious choices are an R2R ladder or a tapped series string. So I guess I will have to build all three types to see if there is a significant difference. Yes three types! A constant impedance stepped attenuator was used in early passive mixers to match the coupling transformers.
Looking at your preamp schematic I see what I think should be corrected. But just to be mean I'll see if you noticed it.
RNM,
I can't imagine who is left at Crown who could talk knowledgeably about their products.
Of general interest I have been looking at relays and their topology for use in a volume control. Now as far as I can tell the two obvious choices are an R2R ladder or a tapped series string. So I guess I will have to build all three types to see if there is a significant difference. Yes three types! A constant impedance stepped attenuator was used in early passive mixers to match the coupling transformers.
Good old Tech support as it is a current model. I told him what I wanted to do and why.... he couldn't send me schematic for current model but found on the schematic the input to the power amp and went back to the feed to it... a funky opamp -- ...2043 type. And then sent me photo of where it is on pcb. and pin number etc. Thats all I need.
I hate working with smd parts.
-Richard
He has it built.
I had my modified versions simmed, and built.
A simplified universal differential or single ended phono preamp
Patrick
Max/Dan,
Let us know when you prove 0.0000001% (That's 1 billionth, -180 dB if you will) of what you claim in your long diatribe above. I'll be thrilled for your Nobel Prize. Until then, please put up or shut up.
Let us know when you prove 0.0000001% (That's 1 billionth, -180 dB if you will) of what you claim in your long diatribe above. I'll be thrilled for your Nobel Prize. Until then, please put up or shut up.
For a man of your background you know that is a sloppy throw away comment. Accept a small tut from across the pond.
- Status
- Not open for further replies.