Re: Power Supply for the TDA1543
Oh yes , of course .
I have unregulated supply at 30 volts. Then a discrete op amp with BJT and a Mosfet into 14 volts (derived directly from passdiy pearl ono stage) and a source follower for 6.5 volts . Big gate resistor used at the gates of the Mosfets to slow down the input capacitance.I like it , but consider I'm playing with resistors at the audio outputs and never tried the op amp.
Elso Kwak said:
Hello Stefano,
May ask what power supply you now are using for the TDA1543?
Oh yes , of course .
I have unregulated supply at 30 volts. Then a discrete op amp with BJT and a Mosfet into 14 volts (derived directly from passdiy pearl ono stage) and a source follower for 6.5 volts . Big gate resistor used at the gates of the Mosfets to slow down the input capacitance.I like it , but consider I'm playing with resistors at the audio outputs and never tried the op amp.
Re: Jitter to its merits
You're basically right but don't forget that jitter on a constant amplitude signal would do absolutely nothing. Jitter moves the sample transitions, hence its effect is proportional to the difference between two adjacent samples. You could say jitter induced problems are proportional to the derivativ of the signal.
Maybe this is why jittery digital can have good bass but always has problems in the midrange/treble ?
The HF bands are also where the details (ambience, etc.) lie.
Also to be on topic : I consider many things important in a DAC design : quality of IV conversion, powersupply, layout and grounding, etc. But if no jitter reduction measures are taken, all other efforts are wasted. Reducing jitter is mandatory, but you still need to address the other parts of the conversion correctly.
Pjotr said:
You're basically right but don't forget that jitter on a constant amplitude signal would do absolutely nothing. Jitter moves the sample transitions, hence its effect is proportional to the difference between two adjacent samples. You could say jitter induced problems are proportional to the derivativ of the signal.
Maybe this is why jittery digital can have good bass but always has problems in the midrange/treble ?
The HF bands are also where the details (ambience, etc.) lie.
Also to be on topic : I consider many things important in a DAC design : quality of IV conversion, powersupply, layout and grounding, etc. But if no jitter reduction measures are taken, all other efforts are wasted. Reducing jitter is mandatory, but you still need to address the other parts of the conversion correctly.
Re: Who said it has good bass?
Of course !
We tried recently the same DAC with and without reclocking -- and the effect was obvious. Without reclocking, as soon as there is more than a few instruments playing, everything is muddied in the center of the soundstage and all detail is lost, it sounds harsh and sibilant ; whereas with reclocking, all instruments kept their position and everything stays "normal" even in the more complex moments of the music...
Jocko Homo said:
Of course !
We tried recently the same DAC with and without reclocking -- and the effect was obvious. Without reclocking, as soon as there is more than a few instruments playing, everything is muddied in the center of the soundstage and all detail is lost, it sounds harsh and sibilant ; whereas with reclocking, all instruments kept their position and everything stays "normal" even in the more complex moments of the music...
While I was looking for a 'repeatable source' that wasn't as annoyingly and teeth-griningly bad as CD performance (for doing loudspeaker design work), I decided to have a go at clocking jitter and see what I could do. One day, the lighting I was looking for struck. I did the dirty deed of correcting the clock in my CD player in the manner that had come to me..and then found that the same CD player (highly modded already) suddenly sounded much like my Linn LP12 turntable. Finally, a nice sounding digital source....
I proceeded to try this unit out against the Linn CD12, and other similiarly priced items. It beat out all of them in terms of sheer musicality and listenability. Everyone who hears it mistakes it for analog and thinks the Turntable is what is being played. It has the liquidity of analog. This, when being used in what is probably one of the most (if not the most) revealing audio systems in existence.
My unit was (and still is!!) a lowly modded NAD 5000. BUT..the clocking circuit has the jitter reduced to vanishingly low levels. From what I am hearing and have been hearing for the past 6 years, I suspect it has lower jitter than has ever been measured. Enough to show how critical it can be to get to low jitter, and how much more quality is hiding in digital reproduction, if the Jitter is fully taken care of. I've heard those $10-20k digital units, and frankly, I'm unimpressed. The clock alone will make a 44.1k cd unit outperform -in terms of sheer musicality- any highly modified/stock/whatever $$,$$$$ SACD unit. That's how important low jitter is in a clock. Hugely important. One of those things - that once you've heard it... there's no going back.
As for the circuit itself, I have been sitting on it for six years, and yes, it most definitely is a new method and approach, completely unlike anything else out there. Still. By far.
I'd love to share it, but sadly cannot. Not until the patent has been filed. Then of course, I will publish and gladly have erery DIY'er who wants to use it, use it. I'm hoping to be able to get to attending to this patent sometime this year.
I proceeded to try this unit out against the Linn CD12, and other similiarly priced items. It beat out all of them in terms of sheer musicality and listenability. Everyone who hears it mistakes it for analog and thinks the Turntable is what is being played. It has the liquidity of analog. This, when being used in what is probably one of the most (if not the most) revealing audio systems in existence.
My unit was (and still is!!) a lowly modded NAD 5000. BUT..the clocking circuit has the jitter reduced to vanishingly low levels. From what I am hearing and have been hearing for the past 6 years, I suspect it has lower jitter than has ever been measured. Enough to show how critical it can be to get to low jitter, and how much more quality is hiding in digital reproduction, if the Jitter is fully taken care of. I've heard those $10-20k digital units, and frankly, I'm unimpressed. The clock alone will make a 44.1k cd unit outperform -in terms of sheer musicality- any highly modified/stock/whatever $$,$$$$ SACD unit. That's how important low jitter is in a clock. Hugely important. One of those things - that once you've heard it... there's no going back.
As for the circuit itself, I have been sitting on it for six years, and yes, it most definitely is a new method and approach, completely unlike anything else out there. Still. By far.
I'd love to share it, but sadly cannot. Not until the patent has been filed. Then of course, I will publish and gladly have erery DIY'er who wants to use it, use it. I'm hoping to be able to get to attending to this patent sometime this year.
I love your screen goo idea ! I'd buy some if it were cheaper...
I can help you with your patent if you like, my father was head of a Patent Office centre in France.
Also I have found a funny way of testing synchronous/asynchronous reclocking !
My DAC feeds its clock to a CD723 player which feeds its SPDIF data synchronously to it. However I have discovered this player is not bit-accurate so I don't use it anymore as a digital source (see my website here) and use my computer instead. Still, I have connected the player's SPDIF out to the computer's SPDIF in, so the computer is in sync with the DAC master clock.
Now if I switch off the player with the remote, the soundcard loses the input signal and reverts to Master Mode. So I can switch between synchronous and asynchronous with a remote control ! If that's not Luxury
My clock is 8.something MHz (originally for CD723). I know Elso has experimented with asynchronous reclocking. I have checked in the scope, it works as I say (when the CDP is on, the BCK coming from the CS8412 in the DAC is synchronous with the master clock, when it's off, they slide quite fast).
And... so far I can't hear a difference... Any hints Elso ?
I can help you with your patent if you like, my father was head of a Patent Office centre in France.
Also I have found a funny way of testing synchronous/asynchronous reclocking !
My DAC feeds its clock to a CD723 player which feeds its SPDIF data synchronously to it. However I have discovered this player is not bit-accurate so I don't use it anymore as a digital source (see my website here) and use my computer instead. Still, I have connected the player's SPDIF out to the computer's SPDIF in, so the computer is in sync with the DAC master clock.
Now if I switch off the player with the remote, the soundcard loses the input signal and reverts to Master Mode. So I can switch between synchronous and asynchronous with a remote control ! If that's not Luxury
My clock is 8.something MHz (originally for CD723). I know Elso has experimented with asynchronous reclocking. I have checked in the scope, it works as I say (when the CDP is on, the BCK coming from the CS8412 in the DAC is synchronous with the master clock, when it's off, they slide quite fast).
And... so far I can't hear a difference... Any hints Elso ?
Pjotr said:Hi,
First let me correct a typo error, I wrote “1 / (44.1e3 x 16^2) = 350 ps “. This must be of course 1 / (44.1e3 x 2^16) = 350 ps
Back to:
Apart from theoretical issues, from experience I can say that a jitter reduction (at the clock signal that is reponsible for the actual conversion) from say 200ps to 20 ps is clearly audible. It brings more detail, a more relaxed sound and etc. etc. So the preoccupation with jitter is not overdone IMHO. But I doubt a reduction from 20ps to 2 ps will be a big improvement with regard to the whole DA conversion process, if it is audible at all.
Another important point is PSU noise, or more precisely noise on the DAC’s reference voltage. The output of the DAC is related to its Vref, whether it is a parallel current out DAC or a Delta-Sigma DAC. In some sense noise on Vref has more or less the same influence as jitter, it modulates the output voltage of the DAC. Since with most audio DACs Vref is taken from the PSU we can also put up the question:
“Why the preoccupation with PSU noise?”
Cheers
Hi all
can anyone explain the huge gap between calculated jitter and much lower jitter values still audible ?
DACs and supply noise: Vref modulation is different from jitter inducing distortion
Ciao
Hi Guido,
The ear is a mysterious sensor and the brain an even more complex “sound processor”, most is not understood very well.
Did wrote "more or less". It is in the light that jitter as well as Vref noise modulates the signal. Regarding signal related jitter no it is not the same. Although, if you look at supply modulation due to the output loading ... ? Jitter is rather complex. There are better explanations around on the web then can be posted here and you know very well.
Cheers
Guido Tent said:
The ear is a mysterious sensor and the brain an even more complex “sound processor”, most is not understood very well.
Did wrote "more or less". It is in the light that jitter as well as Vref noise modulates the signal. Regarding signal related jitter no it is not the same. Although, if you look at supply modulation due to the output loading ... ? Jitter is rather complex. There are better explanations around on the web then can be posted here and you know very well.
Cheers
New clocking method
Prompted by KBK's post re. a new clocking method, I've been letting my imagination wander ....
How about using a sample and hold type circuit after I/V conversion from an R2R dac, and clocking that from the Word Clock? Any jitter in the filter and DAC chip would be nullified. The Word Clock itself could of course be derived from a stable master oscillator.
What do you think? This idea has just popped into my head, so please be kind in case I have missed something obvious.
Prompted by KBK's post re. a new clocking method, I've been letting my imagination wander ....
How about using a sample and hold type circuit after I/V conversion from an R2R dac, and clocking that from the Word Clock? Any jitter in the filter and DAC chip would be nullified. The Word Clock itself could of course be derived from a stable master oscillator.
What do you think? This idea has just popped into my head, so please be kind in case I have missed something obvious.
Guido
That is an interesting question. I guess it depend on the nature of the jitter and if the jitter is constant value or random or has wander. If the jitter is constant, that should make it more discernable and random jitter.
As for Jitter vs. VRef noise; Jitter can effects all bits while reference noise should only effect the lower order bits. I do not know if there has been a study that has looked at the minimum audible jitter, with so many variables what a task. Knowing that jitter is audible and getting rid of it good seems to be the best we can do.
In Telecom, I have measured jitter performance and how it affects some of the end application and systems we run. However, the effect of course is different on different types of systems because of the quality various receivers. Again the best we can do is beat it down to
That is an interesting question. I guess it depend on the nature of the jitter and if the jitter is constant value or random or has wander. If the jitter is constant, that should make it more discernable and random jitter.
As for Jitter vs. VRef noise; Jitter can effects all bits while reference noise should only effect the lower order bits. I do not know if there has been a study that has looked at the minimum audible jitter, with so many variables what a task. Knowing that jitter is audible and getting rid of it good seems to be the best we can do.
In Telecom, I have measured jitter performance and how it affects some of the end application and systems we run. However, the effect of course is different on different types of systems because of the quality various receivers. Again the best we can do is beat it down to
Jitter Sonics
Originally posted by jewilson
Guido,
That is an interesting question. I guess it depends on the nature of the jitter and if the jitter is constant value or random or has wander. If the jitter is constant, that should make it more discernable than random jitter.
More the other way around I reckon.
A perfectly stable clock of course becomes inaudible.
A clock with constant frequency jitter adds a single kind of dirt.
A clock with dynamically varying frequency is the most audible.
Depending on the nature of this dynamic variation, the resulting audio output can be disturbing, or musical and satisfying, but never as clean as a really stable clock.
Dead clean can be sterile and uninvolving, hence different preferences for different clock/clock psu circuits - ie none of them are perfect.
Eric.
Originally posted by jewilson
Guido,
That is an interesting question. I guess it depends on the nature of the jitter and if the jitter is constant value or random or has wander. If the jitter is constant, that should make it more discernable than random jitter.
More the other way around I reckon.
A perfectly stable clock of course becomes inaudible.
A clock with constant frequency jitter adds a single kind of dirt.
A clock with dynamically varying frequency is the most audible.
Depending on the nature of this dynamic variation, the resulting audio output can be disturbing, or musical and satisfying, but never as clean as a really stable clock.
Dead clean can be sterile and uninvolving, hence different preferences for different clock/clock psu circuits - ie none of them are perfect.
Eric.
Mr. Feeback
I don't quite agree with your statements. However we are all on the same page.
While you call it dirt I call it non-linear distortion.
In addition, I have no interest in listening to clocks, just music.
Now Jocko got it down to a science, he can here the modulation I just know when it sucks.
I don't quite agree with your statements. However we are all on the same page.
While you call it dirt I call it non-linear distortion.
In addition, I have no interest in listening to clocks, just music.
Now Jocko got it down to a science, he can here the modulation I just know when it sucks.
Re: Yes....
I meant the jitter noise in one ear and music subject to the same jitter in the other ear in order to learn/hear correlations.
I have done/do sort of this this at the local community radio station - with one cocked towards the sounds spilling from the on-air studio and the other ear cocked toward the foyer off-air monitor speakers I can do a real time comparison and this gives me a good gauge when fine tuning the final agc/compressor box in the equipment room.
The final test is when I get home and listen to much better gear than in the station.
Eric.
Can you expand on that - ta.Jocko Homo said:
I meant the jitter noise in one ear and music subject to the same jitter in the other ear in order to learn/hear correlations.
I have done/do sort of this this at the local community radio station - with one cocked towards the sounds spilling from the on-air studio and the other ear cocked toward the foyer off-air monitor speakers I can do a real time comparison and this gives me a good gauge when fine tuning the final agc/compressor box in the equipment room.
The final test is when I get home and listen to much better gear than in the station.
Eric.
Re: Jitter Sonics
Eric,
I agree with your observations, the same holds for distortion of tube amps - whatever one prefers, but I tend to go for what you call "clean" sound. A clean clock by the way contains only noise at the lowest level, no spurious. Unfortunately lots of clocks suffer from 50 and 100 Hz.....
Back to audibility: Jitter can affect the analog output amplitude in a much more subtle way compared to the lsb of the DAC. This to me is the reason why the calculation of audible jitter based on lsb is worthless.
You may assume a 120 dB linearity requirement and come up with < 1ps jitter requirements.......
Ciao
mrfeedback said:
Eric,
I agree with your observations, the same holds for distortion of tube amps - whatever one prefers, but I tend to go for what you call "clean" sound. A clean clock by the way contains only noise at the lowest level, no spurious. Unfortunately lots of clocks suffer from 50 and 100 Hz.....
Back to audibility: Jitter can affect the analog output amplitude in a much more subtle way compared to the lsb of the DAC. This to me is the reason why the calculation of audible jitter based on lsb is worthless.
You may assume a 120 dB linearity requirement and come up with < 1ps jitter requirements.......
Ciao
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.