implementation of the PCM63
Because just one schema can be attached, here the reclocking scheme.......
Because just one schema can be attached, here the reclocking scheme.......
reclocking just before the PCN63
I tried to find the schematics (a block diagram would satisfy) of the D1V3 DAC but could not find it. I saw al kinds of dedails about the printed circuit board and modifications but NO OVERVIEW!
Many people operate this DAC and gave reports and measurements about the differences between the PCM63P, the PCM63P-K, the PCM63P-K2 and last but not least the PCM63P-Y.... I would like to have an idea about the D1V3 to come to conclusions.
Could anybody be of help?
I tried to find the schematics (a block diagram would satisfy) of the D1V3 DAC but could not find it. I saw al kinds of dedails about the printed circuit board and modifications but NO OVERVIEW!
Many people operate this DAC and gave reports and measurements about the differences between the PCM63P, the PCM63P-K, the PCM63P-K2 and last but not least the PCM63P-Y.... I would like to have an idea about the D1V3 to come to conclusions.
Could anybody be of help?
Hi PAOSU,
you can look here...
http://www.diyaudio.com/forums/showthread.php?s=&threadid=97504
about D1V3 archive 😀
all the best,
a'af
you can look here...
http://www.diyaudio.com/forums/showthread.php?s=&threadid=97504
about D1V3 archive 😀
all the best,
a'af
D1V3 schematics
Hi A'af,
NO !!
I digged them deeply...... no schmatics... but I get help from Stan.
Hi A'af,
NO !!
I digged them deeply...... no schmatics... but I get help from Stan.
Compairing the different types PCM63 with the D1V3 DAC
Thanks Stan for the schematics!!!
Investigating the schematics of the D1V3 (NP D1 clone V3) DAC, I must conclude that the buffers U9 and U10 are buffers indeed (XOR's).
So, the noisy, jittery signals from the digital filter are transported right into the PCM63's .........
The best of all updates which could be made to this design is:
REPLACE U9 and U10 WITH 74HC175's and clock them with the most jitter-poor clock signal there could be found on the board.....
This update will give a TREMENDOUS enhancement on the total performance!
For details, look at the schemes of the TentLabs DAC I have sent before.
Why all these comments?
Well, investigating the doc's of the PCM63, I think the different versions of the PCM63 wil NOT give such great differences in performance.
My plans:
I ordered for two PCM63P-K's and for two PCM63P-Y's and wil investigate (measuring and listening with an experianced panel!) them.
Keep in touch.
Within a week or two I will report the results....
By the way: what about the so called 'stopped clock' to the PCM63? Read the documentation gentlemen.
Thanks Stan for the schematics!!!
Investigating the schematics of the D1V3 (NP D1 clone V3) DAC, I must conclude that the buffers U9 and U10 are buffers indeed (XOR's).
So, the noisy, jittery signals from the digital filter are transported right into the PCM63's .........
The best of all updates which could be made to this design is:
REPLACE U9 and U10 WITH 74HC175's and clock them with the most jitter-poor clock signal there could be found on the board.....
This update will give a TREMENDOUS enhancement on the total performance!
For details, look at the schemes of the TentLabs DAC I have sent before.
Why all these comments?
Well, investigating the doc's of the PCM63, I think the different versions of the PCM63 wil NOT give such great differences in performance.
My plans:
I ordered for two PCM63P-K's and for two PCM63P-Y's and wil investigate (measuring and listening with an experianced panel!) them.
Keep in touch.
Within a week or two I will report the results....
By the way: what about the so called 'stopped clock' to the PCM63? Read the documentation gentlemen.
Re: Compairing the different types PCM63 with the D1V3 DAC
Well, in theory, the better the implementation, the greater the differences, not the other way round...
In practice, if there are no bugs, even a standard dac / player will reveal all the differences.
I think enough people have acknowledged my earlier findings...
Happy spending $$$ 😀
PA0SU said:
Well, in theory, the better the implementation, the greater the differences, not the other way round...
In practice, if there are no bugs, even a standard dac / player will reveal all the differences.
I think enough people have acknowledged my earlier findings...
Happy spending $$$ 😀
Coparing the different types of the PCM63
Liebe Bernhard,
Be patient. I wil do the tests as soon as I got the chips.
Do'n worry: Ich werde ehrlich sein!
I am a researcher for 'all my life'. Such people are only interrested in the truth... and I will tell you my experiances, also if I'm wrong!
Tschüß,
Herbert.
Liebe Bernhard,
Be patient. I wil do the tests as soon as I got the chips.
Do'n worry: Ich werde ehrlich sein!
I am a researcher for 'all my life'. Such people are only interrested in the truth... and I will tell you my experiances, also if I'm wrong!
Tschüß,
Herbert.
Because of the low number of devices under test, your results will be random.
But another piece of the puzzle. 🙂
But another piece of the puzzle. 🙂
Re: Compairing the different types PCM63 with the D1V3 DAC
You mean U4 and U11 (at least that's what they are in my version of the schematics).
Both the sm5842 and pmd100 apply stopped the clock or silent conversion or whatever it is called. When LE activates the conversion, the clock/data lines are not changing state. Loading the data (burst clock) is done long before (relatively speaking)
So at that time, the influence on the buffers is not there (clock/data influencing latch output of buffer).
One can argue if jitter on the data/clock lines during the loading of the data influences the output current on the DAC, which should be steady between LE changes.
What matters is the clock fed to the dig. filter and the frequency (at least for the pmd100, where the lower freq gives less jitter).
Can't speak for the sm5842, but the pmd100 is designed to have a low jitter latch output.
So i doubt reclocking has such a huge effect. As for the tent dac, i think the reclocking is mainly done, because they used a pll and/or the sm5842 has jittery latch output (datasheet with jitter free mode is not that clear on this).
Read the documentation (?)
PA0SU said:Thanks Stan for the schematics!!!
Investigating the schematics of the D1V3 (NP D1 clone V3) DAC, I must conclude that the buffers U9 and U10 are buffers indeed (XOR's).
By the way: what about the so called 'stopped clock' to the PCM63? Read the documentation gentlemen.
You mean U4 and U11 (at least that's what they are in my version of the schematics).
Both the sm5842 and pmd100 apply stopped the clock or silent conversion or whatever it is called. When LE activates the conversion, the clock/data lines are not changing state. Loading the data (burst clock) is done long before (relatively speaking)
So at that time, the influence on the buffers is not there (clock/data influencing latch output of buffer).
One can argue if jitter on the data/clock lines during the loading of the data influences the output current on the DAC, which should be steady between LE changes.
What matters is the clock fed to the dig. filter and the frequency (at least for the pmd100, where the lower freq gives less jitter).
Can't speak for the sm5842, but the pmd100 is designed to have a low jitter latch output.
So i doubt reclocking has such a huge effect. As for the tent dac, i think the reclocking is mainly done, because they used a pll and/or the sm5842 has jittery latch output (datasheet with jitter free mode is not that clear on this).
Read the documentation (?)
Re: Coparing the different types of the PCM63
Herbert,
It is all about this, isn't it?
Grossartig und vielen Dank dafür (fantastic & many thanks) !
IY.
PA0SU said:
Herbert,
It is all about this, isn't it?
Grossartig und vielen Dank dafür (fantastic & many thanks) !
IY.
Dear Herb,
I just wanted to reply to You, but my mail bounced back.
There are fantastic things on Your website, and would be really interesting to exchange ideas! [about the tent dac]
Thanks, George
I just wanted to reply to You, but my mail bounced back.
There are fantastic things on Your website, and would be really interesting to exchange ideas! [about the tent dac]
Thanks, George
Re: actual on ebay:
OK, here is the news: The PCMs (PCM63P-K) are indeed genuine BB, used, being taken from old CD Players. Most important: one can look at the attached picture (ebay).
http://cgi.ebay.com/Burr-Brown-DAC-...ryZ14980QQrdZ1QQssPageNameZWD1VQQcmdZViewItem
I corresponded with the seller; he seems to be OK. He wrote to me that he knows it is not cheap. About 86 US $ for a pair, but genuine.
irgendjemand said:
spzzzzkt said:
OK, here is the news: The PCMs (PCM63P-K) are indeed genuine BB, used, being taken from old CD Players. Most important: one can look at the attached picture (ebay).
http://cgi.ebay.com/Burr-Brown-DAC-...ryZ14980QQrdZ1QQssPageNameZWD1VQQcmdZViewItem
I corresponded with the seller; he seems to be OK. He wrote to me that he knows it is not cheap. About 86 US $ for a pair, but genuine.
Attachments
The price seems to be too high and I suggest buyer try to contact Ruach - Singapore for a NOS PCM63PK chip. I get my new old stock from him and it is non used and later datecode in 1997.
I offer the used chip with the D1V3 for only US$20 each and the source is also from China, so you know the resonable market price here!!
I offer the used chip with the D1V3 for only US$20 each and the source is also from China, so you know the resonable market price here!!
Jitter....
Dear Bernhard, GuidoB and others...
First of all: jitter is crucial in the field of AD- and DA-conversion!
With a couple of friends we are investigating how much jitter will be permitted. My good friend Henk ten Pierick (remember this name!) is owner of a WaveCrast with which he can measure jitter down to 3 ps (pico seconds!). Our experiance is that two different clock oscillators each with a jitter of less than 3 ps may perform different! We are investigating for already two years what could be the cause. We have the idea that the frequency spectrum of the jitter is of great importance. With this in mind I have built me a DC-receiver with which I can investigate noise spectra down to -130 dBm on a distance of 10 Hz from the carrier. But this is out of the scope of this thread......
When you investigate the jitter of the latch from whatever digital filter, you will find figures in the range of 10 to 30 ps. So, (GuidoB) reclocking is absolutely necessary!!!!
One could restrict to the reclocking of the latch indeed. This is right theoretically, but during the devopment of the 'TentDAC' guys have found already many years ago that the reclokking of ALL signals to the DAC-chip gives mutch better results.
Why?
We are not sure but think that cross talk between the different digital signals is the source. So we reclock 'everything' because it costs no more effeort.....
When you do not create this crucial steady environment, you can not compare DAC's with each other (Bernhard!). You wil not find any correlation between measurements and sound impressions.
What I THINK about the different PCM63 types. I DID NOT DO any measurement up till now!
The BB-docs say that the MSB of the PCM63P could be tuned. If you do not want to do this, buy yourself a PCM63P-K. Keep in mind: the Most Significant Bit. Nothing is said about the LSB's.
The LSB comes in sight when we look at the dynamic range. The worst figure I can find in the doc's is: -36 dB at a level of -60dB.
Gentlemen, I made many recordings in the past 15 years and NEVER met a concert hall with a 'noise level' better than -60dB. In this circumstances the distortion is WORST CASE still 36 dB down! What are we talking about?
What I THINK to find in the coming weeks is that the K sounds slitely better in the peaks (of a piano). What the Y's will do, is the question, but I do not THINK to find so different performances if they meet the specs.
Not to say that I'm very curious about the different chips IN A REALLY LOW JITTER ENVIRONMENT, but as I said before I will be onnest all the time.....
Dear Bernhard, GuidoB and others...
First of all: jitter is crucial in the field of AD- and DA-conversion!
With a couple of friends we are investigating how much jitter will be permitted. My good friend Henk ten Pierick (remember this name!) is owner of a WaveCrast with which he can measure jitter down to 3 ps (pico seconds!). Our experiance is that two different clock oscillators each with a jitter of less than 3 ps may perform different! We are investigating for already two years what could be the cause. We have the idea that the frequency spectrum of the jitter is of great importance. With this in mind I have built me a DC-receiver with which I can investigate noise spectra down to -130 dBm on a distance of 10 Hz from the carrier. But this is out of the scope of this thread......
When you investigate the jitter of the latch from whatever digital filter, you will find figures in the range of 10 to 30 ps. So, (GuidoB) reclocking is absolutely necessary!!!!
One could restrict to the reclocking of the latch indeed. This is right theoretically, but during the devopment of the 'TentDAC' guys have found already many years ago that the reclokking of ALL signals to the DAC-chip gives mutch better results.
Why?
We are not sure but think that cross talk between the different digital signals is the source. So we reclock 'everything' because it costs no more effeort.....
When you do not create this crucial steady environment, you can not compare DAC's with each other (Bernhard!). You wil not find any correlation between measurements and sound impressions.
What I THINK about the different PCM63 types. I DID NOT DO any measurement up till now!
The BB-docs say that the MSB of the PCM63P could be tuned. If you do not want to do this, buy yourself a PCM63P-K. Keep in mind: the Most Significant Bit. Nothing is said about the LSB's.
The LSB comes in sight when we look at the dynamic range. The worst figure I can find in the doc's is: -36 dB at a level of -60dB.
Gentlemen, I made many recordings in the past 15 years and NEVER met a concert hall with a 'noise level' better than -60dB. In this circumstances the distortion is WORST CASE still 36 dB down! What are we talking about?
What I THINK to find in the coming weeks is that the K sounds slitely better in the peaks (of a piano). What the Y's will do, is the question, but I do not THINK to find so different performances if they meet the specs.
Not to say that I'm very curious about the different chips IN A REALLY LOW JITTER ENVIRONMENT, but as I said before I will be onnest all the time.....
Dear Herb,
I hope it works now. By the way, your post above contains all that had been written in my mail.. 😀
To all: I DO agree completely with what just had been said - and I think it would be great loss on your side if you would not consider carefully the words of PA0SU.. 🙂
Spencer,
Thank You very much - I got the package all nice!! You are really great!
Ciao, George
I hope it works now. By the way, your post above contains all that had been written in my mail.. 😀
To all: I DO agree completely with what just had been said - and I think it would be great loss on your side if you would not consider carefully the words of PA0SU.. 🙂
Spencer,
Thank You very much - I got the package all nice!! You are really great!
Ciao, George
Re: Jitter....
In a given environment the better DAC chip will always perform better, regardless of a little bit more or less jitter.
The chips are measured relative to each other, I personally don't care wether the absolute numbers are a fraction of a dB off or not.
And I found the correlation between measurements and sound impressions, that's how it all developed for myself...
The MSB of the PCM63 is of no interest as it doesn't affect the low level performance, which we are discussing here.
IMHO Jitter is detoriating sound in another range, at -90dB or so while the harmonics of the low level signal are around -60dB.
Add -90 to -60, that really doesn't matter.
Unfortunately I have no instruments up today to measure jitter.
I use Accuphase DP70 as a transport and after putting in a Kwak Clock, things improved significantly.
But the changes in sound quality from the low jitter clock have nothing to do with low level performance of DAC chips.
Two totally different things.
PA0SU said:
In a given environment the better DAC chip will always perform better, regardless of a little bit more or less jitter.
The chips are measured relative to each other, I personally don't care wether the absolute numbers are a fraction of a dB off or not.
And I found the correlation between measurements and sound impressions, that's how it all developed for myself...
The MSB of the PCM63 is of no interest as it doesn't affect the low level performance, which we are discussing here.
IMHO Jitter is detoriating sound in another range, at -90dB or so while the harmonics of the low level signal are around -60dB.
Add -90 to -60, that really doesn't matter.
Unfortunately I have no instruments up today to measure jitter.
I use Accuphase DP70 as a transport and after putting in a Kwak Clock, things improved significantly.
But the changes in sound quality from the low jitter clock have nothing to do with low level performance of DAC chips.
Two totally different things.
Bernhard,
In fact there is a point that should be clarified here. Your measurements are valid, repeatable. They show a weak point of the multibit converters: low level [non]linearity.
A side-note to PA0SU: The chips are showing the same type of difference
in low lev. nonlinearity also in my tent dac, that is, in a quite controlled jitter environment - so clearly this is a different phenomenon from jitter.
But. The title of the topic is: real or fake PCM63? Low level nonlinearity has nothing to do with factory grading - no wonder that we did not find correlation.
For real K or superior level grades we should test at FULL output level, and look for THD ~-96dB, for a K. Problem is, that You will not be able to do that with just an average sound card, and neither Your analyzer, Bernhard. Maybe mine, a HP3585, but I will have to control. And also the output stage in the test-bed dac should be up to the task. Better semi-pro sound cards would be good, like probably the EMU series.
Then the question is: does it really have a strong correlation with good sound, the better low level linearity? For me, I could not confirm that in my setup, but will have to spend more time on it. But still, if one feel it's important, then why don't just use a modern sigma-delta dac? They have excellent low level linearity, by definition.
Or, use a digital filter scheme before the multibit converter with correct dithering, like the PMD100 does, as I had shown here.
Or use an adjustable old converter, like the PCM58, and adjust the hell out of it.. By the way, on the contrary to some opinion here, I still enjoy very much the sound of my pcm58, I find it more relaxed and natural, then the pcm63... Though less attraction and fireworks...
Ciao, George
In fact there is a point that should be clarified here. Your measurements are valid, repeatable. They show a weak point of the multibit converters: low level [non]linearity.
A side-note to PA0SU: The chips are showing the same type of difference
in low lev. nonlinearity also in my tent dac, that is, in a quite controlled jitter environment - so clearly this is a different phenomenon from jitter.
But. The title of the topic is: real or fake PCM63? Low level nonlinearity has nothing to do with factory grading - no wonder that we did not find correlation.
For real K or superior level grades we should test at FULL output level, and look for THD ~-96dB, for a K. Problem is, that You will not be able to do that with just an average sound card, and neither Your analyzer, Bernhard. Maybe mine, a HP3585, but I will have to control. And also the output stage in the test-bed dac should be up to the task. Better semi-pro sound cards would be good, like probably the EMU series.
Then the question is: does it really have a strong correlation with good sound, the better low level linearity? For me, I could not confirm that in my setup, but will have to spend more time on it. But still, if one feel it's important, then why don't just use a modern sigma-delta dac? They have excellent low level linearity, by definition.
Or, use a digital filter scheme before the multibit converter with correct dithering, like the PMD100 does, as I had shown here.
Or use an adjustable old converter, like the PCM58, and adjust the hell out of it.. By the way, on the contrary to some opinion here, I still enjoy very much the sound of my pcm58, I find it more relaxed and natural, then the pcm63... Though less attraction and fireworks...
Ciao, George
I think one question we never answer is how does BB/Ti sort the PK, K2 and Y grade PCM63 and what parameters are measured to grade the chip? Anyone has any idea?
Personnally I find the K2 or KY has more extended high and low frequency. The sound stage is also more stable and separation of insturment is wider relative to PK. Music is more relax and a little bit more harmonics also. I do not know how to relate that back to measurement and measure what?
So far I find a real K2/KY will sound better than a normal PK and also the Taiwan PK chip is worst in all. My findings is that the sound is correlated to the grade of the chips.
Personnally I find the K2 or KY has more extended high and low frequency. The sound stage is also more stable and separation of insturment is wider relative to PK. Music is more relax and a little bit more harmonics also. I do not know how to relate that back to measurement and measure what?
So far I find a real K2/KY will sound better than a normal PK and also the Taiwan PK chip is worst in all. My findings is that the sound is correlated to the grade of the chips.
Re: Jitter....
Yes, even the pmd100 datasheet recommends reclocking for better performance (depending on fs and dac used). But also near the dac's, so just replacing the buffers is probably not the way to do this. Separate '175's (or similar stuff) per dac, near the dac's (maybe even picogates; separate chips per signal) are required, i guess. And their powersupply becomes important. So not just one near the DF.
In short, i doubt "TREMENDOUS" improvements. And the DF's already use stopped clock operation, hence the remark on reading documentation. I'm now using a pmd100 myself without reclocking (yet) and it sounds already very good in the test setup.
I know Henk, had long e-mail discussions with him on the inner workings of the tda1541. Also discussed here. Nice guy 😎
PA0SU said:
Yes, even the pmd100 datasheet recommends reclocking for better performance (depending on fs and dac used). But also near the dac's, so just replacing the buffers is probably not the way to do this. Separate '175's (or similar stuff) per dac, near the dac's (maybe even picogates; separate chips per signal) are required, i guess. And their powersupply becomes important. So not just one near the DF.
In short, i doubt "TREMENDOUS" improvements. And the DF's already use stopped clock operation, hence the remark on reading documentation. I'm now using a pmd100 myself without reclocking (yet) and it sounds already very good in the test setup.
I know Henk, had long e-mail discussions with him on the inner workings of the tda1541. Also discussed here. Nice guy 😎
Guido,
Maybe I should not speak up for PA0SU, but I think he was thinking of the problem of reclocking the BCK signal in case of the stopped clock operation, which would mean to apply that trick from the Tent Dac. So it's absolutely not only squeezing in a '175.
Ciao, George
Ps: When I was doing that digital PLL mod to my dac:
http://www.diyhifi.org/forums/viewtopic.php?p=26280#p26280
Then it was quite an amount of change in the sound, so maybe that "tremendous" is not so far from reality..
Ciao, George
Maybe I should not speak up for PA0SU, but I think he was thinking of the problem of reclocking the BCK signal in case of the stopped clock operation, which would mean to apply that trick from the Tent Dac. So it's absolutely not only squeezing in a '175.
Ciao, George
Ps: When I was doing that digital PLL mod to my dac:
http://www.diyhifi.org/forums/viewtopic.php?p=26280#p26280
Then it was quite an amount of change in the sound, so maybe that "tremendous" is not so far from reality..
Ciao, George
Attachments
- Status
- Not open for further replies.