Reclocking balanced PCM63

Jump to Latest
#141
PA0SU said:


You are an impulsive man! Next time 'count till 10' as my mother often said when I was angry..... Also a good method is: ask the writer what he did mean precisely before blaming him.......
Do not forget that many people here (and I am one of them) write in a foreign language, so it always could be that a statement is annoying or even insulting for a maiden speaker as you.

But, I accept your apologies.

Herbert.
Click to expand...


Herb,

Impulsive - perhaps, Frustrated - most definitely. I asked very politely for Dr. H and Yourself to start an new thread to continue your very off topic discussions of your "principes" and the methods of connecting transports and dacs. Neither of you have had common courtesy to do so. My irritation that you had both plowed on regardless of my request was major factor in my response. So perhaps you should also take something from this incident - hijacking threads is extremely poor forum etiquette. If you want to discuss something removed from the subject of the thread start another thread for your discussion.

added:
wikipedia recently deleted this entry on Thread Hijacking but neatly summarizes, and from my perspective the final line is particularly true.

Thread hijacking in internet forum communication is the act of steering a discussion off-topic by discussing a subject entirely unrelated to the subject at hand.

While this can be an intentional act of trolling, it is often accidental - caused by participants in the discussion responding to a throwaway remark, thus taking the thread off at a tangent to the original subject matter. The results often provoke a feeling of resentment from the author of the original post.
Click to expand...

or even better the DIYAudio forum rules:


3) No threadjacking. Threadjacking is the practise of stealing another's thread by posting off-topic replies such that the original topic becomes diluted or lost. Off-topic posts, and even more importantly replies to off-topic posts, are welcome, but should also address the original thread topic.

Some of the greatest discussions have come as a result of off-topic replies. If something interesting does arise that warrants further discussion then continue your discussion privately over email, or even better start a new thread and link to it. This rule will of course be used with discretion.
Click to expand...
 
#142
Re: the DIR9001 PRO thing

caine28 said:
Hi,

But: the much better way seems to be not to use the SPDIF in the normal way. Taking a better clock is everyway much better.
I cannot say if a VCXO is better then a clock-signal from the Transport to the DAC, or from the DAC to the Transport, what
also would be a Idea...
Click to expand...

Stick to one clock: an XO, whatever you do. A VCXO is nearly always worse than a good XO.
The only disadvantage of putting the XO in the DAC (and connect it back to the transport) is that sooner or later the transport will be powered on without a clock signal. Many transports do not survive this.

I would say with the original "bad" Clock in the Transport and sending it to the dac you probably have no improvement over using two "bad" clocks...
Click to expand...

I agree for sure

But: I havent done this way ........... on the market who
sell **** for gold, Tent sells things they are let´s say worth the price. What can we want more...
Click to expand...

The only way to judge is to measure the jitter or phase noise of the oscillator in question. The needed measuring equipment is very expensive. Some people here have the possibility of doing this with their home brew equipment (as I do) or with professional equipment of their employer (as I do, to verify my home brews).
Audio tests are often very deceptive....

One thing I want to think about:
Is there a generally thing that 256fs is less problematic then 384fs, cause the frequency is lower.
What can we say about this basics ?
Click to expand...

Not really. The difference in frequency is small. Some CD-players (combinations with DVD) run on 24 MHz or higher. This will be more problematic.

How do you use the SM5842APT ?`
* jitter-free or normal mode ?
* Dither on or Dither off ?
Whats your oppinion on the differencies in sound ?
Click to expand...

I copied the diagram from the TentDAC (as I call it) and they use DITHN = low, so dithering = on and SYNCN is not connected and so is high (internal pull up resistor) which means: jitter-free mode.

I hope the threesome did a good job. Some solutions you should believe from others, otherwise your life will be too short......
 
#143
spzzzzkt said:

Herb,

Impulsive - perhaps, Frustrated - most definitely. I asked very politely for Dr. H and Yourself to start an new thread to continue your very off topic discussions of your "principes" and the methods of connecting transports and dacs.
Click to expand...

Sorry, I have missed this totally.......

Neither of you have had common courtesy to do so. My irritation that you had both plowed on regardless of my request was major factor in my response. So perhaps you should also take something from this incident - hijacking threads is extremely poor forum etiquette. If you want to discuss something removed from the subject of the thread start another thread for your discussion.
Click to expand...

I stepped in at posting 10 (because of an e-mail I got). If you look at your own postings up till 17 than the question arises: who was hijacking the thread? You did more or less yourself in my opinion.....

Please tell me in which posting you asked us to start another thread.
 
#145
New thread

To All,

I started a new thread, called "From Transport to DAC - the different modes", to coinside with Paul' wish to have this Thread only for the original topic.

Dr H. (Ryan), PA0SU (Herb), caine28 (Manfred) and all others - please continue discussion in the new thread.

Many Thanks & Greetings,

IJ.
 
#148
knock it off guys!

At least until VARIAC will take a look into this thread, and as I am not allowed to use the captain or the cop hat (what a pity!) how about the following suggestions:

From now on, PLEASE only diy infos concerning the topic. OK? many thanks!

:grouphug:

p.s. this is all about fun :cheers: & music :Piano:

yours, short before :xmastree:,

IJ
 
#150
please see the other thread

Hi Telstar,

we are trying to separate the subjects "Reclocking Balanced PCM63" from the other topics which are not directly belongs to it – like Herb’s excellent Principia4 (XO, XTI/digf. etc.).

Please see Post #145 above. I also allow myself to copy your mail into the new thread: "From Transport to DAC - the different modes". I am sure that herb will answer your post there. Many Thanks & Greetings, IJ
 
#153
thanks for all the posts and pity about the confusion.

Paul, would be interested to see your progress on this topic, but hoped we could find an optimal clock set-up in the same thread.

PS: Paul, since your transport is also at 16.xxx and as Manfred points out, the 5842 can take the 16.xxx if CKSLN is pulled high (+5v), why not try this option as opposed to the VCXO?

PPS: Hope you're not p*ssed abou the PS, just not sure if you're going to read the other thread...
 
#154
Telstar said:
Hi Herb,

I think I could just use Tent vcxo:
http://www.tentlabs.com/Products/DACupgrades/XODAC/index.html
in my dac.
And the tent link:
http://www.tentlabs.com/InfoSupport/page25/page25.html
between it and the reclocked transport.

It should come close to your solution, right?
Click to expand...

Telstar, sorry that I totally have forgotten to answer you. :worship:

Yes, the TentLink could even be better because the XO is in the DAC, but you need an (extra) VCXO-PLL in the transport (to be sure the transport never will be powered on without a clock).
If the coax-connection for the clock is not too bad, my solution is a bit more simple.......
 
#155
As promised, I tried the reclocking:

Using the 16.xx from the transport fed to XTI on SM5842 and inverting the clock twice as a clock input for 74HC174, I reclocked all the inputs to the PCM63.

I tried just one channel and can now compare (in mono) with the other channel.

Verdict: A really great mod! 😎
Much more open sound, much more detail in high frequencies.More results to follow.

Thansk to Hern, Paul, Manfred and IJ for sugeestions/guidance.😀
 
#157
Tent-Link

PA0SU said:
Yes, the TentLink could even be better because the XO is in the DAC, but you need an (extra) VCXO-PLL in the transport (to be sure the transport never will be powered on without a clock). If the coax-connection for the clock is not too bad, my solution is a bit more simple.......
Click to expand...

Dear Herb,
I have some questions about this. For now, I placed them in the other thread - "From Transport to DAC - the different modes". May be you could please take a look there. Thanks!

http://www.diyaudio.com/forums/showthread.php?postid=1684697#post1684697

IJ.
 
#158
Hi Herb,

I am modding the same DAC that Paul is using, namely the D1V3 DAC. It uses PCM63 dacs in balanced mode. In broad terms,
CS8412
SM5842
74HC86
PCM63
Discrete IV stage


No schematic with me now, but it's quite simple:

1. The transports clock (16.xxx MHz) is inverted twice and fed to the DAC.
2. At the DAC, the clock is fed to pin XTI of SM5842. While this digital filter will normally use 11.xxx MHz, it can be made to use 16.xxx by setting the CKSLN pin HIGH. Note that MCK from the CS8412 receiver is not being used anymore.
3. The same clock signal is then fed to the clock input of a 74HC174.
4. The BCK, data and LE signals from the filter (SM5842) are fed to the 74HC174 and from there to the PCM63's.
5. Note that I have done 2 flip flops for each signal.
 
#159
Dr.H said:
Hi Herb,

the D1V3 DAC. It uses PCM63 dacs in balanced mode. In broad terms,
CS8412
SM5842
74HC86
PCM63
Click to expand...

Originally no VCXO-PLL in it? You will not need it any more, but just for the record.


..........................
5. Note that I have done 2 flip flops for each signal.
Click to expand...

You mean two flip flops in sequence? Did you also separate the data flip flops from the clock flip flops phisically?
 
#160
The original DAc did not use a secondary VCXO-PLL. I have ordered a VCXO (11.xxx MHz) from Tentlabs, but now since the direct link from the transport works, it seems like a waste.


I used a 74HC174, which allowed me to do the following for each signal:

D creates Q which is fed to D1 whoch creates Q1. I then take the output from Q1.

I did this for all 3 lines in one package. Is this the best way or should I run 3 seperate 74HC74's, once for each line?
 
You must log in or register to reply here.
Top Bottom