PCM63 DAC mods

Dr.H

Member
2002-04-08 6:41 pm
Not sure how many of you use this excellent DAC. I have 4 in my Theta Gen Va (2 per channel, running in balanced mode).

I have started the process of getting the most from the chip and yesterday linked up an Andy Weeks (ALW) super regulator to the +5V analog supply of the chip (obviously disconnected the old one).

Big difference in bass response, which has become cleaner and less bloated.
Also, the "space between the music", i.e. the silences are quieter.

My 2c
Ryan
 

euro21

Member
Paid Member
2004-09-17 12:16 pm
Budapest
Mods of my Lite DAC-50 (PCM63P-K DAC, 6922 SRPP):

- HEXFRED diodes in rectifying stages;
- 5H choke in HV stage (C-L-C);
- Power supply bypass capacitors (MKP10);
- NOS Tesla E88CC in SRPP;
- ClarityCap coupling capacitors
- Coupling capacitor bypass by 0.1uF VCap (teflon);

and the best:
- I/V resistor change to Sowter 8347 transformer (1:18, 2K0 Caddock resistor in secondary).
 
DF1700 ??

Hi Eddy,

I don't get it: Would NOS + DF1700 become at the “end results” the better sounding configuration??

The DF1704 is such an excellent Digital Filter! The DF1700 is really nothing great; there is no any comparison between them.

"Going backwards" with the Digital Filter should have a very good reason for doing so.

Meanwhile, any news about your "K2" ?

Regards, IY.
 

eddylws

Member
2006-08-31 6:07 am
Hi irgendjemand,

thanks for you info.

You mean i should go for DF1704 instead of DF1700?
I am help my freind to build up a DAC, do you think Lite DAC50 is a goo idea? Appreciate your Advise.

That is some technical problem still need some time to figher out. I will post it once i have review on it, but i am very sure that the different is more then 70% if compare with PCM63P-K. It sound very dynamic, batter sound stage & background sound very detial.

Thanks

Eddy Lim
 
Eddy,

I don't know the Lite Dac50, I am sorry. May be Manfred Hiller (you know him I think) will be able to tell you more, also about the combination of NOS & DF. By the way, he is the one who gave me the information about using the DF1704 correctly (in SLOW mode!) instead of the 1700.

Meanwhile have success with the PCM63P-K2 & please keep us posted.


Dr.H,

The steps between K and K2 (and Y) are so big that it might be hard to tell, what will be the actual impact; it would be interesting to try it however indeed by all 63’s categories.

As eddy just wrote above, his K2 sounds very dynamic, the sound stage is better & background sound is very detailed. I would add to this your remark – “the silences are quieter” by K2 and also – there is more body in the sound (the all performance of the K2 is somehow similar to the "Y").

My own impression: The higher the selection level goes, the bigger, cleaner and detailed (yes, all the phenomenon comes here together!) the sound becomes, also less aggressive.

What will be hard to tell is the exact “amount” in this combination of “more body” and “cleaner”. With the 63’s “Y” the sound is the biggest, but paradoxally, I don’t feel the necessity for a cleaner / less bloated sound. The Chip seemed to me to be in perfact balance.

But, as you enjoy such an improvement with the regulators – this is all what counts actually.
 
Yes, I am sell the -K2 and -Y of PCM63P here :)

I will try to make a comment as objective i can be.

I already try 4 version (better to say grade) but nothing from BB or TI about performance, so better I still say version.

The plain of PCM63P in early '90 makes me, wow this chips are a trully good DAC, but in this era the price is quite high!!
(but i still noticed, thats digital!)

Found the -K version 7 years later, and realized this -K have much more dimension and detail, thought this end and the digital music only can reach final step here.
(satisfied and tweak the circuit, power supply, analog stage and others to hope better result)

Know BB have the -K2 and -Y version but never find anywhere, even the info about this.

Then BB acquisition by TI, thought all is over. Hopeless to effort to look and search about this two version (-K2 and -Y)

Just got a small info and overheard, this 2 version are for profesional equipment. :bigeyes:

Later (around a year ago) i found this 2 DAC (-K2 and -Y) and try this on my own rigs.
Can not believe my ears this 2 chips sound AMAZING!!
Sounds like "not" PCM63!!

The details, dynamic, harmonics, character of the sound is different!!

After almost 20 years of tweaking my rigs contain PCM63. Only this, make very worthed modification to my DAC.

I know this PCM63 is a good DAC, but with -K2 and -Y. I can say this an excellent DAC!!

all the best,
a'af
 
Better then "K", for sure

Dr.H,

A'af said:

Can not believe my ears this 2 chips sound AMAZING!!
Sounds like "not" PCM63!! The details, dynamic, harmonics, character of the sound is different!!
After almost 20 years of tweaking my rigs contain PCM63. Only this, make very worthed modification to my DAC.

This correspond to the remark on Post #9; the "Y" is surely miles away from the “K”. If you can put your hand on them, go for it.
 

Dr.H

Member
2002-04-08 6:41 pm
More good news:

Pin1 benefits frm bypassing with at least 100uF. I have used a 180uF OSCON in the position and once again, small details that I have NEVER heard before start to make their presence kown. The benefit seems to be focused in the range 3k-10k, but I am just estimating.

Theta chose to use just 100nF in that position.

I've also added a further 180uF to the 100uF Theta uses on pin 4, but I am undecided as to whether there is an advantage in doing so. Will reoprt soon.

There remains just 2 tweaks left on the PCM63:
1. Feed the -5V analog power supply from an ultra low noise supply (like the ALW)
2. Feed the digital +5V line from a low noise supply.

Not sure if any of these wll bring benefits, but in the data sheet for the DAC, burr brown say that there is no benefit in seperating the analog and digital supplies (which Theta does anyway), but that the most important thing is to make sure that the supplies are as clean as possible.

My 2c
Ryan
 

Dr.H

Member
2002-04-08 6:41 pm
More good news!

I simply fed the clean analog +5V to the digital +5V pin and once again, the noise floor has dropped even further. Even my wife noticed the fact that the piano on the dirty digital supply sounded wooly (all relative of course!) compared with the clean supply.

In addition, the sound seems more layered and it was eaiser to follow various lines in the music.

I'm not boasting, but the system is now resolving things like the creaking of the pianists chair (David Benoit, Letter to Evan, track 9) and the sound of violin stops/sharp changes of bow direction (same album, track 5).

FWIW, the rest of the system is:

1. Theta Data III drive, Wydeye balanced digital cable (which is marginally better than the Madrigal Fat boy)
2. Theta Gen Va DAC (highly modified with ultra low noise ALW (Andy Weekes) power supplies on the digital inputs, Van den Hul the Second Balanced cable.
3. Carlosfm's implementation of the AD815. Replacing my old pre (627 with BUF634) with the AD815 was one of the best upgrades I've had in my system.
4. The balanced to single end converters used in the pre are THAT1200, the best I've heard and far outperforming any competitors (SSM, Analog, etc). The THAT1200 was created in collaboration with Jensen.
5. The volume pot in the pre is based on the Lightspeed Preamp, and uses light depedant resistors. A major leap forward in transparency!
6. Modified Royd Doublet speakers (new 30kg bases added, 5 9's silver internal wiring).
7. Speaker cable is a solid core, 28 gauge diy design
 

ims

Member
2008-01-17 8:55 pm
Hi Dr.H

Glad to hear you made some progress on your tweakings. :up:

I believe you got good results with ALWSR implementation in your DAC. Being me, I have 3 pcbs with populated components waiting to solder them and want to try to use them too to my CDPro2 and my DAC. ;)

But have you ever consider to use passive error correction which is used at another regulators like TeddyReg instead of ALWSR (using opamp as active error amplifiers) to your DAC? I know that TeddyReg needs longer time to settle but I hope I can hear other opinions from you when you decided to try ALWSR instead of TeddyReg.

Instead using 150uF Sanyo OS-CON at pin3, have you consider to use lower value such as 47uF or another value below 100uF as indicated it at datasheet? If I were you, I will use another caps instead of OsCon in that pin. Pin3 is Buried Zener Reference, and I assumed that you want to decouple that pin, so maybe better use low impedance caps like Elna Silmic RFS or Pana FM. Even Pana FC is better than Oscon for this job. IMO.

Thank you for sharing your findings and hope we could continue hear from another good news from you. :)

Happy Tweakings!
 

samoloko

Member
2006-06-23 10:45 am
Mods of my Lite DAC-50 (PCM63P-K DAC, 6922 SRPP):

- HEXFRED diodes in rectifying stages;
- 5H choke in HV stage (C-L-C);
- Power supply bypass capacitors (MKP10);
- NOS Tesla E88CC in SRPP;
- ClarityCap coupling capacitors
- Coupling capacitor bypass by 0.1uF VCap (teflon);

and the best:
- I/V resistor change to Sowter 8347 transformer (1:18, 2K0 Caddock resistor in secondary).

would anybody explain why there Is a need for double gain here - with step up transformer and after this with srpp
also would anybody point out to best pcm63 implementation diy dac