Simple, good quality DAC

[Peter Daniel]

I just modified my Marantz CD-94 and as a stand alone unit it is better than used as a transport feeding my recent CS8420/PDF1704/parallel PDF1704 DAC. I'm still using original PS inside the player, while my DAC had 8 transformers and at least 13 local PS based on LT 1035/1085 (or whatever numbers) chips.

The only thing I did to Marantz was bypassing digital filter, simplifying output stage to a single OPA627 per channel, removing DC reducing resistors network right at DACs output, changing all electrolytics to HFQ and Cerafines, adding local bypassing (Goudreu triplets). Well, the results are impressive and that player is producing the best bass I've heard so far. Sometimes simple ways are better indeed.

[miguel2]

Not at all. The CD48 uses the 4 X Oversampling Digital Filter of the CD Decoder Chip. The Signal after that Decoder/Digital Filter is still I2S, but at 176.4KHz and with a Digital Oversampling Filter already applied. If you where a cynic you could use a 192KHz capable DAC Chip with internal 4/8 Times oversampling from BB/TI or Cirrus Logic (aka Crystal) and claim you modded CD Player does syncronous upsampling to 176.4KHz and then further 4/8 Times oversampling.

Hey Kuei,

I understand that if the I2S signal is already oversampled, a simple TDA1543 design is out of the question. And as I am not a cynic (I guess) I have to ask you if using a 192KHz chip would increase sound quality significantly. And which chip in which configuration (a schematic here would be just great)? Do you remember the exact chip of Decoder/Digital Filter, so I can spot it on the cdp and see where I can get the I2S?

Miguel

[jnxw2]

Thanks for all the suggestions!


The Dackit looks pretty good but I think i would prefer to use the TDA1543 if only because I reckon I could learn more..
I had a look at Peter Daniels Schematic of the TDA1543 circuit and PS, and I think I could do that easily enough, the only thing i dont know is where do I get the input signal from?


Thanks

jnxw2

[Kuei Yang Wang]

Koinichiwa,

Originally posted by jnxw2

The Dackit looks pretty good but I think i would prefer to use the TDA1543 if only because I reckon I could learn more..

Ahhm. The DACKIT is based on the TDA1543 and uses in fact the same circuit. I did not recommend buying the finished product, just the PCB and maybe chips... ;-)

Originally posted by jnxw2

I had a look at Peter Daniels Schematic of the TDA1543 circuit and PS, and I think I could do that easily enough, the only thing i dont know is where do I get the input signal from?

Peter AFAIK works without receiver, which means you need tap off the I2S signal inside the Marantz CD6000. Get a Service Manual and work it out if you like.

The DACKIT PCB includes space for an S/P-DIF receiver, in this case you simply plug it into the Digital (Coax) Output of your CD6000.

Sayonara

[jnxw2]

ahhhhhh, hence you see why I didnt think I was up to building a DAC

jnxw2

[Kuei Yang Wang]

Koinichiwa,

Originally posted by miguel2

I understand that if the I2S signal is already oversampled, a simple TDA1543 design is out of the question.

No, but you already have in place the TDA1545, which IMNSHO is a notably better chip than the TDA1543. If you like you can passive I/V conversion on the TDA1545 as well.

Originally posted by miguel2

I have to ask you if using a 192KHz chip would increase sound quality significantly.

As a cynic (and based on what I have heard so far) I'd have to say that I rather doubt that, unless you go for a fullblown implementation of PCM1704 or PCM1738 or a suitable Cirrus Logic "High End" Chip, with all the relevant effort applied to Powersupplies, Analogue Stages etc....

As for the TDA1545, just to put some cats among the pidgeons, I came across an interesting circuit from a French On-Line Mag, originated in the "dustcircle" of Pierre "Mr. MDI" Johannet for modding the Philips and Marantz Players using TDA1545 output Stages (Philips CD71X/72X and Marantz CD3X/4X) which I have attached.

You could use three "D-Cell" NIMH Batteries in this circuit, taking care to have a suitable "standby/float plus fast" charger which is simply disconnected when the Player is unmuted. Given the very low current Draw of this circuit you would likely get 2 - 3 Month battery life out of three non-rechargable high performance batteries, so completely omiting the charger and all may also be an option.

This would likely make for a VERY cost effective modification for the Philips and Marantz Players of this type, elimiating any active output Stage. Not shown in the Schematic are the Output Coupling Capacitors.

Oh, before I completely forgte, for adventerous souls, you could use a simple Silicon Diode (1N400X for example) string where each battery is replaced with three diodes as supply. Simply feed this diode string with a few 100mA current (from an LM317 connected as current source and a seperate mains supply) and things should work quite well to, you could even seek out adjustable references that can handle a few dozend mA and can supply 1.5 to 1.8V. In all cases I'd probably place one each BG NX-HiQ or Sanyo Os-Con on the Reference and Supply pin of the TDA1545 to ground.

Anyway, many options.

Sayonara

[Elso Kwak]

Hi Kuei,
I think the resistors should be connected to ground as is the minus end of the string of batteries. And 33k should read 3k3.
A problem is the supply is 3x 1.5 = 4.5 V with alkaline or Zinc carbon batteries. Fedde would use four batteries giving 6V.

[Elso Kwak]

Hi, Nice thing is it works and not worse than this supply:
http://www.diyaudio.com/forums/showt...882#post181882
Helas, sounded not better either.......
Edit: oops my result is only for TDA1543. Forget my remark about the connections. TDA1545 is different in this respect. Kuei your picture is correct for TDA1545. I was concentrated on the TDA1543 as I don't like the 1545.

[jean-paul]

I was concentrated on the TDA1543 as I don't like the 1545.

You must be the only person that thinks so on this planet !

[Elso Kwak]

Originally posted by jean-paul


You must be the only person that thinks so on this planet !

Jean-Paul, Weird feeling I can tell you........

[PMiczek]

The only thing I did to Marantz was bypassing digital filter, simplifying output stage to a single OPA627 per channel, removing DC reducing resistors network right at DACs output, changing all electrolytics to HFQ and Cerafines, adding local bypassing (Goudreu triplets).

I searched back to find the post by Pete, I hope I am not quoting the wrong one (way too many people called Peter here<!>), but I am not sure what is being bypassed. I am guessing the analog supply pin of the DAC?

Peter Goudreu:

A Panasonic HFQ 'lytic, 120uF/25V paralleled with an AVX 0612 270nF/25V Z5U chip cap paralleled with an AVX 0612 10nF/50V X7R chip cap. Note that these chip caps are 0612s, not 1206s. This yields a nearly perfectly resitive driving point Z out to about 90MHz or so. The residual inductance is on the order of 180pH which is swamped by the internal parasitics of the bypassed device's leadframe and bond wires. The 10nF should be mounted to the board first with the 270nF on top so that inductance at high frequencies is minimized.

And btw, was the filter bypassed as in disabled, just on principle (real audiophiles don't need no stinkin' filters...), or was there something wrong with this part of the CD-94 design??

PM

[Wombat]

Another idea is to buy a used DAC and simply improve some parts.
I have a Sony SDP90 and for~100Euro (as you are from europe)
you can get one. Change some parts (20Euro) and you have a
digital preamp that plays in upper leagues.
I think this is easy for beginners as me and you.

[Peter Daniel]

Originally posted by PMiczek


I searched back to find the post by Pete, I hope I am not quoting the wrong one (way too many people called Peter here&lt;!&gt, but I am not sure what is being bypassed. I am guessing the analog supply pin of the DAC?




And btw, was the filter bypassed as in disabled, just on principle (real audiophiles don't need no stinkin' filters...), or was there something wrong with this part of the CD-94 design??

PM

I bypassed all 3 supply on TDA1541 in that way.


That part was fine (filter), but I wanted to try non-oversampling. I didn't remove the filter completely as I want digitalt output as well.

[PMiczek]

Wombat,

SDP90 I am not familiar with, and do not know what is inside.

I _have_ heard the sdp-E800, quite impressive for movies and surround sound, but pretty much ruined 2-channel audio for my taste, just like the HT setup I own, unless I turn off all the features and use it in stereo mode. Did not think you could do much with the E-800, but never looked closely or asked anyone. Checking today, E800s sell for about $100 on ebay.

Peter,

Q: why bypass all supply pins in such an elaborate manner? Was this really necessary, or one of those "preventive" measures?

Btw, Am relieved to hear you did not jettison your filters completely, I am rather fond of mine.

PM
(I think we have departed from the "simple" part of the original thread now...)

[Peter Daniel]

Originally posted by PMiczek

Q: why bypass all supply pins in such an elaborate manner? Was this really necessary, or one of those &quot;preventive&quot; measures?

I didn't compare ea. supply before and after bypass and preffered do them all at once. So we could call it preventive measures