Schematics for the Pioneer PD s507

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Krishu:
I beleive it is. But that means you loose resolution when you attenuate the signal. It might not be a problem if the DAC uses all 24 bits but i think at least the noise floor hits you if you do.

Anyway i said that i was fiddeling with the analog circuit in simulation and i have created something that should work without the dc blocking cap.

The thing that bothers me is that everyone does seem to use dc blocking caps despite the potential for sound degradation. I assume it has to do with either the problem of having to manually adjust the DC offset sometimes or the fact that you need to have two active components in the signal path where you could get away with 1 or 0 as Tor M suggested.

Anyway this is how i thought you could do it without DC blocking cap. The implementation is perhaps a bit naive but feel free to comment or suggest changes. Will build this when i get the time, if the protests are not too bad.

The filtering is not more than enough to make it stable.
The thing is that i will probably have to add another powersupply for the analog parts since it feels like the +-10v is not really enough.

There is a possibility to remove the last opamp if a inverted signal is accepted, does require a change in gain and currentsource.

There should be a trimpot in the currentsoure to adjust the DC offset but it is left out for clarity.
 

Attachments

  • analogstage2.png
    analogstage2.png
    13.1 KB · Views: 534
Konnichiwa,

Just a few quick comments.

First, a high quality Film Cap as coupling cap is much less soncially damaging than the other coupling cap, namely Cextl/Cextr, especially if you put a cheap 'lytic in there.

Why am I calling this a coupling cap? Because the DAC has a differential output the ouput circuit is an Op-Amp used as balanced to SE Converter with lowpass filtering thrown in for good measure (see attached detailed scheme from the earlier but similar design PCM1710).

BTW, you do not want to know what quality these on-chip silicon oxide resistors and silicon based capacitors are, you really don't.

Basically, as mentioned previously, I feel the sort of DAC like the PCM1716 is cripled at birth sonically sufficientl (CMOS Op-Amp, on board RC lowpasses, DAC itself) that it qualifies as karmic waste of space.

If you absolute MUST keep that dreadful thing, use decent Film cap's as Cextl & Cextr. You can always DC couple any buffers following the DAC and on board Op-Amp (be it a buffer with a gain of 2 or be it a J-Fet or Valve follower with no gain and an output 6db below normal which in most cases is not an issue) if you feel a buffer is needed.

If the load is low capacitance and at least 10 - 22k I would leave all this out, just use the best capacitor you want to afford for the Player for Cextl/Cextr and the two output coupling Cap's and make them at least 4u7F in value, this way you will be okay with loads down to 10k. On the output add a 51 Ohm "build out" resistor (to stop oscillation into capacitive loads) and a 100K Resistor to ground. Finished. Anything else you do cannot undo the cheap and nasty design on chip anyway.

If you do use the direct output I'd think it worth feeding the Vcc inputs via TL431 Shuntregs fed from +12V via a CCS.

But really, we are trying to dress up mouldering, decomposing pigs ears as silk purses, what's the point?

If pin 28 is high (+5V) then your PCM1716 gets an I2S input datastream, so simply fit a bunch of TDA1543 with reclocker and clock send back to the CD-Player (zero jitter operation), if Pin 28 is low you should be able to use TDA1543A or TDA1545 the same way, all no oversampling and the usual schematics & pcb's.

Sayonara
 

Attachments

  • pcm1710.jpg
    pcm1710.jpg
    52.2 KB · Views: 503
You really do not like this circuit!
But really, we are trying to dress up mouldering, decomposing pigs ears as silk purses, what's the point?
:D
I see the challenge in trying to do the best i can with the given circumstances, it is a cumbersome filosofy but that is what i am.

If i understand the biggest changes between 1710 and 1716 is that there is a 8 level quantizer instead of five and the pins that you could have used to extract current output is NC:bawling:

As money shortage and time is my main concern i will pursue this modification first and if i am happy with it i will live with the inferior DAC.

Any thoughts of actively stabilizing the EXTx pins with some form of circuit, plus film caps, to vdd/2?
 
Konnichiwa,

hjelm said:
You really do not like this circuit!
:D

Nope.

hjelm said:
I see the challenge in trying to do the best i can with the given circumstances,

So do I.

I once had a little cheap DAC Kit with PCM1716. I found that the best I could do was to fit TDA1543's into the Op-Amp position, cut the relevant traces and hardwire a "Kusonoki" DAC (I used a stack 4 high).

I did try all the other options, including better Op-AMp's, PSU's, direct output, fiddeling with Extl/r caps etc. At the best I could get it to it was still completely dreadful next to a killer modified Marantz CD67 (until I fitted TDA1543 Non Oversampling that is). So I concluded simply that the PCM1716 is useless for quality audio.

BTW, the kit came with a NE5532 Op-Amp and the Kit Vendor contended that using a "better" Op-Amp would not improve the sound. They where right.

hjelm said:
If i understand the biggest changes between 1710 and 1716 is that there is a 8 level quantizer instead of five and the pins that you could have used to extract current output is NC:bawling:

Yup. Looks to me like that.

hjelm said:
As money shortage and time is my main concern

And a TDA1543 alone will cost around $/€1 and a LM7808 $/€ 0.50 plus a handfull of resistirs and quality output cap's that you need anyway, so why not knock something up on veroboard?

hjelm said:
Any thoughts of actively stabilizing the EXTx pins with some form of circuit, plus film caps, to vdd/2?

Sure. Your "active stabilisation" becomes the signal coupling element. If you can make better regulators than I can buy capacitors for the same money and effort invested, more power to you.

Sayonara
 
:bawling: But i wanna make the circuit work. :bawling:

Maybe i shouldn't even try.

The whole thing i wonder is how i can stand listening to the cd player at all, i cannot.
******* *******. The problem is that i think the setup of the DAC is done by a processor so i am not sure if i can use a 1543 directly.

Kind of makes me wanna go out and buy a DF1704 and PCM1704 and hook that up instead.
 
Hello Hjelm, Hallo Thorsten,

I built a Doede DAC (www.dddac.de) using 8 TDA1543 in parallel but still it is oscillating and I have a kond of Karaoke effect (every instrument but nearly no vocals ... must be kind of out-of-phase). I will send it to Doede and he will fix it for me.

I just wanted to look what I can get out of this thing. This player was recommended by some friends and by Stefan Horwege . I paid far less than $200 for everything and this player at least plays at the level of my NAD C 541i, despite not being able do read HDCD.

First, I wanted to use this player as a transport only plus the DDDAC via coaxial connection. This would be nearly the way KYW suggested few posts above. This is still the plan but I wanted to tweak the original DAC/analogue output, too.

Apart from that I am planning to modify a Pioneer CD/SACD/DVD-A player. Schematics are available at www.schaltungsdienst.de, I just have to decide which one to choose... If I am successful I will sell the PD-S507.

KYW, do you mean the size of the caps at EXTL/R are not that important? I have some 2.2uF Wima MKS on hand. Would these be OK? (I have low-ESR Panasonic FC || EMZ KP72 15500pF there at the moment.)

But: I felt that there definitely were differences in sound between opamps I tried. And AD8620 sound way better than the original one in this application. But maybe the diference is more from the lower output impedance.

CSS + TL431 for symmetric 12V would need at least +/-15 V what would mean there is the need for a new transformer.

BTW, i havent been succesful in figuring out a circuit for this. Do you have a hint for me? (CSS, TL431)


Cheers
Christian.
 
Konnichiwa,

hjelm said:
:bawling: But i wanna make the circuit work. :bawling:

Good luck. I tried on & off for nearly 6 Month and then decided to cut my losses.

hjelm said:
Maybe i shouldn't even try.

By all means try, consider it a learning experience.

hjelm said:
The whole thing i wonder is how i can stand listening to the cd player at all, i cannot.

Really? I fond the 1716 sound always fairly unpleasant and uninvolving. Okay for background music, but not for "listening".

hjelm said:
The problem is that i think the setup of the DAC is done by a processor so i am not sure if i can use a 1543 directly.

As said, check what the Dataformat Pin's are like (voltage at operation) and take it from there.

hjelm said:
Kind of makes me wanna go out and buy a DF1704 and PCM1704 and hook that up instead.

Don't bother. My current CD-Player has PCM1704 X 2 in parallel & PDM200 HDCD Filter, OPA627 Op-Amp's and Valve Buffer. I prefer the sound from the Ack!dAck (non oversampling TDA1545, simple Op-Amp I/V & sealed lead acid battery supplies) by no small margin.

Sayonara
 
Ditto on the PCM1716. BTW : a Sony hack Q.

I also tried a few solutions for this chip. It is in my Sony dvp-s7700. I found the most I could get out of it was with a 1.5uF TRW 30v cap right off of the output pins. I tried lots of NOS caps,real cheap from a salvage place (sprague Vit-Q, gudeman & etc). I have no why this Trw cap is a stand out.

Still its is not pleasing. Hollow sounding.

I repaired my dvp-ns900v. I screwed up doing another hack.

But this time I removed all the caps in the audio path. Used a lm6172(w/proper v+ v- caps) as the buffer with 4 10K resistors matched within 1 ohm. (I used the INA2134 as a model. It sounded bad BTW) CDs have much more detail and better soundstage.But are fatiguing. SACDs are much better now BUT not fatiguing. Hmmmm?

Question for Kuei (and others)

Is it possible to use the unused 6 Ch DAC in this player in parallel with the 2 CH. That is have the CDdata bypass to the 6 ch and output all 8 (v+, v-) to the buffer.

These are ak4383 (2CH) and ak8357 (6ch). I seems I would need to use >8k resistors as the datasheet has the min dc imp. at 8K.

I do not think the DSD needs the extra DACs.

Lewis
 
Re: Ditto on the PCM1716. BTW : a Sony hack Q.

Konnichiwa,

ljordan said:
Is it possible to use the unused 6 Ch DAC in this player in parallel with the 2 CH. That is have the CDdata bypass to the 6 ch and output all 8 (v+, v-) to the buffer.

I am sure there are ways to accomplis that, but I doubt it would be easy or straightforward. I hate to sound like a broken record, but if you don't like the internal DAC's try a TDA 1543 or 1545 non oversampling DAC. It leaves you with your working DVD/SACD Player and gives you good sounding CD. Non of the recent batch of Delta Sigma DAC's sound in any way outstanding or at least good as the old NPC SM5872, whatever the specs may be. The TDA 1543 No Oversampling manages to nudge out the SM5872 to my ears. TDA 1545 and 1541 deliver another grade up. Considering the ease of implementation of 1543 & 1545 I always ask "What is your excuse?".

Sayonara
 
Hi!:)

I`m currently making a new dac for my cd player (PDS707) with tda 1543A. I hoped to have it finished yesterday, but the women in my family always seems to find me more "useful" stuff to do.

The decoder in my player is sony cdx2507aq, and I took for granted that this chip output Sony EIAJ, but now the I checked the dac in the player, but this is set for I2S input (and I wish I had used an ic socket and not soldered directly to dac:( ). The Pioneer High Bit filter, wich I planned to remove to get nonoz, is in between.

Does the high bit chip convert the signal to I2S, or has the cdx2507aq I2S datastreem?

I noticed that the "Data" line has different dc offset before and after the high bit filter, 2.5 V before and 1.6 V after.

I don`t know much about the differences between EIAJ and I2S, so I can`t tell from the datasheet. Something to learn:)

http://www.promelec.ru/pdf/cxd2507a.pdf

Anybody know if this chip is EIAJ out?

Thanks

Tor Martin
 
Hi!

For me the main reason is that it is much cheaper! Cost next to nothing, not more than 20 euro including caps. The input reciever is much more expencieve than the rest of the parts, at least in Norway, and now you don`t need it. Second it is much easier for me, don`t have to add a coaxial output.
Third the I2S or sony format is known to be a much better format than the spdif and you get fewer parts in the signal path.

But I`m not experienced in this area, so theese thoughts are not based on real life experiences.:)

I looked closer at the cdx2507 datasheet and it looks like the wsbd line change when the data line change to MSB.
I2S change the wsbd when the data line go to LSB :confused:
This means that the datastreem from the chip is EIAJ and that the highbit chip converts this to I2S?:confused: :confused:

Regards
Tor Martin
 
Hi!

My cd-player hates me!!!:bawling:

I finished my dac today. Filled with joy I powered up my player, and what did it do?? It went black!:bawling:

So I disconnected the power for the newdac board again, still black. Then I disconnected the power jumpers between power board and digital board. And now the player lighted up again.
I measured for shorts, couldn`t find any. Then I started disconnecting the jumpercables one by one. The RST line is causing me trouble. The player has an ic reg called PQ05RR12 with an reset line. When this is connected the player won`t power up.
I have disconnected all power to the unused chips, dac and a buffer for the clock signal and removed the High bit chip. Both the dac and the high bit chip is connected to this rst line. Could this cause trouble?

Unfortunally I have to go to work now! No time for faultfinding.

If anybody would like to take a look at the schematics, the service manual for the player (PDS707) is to be found on that nice link ergo posted in the post above.

Dissapointing day!:(

Tor Martin
 
Hello again...

yesterday I installed a Kwak Clock 7 and it works gorgeous.

Before, a friend and me compared the modded PD-S507 against an unmodified NAD C 541i as well as against DDDAC1543 (only one TDA1543, source was the NAD cd player).

The Pioneer was by far the best.

I am looking forward to get my digital out pcb for the Pioneer to see how doede's dac sounds then.


Cheers
Christian.
 
Hi!
Glad to hear you like your player!:)

I got mine playing with the tda1543A nonoz dac yesterday, and I must say I like what I hear!!;)
The output from the sony decoder is EIAJ, so the A version is the one to get.

There is much more air in the music, and the soundstage expanded a great deal. Espessialy in the hight, going close to the roof! And this is without any break in time.

There is a little bit of noise though. Probably because I didn`t have any suitable coax cable for the digital signals, so I used plain magnet wire.

The player is unfortunally not back in the system yet. The accident I had when I installed the dac the first time has made the player so that it won`t close the transport. It opens fine, but I have to close it by hand, means no top cover. Hopefully the opamp wich handels the signals are broken.

I don`t know what caused this accident, but I belive it was because I disconnected the powersupply to the PCM1716 dac without disconnecting the 5V reset line. It made two ICP-N10 fuses break!!

But the sound is NICE!!

Regards

Tor Martin:)
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.