Marantz CD63 & CD67 mods list

[SimontY]

I have the discrete output as designed by Ray, using FETs, 2w "Kiwame" carbon resistors and polystyrene filter caps, powered by S Powers @15v+- and decoupled by Elna Cerafines, power taken from the original 20v+- "KI" toroid with schottky diodes and smoothed by 1000uF BG standards.

It sounds good. Compared recently to another good CD63KI with 4562 mine sounded richer with much better dynamics and a punchier bass. Surprisingly similar tonally, although it was a little better.

I also think with each in their current state of "tune", my CD63KI was a little better than Brent's "not-completely" modified CD17KI. The 17 has RIDICULOUS bass weight though

Simon

[RCruz]

Originally posted by SimontY
I have the discrete output as designed by Ray, using FETs, 2w "Kiwame" carbon resistors and polystyrene filter caps, powered by S Powers @15v+- and decoupled by Elna Cerafines, power taken from the original 20v+- "KI" toroid with schottky diodes and smoothed by 1000uF BG standards.

Hi Simon

Do the FET´s need to be "paired" ?

[Andre Visser]

Originally posted by 6h5c
This concludes the analog electronics lesson for today!

Ray

Ray, what do yo think about fitting a 6V2 Zener // 330nF cap to Q1 emitter, change R26 to 1.2K, take output between zener and R26.

R25 can be a trimpot to fine-tune the output to 0V.

André

[magilla]

Experienced builders are often very good at making neat stripboards, so any hints?

[pzogal]

Ray , thanks a lot!

[6h5c]

Originally posted by Andre Visser
Ray, what do yo think about fitting a 6V2 Zener // 330nF cap to Q1 emitter, change R26 to 1.2K, take output between zener and R26.

R25 can be a trimpot to fine-tune the output to 0V.

André

I don't think it's a good idea to fit a non-linear semiconductor in series with the signal path, if you know what I mean. Apart from that, the voltage drop of the zener and the offset of the output stage are changing when the circuit heats up, because there is no feedback. So the compensation will never be exact.

Ray

[poynton]

Hi Andre,

I think you are too concerned with getting rid of the output capacitor.

Use the best quality cap. you can with a value of 4 - 10uF.


Andy

[RCruz]

Originally posted by 6h5c
You're welcome! I just kept on writing, I don't know what got into me...

Hi Ray
Just love the way you do it.... always appearing when needed.

After all my mods (including 4 additional TX an two clocks), I really believe I reached the opamps limits so I am considering to build a discrete output.

My first doubt is about the FET´s you use... can I use any kind that I get on Farnell or do they need to be "paired" ?

Second question is related with the PSU.

Recently I got some +-15v Spowers and placed them on the opamps.

I nedded to do it because the +-12v rayregs where overheating due to the PSU (The TX I am using is a 30VA 2x18v followed by 2x10.000uF.)

I would like to use this +-15v PSU to feed the discrete.. is it feasible ?

Regards

Ricardo

[Thomo]

Ricardo, that psu will be perfect for the discrete stage.

You've got mail about those Mundorf caps, btw.

Lee.

[Andre Visser]

Originally posted by poynton
Hi Andre,
I think you are too concerned with getting rid of the output capacitor.

Use the best quality cap. you can with a value of 4 - 10uF.
Andy

Hi Andy,

There is a few components that I hate to use in the audio path and one of them is a capacitor, I will rather try a discrete cct with servo feedback sometime.

André