Marantz CD63 & CD67 mods list

Jump to Latest
#6,641
I noticed that Ray's filter circuit has a much lower input impedance than the stock player. Around 1/10th. This is similar to the ultranalog SACD enhancer it is based on, but that is for an AKM DAC.

I'm wondering if this has any efffect - good or bad - on audio performance.

Also, can anyone tell me what we are filtering out here. Obviously we want to pass up to 20kHz, but where does the rubbish start? Is it all above 300kHz?
 
#6,643
Glenn2 said:
I noticed that Ray's filter circuit has a much lower input impedance than the stock player. Around 1/10th. This is similar to the ultranalog SACD enhancer it is based on, but that is for an AKM DAC.

I'm wondering if this has any efffect - good or bad - on audio performance.
Click to expand...

Well it sounds nice 😀

I will order the transistors this afternoon.... Hopefully can get it built soon and so have some more insightful comments to make!!

Simon
 
#6,646
Not without a bigger value inductor I think.

I played around with it a bit in a simulator but without changing L, only R and C. It rolls off a bit too late really. Flat-ish (or even a little peaky) out to 100kHz. Not bad though. I think it was about 55dB down at 8fs (352.8kHz).
 
#6,647
opamp

Hi there!

back with some stupid questions, first some more opinions about caps
http://www.diyaudio.com/forums/showthread.php?postid=1206241#post1206241
even if I read some said the Philips MKC sounds very good so it seems my integrated amp is perfect ! It's a jungle to find what is the best for what... I'll get capsed soon!
About the opamps, is there something that came makes the AD8599 sounds not "happy" but "beyond excited"? The caps? I've beside them the old Ray's list stuff, Elna RJH. Maybe time for BG.
I've tried the impact of the AD8599 on the HPhones socket: OMG! No way to use it anymore! The low mids and mids seam lost. Highs are over present. I'm scared I'll loose the bass and the beatif fit back the LM4562. Very hard to say! The bypass cap? Again, some said a cap between 4&8 is stupid, 4&GRND and same for 8 is the way. My 4&8 cap is a 0.22µf Wima MKS4.
I really want to change the opamp asap, but it is wise to put tulip socket there.
I really something is wrong on my setup makes the AD sounds like this.
AVR, what is your caps setup around there?
Again, what is the real size forthe socket? Farnell shows thousand!
http://fr.farnell.com/jsp/Supports+...ARWIN/D2608-42/displayProduct.jsp?sku=1023046
Is this one ok?

Thanks 😉

Note: I know that a guy talking opamps when the mood is at discret... sorry, lazyness...
 
#6,648
#6,649
Glenn2 said:
Just wondered if making the DAC analogue stage work harder was detrimental at all.
Click to expand...

Hi Glenn,

It's not an analogue stage! The output signals are square waves.
Look at the block diagram, there's a PWM modulator buffered by two drivers. The output stage is probably something like a totem-pole: on or off, with saturated transistors. So I don't think the output voltage will drop much by the increased load at the output. Well, unless you make it too low of course 😀. It's the timing of the pulses that's important, that makes the audio-signal, not the actual amplitude.

In the datasheet the output voltage is spec'd AVdd-0.3V at 1mA output current and that would be about 5k impedance.

Regards,

Ray
 

Attachments

  • delta_sigma_output.jpg
    delta_sigma_output.jpg
    97.9 KB · Views: 306
#6,650
Glenn2 said:


Also, can anyone tell me what we are filtering out here. Obviously we want to pass up to 20kHz, but where does the rubbish start? Is it all above 300kHz?
Click to expand...

Without going into sampling theory, the filter roll-off is at just above 20KHz deliberately. This is approximately half the sample rate of 44KHz. The noise starts here, mainly in the form of alias signals which go all the way up to many KHz (in theory to infinity)

Similar filters are used in the recording process.



Andy


For more info, Google 'Nyquist Theory'.
 
#6,651
poynton said:


Without going into sampling theory, the filter roll-off is at just above 20KHz deliberately. This is approximately half the sample rate of 44KHz. The noise starts here, mainly in the form of alias signals which go all the way up to many KHz (in theory to infinity)

Similar filters are used in the recording process.



Andy


For more info, Google 'Nyquist Theory'.
Click to expand...


Hi Andy,
I just wasn't sure what the situation was with a PWM output.
Wasn't the original point of oversampling that it took the grunge up to a much higher frequency so that gentler filtering could be used? Though this was with R2R dacs.
Glenn
 
#6,652
6h5c said:


Hi Glenn,

It's not an analogue stage! The output signals are square waves.
Look at the block diagram, there's a PWM modulator buffered by two drivers. The output stage is probably something like a totem-pole: on or off, with saturated transistors. So I don't think the output voltage will drop much by the increased load at the output. Well, unless you make it too low of course 😀. It's the timing of the pulses that's important, that makes the audio-signal, not the actual amplitude.

In the datasheet the output voltage is spec'd AVdd-0.3V at 1mA output current and that would be about 5k impedance.

Regards,

Ray
Click to expand...

Thanks Ray. The output stage of the DAC seems to be sensitive to regulation and decoupling changes so it would not have shocked me if it was sensitive to loading too.
If it works as you describe, it makes sense that they design it to have differential outputs for noise rejection.
By the way, do you fine-tune your circuit to maximise common-mode rejection? Or just match everything to such a degree that it is not necessary?
Cheers,
Glenn
 
#6,654
Glenn2 said:
Thanks Ray. The output stage of the DAC seems to be sensitive to regulation and decoupling changes so it would not have shocked me if it was sensitive to loading too.
If it works as you describe, it makes sense that they design it to have differential outputs for noise rejection.
By the way, do you fine-tune your circuit to maximise common-mode rejection? Or just match everything to such a degree that it is not necessary?
Cheers,
Glenn
Click to expand...

I didn't want to use any trimmers in the circuit, so I matched the components where it was nescessary. The Holco resistors I used are very close in tolerance already, so it wasn't needed for those. But I matched the filter's inductors and caps and everything in the differential input stage.

Regards,

Ray
 
#6,655

Attachments

  • img_8131.jpg
    img_8131.jpg
    58.6 KB · Views: 383
#6,657
raygulator

Hi guys, i am trying to build Raygulator as i am a beginer- my question is for the 12V+ using the LM317-do i need to bend the pin at all?as i see the pins are not matched as 78M12? or am i missing something. I have looked at the pictures from Ray web site but still unsure esp even when the LM317 reverse in position . Thanks Quan.
 
You must log in or register to reply here.
Top Bottom