As a request from a forum member about some more mods, instead of mail to that user I’m going to make these mods public and hope that anyone tries them🙂 as to the first part of the mods you can look on the forum, also in that thread you can find a link to schematic of the analog part of the player, and read some big discussion 
The schematic is essential as I’m going to refer to it.
Let's start
Analog power supply:
Change 2115 and 2116 electrolytic caps to 4700uf 25volts also electrolytic, this improves noise performance, ripple and output impedance
On the 6000 (not the OSE and KI) replace 2113 and 2114 electrolytic caps to 4700uf 35v, this also improves ripple performance of the power supply.
Remove resistors 3711, 3712, 3713, 3714, 3249, 3250, 3251, 3252 and replace with with a shunt. These resistors forms a R/C filter with the bypass caps 2223, 2224, 2225, 2226, 2711, 2712, 2713, 2714 .
They presumably reduce noise but they kill the regulation because they increase impedance of the power supply especially on the lower frequencies where the cap impedance starts rising. Connect a scope after de R\C filter and you will see music on your power supply😱
I also recommend removing the bypass caps as caps in parallel create an RLC network and create resonances on the power supply.
Bypass caps work in digital, in analog they don’t! that’s my view.
And the final mod, this one is very important🙂
This player uses the SM5872bs, it has tree power supply pins, two digital and one analog.
The digital ones are decoupled with, again, R\C filters.
The R\C on the digital supply(DVDD pin16) creates a lot of problems and the noise there is very high so remove the resistor 3313, 3314 and place there a shunt as it lowers a lot the noise.
the performance of this cd is very good but it cant get even near a cd with a multibit dac.
Best regards
Ricardo
The schematic is essential as I’m going to refer to it.
Let's start
Analog power supply:
Change 2115 and 2116 electrolytic caps to 4700uf 25volts also electrolytic, this improves noise performance, ripple and output impedance
On the 6000 (not the OSE and KI) replace 2113 and 2114 electrolytic caps to 4700uf 35v, this also improves ripple performance of the power supply.
Remove resistors 3711, 3712, 3713, 3714, 3249, 3250, 3251, 3252 and replace with with a shunt. These resistors forms a R/C filter with the bypass caps 2223, 2224, 2225, 2226, 2711, 2712, 2713, 2714 .
They presumably reduce noise but they kill the regulation because they increase impedance of the power supply especially on the lower frequencies where the cap impedance starts rising. Connect a scope after de R\C filter and you will see music on your power supply😱
I also recommend removing the bypass caps as caps in parallel create an RLC network and create resonances on the power supply.
Bypass caps work in digital, in analog they don’t! that’s my view.
And the final mod, this one is very important🙂
This player uses the SM5872bs, it has tree power supply pins, two digital and one analog.
The digital ones are decoupled with, again, R\C filters.
The R\C on the digital supply(DVDD pin16) creates a lot of problems and the noise there is very high so remove the resistor 3313, 3314 and place there a shunt as it lowers a lot the noise.
the performance of this cd is very good but it cant get even near a cd with a multibit dac.
Best regards
Ricardo
You very wisely avoid the subjects of clock, opamps and damping
Views can be interesting but please enlighten us with the theoretical proof of your findings.
Views can be interesting but please enlighten us with the theoretical proof of your findings.
The explanation is there...
About those subjects I said what I had to say... Ne5534 rules (if you have courage, do a null test to the various opamps and post a thread on it)
About those subjects I said what I had to say... Ne5534 rules (if you have courage, do a null test to the various opamps and post a thread on it)
I had the courage and did the test and it wasn't NE5534 that was best and I am not the only one that thought so. It's distortion figures are very good but that is not the most important parameter. Maybe you should show some courage and do the test yourself ?
your new cd6000 mods
Rickpt, thanks a lot for your mods part2!
These new ones are very clear (the use of the schematics is a very good idea) but there're some questions about some of your "mods part1":
1. I don't use the headphones out: which tracks and jumpers I have to cut?
2. How should I replace the hdam and use opamps instead? Should I also replace or just remove the output hdam, how?
As you have been modifying your cd6000 since several months:
3. Can you recommend any other modifications?
Thanks again Rickpt 😉
Rickpt, thanks a lot for your mods part2!
These new ones are very clear (the use of the schematics is a very good idea) but there're some questions about some of your "mods part1":
1. I don't use the headphones out: which tracks and jumpers I have to cut?
2. How should I replace the hdam and use opamps instead? Should I also replace or just remove the output hdam, how?
As you have been modifying your cd6000 since several months:
3. Can you recommend any other modifications?
Thanks again Rickpt 😉
I don’t see who lacks the courage around here 😉 ; if you did the tests, show the results to the forum and also your implementation of the test. And please don’t run way this time😀
So explain what are the parameters that are most important to you?
Best regards
So explain what are the parameters that are most important to you?
Best regards
rickpt said:
To me an important parameter is respect for other peoples opinions and another important one is behaving like a grown-up.
"Force Feeding" ;-)
Hi Ricardo,
May I suggest reading Application Note 202: An I.C. Amplifier Users' Guide To Decoupling, Grounding, And Making Things Go Right For A Change?
http://www.analog.com/Analog_Root/sitePage/direct/applicationNotes/
Hi Ricardo,
May I suggest reading Application Note 202: An I.C. Amplifier Users' Guide To Decoupling, Grounding, And Making Things Go Right For A Change?
http://www.analog.com/Analog_Root/sitePage/direct/applicationNotes/
Swapping 2115 and 2116 to 4700uF will quite marginally improve the ripple simply because you’ll hardly find there any ripple to improve. Also, bigger caps are generally unfavorable for the noise suppression, and are undoubtedly worse regarding the high frequency impedance.
Applied in the Marantz CD players with SM5872 DAC, the difference between NJM2114 (similar to NE5532) and LM6172 (which is still not a sacred cow) is night and day.
Pedja
Applied in the Marantz CD players with SM5872 DAC, the difference between NJM2114 (similar to NE5532) and LM6172 (which is still not a sacred cow) is night and day.
Pedja
I respect your opinion but that doesn’t mean that it is the absolute truth!! And you have a certain difficulty backing up your opinions.
And look at your first post and this quote applies exactly for you!
Best Regards
Ricardo 🙂
And look at your first post and this quote applies exactly for you!
Best Regards
Ricardo 🙂
Hi Elso
I did read that aplication note years ago and i didnt agree with everything on it. the reasons you can read on my first post.
I made my own tests and got to that conclusion. you should do it to.
Regards
I did read that aplication note years ago and i didnt agree with everything on it. the reasons you can read on my first post.
I made my own tests and got to that conclusion. you should do it to.
Regards
Hi pedja
But you agree it lowers noise? i did some mesurements and it did!
And big caps arent what they were years ago!😀😀😀
And de diference is to better or worse objectivly? 😀😀😀😀
Best regards
Ricardo
But you agree it lowers noise? i did some mesurements and it did!
And big caps arent what they were years ago!😀😀😀
And de diference is to better or worse objectivly? 😀😀😀😀
Best regards
Ricardo
back to bussiness
Olá Pedro
the easiest way to remove the headphones out is removing the opamp 7235 but that approach is sub optimal, as you leave all those tracks connected and they will act as antennas(even worse as you have a xo clock). Have a good look at the schematic and the PCB and disconnect the power supply lines and the signal lines close to their source that lead to that opamp.
About the HDAM if you aren’t 100% sure of what to do... simply dont do that mod and keep the HDAM.
Other mods? I recommend that you remove the xoclock and have a listen to the original one, if you still prefer the xo, at least you didn’t spend your money in vain...
Cumprimentos e um abraço
Ricardo
Olá Pedro
the easiest way to remove the headphones out is removing the opamp 7235 but that approach is sub optimal, as you leave all those tracks connected and they will act as antennas(even worse as you have a xo clock). Have a good look at the schematic and the PCB and disconnect the power supply lines and the signal lines close to their source that lead to that opamp.
About the HDAM if you aren’t 100% sure of what to do... simply dont do that mod and keep the HDAM.
Other mods? I recommend that you remove the xoclock and have a listen to the original one, if you still prefer the xo, at least you didn’t spend your money in vain...
Cumprimentos e um abraço
Ricardo
Hola Ricardo!
I'm going to do the mods.
Just one thing: I have tried to replace the original clock and then, again, Guido's XO clock.
With Guido's clock, cd6000 sounds different. I thought it was a kind of subjetivism: violins sounds something more close to the instrument original sound. But I've tested (in both cases and not reveling when the clock was the original and when the XO) with a friend violinist and, in every test, she perceived the difference and choose the one with XO clock.
Sorry and muchas gracias!
Pedro.
I'm going to do the mods.
Just one thing: I have tried to replace the original clock and then, again, Guido's XO clock.
With Guido's clock, cd6000 sounds different. I thought it was a kind of subjetivism: violins sounds something more close to the instrument original sound. But I've tested (in both cases and not reveling when the clock was the original and when the XO) with a friend violinist and, in every test, she perceived the difference and choose the one with XO clock.
Sorry and muchas gracias!
Pedro.
rickpt said:
rickpit,
I don't quite understand what your problem is. You are happy with your ruling chip, so why are you interested with other's results. We all know those results, and we don't have to go through all that again. If you continue posting in that way (which is quite rude in some way), I will have no choice, but move most of your posts into "Modeling Emotional Engagement" thread, in Off Topic section.
Peter Daniel
Moderator
Op-amps again
Ricardo,
I don't want to start a big discussion here, but the NEs behave better than others on bad implemented circuits.
If you don't bypass the psu properly near the op-amps, the NEs may play better than OPAs, etc.
But bypass it properly and you can put OPAs, or even LM6172, which I like a lot too.
And the sound of OPA2604/604, OPA2132/132, OPA627, LM6172 is miles ahead of the NE5532/4.
It seams you're doing things the wrong way.
Sorry to say, but I had a good laugh reading the opening note on this thread where you described the changes.
😱 😱 😱 😱 
rickpt said:The explanation is there...
About those subjects I said what I had to say... Ne5534 rules (if you have courage, do a null test to the various opamps and post a thread on it)
Ricardo,
I don't want to start a big discussion here, but the NEs behave better than others on bad implemented circuits.
If you don't bypass the psu properly near the op-amps, the NEs may play better than OPAs, etc.
But bypass it properly and you can put OPAs, or even LM6172, which I like a lot too.
And the sound of OPA2604/604, OPA2132/132, OPA627, LM6172 is miles ahead of the NE5532/4.
It seams you're doing things the wrong way.
Sorry to say, but I had a good laugh reading the opening note on this thread where you described the changes.
😱 😱 😱 😱
Re: Op-amps again
About bypassing (analog) it doesn’t improve anything... been there, done that.
Doing things the wrong way? Don’t think so? Have a look at the service manual of Pass labs. Do you see any bypass on the analog part? Some schematics even have discreet opamps (ono)
So pass labs products are badly design and Mr. Nelson Pass is a poor designer...
carlosfm said:
Ricardo,
I don't want to start a big discussion here, but the NEs behave better than others on bad implemented circuits.
If you don't bypass the psu properly near the op-amps, the NEs may play better than OPAs, etc.
But bypass it properly and you can put OPAs, or even LM6172, which I like a lot too.
And the sound of OPA2604/604, OPA2132/132, OPA627, LM6172 is miles ahead of the NE5532/4.
It seams you're doing things the wrong way.
Sorry to say, but I had a good laugh reading the opening note on this thread where you described the changes.
😱 😱 😱 😱
About bypassing (analog) it doesn’t improve anything... been there, done that.
Doing things the wrong way? Don’t think so? Have a look at the service manual of Pass labs. Do you see any bypass on the analog part? Some schematics even have discreet opamps (ono)
So pass labs products are badly design and Mr. Nelson Pass is a poor designer...
Resonate or not resonate here is the question...
For some people that like to try things like I do, here is a simply test that anyone can try with simple test tools...
Some basics:
A cap should be a shunt to all frequencies right? That would be great, but reality is far worse... As the internal inductance of the electrolytic cap rises it's efficiency on the higher frequency starts degrading and the impedance starts rising...normally to resolve this problem, it's used the bypass technique to continue when the big cap starts loosing efficiency and keep the impedance low...but the problem is that any technique isn’t 100% effective...as the small cap and the big cap will start resonating the impedance will start rising at the resonating frequency...If this resonance exists this test will measure it...
To the test...
"A cap should be a shunt to all frequencies right?" so what we do is to a signal generator feeding a cap in series through a 1kohm resistor (to prevent overloading the signal generator as it will be driving a shunt) and connect an oscilloscope in parallel with the cap so that we can measure the residual of the cap.
Change the frequency of the signal generator and see what’s happening on the scope...
The cap (or association of caps) that will show the lowest residual in the highest bandwidth is the best cap!
In my tests the single cap won but don’t believe in my word or anyone else, do this test and believe in you...
So Easy, So true!! Like the null test 😀😀😀😀😀
For some people that like to try things like I do, here is a simply test that anyone can try with simple test tools...
Some basics:
A cap should be a shunt to all frequencies right? That would be great, but reality is far worse... As the internal inductance of the electrolytic cap rises it's efficiency on the higher frequency starts degrading and the impedance starts rising...normally to resolve this problem, it's used the bypass technique to continue when the big cap starts loosing efficiency and keep the impedance low...but the problem is that any technique isn’t 100% effective...as the small cap and the big cap will start resonating the impedance will start rising at the resonating frequency...If this resonance exists this test will measure it...
To the test...
"A cap should be a shunt to all frequencies right?" so what we do is to a signal generator feeding a cap in series through a 1kohm resistor (to prevent overloading the signal generator as it will be driving a shunt) and connect an oscilloscope in parallel with the cap so that we can measure the residual of the cap.
Change the frequency of the signal generator and see what’s happening on the scope...
The cap (or association of caps) that will show the lowest residual in the highest bandwidth is the best cap!
In my tests the single cap won but don’t believe in my word or anyone else, do this test and believe in you...
So Easy, So true!! Like the null test 😀😀😀😀😀
- Status
- Not open for further replies.