Update the DOS.... I was not expecting this... Here I go again
Would you send us a revised schematic with real resistor values ?
What about some pics of the underside of your DOS ?
Be patient, Ray will soon have the information on his webpage for the CFP.
Ray, i need help!!!
First hello to everyone.
Quincie is my name (not realy:its Ruud)
I`m a newbee with electronics, but i love great music.
I have a Marantz cd67 MKII OSE and i want to upgrade this beatiful player.
I looked around youre site and i love it, but here is my problem.
I don`t understand a lot of your schema`s etcetc, but i want still upgrade my player.
Is there anyone, or do you know someone who can do that for me.
Please help meeeee!!!!
Quincie.
Hi Ruud, and welcome to this forum. Sorry for replying a bit late, but your post got lost in my DOS-tweak-upgrade-enthousiasm.
We will be very willing to help you to upgrade your CD67, but the first question is of course: can you solder?
Ray
Okay, I have done things slightly different to the posted schematic in my 8260. I changed the values of the resistors R5 and R6 to 2.2k and R9 to 150R.
I also halved the gain resistors, R7 and R8. This will double the current into the CFP and is what I have done.
However, keeping the current the same will work just as good. I found that the most important thing is the resistor into the 170BL, and it needs to be around 6K.
Oh yes, you can also use a BC557 transistor if you don't have a 560, and the Hfe of the transistor is not important for this application. In my prototype model I used both the above transistors as I didn't have enough of each. I think my finished board uses 557's.
If you don't change any of the supply current resistors, it's pretty easy to solder up the CFP under the board. Just follow the schematic point to point.
I was quite surprised at both implementations of the CFP. The first implementation really opened up and uncompressed the music signal. Then after Ray did some simulations I tried a large resistor on the 170BL, and was completely shocked at the depth of the bass, and how loud it came out. However, I have to admit that I have a subwoofer that is flat to 20hz., which helps a bit. But most of the bass is realized thru the main speakers, which are just 7" two ways with good bass.
It will be interesting to hear other viewpoints once other have implemented the CFP in their DOS's. It made a significant difference for me, and I am assuming the same for others as well.
I was at the market, but it's not easy to find those BC560's in all that junk
And i found that i need to go to the shop for more parts. For some reason i ended up with 7815 and 7915's on the pcb. Ooops. Maybe that's where the highish distortion comes from...
Not that much interesting stuff on the market, apart from one part that made my day: a 7A22 plugin for a tek mainframe. Very good condition and a reasonable price. It's ideal for audio: differential inputs going down to 10uV/div. DC offset compensation and high and low cutof filters. Happy
Ricardo,
Ray knows about the current changing resistors.
He also knows about the 557/560 transistors, it makes no difference.
The 6.8k resistor into the 170BL is the way to go. You basically have a choice between 5.6k or 6.8k, as these are the most common values, and either one will work fine as both are close to the 6k desired value.
FWIW, my proto model had 5.38k of resistance, and it was phenomenal, but that had a string of resistors in it.
When I have some time I will write up some quick instructions for soldering up the CFP. I've taken my proto all apart, but still have the resistor and transistor soldered together. It's not all that hard, but I can't seem to do it off the top of my head, so I'll get back to it later today, as time allows.
Steve
Okay, here we go on soldering up the CFP:
1. Solder the 6.8k resister across P1 and P2 of the 557/560. Leave resistor leads long.
2. Solder the resistor lead that is soldered to P1 of 557/560 to P2 of R5.
3. Solder the resistor lead that is soldered to P2 of 557/560 to P1 of T1
4. Solder P3 of 557/560 to P1 of R7.
Do the same on the other side of the T2 side, just remember that it is a mirror image.
Adding these 4 parts to the circuit board requires a complete rework of the pcb. Doing it underneath like above will work just fine. Just keep good spacing with the leads and keep things as neat as possible.
1. Solder the 6.8k resister across P1 and P2 of the 557/560. Leave resistor leads long.
2. Solder the resistor lead that is soldered to P1 of 557/560 to P2 of R5.
3. Solder the resistor lead that is soldered to P2 of 557/560 to P1 of T1
4. Solder P3 of 557/560 to P1 of R7.
Do the same on the other side of the T2 side, just remember that it is a mirror image.
Adding these 4 parts to the circuit board requires a complete rework of the pcb. Doing it underneath like above will work just fine. Just keep good spacing with the leads and keep things as neat as possible.
Just thought I'd drop ths into the mix
NewClassD Discrete OPAMPs
I guess that's uncouth of me but it's not often I find something.
NewClassD Discrete OPAMPs
I guess that's uncouth of me but it's not often I find something.
Hey guys!
I'm new here, won't bother you for a long time. I'm Lars. From The Netherlands.
I've got a complaint about you ;-)
Just want to say, that after a long day of working [since I have a few days off ;-) ] I see at 01:15 AM local time (I can't get asleep) that my collegae at work Ray, has a thread with 14290 posts.... Unbelievable. And because I bought 6 (!!) second hand Elektor Crescendo amps last week (tweaking Crescendo's is another project of Ray), I'm now sure I can't impossibly get asleep for the rest of the WEEK, because YOU ALL keep my friend Ray to busy to help me!!
Regards (and kind greetings to Ray) from The Netherlands,
Lars (#14291....)
I'm new here, won't bother you for a long time. I'm Lars. From The Netherlands.
I've got a complaint about you ;-)
Just want to say, that after a long day of working [since I have a few days off ;-) ] I see at 01:15 AM local time (I can't get asleep) that my collegae at work Ray, has a thread with 14290 posts.... Unbelievable. And because I bought 6 (!!) second hand Elektor Crescendo amps last week (tweaking Crescendo's is another project of Ray), I'm now sure I can't impossibly get asleep for the rest of the WEEK, because YOU ALL keep my friend Ray to busy to help me!!
Regards (and kind greetings to Ray) from The Netherlands,
Lars (#14291....)
That sure sounds VERY bossy, Lars
Any of those amps up-and-running yet? We're into the DOS-CFP-thing over here.
Ray
Hey Ray, no amps running yet (to busy with administration over here). Must be my lack of education, but what is DOS-CFP-thing? I only know my own BOS (Bossy Operating System)
Regards,
Bossy.
Of course, here it is (english version). You can also click on my public profile, it is listed under contact information.
Ray
Hey Ray, no amps running yet (to busy with administration over here). Must be my lack of education, but what is DOS-CFP-thing? I only know my own BOS (Bossy Operating System)
Regards,
Bossy.
DOS is short for Discrete Output Stage. It's an output stage for CD-players that replaces the opamp filter that is often used after the DAC-chip. I have a page dedicated to it here.
The CFP-part stands for Complementary Feedback Pair, a tweak to the DOS circuit that replaces the input transistors by two pairs in CFP arrangement. I've just started a page for this CFP-tweak, there's not much info there yet, but I did place an interesting link about this subject.
Ray
OK, got it. Thanks. Besides this, why so much to talk about these Marantz series CD players? Is there more interesting quality in it than the TDA1541 DAC that is applied to some (most?) of them?? That can't be the main reason, since you build your own DAC's...
Lars
Yes, there is
Guys, here is a man that never heard a properly tweaked CD63/67 before. These players are very tweakable, and sound very good once you put in some decent components. The DAC chip is one of the few that has PWM outputs, without any (sound murdering) analog filters or opamps integrated on the chip. And the transport and laser mechanism, the well known CDM12.1, is simple and effective, cheap, and still available today.
You are invited to pick up a CD67 here in Heerhugowaard some day Lars, and do some listening!
Regards,
Ray
Guys, here is a man that never heard a properly tweaked CD63/67 before
. These players are very tweakable, and sound very good once you put in some decent components. The DAC chip is one of the few that has PWM outputs, without any (sound murdering) analog filters or opamps integrated on the chip. And the transport and laser mechanism, the well known CDM12.1, is simple and effective, cheap, and still available today.You are invited to pick up a CD67 here in Heerhugowaard some day Lars, and do some listening!
Regards,
Ray
Hi everyone,
Remember I was testing Brent's C2 clock? Well, testing is finished, and the final results are in. Slightly delayed, due to family expansion over here, but no less interesting because of that.
To get right to the point: the performance is very good! 
First a disclaimer: measuring jitter in the pico-second range is not a problem nowadays, with a suitable (expensive) analyzer. Unfortunately, the possibilities in my case were limited by the available equipment: an Agilent Infiniium oscilloscope model 54835A. Because it's functionality regarding jitter is limited, it is not possible to filter the jitter-signal. The scope will therefore display the entire jitter-spectrum, while the area of interest for audio is actually only the low-frequency portion. The results will be a bit higher because of this, and that's why this measurement is only a relative comparison.
Brent has kindly donated a C2 for testing purposes, thanks again Brent! I used The Flea with a 24.576MHz low-jitter Tentlabs XO module from my ezDAC for comparison and the C2 with both 16.9 and 33.8MHz oscillators. The result: the performance of all three is very good, AND very close!
From the C2, the lowest jitter clock signal can be obtained directly from the oscillator output, at point T1 on the PCB. This is no surprise of course, because at this point there is no extra logic in the signal path that can cause jitter. The outputs F/1 through F/4 show slightly more jitter, but only a little bit. The extra divider logic has an extra benefit: a nicely squared-up waveform with steep flanks, and this keeps the rise in jitter limited.
I also tested the C2 in my CD63mkII-KI and clocked the servo with the F/2 output. The result was clearly audible: a more controlled low-end, more quiet background and increased soundstage depth, to name a few.
So my conclusion is: the C2 is a highly recommended upgrade!
Regards,
Ray
Remember I was testing Brent's C2 clock? Well, testing is finished, and the final results are in. Slightly delayed, due to family expansion over here, but no less interesting because of that
.To get right to the point: the performance is very good!
First a disclaimer: measuring jitter in the pico-second range is not a problem nowadays, with a suitable (expensive) analyzer. Unfortunately, the possibilities in my case were limited by the available equipment: an Agilent Infiniium oscilloscope model 54835A. Because it's functionality regarding jitter is limited, it is not possible to filter the jitter-signal. The scope will therefore display the entire jitter-spectrum, while the area of interest for audio is actually only the low-frequency portion. The results will be a bit higher because of this, and that's why this measurement is only a relative comparison.
Brent has kindly donated a C2 for testing purposes, thanks again Brent! I used The Flea with a 24.576MHz low-jitter Tentlabs XO module from my ezDAC for comparison and the C2 with both 16.9 and 33.8MHz oscillators. The result: the performance of all three is very good, AND very close!
From the C2, the lowest jitter clock signal can be obtained directly from the oscillator output, at point T1 on the PCB. This is no surprise of course, because at this point there is no extra logic in the signal path that can cause jitter. The outputs F/1 through F/4 show slightly more jitter, but only a little bit. The extra divider logic has an extra benefit: a nicely squared-up waveform with steep flanks, and this keeps the rise in jitter limited.
I also tested the C2 in my CD63mkII-KI and clocked the servo with the F/2 output. The result was clearly audible: a more controlled low-end, more quiet background and increased soundstage depth, to name a few.
So my conclusion is: the C2 is a highly recommended upgrade!
Regards,
Ray
Last edited: