Hi all
I`m currently in the process of planning som tweaks for my cdp.
And I wondered if some of you have an idea or know any ways to build a general dac output stage using opamps.
My old marantz 63mkII uses a npc sm 5872 dac and it does not have any analouge filtering like many other dacs.
Any ideas and advice will be fine and recived with gratitude
I`m currently in the process of planning som tweaks for my cdp.
And I wondered if some of you have an idea or know any ways to build a general dac output stage using opamps.
My old marantz 63mkII uses a npc sm 5872 dac and it does not have any analouge filtering like many other dacs.
Any ideas and advice will be fine and recived with gratitude
CD 63 issues
Hi Chris,
Please do a search for Marantz cd63 on the forum. We covered a lot of issues earlier and a service manual was also available in pdf form. These should give you lots of information. You will then be able to put up some more questions in detail.
I have modified my CD67 output. But have a look at the threads. TNT audio also has a very detailed upgrade proceedure for these machines. I can't remember the URL. You can get it by typing in Marantz cd63 upgrades and look for the tnt URL in GOOGLE.
have fun,
Cheers.
Hi Chris,
Please do a search for Marantz cd63 on the forum. We covered a lot of issues earlier and a service manual was also available in pdf form. These should give you lots of information. You will then be able to put up some more questions in detail.
I have modified my CD67 output. But have a look at the threads. TNT audio also has a very detailed upgrade proceedure for these machines. I can't remember the URL. You can get it by typing in Marantz cd63 upgrades and look for the tnt URL in GOOGLE.
have fun,
Cheers.
Chris said:
I guess you're aware of the LC-audio stuff aswell, if not take a look here:
http://www.lcaudio.dk/zapfilter.htm
http://www.lcaudio.dk/ad825.htm
Bessel type analog filter for CD63
CD63 already has an output stage, actually analog filters made with opamps. Strictly talking its output stage (HDAM) is made with bjts, but I don’t see any point in swapping them with the opamps.
First, it is really worthwhile to change opamps to something better as LM6172.
Second, I can suggest you one thing that was mentioned earlier, but concrete solution, as I know, was nowhere explained. It is about changing filter type to Bessel instead of the original Chebychev. There is a 2nd order filter around the second opamp. It has peak at treble to compensate treble roll-off (0.7dB at 20kHz) which is the consequence of the RC filter around the first opamp. This filter has ringing of some 10%. I swapped this to Bessel type changing C605 and C606 to 390pF. This also forces changing of the L601 and L602 to (ideally) 520uH to preserve frequency of the notch filter at ~352kHz. Unfortunately, all this means further treble roll-off of 0.5dB (1.2dB total). You can also model this filter to move roll-off further above the audible range, but effectiveness of such filter will be dubious.
Pedja
Chris said:
CD63 already has an output stage, actually analog filters made with opamps. Strictly talking its output stage (HDAM) is made with bjts, but I don’t see any point in swapping them with the opamps.
First, it is really worthwhile to change opamps to something better as LM6172.
Second, I can suggest you one thing that was mentioned earlier, but concrete solution, as I know, was nowhere explained. It is about changing filter type to Bessel instead of the original Chebychev. There is a 2nd order filter around the second opamp. It has peak at treble to compensate treble roll-off (0.7dB at 20kHz) which is the consequence of the RC filter around the first opamp. This filter has ringing of some 10%. I swapped this to Bessel type changing C605 and C606 to 390pF. This also forces changing of the L601 and L602 to (ideally) 520uH to preserve frequency of the notch filter at ~352kHz. Unfortunately, all this means further treble roll-off of 0.5dB (1.2dB total). You can also model this filter to move roll-off further above the audible range, but effectiveness of such filter will be dubious.
Pedja
would this work? (schematic of the cd63mkII dac outputstage with possible changes in red)
Sorry for the bad quality
The npc sm 5872 dac need some filtering at its output to work.
Please have a look at this website:
http://thunderstoneaudio.nav.to/
and look up the following project:
Thermionic Valve Analogue Stages for Digital Audio
Sorry for the bad quality
The npc sm 5872 dac need some filtering at its output to work.
Please have a look at this website:
http://thunderstoneaudio.nav.to/
and look up the following project:
Thermionic Valve Analogue Stages for Digital Audio
Attachments
Pedja said:
Thats true.
My initial goal with this thread was to ask if anybody had any idea on how to make a new/better outputstage for the particular dac that marantz cdp 63 use.
The point with the link was to give you some information about the npc sm5872 dac.
Having done some searches on the web and reading a few datasheets I see things a bit more clearly now.
Hi Chris!
I started this thread some time ago: http://www.diyaudio.com/forums/showthread.php?threadid=10140&perpage=15&pagenumber=1
My Onkyo uses SM5861AP which is very close to your SM5872.
I am working on my last suggestion, though it will be a little different.
Why do you think SM5872 needs filtering?
My Onkyo sound much better when I take the output just after the first opamps.
I started this thread some time ago: http://www.diyaudio.com/forums/showthread.php?threadid=10140&perpage=15&pagenumber=1
My Onkyo uses SM5861AP which is very close to your SM5872.
I am working on my last suggestion, though it will be a little different.
Why do you think SM5872 needs filtering?
My Onkyo sound much better when I take the output just after the first opamps.
Output
Hi Henrik
The sm 5872 is what Thorstein call a "Timeslicing (or Delta Sigma) DAC" an futher he sais it; "require lowpass filtering and (if we insist on being boring and conservative) a balanced to SE converter in case of most high performance devices".
AND more:
"The two outputs (from the dac) however need to be combined into a single-ended output and then further lowpass filtering is needed as the signal is overlaid with large quantities of ultrasonic noise. This is actually much more problematic than the already problematic situation with Multibit DAC’s."
The picture you have attached is from the datasheet is the manufacturers (npc) suggestion for the needed lowpass filter And its also the same type of filter the marantz cd63 uses.
Burr Brown dacs like the PCM 1710,1716,1728 and 1732 has integrated this in the dac chip so that you can do anything about this.
Also in the datasheet you can find fig 19 output interface 1 wich is a balanced to SE config for this dac. It all explained at this website:
http://thunderstoneaudio.nav.to/
Under: Thermionic Valve Analogue Stages for Digital Audio.
So what i have realized is that you need this filter at the output of the dac and that you cannot remove it.
A friend of mine bought the zapfilter for his Marantz 63KI and to make it work he had to connect the zapfilter after the lowpass filter. He could not simply connect it to the dac ouputs like its intended.
I did not know this when i started this thread and I`m sorry for the confusion that may have caused. My inital idea was to make a simpler output with fewer components in the signal path.I realize now that it will be bit more difficult.
Hi Henrik
The sm 5872 is what Thorstein call a "Timeslicing (or Delta Sigma) DAC" an futher he sais it; "require lowpass filtering and (if we insist on being boring and conservative) a balanced to SE converter in case of most high performance devices".
AND more:
"The two outputs (from the dac) however need to be combined into a single-ended output and then further lowpass filtering is needed as the signal is overlaid with large quantities of ultrasonic noise. This is actually much more problematic than the already problematic situation with Multibit DAC’s."
The picture you have attached is from the datasheet is the manufacturers (npc) suggestion for the needed lowpass filter And its also the same type of filter the marantz cd63 uses.
Burr Brown dacs like the PCM 1710,1716,1728 and 1732 has integrated this in the dac chip so that you can do anything about this.
Also in the datasheet you can find fig 19 output interface 1 wich is a balanced to SE config for this dac. It all explained at this website:
http://thunderstoneaudio.nav.to/
Under: Thermionic Valve Analogue Stages for Digital Audio.
So what i have realized is that you need this filter at the output of the dac and that you cannot remove it.
A friend of mine bought the zapfilter for his Marantz 63KI and to make it work he had to connect the zapfilter after the lowpass filter. He could not simply connect it to the dac ouputs like its intended.
I did not know this when i started this thread and I`m sorry for the confusion that may have caused. My inital idea was to make a simpler output with fewer components in the signal path.I realize now that it will be bit more difficult.
Hi Chris!
At the moment I am building the analog stage as shown in the lower scematic, which is based on Nelson Pass supersymmetry or X. I don´t think I need the lowpassfilter at the input, since the input capacitance of the mosfets will do the job. I hope that I can run this thing balanced, if not I will remove the X feedback and let it convert the LO and LON into a singleended output.
At the moment I use the red Left+ as SE output as shown in the attached scematic, and this is so so much better, hard to believe. I also tried the red Left+ and Left- as balanced output, but it turnd out awaful, no soundstage, everything bumped into the center. So my answer is yes to your question.
LO and LON.
At the moment I am building the analog stage as shown in the lower scematic, which is based on Nelson Pass supersymmetry or X. I don´t think I need the lowpassfilter at the input, since the input capacitance of the mosfets will do the job. I hope that I can run this thing balanced, if not I will remove the X feedback and let it convert the LO and LON into a singleended output.
Tube stage
Hi Henrik,
You must have gone to the link and seen
"Thermionic Valve Analogue Stages for Digital Audio".
The second circuit is probably what you are looking for.
The DAC will see a balanced load because of the 'balanced' filter. The tube will hardly load the filter and so the 'balance' will not be disturbed. In addition you get enough voltage to drive almost any amp.
I have used a tube and OPA2134 interchangeably after a passive filter on a DAC. The tube sounds much better , though both sound good.
If you were a tube purist you may want to implement the circuit without a solid state device. In any case let us know how your implementation worked out.
Cheers.
Hi Henrik,
You must have gone to the link and seen
"Thermionic Valve Analogue Stages for Digital Audio".
The second circuit is probably what you are looking for.
The DAC will see a balanced load because of the 'balanced' filter. The tube will hardly load the filter and so the 'balance' will not be disturbed. In addition you get enough voltage to drive almost any amp.
I have used a tube and OPA2134 interchangeably after a passive filter on a DAC. The tube sounds much better , though both sound good.
If you were a tube purist you may want to implement the circuit without a solid state device. In any case let us know how your implementation worked out.
Cheers.
ashok
Yes I am goning with the second scematic.
I got the missing information on how to deal with this modification from Thorsten Loesch´s article "Thermionic Valve Analogue Stages for Digital Audio".
The good thing about his design, besides from the tube, is that he uses the inductors, wich only reduces the voltage level very little, so one only need a follower.
I have red the recommended filtering from the datasheed of the SM5864 (which is close to te one in my CD) without inductors, but the cost is you miss half the output voltage swing, but I will play a little with this idea, and see what happens, I have attached the test scematic.
And I will be back with some comments on the test.
Yes I am goning with the second scematic.
I got the missing information on how to deal with this modification from Thorsten Loesch´s article "Thermionic Valve Analogue Stages for Digital Audio".
The good thing about his design, besides from the tube, is that he uses the inductors, wich only reduces the voltage level very little, so one only need a follower.
I have red the recommended filtering from the datasheed of the SM5864 (which is close to te one in my CD) without inductors, but the cost is you miss half the output voltage swing, but I will play a little with this idea, and see what happens, I have attached the test scematic.
And I will be back with some comments on the test.
Attachments
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.