MF X-DAC Output stage SCHEMATICS

May I know what does the 2 opamps do (between the active i/v stage and output buffers ?)

Can I remove them ?

Will
 

Attachments

  • mf-xdac-output schematics.gif
    mf-xdac-output schematics.gif
    17.4 KB · Views: 1,228
Werner said:
Am I correct in that they use a dual opamp with one half for
the IV conversion, and the other half for the output buffer?
I knew recent MF stuff was of low design quality, but this ...



All in all they used 4x dual op-amp for left and right channels !!

This dac design really suck, especially on the PSU section where they use a 12vac and convert to a dual +/- 15 v regulated for the op-amp, thru the implementation of half wave rectification.



will.
 

jean-paul

diyAudio Moderator Emeritus
2002-09-20 7:20 am
Germany
I wouldn't remove them as they function as a lowpass filter with R101. This filter ( forgot the name of it but acts more or less like an inductor ) is connected in parallel with the signal and not in series as usual. It was used by many manufacturers in their higher range cdplayers in the past.

It could be a drawing error but normally the output pin 7 of U100b would be connected to the non inverting input at pin 5. A 100 k resistor to ground from pin7 to avoid plops when cables are disconnected and then the stop-resistor R107 from pin7 to output. I also fail to see why C109 has to be an electrolytic of 10 uF with R106 being 10 k when a quality film cap could have been used with a higher value resistor !?!?

It could be that my theory is getting rusty, I know.
 
I have removed this kind of filter from many many CD players, and it improves the sound every time.

Even those DAC's with a pure 0-5V output like SONY Pulse, you can get away with removing the GSC filter (the filter in question). I never experienced any problems with excessive glitch noise on the output.
And in this case the two other filter stages are enough.

I would remove it any day of the week.

Best regards

Lars
 

Nicke

Member
2001-06-20 11:27 am
Sweden
This kind of filter was used much more about 10-15 years ago.
You can see it in some older DAC datasheets like PCM63 and some others.

Lars Clausen said:
I have removed this kind of filter from many many CD players, and it improves the sound every time.

Even those DAC's with a pure 0-5V output like SONY Pulse, you can get away with removing the GSC filter (the filter in question). I never experienced any problems with excessive glitch noise on the output.
And in this case the two other filter stages are enough.

I would remove it any day of the week.

Best regards

Lars
Lars,didn´t you use GIC filter in your first DAC,the combined DAC/preamp AIR?
 

jean-paul

diyAudio Moderator Emeritus
2002-09-20 7:20 am
Germany
Lars Clausen said:
GIC .. thankyou !!! :D 2 out of 3 .... Yes we did, it was the hot thing to do in 1990. ;) Later it turned out we got better sound from just bypassing it, and the PCM63 gives out relatively little glitch noise. So it gave no side effects.

Sorry Lars but it seems to me that when a design is prototyped this is discovered in the first stage with listening tests and measurements. Why using excessive parts when you could have done without them ? Blindly following datasheets or what competitors do ?

Proof of my personal theory that listening tests are mostly done by customers.
 
I have managed somehow to cut the original file size from 700 to < 100Kb.

This is the analog stage in my Denon 1650SR. It is differential I / V to single ended filter and buffer. It is very complex and by far the weakest part in my CD player. I also tried to remove the filter in question and did not like the sound. It did not sound "correct" and the bass definition was smeared - I was under impression that bass timing was off and the mids were actually harsher without this filter. Replacing OP275 for AD826 at this place brought much better timing in low and mid range.

I did improve the sound of the whole analog stage. I to V converter would benefit if very low input bias current (and fast!) IC is implemented. I used AD8066 with excellent results (power supply is + and - 12V). The other IC's are AD826, except the buffer which is still OP275.

Each buffer IC supplies one RCA connector and one channel of the headphone stage. I disconnected the headphone stage and "paralleled" OP275 witch also improved sound a lot (I use 1m ribbon interconnect cables with highish capacitance load on the buffer stage).

It seems that over-complexity payed off in one area – the sound appears correct in time domain.

You could try for yourself. These are just suggestions and they may not work okay in your DAC.

Regards,
Extreme_Boky
 

Attachments

  • analog.jpg
    analog.jpg
    98.4 KB · Views: 448

Nicke

Member
2001-06-20 11:27 am
Sweden
Lars Clausen said:
GIC .. thankyou !!! :D 2 out of 3 .... Yes we did, it was the hot thing to do in 1990. ;) Later it turned out we got better sound from just bypassing it, and the PCM63 gives out relatively little glitch noise. So it gave no side effects.
Then I remembered right ;)
I actually built it myself...but the DAC was not that good that I thought when I built it.I never tried the upgrade with the AD844 you upgraded it with.

BTW you must have been pretty early with the 844 used that way(same as Pedja does now..)?