Discreet output stage for Sony c222?
I have a Sony SCD C222ES. I would like to build a discreet output stage, taking the output from the DAC. I've seen a couple of examples on the net for other players (i.e., www.ultranalog.com), but those appear to require a balanced input. As far as I can tell, the DAC output on this player is single ended. Anyone have suggestions for a good anlalog output stage. Doesn't need to be anything exotic.
Sheldon
I have a Sony SCD C222ES. I would like to build a discreet output stage, taking the output from the DAC. I've seen a couple of examples on the net for other players (i.e., www.ultranalog.com), but those appear to require a balanced input. As far as I can tell, the DAC output on this player is single ended. Anyone have suggestions for a good anlalog output stage. Doesn't need to be anything exotic.
Sheldon
rfbrw said:
Ah, yes. Pretty obvious oversight on my part. Anyway, the DAC is a CXD 9658N. It's a combined digital filter and DAC. I haven't been able to find out a whole lot about it yet. I'm guessing it's a voltage output. Each channel feeds a njm2114m-te2 op-amp. Nothing in between except some filtering.
Sheldon
Hi Sheldon,
It appears your DAC is voltage - out DAC.
You could also try better quality OP amplifiers like AD826 (simple swap), or excellent AD 8066 (surface mount dual, applicable with adaptor board and ONLY if the voltage is + and - 12V max!) Try also better quality capacitors like WIMA polypropylene for analogue filtering and Black Gates FK for analogue power supply decoupling, or even separate power supply for analogue section!
I had some excellent results with dual triodes, but this requires a lot of hassle - high plate + separate heater and biasing voltages, and large coupling capacitors..... However, if you want to experiment, give them all a go and decide what suits your taste the best!!! Some members report an excellent result obtained with simple transformers – I haven’t tried them yet.
Regards,
Extreme_Boky
It appears your DAC is voltage - out DAC.
You could also try better quality OP amplifiers like AD826 (simple swap), or excellent AD 8066 (surface mount dual, applicable with adaptor board and ONLY if the voltage is + and - 12V max!) Try also better quality capacitors like WIMA polypropylene for analogue filtering and Black Gates FK for analogue power supply decoupling, or even separate power supply for analogue section!
I had some excellent results with dual triodes, but this requires a lot of hassle - high plate + separate heater and biasing voltages, and large coupling capacitors..... However, if you want to experiment, give them all a go and decide what suits your taste the best!!! Some members report an excellent result obtained with simple transformers – I haven’t tried them yet.
Regards,
Extreme_Boky
Sheldon said:
It is a single-ended voltage out dac as used in XB770 SACD player.
This one may need a little modifying as it was meant for a current output dac
http://www.diyaudio.com/forums/showthread.php?s=&threadid=13873&highlight=
You could also try VSE's SLCF www.vacuumstate.com
Extreme_Boky said:
Thanks,
The audio ps in this unit is +/- 7 volts. Most of the discreet type of buffers, preamps tend to run 12-15 volts. Is there a general preference for the higher voltages? I have no problem making a dedicated ps.
Hmm, by using simple transformers, do you mean instead of the output coupling cap?
sheldon
Transformers do I to V conversion, natural aliasing filtration, gain and impedance matching, but I question a good impedance matching!!!. Check this site re transformers:
http://www.lundahl.se/
I like higher voltages because i prefer true-to-life sound, a bit like studio monitoring. Higher voltages almost always produce higher slew rates / faster settling times.
However, lower voltages give that warm sound that many people prefer.... not for me.
Power supply is 60 - 70% of global sound quality, yet it's quite often neglected. Try good quality transformer, soft switching fast recovery diodes (RURP3060), Black Gates FK for main storage + storage after voltage regulators. I do not like to bypass these caps with lower value (0.1uF) film types. High frequency decoupling should be done only as close to ICs (transistors) as possible and for this I use 0.1uF / 50V NX Black Gates, again, without any caps in parallel. This is possible due to very high quality of both of these capacitors’ types. Experiment a lot and see for yourself....
AD 8066 is outstanding amplifier and it should work no problems with + and -10V DC. I use them as I to V converters in dual differential PCM1704 configuration, with + and - 12V DC with addition of little heat sinks on top of them, to keep input bias current / conversion error to min. They tend to run hot as you increase voltage to their maximum of 24V DC.
You can try discrete anti aliasing (analogue) filters and see if you like them....
Many people report very good results with simple resistor as I/V converter and discrete amplification with passive filtration, or transformers.
Cheers,
Extreme_Boky
http://www.lundahl.se/
I like higher voltages because i prefer true-to-life sound, a bit like studio monitoring. Higher voltages almost always produce higher slew rates / faster settling times.
However, lower voltages give that warm sound that many people prefer.... not for me.
Power supply is 60 - 70% of global sound quality, yet it's quite often neglected. Try good quality transformer, soft switching fast recovery diodes (RURP3060), Black Gates FK for main storage + storage after voltage regulators. I do not like to bypass these caps with lower value (0.1uF) film types. High frequency decoupling should be done only as close to ICs (transistors) as possible and for this I use 0.1uF / 50V NX Black Gates, again, without any caps in parallel. This is possible due to very high quality of both of these capacitors’ types. Experiment a lot and see for yourself....
AD 8066 is outstanding amplifier and it should work no problems with + and -10V DC. I use them as I to V converters in dual differential PCM1704 configuration, with + and - 12V DC with addition of little heat sinks on top of them, to keep input bias current / conversion error to min. They tend to run hot as you increase voltage to their maximum of 24V DC.
You can try discrete anti aliasing (analogue) filters and see if you like them....
Many people report very good results with simple resistor as I/V converter and discrete amplification with passive filtration, or transformers.
Cheers,
Extreme_Boky
Extreme_Boky said:
Thanks for taking the time to respond. On the transformers, I was curious if they were sometimes used in applications like this, not for impedence matching, but for coupling instead.
I have some questions about the filtering. It would be helpful if I were to describe what is present. I haven't yet mastered any of the schematic drawing programs, so here is a description of the present analog filter system:
In order, after the DAC output to the non-inverting input: 47uf, 10k to ground, 6.8k, 680p to ground, 5.1k, 330p bypassing op-amp, 5.1k, 47pf to ground, then to the input.
From the output: 10K to inverting (with 6.2k to ground for the feedback loop), 47uf (bypassed with 15n), then two 330ohm and one 100 ohm resistors to the output, with muting transistors to ground between the resistors.
Given that background, if I make a discreet stage, any recommendations on filtering? I would probably leave the present output intact and make separate outputs for the new stage.
Cheers and Merry Christmas to some, Happy Holidays to all,
Sheldon
From what I could see, it looks like that first stage is a filter and second low gain amplifier. There are many things I don't like about this, mainly two coupling caps (47uF), 680pF cap loading DAC output to ground, and muting. This is very simple way of getting rid of high frequency rubbish in "one go".
There are many ways of improving the analogue stage, which I described above in my replies. SONY likes to implement high over sampling, but uses pretty good digital filters as well. One of Lundhal transformers (with an addition of AD8066 as a buffer, maybe...) might work fine with your CD player.
As I mentioned before, you have to try as many options as you can get your hands on, and then decide... Good analogue CRO is very handy when experimenting with different options.
Another thing: AD8066 might be too fast and accurate for some systems. If you find this was true, try OP275.
Extreme_Boky
There are many ways of improving the analogue stage, which I described above in my replies. SONY likes to implement high over sampling, but uses pretty good digital filters as well. One of Lundhal transformers (with an addition of AD8066 as a buffer, maybe...) might work fine with your CD player.
As I mentioned before, you have to try as many options as you can get your hands on, and then decide... Good analogue CRO is very handy when experimenting with different options.
Another thing: AD8066 might be too fast and accurate for some systems. If you find this was true, try OP275.
Extreme_Boky
Extreme_Boky said:
Thanks, I know the general filter stuff and have cook bood articles on filter design. From those , I can concoct a basic filter and buffer. What I don't know is the noise profile from this type of output. So I apologise for making the question too general. Maybe I should have framed the it this way; for this specific application, what is a typical filter type, order and corner frequency for a good starting point? Or, what are the frequencies to worry about here, and at what level do they occur? I'd like the LP filter to have as high a corner frequency as practicable, and as low in order as practicable, but still good enough to do the job. I can take a guess or I could base it on what Sony has in place. Maybe someone has a better recommendation, though? I don't mind looking on my own, but if someone has a good example or a good site to start with, even better.
Sheldon
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.