Purpose of the resistors and capacitors on output here?

[Garbz]

Quote:

Originally Posted by rickcr42 and .....

Racing stripes, can't forget the racing stripes

Actually I already have a plan for this dac. Initially the first upgrade will be Jung Super Regulators controlling all aspects of the DAC's power. If it doesn't consume 1000% what it puts out it's not good enough yet.

Then I'll try playing with all passive output stages. I've already made all the necessary arrangements on the board to allow me to sit another PCB on top of it.

Then getting creative with the transformer coupling of the S/PDIF coax inputs, then I make it up from there.

The downside to this is that I'll end up retiring my wonderful Cinemag transformers since my DAC will already be fully balanced

 

[rickcr42]

Quote:

Then I'll try playing with all passive output stages. I've already made all the necessary arrangements on the board to allow me to sit another PCB on top of it.

Quote:

The downside to this is that I'll end up retiring my wonderful Cinemag transformers since my DAC will already be fully balanced

Passive resistor IV>jfet (or triode) gain stage>output trafos

Totally passive would mean a 1:10 or 1:20 step up transformer in combination with the IV resistor so a microphone transormer and again $$$$.You really don't want to try and force the DAC to produce voltage outside its "sweet spot" which depending on chip is in the 100mv-500mv range which can then be followed by a X10 active stage.Totally a by ear adjistment for your particular DAC and systme gain requirements.
Some shoot for 1 volt across the board for all sources followed by an active gain stage/preamp,some shoot for 2 volts across the board for all sources and use either a full passive preamp or passive /buffered.
I personally have 1 volts at all source outputs followed by another +6dB at the preamp stage followed by all low gain one or two stage final amplifiers (speaker or headphone amps).Any additional gain I need is punched in at the tone controls/eq/filter box.
All about system sensitivity and not some "standard" figure


Quote:

ctually I already have a plan for this dac. Initially the first upgrade will be Jung Super Regulators controlling all aspects of the DAC's power. If it doesn't consume 1000% what it puts out it's not good enough yet.

the Jung regulators would be misused for digital chips.Far better to do the +8 volts pre-regulator buss (I use a zener referenced slow turn on LM-350K for all my pre-regs) then individual shunts for each digital chip power pin with a LC decoupling filter for each.Total isolation of each power feed.The output stage power source would be the place for the jung.
The ultimate ? Totally battery powered to replace the initial pre-reg AC supply.Would preclude triodes in the output stage unless you want to use a low voltage type such as the 6GM8 but still in range of the jfet run at +24 VDC.You would then add you output trafos to this stage and have both SE and BAL outputs.Consider the trafos as a final low pass anti-alias filter and digital ground noise isolator.
At least that is how I would do 'er.

BTW-what DAC are we talking about ? Surely not the above which has (as far as I can tell) only voltage outputs in which case RC filters to a 1:1 output trafos would be my call.

 

[Garbz]

The above? Ewww I shudder at the thought. No i'm designing my own from the ground up. It was a race between me an Daki][er but I ran into cashflow problems and now manufacturing problems so he won

Basically it will be a CS8416 receiver -i2s> PCM1730 -current> OPA637 -voltage> single 2sk389 BJT follower biased in class A. with an analogue filter somewhere in there. I've actually already bought all the parts and i'm just waiting for sparkfun to resume operations so I can get the damn pcb printed.

The jung regulators were mainly going to run the 5V analogue supply on the DAC chip, and the 3.3V digital supply for the phased locked loop on the CS8416. Everything else was going to run from a standard lm317. Not sure what I'll do with the output stage yet though. I'd like to go overkill there too but even at this stage I think the SNR will be better then the 98dB on my integrated amp.

If I used the passive I/V -> JFET gain stage approach then I wouldn't need the output trafos would I. Kiss principle

Also as far as actual output volume goes I'll have to tweak it to exactly the output of my cdplayer which is currently perfect. My quietest song only just gets to the 10 oclock mark on the volume control so at full volume it's only just uncomfortable. At a later date i'll establish what rms voltage this value actually is.

 

[rickcr42]

Quote:

If I used the passive I/V -> JFET gain stage approach then I wouldn't need the output trafos would I. Kiss principle

I personally can not stand digital from any source without a transformer isolating the digital ground noise that for some reason I am extremely sensitive to (and what some call "detail"

) so use one in all my digital designs.Since this is a new design have you considered a transconductance amp for the IV stage ?

http://www.maxim-ic.com/appnotes.cfm/appnote_number/692

http://pdfserv.maxim-ic.com/en/ds/MAX435-MAX436.pdf

http://www.diyhifi.org/forums/viewtopic.php?t=347

Have the chips but have not yet actually tried it.i will sooner or later though

Quote:

The jung regulators were mainly going to run the 5V analogue supply on the DAC chip, and the 3.3V digital supply for the phased locked loop on the CS8416. Everything else was going to run from a standard lm317. Not sure what I'll do with the output stage yet though. I'd like to go overkill there too but even at this stage I think the SNR will be better then the 98dB on my integrated amp.

don't get hung up on the specs man ! Way better to design for sonic content and i defy anyone to tell me they can hear the difference between 90dB and 98 dB SNRs

if you are smart you will do a modular build so you can upgrade individual sections as you see a path to improvement.Otherwise you will be locked in and have to do an entire new design to implement a single idea.Just a thought............

 

[Garbz]

Way ahead of you man. Even have screw holes so I can bolt new boards on top of the old one feeding of the old dac, or the old i2s line. Powersupply is completely external so that leaves a lot to play with. What can I say. I love messing with things. No doubt you do too

Yeah specs are well useless in a real world situation. Even using me dodgy tube as a preamp with it's ludicrously high noise (76dB or something) I couldn't hear noise over the environment like the wind outside or cars etc. That's why I guess the jung reg is most important for the PLL where it matters.

I've never even heard of a Transductance amp. I'll read up sometime I've had a little less to drink. Atm the board will be done as mentioned above, mainly becasue the board desing is finished and the designs have been sent off. But it's something to look at once it's working and I start playing.