If it is about analog, this goal seems not very important, depending of the PSRR of the amplifying circuits. As long as you cannot hear the hisses and rumbles, why fighting for more?
While the question can be: do this noise induce distortions (like in some DA converter) ?
The dynamic behavior of the PSU seems more important to me, as it can have a direct impact on the sound quality.
And, i will finish to look like a monomaniac, isolation
If it is about analog, this goal seems not very important, depending of the PSRR of the amplifying circuits. As long as you cannot hear the hisses and rumbles, why fighting for more?
While the question can be: do this noise induce distortions (like in some DA converter) ?
The dynamic behavior of the PSU seems more important to me, as it can have a direct impact on the sound quality.
And, i will finish to look like a monomaniac, isolation
Christophe, I don't hear noise and hiss in good part because I take a lot of trouble and cost regarding the power supplies. It is not easy to now work backwards to see what contributes more, and whose contribution is more important.
For example, I like to throw in a FET between the base of a zener regulated CCS transistor and ground. According to what someone posted right here several thousand messages ago, that particular connection should have a CMRR of at least -82 dB. I throw in a 47 Ohm and a 220 uF cap before by input stage diff amp(s). I sprinkle 10uF and 100 nF caps around the circuits, which are fed with my favorite zener diode assisted cap multiplier ("Virtual Battery", according to Technics/Panasonic). Plus a few other things, all resulting in deathly silence. And, I admit it, increased cost and size as well, but it's really not too bad.
My own "bean counter", my wife, is very tolerant of my small quirks.
And if she likes 'em, she takes 'em. I actually owned my HK 680 integrated amp for about 5 minutes after I got home with it, after that it's sitting in her room, powering her speakers, and I get to just look at it. Tant pis!
Better than compromising on the Zout imo
Something that is bothering me, is whether a decision to use low noise BC560C is useful in low PSRR CFA. With this amplifier, it is easier to get single digit ppm THD (3W) with higher supply rail. Alas, I'm not sure how much rail voltage is save for the BC with Vce max of 45V.
I have no idea if the low noise characteristics of the BC is useful at all.
Same here. Comparing circuits (by ears) is a simpler way for me than doing technical analysis. Once I'm satisfied with one scheme, I tend to repeat using it again and again.
How critical is supply voltage for preamp stages...really ?.
Op amps perform perfectly fine with supply series 10 ohm/shunt 100uF rail decoupling.
IME, rectifier stage followed by multiple RC-RC-RC (large cap values) filtering gives close to perfectly quiet supply rails, more than adequate for audio stages.
Sure, this scheme is prone to longer term variations in AC supply voltage, but in practice I find this is not a problem.
Dan.
Op amps perform perfectly fine with supply series 10 ohm/shunt 100uF rail decoupling.
IME, rectifier stage followed by multiple RC-RC-RC (large cap values) filtering gives close to perfectly quiet supply rails, more than adequate for audio stages.
Sure, this scheme is prone to longer term variations in AC supply voltage, but in practice I find this is not a problem.
Dan.
What do you mean? Of course it is very critical as probably every different solutions give audible result.
More than adequate in term of noise, similar to what you can get with 78XX/79XX IC.
I have opened a dedicated thread on this subject here:
http://www.diyaudio.com/forums/power-supplies/265435-tandem-regulators.html
Really? How silly do you want to get?
The revival of shunt regulators was partly inspired by listening tests, then there are other listening tests I know you are aware of.
Of course there are lots of easy measurements some of which actually correspond with listening tests.
An AC fed power supply is a quite complicated device. It has to not put noise back into the AC line, reduce incoming line noise, be reasonably ripple free and produce a constant voltage with low added noise.
John wants to start with the regulator side, previously I have shown lots of issues with the AC to DC side. Eventually I will finish my article so Jan can spread it out over a dozen bookzines or so...
The more I look at the issue the more I see.
ES
Do we ever come to some standards to use in this audio industry or do things always have to look so convoluted. You would think by now that a few very effective power supply topologies would have been perfected by now that would work with most pre and power amp circuits? I understand then you throw in variables such as transformer type, toroid vs R-core and other types and whether a CRCRC type of cap bank or whatever. Do we need 50 types of power supply for analog circuits or can we actually develop some universal designs that can be scaled for voltage and current depending on demand? So many ways to skin the cat it seems.
probably every = some!
Your turn!
Ed, it will please you that I have no opinion on this matter.
- Status
- Not open for further replies.