skp112233 said:
actually one of the lm3xx datasheet had an amp based on a regulator.
yeah, I would think slew rate matters.
I doubt it in the context we're talking about here.
It would be one hell of a preamp to be able to present a load in which slew-rate becomes relevant
Andy.
Yeah, I know.
It is fun though, and it gets people thinking and since few, if any, will have considered this, it should keep people busy for a while afterwards proving me wrong.
By which time I'll have fled the country 😉
Andy.
I shall not bury anyone, you're not right😉
It's certainly a relevant parameter though, but fairly large variations in it though are not subjectively dominant, within the context we're discussing here (i.e. preamps, primarily class A).
Andy.
Hi,
Thanks for the lecture Milwood.
That's what I meant when I said "in theory".
Unfortunately, my ears don't always agree with theory.
Cheers,😉
Thanks for the lecture Milwood.
That's what I meant when I said "in theory".
Unfortunately, my ears don't always agree with theory.
Cheers,😉
fdegrove said:
I think that's more than theory. a typical opamp by-passing for example requies both a large electrolytic and a small ceramic cap; and multiple capacitors are used and located throughout motherboards for precisely the same reason.
Hi,
My motherboard doesn't sound all that great though...
I'll tell you once more, I'm well aware of the theory behind the mixing of the caps and I also know what different technology caps do etc.
Knowing that, I also know that it doesn't necessarily make for a
good sounding PSU.
In fact, I never apply the theory of decoupling caps to lower ESR, and if I need multiple caps on the same rail I use exactly the same value and type throughout.
According to Peter Daniels' experiments with his gainclone, a single cap of appropriate value sounds better to him.
From past experience, I'd agree with him.
Just my experience, YMMV.
Cheers,😉
My motherboard doesn't sound all that great though...
I'll tell you once more, I'm well aware of the theory behind the mixing of the caps and I also know what different technology caps do etc.
Knowing that, I also know that it doesn't necessarily make for a
good sounding PSU.
In fact, I never apply the theory of decoupling caps to lower ESR, and if I need multiple caps on the same rail I use exactly the same value and type throughout.
According to Peter Daniels' experiments with his gainclone, a single cap of appropriate value sounds better to him.
From past experience, I'd agree with him.
Just my experience, YMMV.
Cheers,😉
Very true in the case of motherboards, really not applicable to quality audio. Bypassing with ceramics?! Most of the better sounding gear uses bypassing carefully; you may achieve improvement in sound but it takes a lot of effort. Using quality caps in appropriate locations is a much cleaner solution if you care about the sound. All these different resonances of multiple caps you mention can really mess up the natural timbres.
Talking of which, has anyone tried the bypassing method Audio Research uses, with a small inductor in series with the bypasing cap?
fdegrove said:
maybe you do. But you didn't come out that way, 🙂
fdegrove said:
maybe if you follow the theory next time, your amp will sound better. If you don't try, how do you know that it doesn't work for you, right?
Requires?
"I think that's more than theory. a typical opamp by-passing for example requies both a large electrolytic and a small ceramic cap; and multiple capacitors are used and located throughout motherboards for precisely the same reason."
I think you might find dozens of people who disagree with that and have built and sold products in contradiction to this. I avoid ceramic caps whenever possible as they usually sound nasty. Indiscriminate use of multiple caps leads to resonant behavior that can result in resonant frequencies at undesirable frequencies. The use of very low ESR electrolytics in the 10uF to 1000uF range for op amps in the 30MHz and under GBW is becoming very common. If you are going to HF bypass, I like 0.1 uF metalized polypropylenes. I often see people use things like 0.01 uF foil and film caps which can sound pretty bad.
Every cap and combination of caps will have a (several for more than one cap) resonant frequency and where that resonance lies can be critical in both analog and digital design. Don't take my word for it.
http://www.sigcon.com/pubsIndex.htm#bypass capacitors
"I think that's more than theory. a typical opamp by-passing for example requies both a large electrolytic and a small ceramic cap; and multiple capacitors are used and located throughout motherboards for precisely the same reason."
I think you might find dozens of people who disagree with that and have built and sold products in contradiction to this. I avoid ceramic caps whenever possible as they usually sound nasty. Indiscriminate use of multiple caps leads to resonant behavior that can result in resonant frequencies at undesirable frequencies. The use of very low ESR electrolytics in the 10uF to 1000uF range for op amps in the 30MHz and under GBW is becoming very common. If you are going to HF bypass, I like 0.1 uF metalized polypropylenes. I often see people use things like 0.01 uF foil and film caps which can sound pretty bad.
Every cap and combination of caps will have a (several for more than one cap) resonant frequency and where that resonance lies can be critical in both analog and digital design. Don't take my word for it.
http://www.sigcon.com/pubsIndex.htm#bypass capacitors
Recently I tried to bypass a cap in a tweeter network with silver mica cap, as someone mentioned it some time ago as an easy way to improve the sound of a speaker. I couldn't believe the sound it made. I quickly removed silver mica cap and will never try it again😉
I also don't think that millwood is really qualified to discuss sonic issues of PS, as he doesn't even believe that resistors sound differently.
I also don't think that millwood is really qualified to discuss sonic issues of PS, as he doesn't even believe that resistors sound differently.
Re: Requires?
and if we judge a product on the basis that someone is selling them, we would all be drinking snake oil by now, 🙂
they may indeed sound nasty to you. But I just haven't seen any statistically significant study on their sonic characteristics.
indiscriminate of anything, no matter how good they are, will result in bad things. Isn't that the definition of "indiscriminate"?
Fred Dieckmann said:
and if we judge a product on the basis that someone is selling them, we would all be drinking snake oil by now, 🙂
Fred Dieckmann said:
they may indeed sound nasty to you. But I just haven't seen any statistically significant study on their sonic characteristics.
Fred Dieckmann said:
indiscriminate of anything, no matter how good they are, will result in bad things. Isn't that the definition of "indiscriminate"?
Peter Daniel said:
Peter, first of all, I am not sure if I am talking about sonic issues of PS. 2ndly, I fail to see the link between resistor sound and by-passing caps. Enlighten me please?
Hi,
Maybe it's because I tried already ages ago already.
Never tried it. I assume they choose the L carefully with repect to C?
Worth a try, I reckon.
Although on its' own a siver mica of good quality is said to be amongst the best sounding caps around.
They're often used in RIAA correction ccts.
Ceramics and tantalum caps on the other hand should stay well away from audio circuitry unless you know exactly what you need them for.
Cheers,😉
Maybe it's because I tried already ages ago already.
Never tried it. I assume they choose the L carefully with repect to C?
Worth a try, I reckon.
Although on its' own a siver mica of good quality is said to be amongst the best sounding caps around.
They're often used in RIAA correction ccts.
Ceramics and tantalum caps on the other hand should stay well away from audio circuitry unless you know exactly what you need them for.
Cheers,😉
Inductor?
"Talking of which, has anyone tried the bypassing method Audio Research uses, with a small inductor in series with the bypasing cap?"
For moving the resonance I suppose? Changing the value of the decoupling cap might be an easier way to achieve this.
can you tell us the source of this information?
I have been looking at a small resistance in series for one cap in multiple cap applications.
"Talking of which, has anyone tried the bypassing method Audio Research uses, with a small inductor in series with the bypasing cap?"
For moving the resonance I suppose? Changing the value of the decoupling cap might be an easier way to achieve this.
can you tell us the source of this information?
I have been looking at a small resistance in series for one cap in multiple cap applications.
Whilst mica is a pretty good dielectric, having low DA and low loss at HF, for audio use I would prefer polystyrene, polypropylene or polyphenyl sulphide, where HF performance wasn't critical.
Judicious application could be beneficial in certain areas though.
Andy.
millwood said:
This was my impression. If we are talking about PS for audio applications, I assume that sonic performance would be actually the dirving force behind any PS design. If not, I don't know what this thread is all about indeed.
The link between resistors sound and bypassing caps looks simple to me. For someone who treats resistors as electrical parts only, without any consideration for possible sonic aspect, will talk about bypassing caps in a similar way, from measurements POV, concentrating on best measured performance but without any regard as to sonic aspect again.
Peter Daniel said:
that should give you enough reservation about those sonic tweaks floating around the net.
Fred, Frank, anyone else not auditioning ceramic caps in motherboards
Somewhere in the back of my memory i recall Audio Research had a patent for something remotely sounding like 'distributed PS decoupling". Looking at the PS of a PH3 one sees a 470uF in series with a parallel combo of a 2.7uH and 54R9. The resistor is presumably to damp the inductor and the result is a predictable impedance behaviour (rising) at high frequencies. All this is bypassed with a 2,0uF and 0.01uF. Seems like a reasonable way to achieve bypassing and at the same time decrease the sonic signature of the electrolytic at HF.
cheers
peter
Somewhere in the back of my memory i recall Audio Research had a patent for something remotely sounding like 'distributed PS decoupling". Looking at the PS of a PH3 one sees a 470uF in series with a parallel combo of a 2.7uH and 54R9. The resistor is presumably to damp the inductor and the result is a predictable impedance behaviour (rising) at high frequencies. All this is bypassed with a 2,0uF and 0.01uF. Seems like a reasonable way to achieve bypassing and at the same time decrease the sonic signature of the electrolytic at HF.
cheers
peter
millwood said:
This makes one thing certain for me, I will try everything myself, before incorporating it in a circuit. From what I see both on a forum and in general, not too many people are really into trying things for themselves and doing their own R&D's. 😉
- Status
- Not open for further replies.