Have 2 X 15000 uf and 2x 10000uf filter caps in the amp pwr supply.
Will it benefit, if i put on smaller in parallel as well, improve ESR?
Maybe sound better?
Nad 3400
Cheers
Will it benefit, if i put on smaller in parallel as well, improve ESR?
Maybe sound better?
Nad 3400
Cheers
In reality, no, the benefits of parallel capping like that are nonsense dreamed up by compulsive non-techs and spread across the internet.
Stability purposes in high impedence TUBE circuits.
But the OP is discussing a Solid State amp.
The impedance at the point where the parallel caps are in the power supplies of the meters in post #3 is low.
They date from an epoch when E-caps had severe imperfections (and were not as well understood as today): series inductance and resistance.
As early as the seventies, these flaws were mostly addressed
Wow, no explanation whatsoever, while you were the one that brought up "high impedance TUBE circuits".
@ Elvee: Thanks. That looks like a possible/plausible explanation to me.
You'll have to fight that one out with Dave Jones, I think he wins the argument convincingly though, certainly not nonsense.
This video talks about similar capacitors //, which is OK, and often commendable.
What PFL200 seems worried about is the paralleling of a largish E-cap with a smallish cap, like 6.8nF in one example. This is completely counter-productive.
Some years ago, I made hundreds of systematic measurements of various combination using a VNA, and I posted a number of results on the forum (but due to the impaired search function of the new forum, I cannot locate them).
I could repost them, as I still have them stored somewhere, but the obvious conclusion was that 90% of the combinations produced unwanted dips and peaks in the resulting impedance, some of them very nasty. Around 5% were globally neutral, and 5% showed an improvement.
When the capacitance ratio is >100, the result is always detrimental.
1:20 to 1:50 is generally OK.
These are rules of thumb, valid for a direct paralleling, with negligible distance between the two caps, and no additional damping components.
The situation would be very different if the smaller cap is distant, at the remote end of an umbilical supply for instance.
Paralleling of similar caps is normally OK, and generally beneficial
What PFL200 seems worried about is the paralleling of a largish E-cap with a smallish cap, like 6.8nF in one example. This is completely counter-productive.
Some years ago, I made hundreds of systematic measurements of various combination using a VNA, and I posted a number of results on the forum (but due to the impaired search function of the new forum, I cannot locate them).
I could repost them, as I still have them stored somewhere, but the obvious conclusion was that 90% of the combinations produced unwanted dips and peaks in the resulting impedance, some of them very nasty. Around 5% were globally neutral, and 5% showed an improvement.
When the capacitance ratio is >100, the result is always detrimental.
1:20 to 1:50 is generally OK.
These are rules of thumb, valid for a direct paralleling, with negligible distance between the two caps, and no additional damping components.
The situation would be very different if the smaller cap is distant, at the remote end of an umbilical supply for instance.
Paralleling of similar caps is normally OK, and generally beneficial
Someone correct me if I'm wrong, but aren't most discussions on Youtube of paralleling capacitors made by people who are talking about digital circuits (even if they are not specific about this) operating at high frequencies? They're discussing effects that occur in the tens of megahertz range, not 20kHz.
Oh, I see now.... You're showing me some youtube video of some guy named Jones..
Whom I never had any dealings with, nor do I know of his qualifications, background, schooling, and certificates/dimploma's.
Just some guy on a youtube video...... and I'm supposed to believe in what he preaches?
In my actual, real world Mr Tillotson, the "virtual" world doesn't supply me with the things I need to believe in, or food for my table.
Nor do I prefer to "argue" or fight with or concede to "virtual" beings.
You want to argue, I don't.
Those two caps are not for filtering, they're plate isolation / coupling, not power supply.
The two capacitors are connected between B+ and ground. For me they are part of the power supply. But if you insist in calling them "plate isolation / coupling" instead, than that's fine with me.
I managed to recover some of the results of my measurements:
https://www.diyaudio.com/community/...ps-with-electrolytic-caps.106648/post-2257381
https://www.diyaudio.com/community/...ps-with-electrolytic-caps.106648/post-2257381
I'm building an amplifier. Dual mono. Separate transformers (500VA, 8A) per channel, output stage supply voltage - 40V.
I want to organize the capacitors in the power supply as follows:
First (one channel) there are four 22000uF capacitors next to the transformer, then on the printed circuit board five centimeters from the output transistors there will be two low-impedance ones at 4700uF, and at the very collectors of the output transistors - 10uF per transistor (6 pieces in total - 60uF) - polypropylene .
Tell me, how successful will such a bunch be?
I will use capacitors in this order:
I want to organize the capacitors in the power supply as follows:
First (one channel) there are four 22000uF capacitors next to the transformer, then on the printed circuit board five centimeters from the output transistors there will be two low-impedance ones at 4700uF, and at the very collectors of the output transistors - 10uF per transistor (6 pieces in total - 60uF) - polypropylene .
Tell me, how successful will such a bunch be?
I will use capacitors in this order:
Last edited:
Not in the 1604. It does look that way but the drawing is confusing.
A mica across an electrolytic did make sense in 1959. Electrolytic caps had large internal impedance, especially by 100kcs-1mc, and large dimensions so could not be right AT the connector or load. Modern e-caps have much lower stray impedance and can be mounted much closer to where they are needed.
If by '1604' you mean the GM6014 than please look again in the schematic (see post #3). C504 (8 uF) is paralleled by C505 (100K = 0.1 uF). On the power supply circuit board they are also right next to each other.
Thanks for your possible/plausible explanation, which is in line with Elvee's explanation in post #7. C505 is not mica but polyester, but that probably won't matter much for your explanation.
Thanks for your possible/plausible explanation, which is in line with Elvee's explanation in post #7. C505 is not mica but polyester, but that probably won't matter much for your explanation.
And the discussion was originally about a solid state amplifier.
How and why is it now about tubes?
How and why is it now about tubes?
So capacitors 'know' whether they are paralleled in a solid state circuit or in a tube circuit? And your post #2 (with 'nonsense' and 'compulsive non-techs' in it) already said it all?
But don't worry, I'll stay out of this thread from now.
But don't worry, I'll stay out of this thread from now.