Is a single capacitor power supply possible.

Thread Starter

john1211

Joined Sep 17, 2018
12
Hi thanks for reading.
I built a simple psu for my amplifier, consisting of a 12v 15 amp transformer, followed by a bridge rectifier, then a 47,000 uf cap.
There was noise coming through the speaker so i inserted another 47,000uf and the noise went down. Then I kept inserting caps, the noise kept lowering, until i reached 135,000 uf and the noise disappeared completly, using six caps in total.
I figured 135,000 uf was the perfect value so i took out the six caps and replaced with a single 150,000uf cap which is the closest value i could find, but some of the noise came back.
Why is the noise returning. Is it because the first cap smoothed the ac from the bridge rectifier, the second cap smoothed a little more, until it reached the sixth cap and was further smoothed to clean out the ripple totally, something a single cap cannot do.
Or is it the six caps being paralled connected had lower esr and esl values which aided the smoothing process.
From this is it possible to build a power supply with just one smoothing cap or is this going too far.
The reason i am trying to use a single capacitor in the power supply is the more caps i put into the psu the more the sound changes, becomes inaccurate when compared against a single cap.
Caps add a certain colouration of their own, the more caps the more the colouration, the more inaccurate the music sounds.
Searching around i've read a ton of threads wherein people say things like elna caps sound different / better than nichicon, nichicon sound different to panasonic, panasonic sound different to mundorf. There too much evidence to think i am making a mistake or imaging it.
Just am disliking the sound of multiple caps and all the different colourations messed up the experience for me. Sorry for being so picky.
I think what is happening adding caps in 'bypass' with each other ends up with each cap's frequency impedance being added to the mix, in the end i ended up with a filter with multiple poles - almost like a EQ. That also varies with temperature, model range (capacitance) and the model/make build etc depending on the cap. I suspect this is why the sound is being coloured. The treble goes down considerably compared to using a single capacitor.
What do you think, why is a single cap not effective against lowering the 50hz ac mains noise compared to a bank of caps. Can anything be done about this, or will i be forced to use more than one capacitor.
Also, what do you think about going up in capacitance. I am on 150,000uf at the moment.
Thanks.
 

Ya’akov

Joined Jan 27, 2019
9,070
I don't understand how you think output capacitors on the DC power supply are affecting the frequency response of the amplifier being powered, particularly at higher frequencies.

If you wanted to say that large bass transients caused sags in the supply voltage, I could understand that. But the PS output capacitors have no way to influence the high frequency performance of the amplifier. That are not not in the reproduction path.

As far as the caps themselves, they aren't ideal capacitors with one dimension: C, the also have ESR and ESL (Equivalent Series Resistance and Inductance) and that will affect their performance as filters.

You also need an inductor in the filter formed by the capacitor for best performance.
 

Thread Starter

john1211

Joined Sep 17, 2018
12
Thanks. I have thought about crc filters and inductors, first i would like to find out why a single cap is not smoothing as well as a bank of caps.
 

DickCappels

Joined Aug 21, 2008
10,152
Does your power supply have a regulator? That can cut AC hum by a factor of 1,000 or more.

Your capacitors don't only have capacitance. There is also some series resistance in them. Changing capacitors changes the resistance which affects the power line frequency hum amplitude.
 

Ya’akov

Joined Jan 27, 2019
9,070
Does your power supply have a regulator? That can cut AC hum by a factor of 1,000 or more.

Your capacitors don't only have capacitance. There is also some series resistance in them. Changing capacitors changes the resistance which affects the power line frequency hum amplitude.
Yes, the ESR of the larger cap is most likely higher so the high current at the peak conduction is introducing ripple on the output. But... I don't get how high frequency performance of the signal path is affected by the PS output filter....
 

Thread Starter

john1211

Joined Sep 17, 2018
12
Thanks for the replies, much appreciated.
Yes the caps in the psu can affect the frequency response of the amp.

It reminds me of a quad amplifier that i repaired a few years ago. The stock caps were 10,000uf. I took them out and replaced with 15,000uf and noticed the very low bass started coming through slightly better. I then bypassed the 15,000 uf with 1800uf cap and noticed a slight increase in lower midrange detail. I then added a 100uf cap and noticed a slight increase in treble detail.

Adding different sized caps does seem to be doing something.

It reminds me of a pair of speakers i built, wherein the crossovers required different size caps for the different drive units, bass midrange and treble, the caps capacitive reactance changes with frequency so affects different frequencies in different ways. Small caps block bass and are used in the tweeter. From my experience different size caps in the power supply do affect the frequency response of the amp, guess you would not suspect this unless you have done it.
I do appreciate your point of view and i also would have thought different cap sizes in the psu should not affect the frequency response, but real world test show different.
 

crutschow

Joined Mar 14, 2008
34,283
i would like to find out why a single cap is not smoothing as well as a bank of caps.
It's likely the equivalent ESR of the multiple caps is less than the ESR of a single cap.

Your amplifier apparently has poor power supply ripple rejection.
What amp design is it?

Using a capacitive multiplier circuit or voltage regulator would also significantly reduce the ripple and the observed noise.
 

Ya’akov

Joined Jan 27, 2019
9,070
Thanks for the replies, much appreciated.
Yes the caps in the psu can affect the frequency response of the amp.

It reminds me of a quad amplifier that i repaired a few years ago. The stock caps were 10,000uf. I took them out and replaced with 15,000uf and noticed the very low bass started coming through slightly better. I then bypassed the 15,000 uf with 1800uf cap and noticed a slight increase in lower midrange detail. I then added a 100uf cap and noticed a slight increase in treble detail.

Adding different sized caps does seem to be doing something.

It reminds me of a pair of speakers i built, wherein the crossovers required different size caps for the different drive units, bass midrange and treble, the caps capacitive reactance changes with frequency so affects different frequencies in different ways. Small caps block bass and are used in the tweeter. From my experience different size caps in the power supply do affect the frequency response of the amp, guess you would not suspect this unless you have done it.
I do appreciate your point of view and i also would have thought different cap sizes in the psu should not affect the frequency response, but real world test show different.
Improved bass with added bulk capacitance in the PS makes sense. Bass requires power, and if the supply is not up to providing the needed peak power, large caps can prevent sagging.

Differential effects on frequencies that don’t stress the PS don’t make sense. Reducing midrange performance with larger caps makes no sense with the possible edge case of a time when the bass is produced better by the larger supply of current and the midrange of the speaker is interfered with by the great bass reproduction, but that’s not because of filtering in the PS.

You are conflating capacitors in the power supply with capacitors that function as filters in the signal path. Crossovers have the audio signal pass through them. This is not true of the power supply which can only either not provide enough power (poor bass response) or produce noise that will be superimposed on the signal.

PS filters aren’t in the signal path. Your by ear judgment of “slight improvements” is not “real world testing”, get a calibrated microphone and do a spectral analysis and then we can talk. I have had a lot of experience with audiophile gear and have reasons to completely distrust subjective analysis.

I am not skeptical because of inexperience, it precsiely because of experience in both reproduction and recording that I find no credence in a large swath of ”audiophile” lore that has no measurement behind it.
 

Ian0

Joined Aug 7, 2020
9,668
Improved bass with added bulk capacitance in the PS makes sense
It makes some sense but not much. If you have 20% ripple on the power supply, and double the capacitance so you have 10% ripple, then the point at which the amplifier clips on a bass transient increases by just 1dB.
if the amplifier is not clipping, then it makes no difference at all.

And running it that close to peak output means that for every peak which would have been clipped but is now undistorted, there will be dozens of slightly larger peaks that still get distorted.
 

Ya’akov

Joined Jan 27, 2019
9,070
It makes some sense but not much. If you have 20% ripple on the power supply, and double the capacitance so you have 10% ripple, then the point at which the amplifier clips on a bass transient increases by just 1dB.
if the amplifier is not clipping, then it makes no difference at all.

And running it that close to peak output means that for every peak which would have been clipped but is now undistorted, there will be dozens of slightly larger peaks that still get distorted.
Did you miss that the TS said with sufficient capacitance to fix bass transients midrange was lost? I am happy with the idea that an insufficient supply could cause high demand slumping and so distortion at bass peaks but if you increase the capacitance to the point that bass problems are gone, why would you hear a degradation in the midrange, and further, why would you hear an in improvement in the high end by making the capacitance even more insufficient?

EDIT: to be clear I understood what you meant, but I have seen cases where large woofers caused trouble with the power supply that more capacitance on the PS output helped. It's the rest that baffles me.
 

Ian0

Joined Aug 7, 2020
9,668
It’s interesting that there is just one power supply capacitor. That means that this must be a AC coupled amplifier with coupling capacitor to the loudspeaker.
For a 12V transformer, the power supply is only going to be 17V, which would give only 4.5W into a 8Ω load, so why the 180VA transformer? I would have thought 4700uF would have been plenty if there was adequate PSRR, as the current drawn from the capacitor will only be 260mA.
The ripple at 1500000uF should only be 17mV, but the pulse currents in the transformer wires must be huge, and very short.
 
Last edited:

LowQCab

Joined Nov 6, 2012
4,023
The thread Starter is searching for that ever-elusive "Magic-Tone" for his guitar.
The Amp design is extremely poor ...... on-purpose ......
This is the strange world of "Musical-Instrument-Amplifiers",
which has nothing to do with Logic, or perfect signal reproduction.
It's all about the "character" of the heavily Distorted "Tone".

A pile of multiband EQ's strategically placed between Effects-Pedals will work wonders.
For Lead-Guitar, you need to boost 1Khz at EVERY opportunity, and generally
"roll-off" everything above ~3 to 4Khz, and below 100hz.
Quite often, this is done for you by a really "Low-Fi" Speaker, and Cabinet, combination.

I would suggest that you use a very High-Fidelity, very High-Power, Amplifier that
will never be pushed into Clipping,
along with 2- reasonably High-Fidelity Home-Stereo-Speakers.
These can be purchased super cheap from any Pawn-Shop.

Then, drive the Amp and Speakers with a Cheap Lap-Top Computer which is
running any one of a huge number of Guitar-Amp-Emulator Software Packages.
As far as I'm aware, there are no "bad" ones, only some with more features than others.
This will allow you to experiment with hundreds of presets, or create your own
"Chain-of-Effects" which suits your taste.

This setup works REALLY WELL,
and allows you to have any "Tone" you want with the click of a Mouse,
and there's no Tubes to replace.
Then all you have to choose are the Pick-Ups and Strings on your Guitar,
to get any "Tone" ever created.

Back-in-the-Day I used a Software-Package called "Guitar-Rig",
I think they're still around,
but there have been plenty of new-comers over the last ~20 years.
 

Audioguru again

Joined Oct 21, 2019
6,673
The amplifier that makes "a noise" probably uses vacuum tubes with no negative feedback. Lots of distortion.
The guy replacing its antique worn out capacitors wants the amplifier to have a baaaad unique sound.
 

LowQCab

Joined Nov 6, 2012
4,023
I think it's absolutely laughable when I read or see Videos talking about the absolute "need"
to retain ~30 to ~50-year-old Paper and Wax Capacitors that now have a 150K DC Leakage Resistance,
which trashes every Bias-Value throughout the entire Amp,
and has half the Tubes going for a Nuclear-Melt-Down,
because ONLY the original Caps have the "Magic-Mojo-Juice" inside.
Gimme a break ......
 
Top