Can you really HEAR mkt caps?

[Pooh Bear]

The distortion produced by an electrolytic capacitor will be smaller than
the signal across the electrolytic capacitor. In fact, second-order
distortion will probably decrease with the square of the signal across the
capacitor and third order distortion will probably decrease with the cube
of the amplitude of the signal across the capacitor. Therefore in order to
achieve very low values of distortion, it would suffice to use a large
value electrolytic capacitor, so that the amplitude of the signal across it
is small.

That's true. Using 2 electrolytics back-to-back simply increases THD because
there are 2 sources adding together btw..

Graham

 

[Michael]

So, the amount of capacitance also depends on the amplitude of the signal?

UF value is more than just the reactance/frequency relationship?

I've always just used a high value electrolytic (100UF or more) to get as
much audio through as I can and (it always sounded better...more low end).

I was about to ask why I see values like 3.3uf or 4.7uf on the output of
audio opamps (especially these MKT caps).

How does the level of signal influence the value of capacitance? (not the
working voltage but the value)

I saw the back to back thing in Walter Jung's "How to pick a
capacitor"....hence my using it. (Not a series back to back....I'm putting
them in parallel to create a non-polarized combination).

 

[Pooh Bear]

So, the amount of capacitance also depends on the amplitude of the signal?

Actually, very slightly for an electrolytic, but its an indirect relationship.

UF value is more than just the reactance/frequency relationship?

Biggest uF is best for electrolytics.

Simply doesn't matter for film caps. Film caps simply suffer from big cost/uF.

I've always just used a high value electrolytic (100UF or more) to get as
much audio through as I can and (it always sounded better...more low end).

It's *not just about low-end*. Bigger electrlytics mean smaller signal volts
across the cap. This means less distortion as a result of non-linearity.

Film caps don't have the non-linearity issue but aren't available in practical
sizes for good low-impedance coupling.

I was about to ask why I see values like 3.3uf or 4.7uf on the output of
audio opamps (especially these MKT caps).

I've never seen that but I'm an audio pro and don't bother with audiophool
nonsense.

If you like to use $5 caps suit yourself !

I simply avoid using caps where they aren't needed. Often they aren't
essential. DC coupling is often preferable.

How does the level of signal influence the value of capacitance? (not the
working voltage but the value)

If the signal across an electrolytic exceeds ~ 100 mV they start to distort.

Obviously film caps don't distort ( no relevant mechanism ) but they aren't
available easily in large uF values.

I saw the back to back thing in Walter Jung's "How to pick a
capacitor"....hence my using it. (Not a series back to back....I'm putting
them in parallel to create a non-polarized combination).

Tha sounds intruiging.

I've never tested them 'back-to-back' !

BTW - the THD introduced by electrolytics is small typically and only at low
frequencies ( maybe 0.01% ) but still worth avoiding.

Graham

 

[Michael]

Thanks Graham, I'll print out this for reference.

Yeah, I'm no audiophool myself....it's just amazing the amount of debate
(and contradiction) you can run across.

Hell, it's about what you hear......and don't laugh but I'll take some
harmonic distortion and a tad of "soft compression" when I listen to
music.....and in blind tests, I imagine alot of others would too.

That Walter Jung article is on the web at his website.