Subsonic Filters & Film Caps

[DF96]

SY just beat me to it. Capacitors are often nearer perfection than transformers, and are certainly cheaper. A good rule of thumb is: use the appropriate component for the task in hand.

A good thing to remember is that nothing can add to the quality of a signal, but plenty of things can damage the quality of a signal. The best we can do is to minimise the damage. Some of the things people do to avoid things that frighten them (e.g. DC servos to avoid coupling capacitors) can be worse than the thing avoided.

This whole hobby is built around fretting over parts and components and circuit designs

Some fret over parts. Others fret over circuit designs. If the latter is wrong there is no point in the former.

[AmCan]

Originally Posted by DF96

SY just beat me to it. Capacitors are often nearer perfection than transformers, and are certainly cheaper. A good rule of thumb is: use the appropriate component for the task in hand.

A good thing to remember is that nothing can add to the quality of a signal, but plenty of things can damage the quality of a signal. The best we can do is to minimise the damage. Some of the things people do to avoid things that frighten them (e.g. DC servos to avoid coupling capacitors) can be worse than the thing avoided.


Some fret over parts. Others fret over circuit designs. If the latter is wrong there is no point in the former.

Thanks for the reply. I image you are correct to a certain degree but then there is no one right way to make a circuit otherwise there there would only be one design and not the several that exist. Although I don't doubt that some achieve their objective more efficiently or effectively than others. Subjectivity comes into play. I am not qualified to judge a circuit design. All I know is I like the way my HK 17 preamp sounds. and I want to fret over what coupling cap I use in it because it gives me pleasure to do so

[DF96]

but then there is no one right way to make a circuit

Agreed, but there are a lot more wrong ways than right ways. At least, that is my intuitive feel. It could be that, given a set of objective criteria, it can be shown that a particular circuit best satisfies the criteria - fortunately most of us are not that interested in existence theorems for audio circuits, and I doubt if it is an active research topic in our universities.

[Wavebourn]

Originally Posted by DF96

A perfect cap will modify the frequency response. An imperfect cap will also add some distortion. No cap can add quality. Therefore there is no reason to have the subsonic filter engaged unless your source contains unwanted subsonic frequencies. The only exception to this is if you happen to prefer the sound of the distortion added by a particular capacitor.

No cap can add quality, but a filter on input can minimize distortions caused by other caps, after the filter, attenuating frequencies on which they distort more.

[Wavebourn]

Originally Posted by SY

...except the transformer, which is likely to be far more imperfect than a capacitor.

Let me defend the devil: when transformers softly saturate they add low order distortions to fundamental bass frequencies. Such trick is often used by audio engineers, they call it "Adding the focus". How I understand it, subjective perception of harmonics of fundamental bass notes cause perception of attenuated fundamental frequencies in phase with harmonics. But if to preserve level of fundamental bass frequencies, but add phase shifts, it would sound less focused. And less audible on iPods and in cars.

[DF96]

Originally Posted by Wavebourn

a filter on input can minimize distortions caused by other caps, after the filter, attenuating frequencies on which they distort more.

True, provided those frequencies are present in the input (hence my comment about the subsonic filter). Or, perhaps, they are generated as IM by other frequencies. It is always difficult to phrase a statement so it is clear and simple enough to educate the learner, but sufficiently accurate to satisfy the expert!

Regarding bass, I remember reading about this effect: harmonics of a bass note can give the impression that the (missing) fundamental is present.

[Wavebourn]

Originally Posted by DF96

True, provided those frequencies are present in the input (hence my comment about the subsonic filter). Or, perhaps, they are generated as IM by other frequencies. It is always difficult to phrase a statement so it is clear and simple enough to educate the learner, but sufficiently accurate to satisfy the expert!

Sure, and an expert knows that any filter attenuates, but not eliminates signals.

Regarding bass, I remember reading about this effect: harmonics of a bass note can give the impression that the (missing) fundamental is present.

Also, chair shaker sounds more natural than sounds of the same frequencies in headphones. Ears perceive bass more like inter-modulation of higher frequencies rather than as a sound. But when the body hears the bass it feeels sooo goood!

[DF96]

an expert knows that any filter attenuates

Apart from an all-pass filter.

[SY]

Originally Posted by DF96

Regarding bass, I remember reading about this effect: harmonics of a bass note can give the impression that the (missing) fundamental is present.

That's the LS3/5A in a nutshell.

[Wavebourn]

Originally Posted by DF96

Apart from an all-pass filter.

In Russian it is called Phase Shiftier, because it does not filter itself. But it can be a part of a filter, that actually attenuates. But again, does not eliminate. However, a pair of APFs can be used to create quite deep notch.