PI input cap Question

[stevlech]

There was discussion of Ken Fischer's choosing a .1 cap for the PI input and I can't recall the reasons given. Would anyone be so kind as to refresh my memory?

Thanks in advance

[chuck]

I can't say what was in the interview. But as I've considered and evaluated the Express design, I think the point is to let as much bass through as possible since it's rolled off hard earlier in the amp. The idea being that bass is gradually increased as you move toward the ODing stages. This way you can achieve maximum bottom end for the least amount of farty mush.

The Express preamp is rolled off on bass but the PI and power amp are wide open allowing whatever bass the amp has to be reproduced as cleanly as possible for an overdriven amp.

[rooster]

Yes, I would describe it visually as putting a full length, thick wool coat on a skinny man and sending him out into the cold weather.

Too, I have to question if anybody can actually appreciate Ken's sonic idea here without a 4/12 box of original G12Ms, always something to think about. Listen to Glen's recordings, for example.

Which is to say that if the PI coupling cap seems a little odd to you, just remember that Ken found a unique synergy in the parts - all the parts. You can't look at it from just one aspect, the preamp, for example, or the output section.

[Roe]

cap is way big. a 1n5 is sufficient, as shown by randall aiken. too big a cap makes it easier to get blocking distortion in PI. I use 10n-2n2 myself usually

[katopan]

There is also some dynamic shift happening with that cap, and the time constant of its recovery is influenced by the large value.

[Clyde]

Yes that cap's a little large, same as in the Komet 50. The amp "breathes" as the cap charges.

[stevlech]

To clarify, I read a thread in this forum in which someone mentioned a reason for having such a large cap in that position. The thread was a few years ago and I can't remember which one. I know it's much larger than needed for full bandwidth and I'm sure Mr. Fischer had a specific reason for it, as that was a fairly consistant design choice for him. I just wish i could remember that thread! My feeling is that the input cap size affects the headroom of the power amp. The smaller the cap, even while maintaining full bandwidth, the greater the headroom of the PI/power amp IIRC.

[katopan]

I haven't stumbled on that thread in all my reading here, but would be very interested if it can be dug up.

[JamesHealey]

about 4n7 is sufficient to get full frequency through.

Although 100nF does sound much more sluggish and has a compression to it that comes along with carrying all the sub sonic mush along with it.

Personally I prefer the default value of 100nF here on the Express and Liverpool.