Purpose of the Feedback Loop Capacitor in the NFB

General discussion area for tube amps.

Moderators: pompeiisneaks, Colossal

Post Reply
Decko
Posts: 309
Joined: Sat Feb 23, 2008 8:36 pm
Location: California

Purpose of the Feedback Loop Capacitor in the NFB

Post by Decko »

Hi Gents,

Quick question:

How does the capacitor value affect the feedback loop? For example .1uf cap on the presence pot...does that filter out certain frequencies to ground? If that is the case, how do smaller or larger values affect the NFB?

Sorry if I am totally off, but I would like to understand the circuit here.

Regards,
Danny
Alexo
Posts: 477
Joined: Thu Mar 22, 2007 3:27 am
Location: The Hudson Valley

Re: Purpose of the Feedback Loop Capacitor in the NFB

Post by Alexo »

That caps removes (sends to ground) higher frequencies from the negative feedback loop. Because this is *negative* feedback, removing these frequencies from the loop actually *increases* high frequency content from the output stage, because hey are no longer canceled out.

It has a different effect than just brightening the circuit though. It removes the linearizing effect of nfb on that range, so in some ways, you get a "tubier" sound across the top end, with smoother breakup in that region when the presence is up.
Life is a tale told by an idiot -- full of sound and fury, signifying nothing.

...in other words: rock and roll!
User avatar
martin manning
Posts: 14308
Joined: Sun Jul 06, 2008 12:43 am
Location: 39°06' N 84°30' W

Re: Purpose of the Feedback Loop Capacitor in the NFB

Post by martin manning »

Decko wrote:How does the capacitor value affect the feedback loop? For example .1uf cap on the presence pot...does that filter out certain frequencies to ground? If that is the case, how do smaller or larger values affect the NFB?
Assuming you are talking about a typical guitar amp with a LTP phase inverter, the feedback resistor, PI tail resistor (and/or the presence pot, depending upon the circuit), and the presence cap form a low-pass filter, which reduces the feedback of high frequencies, as described above. The cutoff frequency of this network, where the feedback signal is reduced by 3dB, comes down to 1/(2PiRC), where R is a combination of the resistors (depending upon the circuit), and C is the presence cap. If you leave the resistors as they are, then you can see that the cutoff frequency is inversely proportional to the value of the cap. If you double its value the cutoff frequency is reduced by half, and the range of boosted treble is extended downward one octave.
Post Reply