Solid State Guitar Amp Forum | DIY Guitar Amplifiers

Solid State Amplifiers => Amplifier Discussion => Topic started by: NPreston on February 28, 2021, 09:40:34 AM

Title: Power amplifier capacitor in NFB question
Post by: NPreston on February 28, 2021, 09:40:34 AM
Hi,
Referencing the power amp at https://sound-au.com/project27.htm  I see C4 is an electrolytic cap.  My understanding is that this cap prevents any DC from being amplified by feeding all DC back to the negative input?  Correct me if I am wrong..   Question is, how can this be an electrolytic if the AC signal swings above and below ground??  I am missing something here..   Also, am I correct in understanding, that together with R7 at 22K, R7&C4 make a high pass filter with the corner F at 0.24Hz??  And overall AC gain would be (R7/R6)+1??

Just trying to understand the nitty gritty of power amps..  every time I think I understand something, I find something else that I dont understand.

Thanks.

N.
Title: Re: Power amplifier capacitor in NFB question
Post by: willpirkle on February 28, 2021, 12:01:55 PM
This is a mixed mode amp (voltage and current feedback). Neglecting the current feedback branch (which changes the power delivered to the speaker), the gain of the amp in this configuration is 1 + (R7/(R6 + ZC4)) where ZC4 is the impedance of C4 at a given frequency. At DC, the impedance of C4 is infinite. This causes the (R7/(R6 + ZC4) a term to become 0.0. Then, the DC gain of the amp is unity. As the frequency increases above 0Hz, the impedance of the cap drops, until it is an effective short. Then, the gain of the amp is 1+(R7/R6). Between DC and the short-circuit frequency, the gain rises from unity to 1 +(R7/R6), forming a low frequency shelving filter.

Rod Elliott has a nice discussion of current feedback (aka "constant current amplifier") on that site as well. And, this kind of mixed mode feedback is generally limited to guitar and bass amp applications. And, massive unending arguments among audiophiles.

The C4 cap being electrolytic (and polarized) is non-consequential to the gain equation, and does not limit the voltage swing.

Will