F6 Amplifier

I did a simulation of that circuit and got about the same results as with my Teaser-6 circuit. I question the choice of 470uf-1000uf for the bias coupling caps. In the simulations there is only 30 degrees phase margin at .06Hz. In the actual Teaser-6 builds with 470uf caps, the noise floor and harmonics were seen to "bounce around" significantly at a frequency of about .1-.2 Hz. When I reduced the caps to 100uf, this problem went away; verified by both the simulations and the actual builds.

In the plot, the colors are as follows: blue=100uf, red=470uf, and blue=1000uf.


you can make this one in a meantime

:devilr:

disclaimer - from slide 13 of 19

really don't know what was in later slides :rofl:

Papa still didn't sent me lecture and I doubt it will :clown:
 

Attachments

  • F6ZM-phase-margin1.jpg
    F6ZM-phase-margin1.jpg
    111.6 KB · Views: 564
Official Court Jester
Joined 2003
Paid Member
I'm used to trust in Pa's simple solutions

in those cases , simple means elegant .

even if you can get better results , when circuit is intended for Greedy Boyz , Pa is not going to squeeze last ounce of performance , from simple reason - not every greedy boy is having THD measuring facility

one who is having that - is free to fire up both his brain and measuring equipment


whatever - if your amp is , in any way , in situation to worry about 0.1 or 0.2 Hz , you're in trouble

there is no way that you can find it in motorboating region

edit : how you measured that behavior in vivo ?
 
Last edited:
How can a reduction of the values of the bias caps be considered an increase in the circuit complexity?:scratch2:

I used two separate measurement systems which both gave similar results. The first was Visual Analyzer Visual Analyser 2011 XE, running continuously showing the spectrum for a 1 watt sine signal at 1kHz, updating at about once per second. The second measurement system was my own software, which is similar to Visual Analyzer, also running continuously showing the spectrum with frequent updates. With both measurement systems the noise floor and strength of the harmonics bounced around significantly with 470uf caps. When the caps were reduced to 100uf that problem went away. The problem did not manifest itself as audible motorboating, but I suspect that careful listening would detect it as temporal variations in the levels of the harmonics. In an early Teaser-6 build, with a different bias circuit, I was able to stimulate it into runaway motorboating at a frequency below 1Hz.

I'm used to trust in Pa's simple solutions

in those cases , simple means elegant .

even if you can get better results , when circuit is intended for Greedy Boyz , Pa is not going to squeeze last ounce of performance , from simple reason - not every greedy boy is having THD measuring facility

one who is having that - is free to fire up both his brain and measuring equipment


whatever - if your amp is , in any way , in situation to worry about 0.1 or 0.2 Hz , you're in trouble

there is no way that you can find it in motorboating region

edit : how you measured that behavior in vivo ?
 
Official Court Jester
Joined 2003
Paid Member
remark about simplicity was general , not exactly connected with your question

whatever , what I meant with that is that I'm usually respecting his choices ;)

even if I'm notorious in using everything shunt regulated , while he is not :devily:

however , it's best to ask him about those small details , when you see exact schm and article