Explain how feedback affects DF?

[mfaughn]

I know what damping factor is. I am trying to understand how nfb lowers to output impedance of an amp, thus affecting DF. I am early in the process of working up a triode strapped EL34 PP amp and ran across this thread where it is suggested that a triode strapped EL34 PP biased into class A and with no NFB will have a good DF. How is this so? Is it a matter of triodes, bias, pentodes strapped as triodes?

Thanks for any insight shared with this Orthopteran.

[Loren42]

I don't know that well enough to answer why the impedance of the amp drops, but it does.

However, DF is pretty much overstated. The largest component of the DF is the impedance of the speaker driver coil, the speaker cable, and the crossover. In that regard, the real difference in system performance between a high DF amp and one with low DF does not contribute very much to system performance.

So, while it doesn't answer your exact question I thought I would at least make my claim that it is not as important a consideration when putting together a system as many people or marketing departments would like you to believe.

[mfaughn]

That makes sense. If one is using full range drivers (no xo) and builds monoblocks so the speaker cable R can be negligible then the impact of the amps output impedance should be proportionally greater.

[the_manta]

Hi !
I would suggest to read at least the Wikipedia article first, especially this passage: Negative feedback amplifier - Wikipedia, the free encyclopedia

Explaining this here in extreme detail would fill pages and pages. Just tell what you didn't understand exactly.
You have to understand why there are 4 different types of feedback(series/parallel derived and series/parallel applied) and what they do. With current feedback you can increase output impedance (and thus lower DF) just as simple.

Now that's how you can affect output impedances and thus DF with external feedback. You asked furthermore why a stage triode has a lower DF, than a pentode. The reason is simply their output impedances. Just look in a triode datasheet or look up the curves for the Ri (or Rp in american literature) the value is relatively low and always lower than the OPT's primary impedance which is usually 2 till 4 fold larger. If you look in the AD1 triode's Datasheet, they tell us an Ri of 670 Ohms. They also tell us an optimal load impedance of the OPT of 2300 Ohms. Being an AD1 I see a e.g. an 8 Ohms speaker, but cause of the OPT these 8 Ohms are translated to 2300 Ohms which I have to drive. Being the Speaker and looking into the amplifier I see an Rp of 670 Ohms, but because of the OPT this is transformed into (8/2300)*670 = 2.3 Ohms. So for the amplifier it is easy to drive the larger 8 Ohm load.

A pentode is the opposite, its Ri is much higher, resulting in that the "speaker drives the amplifier"

BTW: The reason the Triode has a lower Ri is nothing more than internal negative feedback from the anode to the grid. A pentode doesn't allow this because the screen is between and shields the Grid from the Anodes effects. And yes, that means a SET no feedback amp is kind of ********, since it already has inherent feedback.

Originally Posted by Loren42

I don't know that well enough to answer why the impedance of the amp drops, but it does.

However, DF is pretty much overstated. The largest component of the DF is the impedance of the speaker driver coil, the speaker cable, and the crossover. In that regard, the real difference in system performance between a high DF amp and one with low DF does not contribute very much to system performance.

So, while it doesn't answer your exact question I thought I would at least make my claim that it is not as important a consideration when putting together a system as many people or marketing departments would like you to believe.

Just remove the feedback from your pentode amp and listen. Do you like the sound of a wobbling bass and a sound like slurry in your speaker boxes ?

regards, simon