Tube Amps / Music Electronics
For current discussions, please visit Music Electronics Forum. New: visit Schematic Hell!
The sunn still shines online!

Listen to great tunes streaming live right now!

ampage archive

Vintage threads from the first ten years

Search for:  Mode:  

Negative Feedback...??

6/28/2003 3:48 PM
Negative Feedback...??
im trying to figure out wether i should have negative feedback on my homebrew or not. all it seems to do to me is make the amp a little more quiet. what is negavtive feeback? what's the purpose of it? and any other info is very welcome. thanks
6/28/2003 6:01 PM
John Culp

Negative feedback does make the amp quieter. It also tends to flatten the amp's response over the frequency spectrum. Helps prevent oscillations. Increases speaker damping for "tighter" bass. Makes the amp's tonal output less dependent on the varying characteristics of different speakers used with it.
7/1/2003 4:22 PM
Jack Plug
As a broad generalisation, the more negative feedback you use the closer your amp will approach the performance of a perfect amplifier.  
This may be a very desirable objective for industrial applications but most of us find such ideal amplifier performance 'unmusical'.  
For guitar amps the rule seems to be use as little negative feedback as you can get away with. As well as other benefits that have already been mentioned, negative feedback will make your amplifier much more tolerant to temperature changes, component values drifting and tube changes.
7/3/2003 3:37 AM
Doug B.

Also, NFB will tend to reduce harmonic distortion and that pesky crossover distortion... as long as the loop is intact. However, when pushed into clipping, a NFB stage will clip more sharply, more like a solid-state circuit.  
- Doug B.
7/3/2003 3:01 PM
I was wondering if the following statement is too simplistic regarding NFB in guitar amps with a Presence control: "If most of the range of the Presence control is un-useable due to harshness in the highs when turned up, you probably have too much NFB." I guess this assumes that the amp is otherwise well-designed. I recently noticed this but only with a couple of my amps, I'm not a tech and don't have much experience except for working on and building my own amps. Decent rule-of-thumb or just a coincidence with my amps? Reducing the NFB in both amps made them sound better and now the Presence control is much more useable.  
7/3/2003 4:28 PM

That's a good question. Might it be better in some amps to leave the amount of feedback alone and just not turn up the Presence so far? Or, as you suggest, might this be gauge of the amount of NFB, and the over application of it? I'd love to hear what other's here think of this.
7/3/2003 5:36 PM
OK. This isn't intented as a complete answer in the least...  
But, turning up the Presence basically opens the loop for frequencies above the circuit's cutoff point - essentially no feedback for the highs. This makes these frequencies more susceptible to artifacts and characteristics of the output tubes and speakers. I read recently that this is why Fender never really adopted the use of the Presence control for all of their amps. The effects of tube wear and differences in speakers made things too unpredicatble. A harsh sounding set of tubes and some brittle speakers could potentially get downright nasty if made more "present"!  
""If most of the range of the Presence control is un-useable due to harshness in the highs when turned up, you probably have too much NFB.""
That's probably not a bad rule of thumb. Reducing the amount of feedback basically "brightens" the entire signal when the Presence is turned all the way down so that the effect of turning it up is less pronounced. However you could also design the presence control to not roll off as much of the highs. Instead of having the HF running open loop you could just change the amount of feedback going to each part of the frequency spectrum. Using different cap and pot values, putting some resistance in series with the cap to ground, or bypassing the the feedback resistor with a cap and resistor in series could all be ways of adjusting the HF response in the feedback circuit.  
Ultimately though, your comment is probably the most straightforward to implement. Hopefully better designers will chime in here too!  

  Page 1 of 1