Thank's,
Johnhenry
NFB Question
Moderators: pompeiisneaks, Colossal
Re: NFB Question
My Bad Novosibir, I was looking at a Fender Delux schematic while i was typin that in, a Super show's the blue wire closest to the pre/amp, a Who lot I know ! LOL !
Thank's,
Johnhenry
Thank's,
Johnhenry
NFB
Hi,
Assuming that feedback is made negative...........
What you have here is voltage feedback and it is done with a resistive divider. As long as the relation is kept the closed loop gain of the poweramp remains the same but the damping factor will change.
That's damping as in the speakers reflection back into the amplifieroutput
If one of the resistors in the dividing network would be changed the ratio of voltagefeedback will change and thus also the closedloop gain of the poweramp.
If the feedback resistor ,in this case 100K would be removed the poweramp would work in openloop mode and then have the loosest response,highest gain and highest outputimpedance aswell as highest distortion levels and narrowest bandwidth. Now one can trade gain for higher bandwidth ,lower distortion, lower outputimpedance and tighter response, but only to the level when the poweramp becomes instabile from feedback inpart becoming positive and it will oscillate.
Now if the circuit would be such that it would be designed with a divider made with a series 100K and a shunt 5K, you can safely increase the 100K or decrease the 5K both making feedbackratio less.
To test this you can mount a potentiometer wired like a reostat in series with the series feedback resistor and if you design the so that total feedback is less than 20dB, in values 47K in series with a 100K linearpot from output and shunt with 5K you can determine the safe zone- remember tubepoweramps have low openloopgain and they don't like heavy feedback.
It might be possible to run higher feedback but this may necessitate other measures to keep the amplifier stable.
Usually a dominant rolloff at 6dB is enough and it can be made with a single capacitor sitting across the phasesplitter outputs, if that is has anode outputs and thus output impedannce equal (or near) to both sides. For circuits that have the singletriode phasesplitter feedback entrypoint must be placed earlier and so must the dominant rolloff.
Sonically you'd likely find that heavier feedback makes tighter bass, but staler response and less feedback makes looser bass and more distortion- somewhere in between there's a happy medium that also is stable.
Just a few thoughts
Assuming that feedback is made negative...........
What you have here is voltage feedback and it is done with a resistive divider. As long as the relation is kept the closed loop gain of the poweramp remains the same but the damping factor will change.
That's damping as in the speakers reflection back into the amplifieroutput
If one of the resistors in the dividing network would be changed the ratio of voltagefeedback will change and thus also the closedloop gain of the poweramp.
If the feedback resistor ,in this case 100K would be removed the poweramp would work in openloop mode and then have the loosest response,highest gain and highest outputimpedance aswell as highest distortion levels and narrowest bandwidth. Now one can trade gain for higher bandwidth ,lower distortion, lower outputimpedance and tighter response, but only to the level when the poweramp becomes instabile from feedback inpart becoming positive and it will oscillate.
Now if the circuit would be such that it would be designed with a divider made with a series 100K and a shunt 5K, you can safely increase the 100K or decrease the 5K both making feedbackratio less.
To test this you can mount a potentiometer wired like a reostat in series with the series feedback resistor and if you design the so that total feedback is less than 20dB, in values 47K in series with a 100K linearpot from output and shunt with 5K you can determine the safe zone- remember tubepoweramps have low openloopgain and they don't like heavy feedback.
It might be possible to run higher feedback but this may necessitate other measures to keep the amplifier stable.
Usually a dominant rolloff at 6dB is enough and it can be made with a single capacitor sitting across the phasesplitter outputs, if that is has anode outputs and thus output impedannce equal (or near) to both sides. For circuits that have the singletriode phasesplitter feedback entrypoint must be placed earlier and so must the dominant rolloff.
Sonically you'd likely find that heavier feedback makes tighter bass, but staler response and less feedback makes looser bass and more distortion- somewhere in between there's a happy medium that also is stable.
Just a few thoughts
Re: NFB Question
If you have access to an oscilloscope then the method I use is to take the 1kHz reference square wave it provides into the primary and then hook up the primary and one of the secondaries into channels 1 and 2.dartanion wrote:How about for an SE OT. I can't really figure out which lead is the start or end of the winding. Is this just a trial and error approach?
The two waveforms will either be in phase or out of phase. If they are in phase then you have your ground leads connected to the start of both windings. If they are out of phase then your primary ground is connected to the end of the winding.
Paul
Re: NFB Question
Hey Paul,
Thanks for the tip. Now I just have to get myself a scope. Any recommendations? I haven't used one since the early 90s, so I'll have to give myself a crash course again.
Thanks for the tip. Now I just have to get myself a scope. Any recommendations? I haven't used one since the early 90s, so I'll have to give myself a crash course again.
Eardrums!!! We don't need no stinkin' eardrums!
Re: NFB Question
You don't need anything too fancy, although from my experience with using a 40MHz one I wouldn't suggest any lower bandwidth than this so you don't end up struggling with brightness at higher frequencies.
You'll want 2 channels to be able to compare inputs to outputs so that would be on the shopping list as well.
A digital storage mode can be useful, but since most of what you're going to be looking at in an amp will be regular waveforms (even parasitic oscillation is pretty regular) it shouldn't be strictly necessary, and will certainly add to the cost.
The one other thing you'll be wanting is probes with a 10x scaling factor on them as you can easily exceed the maximum input voltage on the scope itself with plate voltages.
There always seem to be plenty of Tektronix ones coming up in the US. They're a pretty good manufacturer.
Paul
You'll want 2 channels to be able to compare inputs to outputs so that would be on the shopping list as well.
A digital storage mode can be useful, but since most of what you're going to be looking at in an amp will be regular waveforms (even parasitic oscillation is pretty regular) it shouldn't be strictly necessary, and will certainly add to the cost.
The one other thing you'll be wanting is probes with a 10x scaling factor on them as you can easily exceed the maximum input voltage on the scope itself with plate voltages.
There always seem to be plenty of Tektronix ones coming up in the US. They're a pretty good manufacturer.
Paul
Re: NFB Question
If you can afford it, the Fluke handheld scope is very nice and take's up very little space on a small bench, but i also have 2 old Tektronix 454 scope's that have served me well for the last 25 yr's fixin amps, there are probably some good buy's out there on more modern scope's, nice part about tube's is it's not nessesary to go all out, unless you do Synth and Keyboards, then you would need a highend scope,
Happy Hunting,
Johnhenry
Happy Hunting,
Johnhenry