Calculating NFB resistor

[TurboGuitarMelton]

Hi I've been trying to understand NFB resistors and their relationship to the circuit.


What Im trying to figure out is what is the equivalent NFB resistor value for these two loads. Are they both 22k? Or does the different primary impedance change the value?


What is the equivalent amount of NFB for this


8k primary/4ohm = 22k


2.5k primary/4ohm = ?

[sluckey]

As long as the NFB is connected to the 4Ω secondary, the primary has no bearing on the resistor value. That assumes the primary is connected to a proper impedance, ie, 8K connects to a proper 8K circuit, and the 2.5K does likewise.

[jjasilli]

Typically, the NFB R forms a voltage divider with the lower segment of the cathode resistance at which it is fed back into.  This relationship may be complicated by caps(s) in the NFB loop.


The actual amount of NFB voltage will be determined by the voltage at the OT secondary tap from which the NFB is sourced, as sluckey says.  Typically, manufacturers use the highest Ohm tap which delivers the highest output voltage; but you can use any tap you want.  Taps are typically 2, 4, 8 or 16 Ohms.  But as the nominal output impedance doubles, the output voltage changes by a factor of 1.414. 


You can closely calculate the output voltage of an amp if you know its output W > x Ohm load.  Now you know the source voltage of that amp's NFB.  Then use the voltage divider formula to calculate how much net NFB voltage is actually injected at the point of insertion. 

[TurboGuitarMelton]

Okay! Thanks for the reply! Here is my shot at the math...


Im estimating the output wattage in both cases



Amp A 5F1


sqrt(5W x 4ohms) = 4.47214V
NFB = 22k
Rk = 1.5k
NFB voltage = 0.285V


Amp B 5f1 parallel SE


sqrt(10Wx4ohms) = 6.32456V
NFB =22k
Rk= 1.5k
NFB voltage = 0.404V


Okay, so now that Ive done all that I can see that they are proportional to each other. Which doesnt mean they are equal since both preamp stages are the same. The output wattage is doubled so the amount of NFB is increased.


I think 33k is the correct value. (31.789 is what I calculated)

[jjasilli]

Agreed as to Amp A.


Re Amp B:  I didn't re-do the math.  Note that 2 parallel tubes won't provide 2X output W due to circuit losses.  You might recalculate at 8W & 9W -- maybe not much difference??? or average them.   Also, 2X parallel Power Tubes halves their combined output impedance. Is this compensated for?

[TurboGuitarMelton]

I recalculated for 8W and the NFB R is 28k. 33k is what I have on hand and is close enough for me, so I'll go ahead and try that.

Also, 2X parallel Power Tubes halves their combined output impedance. Is this compensated for?

Yep! I have 2.5k primary on OT. That why I mentioned the 8k/4ohm and 2.5k/4ohm. I initially thought the OT primary might be a factor in the NFB. In hindsight, I dont know why I thought that.

[jjasilli]

    The "straight" math works only if the whole output chain is "matched" from:  net Power Tube output impedance"  > OT Primary Impedance > OT Secondary Impedance > Speaker Load.  (Note that matched does not necessary mean equal.  Voltage matching is usually 1 : 10 from source > input).  If not all matched, the equation must be adjusted accordingly to derive the correct output voltage.


I think this answers the question of how the NFB R functions.  It is the series resistor in a voltage divider circuit.  There are more complicated NFB circuits, but this is the overall framework.


The preamp doesn't matter unless it can't drive the power amp to full power.  Even so, the NFB voltage divider ratio remains the same.  I think this basically sums it up for guitar amps.


*****
Note that we are talking impedance so voltage will vary with frequency.  Usually, measuring is done at the "standard" 1000Hz.  But most of the guitar fundamental frequencies are lower.  Also, NFB might be used to dampen speaker resonance.  This gets complicated, because speaker resonance will typically be at around 100Hz or below.  Also, speakers have significantly higher electrical impedance at and near their resonant frequency.  So 1000Hz performance @ nominal impedance is not applicable.  Caps in the NFB loop may be used to fine tune this; but this introduces phase issues: putting some of the frequency band out of time with the rest of the frequencies.  Though this is not important for guitar amps.

[PRR]

Anyway, NFB on a GUITAR amplifier depends on speaker need for damping, player style and genre, even the room.

Get any rough answer, tack it in, and then adjust on test.