There are a lot of mysterious components within any guitar amp...
many of which remain puzzling even to hobbyists who have built a DIY project or two. In the fifth part of Mojotone’s series What Does This Thing Do? we’re examining negative-feedback loops and their impact on any guitar amplifier’s sound.
An extremely simple network that can have a tremendous effect on any guitar amplifier’s ultimate sound, the negative-feedback loop remains a rather mysterious thing to many new and novice amp builders and modders. Consider that the entire loop usually consists of just two pieces of wire going from point A to point B with a single resistor in between, and it’s incredible that this can be one of the major defining characteristics of some breeds of amp, yet such is the impact of this so-called negative feedback.
Many kinds of negative-feedback loops are used in audio amplification, but where tube-powered guitar amps are concerned a loop positioned around the output stage is far and away the most common, so that’s what we’ll focus on here. It’s important to note that several classic amps have no negative-feedback loop whatsoever, and that in itself plays a big part in their sound and playing feel.
A negative-feedback loop does exactly what the term implies...
but it’s where that happens and the resulting affect it has on the signal that really matter. Look at an original schematic for Fender’s 5F6A tweed Bassman of the late ’50s, and in the top-right corner of the drawing you’ll see a connection that approximates three-quarters of a rectangle, comprising a line running upward from the output transformer’s positive connection to the speaker output jacks, leftward toward the rest of the circuit via a 27k resistor, and downward to connect with a 10k resistor that’s part of the long-tailed-pair phase inverter. What this simple loop does is connect the final output signal to an element of the input of the output stage (which is essentially the phase inverter). Within this connection, however, is a resistor that reins in the amount of signal that is fed back.
Presenting this output signal back to the input of the stage generally results in dampening the stage’s propensity to distort, which tightens up the overall performance, enhances the low end somewhat, and tips the sound toward clarity and articulation, rather than rawness and harmonic distortion. When you’re looking to produce volume and power, as designers of larger amps were striving to do in the late ’50s and ’60s, all of these might be perceived as good things. And they remain good things according to many design goals today. When a lower-powered, looser, and more complex-sounding amplifier is acceptable, however, amp designers often want to adjust this negative-feedback loop, or eliminate it entirely.
Consider, anecdotally, that the lack of a negative-feedback loop is about 50% of the equation behind the marketing spiel when the majority of amp manufacturers promote a model as being “Class A.” Since a genuine Class-A amplifier requires a lot more technical analysis and qualification than most such guitar amps actually allow for, the industry has tended to use this label with any amp that has cathode-biased output tubes, with no negative feedback loop. Whatever you call it, those are two of the main factors behind the sounds of the Vox AC15 and AC30, Fender’s tweed Deluxe, and the countless amps that emulate those designs today.
If a negative feedback loop achieves the sonic ends described above...
then of course the lack of one accentuates characteristics in the other direction. This means that building an amp without a loop enhances a degree of looseness, harmonic saturation, and an early onset of distortion, at the sacrifice of some fullness in the lows and the addition of perhaps a little more rawness in the tone overall. Be aware that good designers and manufacturers can also taper these characteristics with other elements of the circuit architecture and component selection, so those these are usually degrees of difference, rather than extremes.
For the hobby builder, all of this means you can further temper the personality of your amp by altering the amount of negative feedback in one direction or the other, or eliminating it entirely. Be aware that the selection of an appropriate negative-feedback resistor can be confusing: a resistor of higher value means less negative feedback, since it’s blocking more of that signal, and a lower-value resistor means more.
In amps with multi-tap output transformers, providing two or three speaker-output impedances...
it also matters where you connect that resistor. Consider that Marshall copied the 5F6A Bassman to build their first amp, the JTM45, and used the Bassman’s 27k feedback resistor as a result. Marshall, however, used an output transformer with speaker outputs for 4, 8 and 16 ohms, and connected the loop to the latter. The result was almost three times as much negative feedback on the JTM45, which is one of the reasons those amps sound different from the Bassman that inspired them.
All of this talk of varying the amount of negative feedback to alter the sound of your guitar amp is probably putting an idea in the head of some DIY’ers already: as simple as this loop is, and given the fact that it only requires changing one resistor to alter its performance, it’s easy in most amps to install a three-way switch to give you on-the-fly changes. We don’t have room here to provide full instruction—and these mods are pretty easily found on the web—but whether you’re modifying or building from scratch, it’s pretty easy to install a simple three-way switch that offers two different levels of negative feedback, plus none at all.
Even without such a switch, you can easily experiment with resistor values (observing all safety protocols while working inside any amp, of course) to find what works best for your own sound, then either stick with that as a one-and-done scenario, or make it switchable. Either way, it’s worth understanding the surprisingly significant impact the negative-feedback loop, or lack thereof, can have on your amp’s sound.