The Negative Effects of Feedback

The Description

Negative feedback, quite simply, is the application of an inverted portion of an amplifier's output signal back to its input, in an attempt to lower the distortion to zero. This "extra" signal is subtracted from the input, and reduces the effective amplifier gain in the process. Local feedback is generally less harmful to the signal than global feedback. The reason being that this negative signal has less distance to go than it's global relative and is usually contained within a single amplification stage. A single amplification stage is one where only one active device is used (single-ended) or in the case of balanced circuitry two active devices (one for each half of the signal). Wyetech Labs to date has not made use of negative feedback of any sort , be it local or global. Remember our philosophy is to keep it as simple as possible while maintaining the highest linearity and lowest distortion. In doing so there is no need for error correction. This philosophy has been proven to work in single ended tube circuits.

As we said earlier, if one must resort to feedback then local over global is the preferred route. Unfortunately, most amplifier companies always resort to the global kind. This global feedback is usually taken from the last output stage and fed back in to the first input stage 180 out of phase. This is done in the hope of correcting any distortions that occurred as the signal passed through all the stages in the loop. Since the distortion and signal has been amplified, some of it can be fed back to the input in the hope of canceling the distortion to zero. Now theoretically, this would work well providing that the speed of travel of electricity through the wires and circuitry occurred instantly in time (ie. travelling at infinite speed!). And therein lies the problem!

The Problem

Tube amplifiers of the push-pull variety usually use an average of 20 db of feedback, while solid-state amplifiers can require much higher levels of feedback, especially those employing chips or SMD devices (chips = operational amplifiers, SMD = surface mount devices). The extra gain needed to implement feedback involves either adding more gain stages or using active devices (tubes or transistors) that have a higher amplification factor. If we add more gain stages, we increase the complexity of the circuit, require more parts and thus introduce even more distortion that we hope to reduce by using the feedback method. If we use active devices that have more gain (tubes or transistors), we introduce other negative elements, such as much reduced bandwidth of the circuit (frequency response). Hey, didn't we say we're trying to keep things as simple as possible?

The effects upon the signal

This scheme presents a very obvious problem upon circumspection. All amplifiers introduce some delay in passing the signal from the input to the output and back again to the input to complete the feedback route. During this delay period, a phase lag of the feedback signal in relation to the original signal will occur. It is not until this initial delay period is over, that the circuit begins to exhibit its intended operation. There is therefore a constant introduction of "out of date" information into the amplifier.

Under transient music conditions, this results in the presentation of an error correction signal intended to reduce the distortion of an input signal which has already passed through the amplifier and is either already out of the circuit or well on its way out of the circuit. The signal present at the input, by the time the feedback has arrived, may bear no relation to the previous signal and thus will not be properly acted upon by the regenerated information. The current input signal is then distorted once, through the subtraction of an erroneous feedback waveform, and again by the amplifier.

This regenerated signal is then passed through the amplifier and again fed back, with all of the newly created anomalies, to make yet another trip through the circuit, until it is allowed to decay through successive attenuation. Thus, a distortion signal which originally may have lasted only a few microseconds, can pass through the amplifier enough times for its effective duration to have exceeded the threshold of audibility. This mechanism originally designed to reduce distortion, has actually, under these transient conditions, created new forms of distortion. Thus the solution has evolved into the creation of a new signal with distortions that bear no relation to the original distortions that would have been present otherwise. Thus the cure is worse than the original disease.1

The real effects upon the music

If we have a poor circuit to begin with that, if left to itself sounds horrible, we just might be lucky enough to make it sound better by using feedback.2 On the other hand, If we use a circuit that sounds superb to begin with, such as Wyetech amplifiers, we can literally destroy the life of the music by applying feedback. One effect is a shrinkage of the soundstage in all directions as feedback is increased. The other effect is a lessening of the emotional impact upon the listener. The traversing from live sound to increasingly dull sound as we apply more feedback. This result is a totally unappealing, boring and joyless experience.

1. Qualifiedly, it is possible to improve the sound of a weak circuit design by using negative feedback.

2. Thus the confusion reigns over the issue of feedback in general.