Boosters, Gain and Distortion

It is a well known fact that booster pedals produce a range of tones other than simple gain. While equalization and tone shaping can be built into a pedal to alter the frequency response, there is a less understood circuit principal at work which causes some pedals to sound clean and others to have a fat or even distorted tone. This article will delve into a few of the causes of the tonal differences in boosters.

The basic booster pedal, as shown below, is a voltage amplifier that makes the signal larger to drive the amp or other effects harder. The gain stage is followed by an output volume control that allows the amount of drive to varied to suit the gear and playing style. A practical example of a simple jfet booster is also included for reference.
In our example, the input signal coming from the guitar pickups has a peak to peak level of 1 volt. As we know from the AMZ article on pickup signals, that signal rapidly falls to less than 100 mv. but the initial attack when plucked is much larger.

The booster circuit in our example is capable of a voltage gain of about 15. This means that the 1v input signal theoretically will become 15v on the output. However, in the real world where the booster pedal is powered by a 9v battery, it is not possible for the output to reach 15v since it is limited by the power supply to 0v to 9v maximum swing, and in our jfet example it is more like 0 to 8v.

What happens to the signal? Since the gain stage is trying to produce a 15v output but the maximum output is 9v, some of the signal cannot be reproduced accurately by the gain stage... in other words, it is clipped off!

When parts of the signal are clipped off, distortion is heard as a result. Our signal is AC coupled into the booster so the clipping takes place symmetrically on the top and bottom of the waveform.

Note that the initial pick attack is lost (clipped) but the remainder of the signal will be reproduced accurately once the input falls to a level that the booster is able to output. What is heard is distortion and compression during the first few milliseconds while the gain stage is banging against the limits of its power supply rails. The manner in which the circuit begins to go into this non-linear response region (the clipping area) is what determines the character of the distorted sound. It is commonly thought that bipolar transistors begin to clip suddenly and harshly while jfets will enter into this region more gracefully. Further characterization of those ideas is beyond the scope of this article.

The gain circuit discussed above is the most common booster pedal topology in use today. The LPB (bipolar) booster and its clones typify this design style. The mini-booster is also a gain-then-attenuator setup as are many other booster pedals. The clipping on the pick attack contributes to the characteristic sound of each of those pedal designs.

If we want to eliminate the non-linear clipping and distortion from our booster, there are several methods available. One method is to move the attenuator (volume control) to the front of the circuit before the gain stage, as shown here.
The volume control has been adjusted so that the input signal is trimmed down to 0.33v peak-to-peak. When this is multiplied by the 15x gain of the circuit, we have the 5v output just as in the first circuit example.

Since the gain circuit only has to swing from 0v to 5v, there is no clipping distortion and compression. A clean signal is the result. While this may seem like the ideal solution, it has a couple of significant problems.

First, the input impedance is no longer the amp-like 1M as in the first jfet booster example, but it is now the parallel value of the volume control and the gate resistor on the jfet, or, about 91k ohms. Yikes! This relatively low impedance will cause loss of signal strength and a dulling of high frequencies. However, it may be acceptable depending on the other parts of the booster design.

More importantly, the signal-to-noise ratio of the circuit has suffered greatly. The input signal is smaller but the noise contributed by the 15x gain stage is exactly the same as before. Smaller signal, same noise = worse signal-to-noise and a booster that is not optimized for guitar use.

Some vintage pedals use this method of trimming the input signal to control the output level of the boost pedal.

 
A better way to manage the level of the output signal is to do away with the volume control entirely and instead redesign the circuit so that it has variable gain.

The input signal coming from the guitar is still 1 volt but the circuit now has variable gain, essentially 1x to 15x. The gain control can be trimmed back to the 5x position and the output is at the level desired. The input impedance is still 1M and the noise is actually less since any fizz-hiss-pop on the input signal is amplified less than when the booster is running full gain as in the first two examples.

The jfet example of variable gain shown above is not practical since the bias of the jfet (drain voltage) is changing as the gain control is adjusted. A more practical example of a variable gain booster pedal is the AMZ Mosfet Booster.

An alternative method of getting a cleaner sound from a booster pedal is to power the circuit with a higher voltage supply. If the supply voltage is increased to 18v, then the booster in the first example can swing 0v to 17v (approximately) and the situation is much improved. The 1v input is multiplied by the gain-of-15 circuit and the output swings from 0v to 15v, which is within the capabilities of the pedal to reproduce without clipping. This is a common technique employed to give a cleaner sound to boosters and it may be achieved by using two batteries, an external power adapter or an internal voltage multiplier such as the MAX1044 to increase the supply voltage.

Each of these alternate methods have problems of their own: two batteries are costlier, heavier and require more space in the pedal, an external adapter can add hum and requires access to wall power, and the voltage multiplier has a high frequency oscillator that can induce noise into the circuitry. Also, if the output of the guitar's pickups are hotter than average, let's say 1.2v pk-pk, then the output would be required to swing to 18v and it cannot even with the 18v power supply -- we start to encounter the same problems again. On the other hand, the variable gain circuit is able to handle the stronger pickup signal with no problem.

The same circuit configurations apply to opamp circuits as well:
           
This article should give you some insight into why simple booster pedals can sound differently, and why some are "cleaner" sounding than others of similar design. No one design is superior to another and they all have their uses. For cleanest sound you want a variable gain pedal and an 18v (or higher) power supply. For more character, a gain-then-attenuator pedal (first example) can give some fat, compressed boost to the signal, and for even more chunk, trim the supply voltage down to 6v and see if your pedal acquires more character!