[359] Boosters Are Not Buffers

Date: March 28th, 2009 | Comments : [6] | Categories: DIY, pedals.

LPB1 booster

Boosters are not buffers! While it is possible to make a booster that also provides a buffering action, the vast majority of the boosters commercially available ready-made or kits, do not have the proper design to be buffers.

A buffer should have a high input impedance (Z) and a low output impedance. The LPB1, shown at the top of this post, has a fairly low input impedance, which automatically disqualifies it from the buffer category. The mini-booster has a high input impedance but a moderately high output Z, and is not a good buffer either.

LPB1 booster

Another part of these designs that prevents their use as buffers is that they have a volume control potentiometer on the output. The volume pot impacts the impedance of the circuit. If the volume pot is at the 50% position, represented above by R5 and R6, the output pot is acting as a voltage divider. Here we have the 100k pot in parallel with the output impedance of the LPB1 (10k) that is driving the cable to the amp or next pedal.

The cable capacitance (Cc) even on a good cable can be 800pF, as I measured in a previous test, and is shown on the schematic in light blue. The output impedance of the booster and the resistance of the pot are driving the cable and form a low pass filter with the cable capacitance. In this example, the high frequencies are rolled off starting at around 7.2kHz.  This is obviously not what we want from a buffer!

The same problem caused by the output volume control is present, even with an opamp output, as illustrated by the following schematic:

Opamp buffer

This is a typical opamp distortion pedal schematic fragment, and the high frequency limit is similar to the transistor booster example (7.5kHz corner frequency). You should note that as the volume is turned up, the frequency limit rises as well, letting more high frequencies through.

The shunt resistance of the pot, shown as R6 on the examples, is in parallel with the input impedance of the next stage. If that stage is a pedal with a high impedance, much of the advantages of the high Z input will be lost since the resistance of the pot is in parallel with the input and effectively reduces the impedance. This will change how a pedal responds or reacts to an input signal, and will vary with the setting of the volume pot.

It is obvious from these examples that an unbuffered volume pot on a pedal circuit output can be a problem, whether on a booster like the LPB or a distortion pedal like the Fuzzface (which has a 500k pot that makes the problem even more noticeable).

The AMZ Mosfet Booster is a good example of a booster design that does not have a volume control on its output, and its fairly low output Z makes it capable of driving cables with little difficulty – the 2.7k output Z will drive the 20 ft. cable in our example to over 73kHz!

Use the AMZ Filter Calculator to check some of the values of pots and cable capacitance. (0.8nF = 800pF = 0.0008uF)


6 Responses to “Boosters Are Not Buffers”

[639] Alex Says: 7:47 am, March 30th, 2009

Thanks for doing the article, Jack. The subject seems to come up quite often on internet forums.

The issue becomes even more complex when you model the cable as a series of L-C filters rather than a single capacitance. This gives a frequency response that may just be a simple 6dB roll-off, or a resonant peak followed by a much steeper roll-off, depending on how the cable is terminated.

[642] admin Says: 5:38 am, April 1st, 2009

Thanks for your comments… the inductance and capacitance of the connecting cable can certainly introduce additional changes in the frequency response that can be avoided by have a buffered volume control, or a low impedance output as with the AMZ Mosfet Booster.

[681] Nicholas Maris Says: 8:59 am, April 23rd, 2009

Hi Jack,
Great website, been visiting it for a while, some truly good work going on… FYI, as I recall, Craig Anderton codified a set of basic pedal design requirements back when he was doing his GP column and it was also summarized in EPFM (darn it, I’ve misplaced my copy), but the concept has stayed with me, so when I design a pedal it conforms to his basic requirements of hign Zin (I think his min value was 100k) and low Zout (I think his was 100 ohms max or thereabouts). Sometimes it pays to add an emitter follower or source follower output stage. If I can do a loading estimate I basically try to follow a “maximum 10%” rule of thumb so that the attenuation caused by loading is no more than 10% (or -20dB, whatever). Keep up the good work and enjoy your new lab! – Nicholas

[734] chuck Says: 5:11 pm, June 13th, 2009

couldn’t a small cap across lugs 3 and 2 of an out-pot help “fake fix” the loss of highs and allow them to pass?

similar to a bleed mod to a guitar’s volume pot.

Please tell me if that’d work, I’m just curious…

[826] Big M Says: 10:59 am, October 13th, 2009

Would the Zvex SHO make a good buffer?

[827] admin Says: 5:31 pm, October 13th, 2009

>>>Would the Zvex SHO make a good buffer?<<<

Lots of people use it for that purpose with good results. The output impedance is around 5k which is a bit high for my liking though.


AMZ Home Page

Guitar FX PCBs