Audio i2e

  AG2.1    AG1.0 / The PURR

Why a buffer is the most important "effect" pedal !?

A buffer? Important ? It does not really do anything ...!
After all, most buffers are advertised as "neutral", so the sound should not change anything.
But even if, because of their circuitry, they actually pass the signal at their input "neutrally" to their output, they still change something, and this is related to the physics of the electric current and the media that transmit it. And what they change there then has an effect on the sound, which is usually positive.

What is the usual medium for the transmission of the guitar signal?
The guitar cable, and a buffer is the means of choice to get its inadequacies under control.

Which cables exactly is mentioned?
Problematic are all cables that lie in the signal path between guitar and amplifier. So the cable from the guitar directly to the amplifier or when using effects in front of the amplifier, the cable between the guitar and the first effects unit. Then all (patch) cables between each effect and finally the cable from the last effect to the amplifier (often the biggest problem, because the longest ...).

What are the shortcomings?
A guitar cable seems to be a relatively simple component, but as so often the problem lies in the details.
The cable should first meet a variety of "mechanical" requirements:
It should be long enough, and it should be as flexible as possible. It should have good, stable jack plugs, but they also should not be too clunky. The outer material should be robust, you should be able to get around it or even drive over it with a 4x12er cab. It may even feel good, look good and should not smell unpleasant. It should be as insensitive as possible to interference radiation and should not emit its own signals when it is moved or loaded. And it's supposed to keep its properties for a very long time.
Because of these demands, it is already possible to distinguish between "good" and "bad" cables.
But here is another "quality" of a cable to be illuminated, namely, how much the cable affects the sound of my guitar. And why !
It is about influencing the frequency spectrum, specifically the attenuation of higher frequencies and the change of the resonant frequency of the pickups.

The physics of a cable:
A guitar or patch cable is essentially a piece of more or less long wire.
Electrically, this initially represents a (ohmic) resistance, easy to measure by means of an ohmmeter. However, even with long cables, this resistance remains in the range below one ohm to perhaps a few ohms (ohms is the unit of electrical resistance). This is completely negligible compared to all other resistors that occur in the signal chain, ie the internal resistance or the impedance (AC resistance) of the pickup (several 1000 ohms) or the input resistance of an effect device or an amplifier (several 100000 ohms)!

The ohmic resistance of a cable plays in practice absolutely no role and has no effect on the sound.
Really not !

Furthermore, each wire represents an inductance.
Normally inductors in the form of coils are known (e.g., each pickup is essentially an inductor, a very many turns coil). The size of the inductance increases among other things with the number of turns of the coil. For large inductance values, many turns are needed. A wire is practically a single-turn coil, so the size of its inductance is very small. Whether even the small inductance of a cable has an influence on the signal to be transmitted (here: sound signal of the guitar), depends on the frequency of the signal. The higher the frequency, the greater the influence of an even small inductance.

For the combination of guitar cable and audio frequency range (20 to 20,000 hertz), the very low inductance of the cable has absolutely no influence on the signal or the sound of the guitar.
Really not !

So, resistance and inductance are not a problem with the cable. So where does the influence on the sound come from?
"Villain" is the capacitance of the cable.
The component deliberately providing a capacitance is the capacitor.
As you may know, the simplest capacitor consists of 2 metal plates that face each other at a certain distance (without touching each other ...). There is no conductive connection between the plates and still electricity can flow ? This is where the difference between DC and AC comes into play. In fact, no DC current can flow through the plates (a connected battery would not give any current flow), but an AC current, such as the signal of a pickup, causes a constant mutual loading and unloading of the capacitor plates and thereby an actually measurable (alternating) current comes about.
Where is the capacitor in the cable?
If an electrical signal is to be transmitted via a line, 2 conductors are always required: one outgoing and one incoming line. Current flow and thus message transmission is possible only in the "closed" circuit, the current flows via a line from the producer (guitar) to the consumer (amplifier) ​​and back via a second line. The battery has 2 connections and the mains socket in the house also has 2 connections! (the third, namely the "protective conductor", is only for electrical safety ...).
In the case of the guitar cable, this usually occurs in such a way that one of the two conductors sits inside the cable (core) and the second conductor runs as a shield around the outside (coaxial). This shielding conductor is used in the amplifier connected to the housing and thus the "earth", resulting in a more or less good shielding from external interference.
However, this positive aspect causes the two conductors to behave like the 2 plates of a capacitor, resulting in a capacitance between the two conductors. This is the greater, the longer the cable, the closer the two conductors are to each other and also depends on certain characteristics of the intervening (plastic) material.
A typical capacitance value for a coaxial guitar cable is e.g. 100 picofarads (abbreviated pF) per meter of cable length (picofarads or the farad is the unit of measure of a capacity). With a 6m cable come about 600 pF together. What was still an advantage with the (small) ohmic resistance of the cable, namely that the other resistors involved were very large, proves to be a disadvantage in connection with the cable capacitance:
A high resistance in the signal generator (guitar) forms with the capacity of the connected cable a low-pass filter, a so-called RC element (R is the resistor, C the capacitor), which attenuates higher frequencies.
The effects are particularly noticeable when using the volume knob of the guitar:
When the control is fully open, the resistance of the potentiometer is parallel to the pickup and, if chosen large enough (for example, 250kOhm-500kOhm), disturbs the sound only insignificantly, or just enters the basic sound of the instrument. However, as soon as the controller is turned back, a corresponding part of its resistance appears electrically between the pickup and the output jack, to which the cable is ultimately connected. It forms the previously mentioned RC element of potentiometer resistance and cable capacitance, which attenuates the more intense the higher frequencies, the further the potentiometer is de-regulated (increasing resistance ...).
This effect, namely that the guitar becomes dull, the further the volume is turned down, one encounters the fact that one switches a capacitor so parallel that it bridges the resistance for higher frequencies and thus virtually compensates the cable capacitance. This use of the capacitor is also called "treble bleed". Strictly speaking, this only works perfectly depending on the value of the "treble bleed" capacitor used, but only for a certain potentiometer position and a certain cable capacitance. Other positions and other cables give different results.
An example: A guitar is equipped with a volume pot of 500kOhm. If I turn the potentiometer about 1/4 of its way back, so are a little more than 100kOhm in series to the output. The connected cable is a 6m cable and has (good) 600pF of capacitance. The resulting cutoff frequency, which is the frequency at which the signal only has about 0.7 times the value of the original, is about 2.5 kHz! And at the double frequency, namely about 5kHz only about 1/3 of the original voltage remains. A singlecoil guitar does that really hurt.
It gets even heavier with longer cables or cables with higher capacitance:
An old spiral cable has e.g. a capacitance of 1200pF. At 5kHz, only about 1/6 remains, corresponding to about 16%. That's really dull!

Second effect is the following:
Although the pickup itself essentially represents an inductance (coil), it also has a certain capacitance, the so-called winding capacitance, due to the many turns of wire which come very close to each other. This is the coil electrically parallel and the two form a so-called resonant circuit. Practically this has the effect that the output voltage of the pickup does not have the same amplitude everywhere in the generated frequency spectrum, but that at a certain frequency, namely the resonant frequency of the resonant circuit, an increased output voltage. Above the resonance frequency (higher frequencies), the amplitude drops relatively quickly, one speaks of a "low pass with resonance".
The typical resonance frequency of a pickup is somewhere between 2 kHz (humbucker) and 6 kHz (single coil) and the respective position characterizes the basic character of the instrument. How wide the resonance point, how strong is the elevation and how quickly higher frequencies fall off determined by the respective interaction of all pickup parameters: inductance of the coil, resistance of the winding wire, winding capacity, (in addition magnetic strength or coupling of the strings).
But why is the connected cable a problem now?
Seen electrically, the capacitance of the cable is parallel to the pickup, ie also parallel to the resonant circuit, formed from pickup inductance and winding capacitance. And parallel capacitances add up in value. And depending on cable type and length, the capacitance of the cable is often even greater than the winding capacitance itself. As a consequence, the resonant frequency of the pickup changes and, because of greater capacitance, to a lower frequency. And as already mentioned, the frequencies above the resonance drop off very fast. The sound of the guitar will get a different character due to the changed resonance and produce less treble. And all of that depends on the cable used or the cable length!

Phew, that was a lot of theory now. The practice is very simple:
Depending on the connected cable my guitar sounds different and is usually duller.

So far, maybe I have only one cable, which is connected directly to the guitar, as an offender in mind. But depending on the configuration of my equipment it may not stop there:
If I have effect devices, which are in my signal chain in front of the amplifier, further cables are added. The problem is a really desirable property of many effect devices, namely the "true bypass".
In the "true bypass" the effect unit is switched to "transparent" in the switched-off state, ie its input is connected directly to the output. This happens either via additional contacts on the activation switch of the effect or through a relay. Desirable because, with the effect off, there is no effect on the signal by the effect itself.
And why is that a problem?
Electrically, the cable from my guitar to the input of the effect and the cable from the output of the effect to the amplifier are directly in line. They behave like a cable with the added length of both cables. Two 3m cables become a 6m cable and two 6m cables become a 12m cable! Double the length means double the capacitance! Parallel to my pickup, affecting the resonance, or together with the volume pot, forming the RC lowpass!
When the effect is activated, the electronics disconnect the cable at the output electrically from the input, so also from our guitar. Already the frequency relations change again, although the effect might add only an echo? Complicated conditions!

What exactly does a buffer do now ?

The buffer uses active electronics to separate the effects of the cable at its output from the input and thus from the guitar. If the buffer is, as recommended, the first device after the guitar, it will act as "only" the connection cable to the buffer, all subsequent cables, no matter how long or through how many more ("true bypass") effects looped through no influence anymore. And regardless of whether the effects are activated or not. The guitar always looks constant, always "only" the first cable.
The buffer can do even more! It has the lowest possible output resistance and can deliver as much power as possible. This means that no matter how high the connected cable capacitances are (no matter how long the cables are connected), they can no longer form an effective low-pass filter, as this not only requires a large capacitor but also a large resistance. The buffer has a negligible resistance in the output. Ideal conditions!
In addition, there are effects units with relatively low input resistance, which, if the guitar is connected directly, the resonance of the pickup overloaded and may make the guitar appear expressionless, the actual character is lost. The buffer, with its small output resistance and the ability to supply a lot of power, has no problem with subsequent low input resistances. Even the smallest input resistance is negligible in relation to the low-ohmic output of the buffer.
And it goes further ! The greater the resistance that a cable connects, the more sensitive the cable is to interference. Mobile phone rattling, noise, hum or tapping noise can be very noticeable, especially for less good quality cables with less good shielding. An upstream buffer ensures a very low-impedance connection of the cable, disturbances are minimized.
Even (electrically) less good cables can be used at once!
One fact is that the demands on a cable are partially contradictory.
The shield against external disturbances should be as good as possible. However, this may increase the capacitance and the cable may become less flexible.
The plastic used between the shielding and the soul should keep the capacitance as small as possible and also make the cable insensitive to mechanical loads, but this may also result in less flexibility. If everything is mechanically optimized and the capacitance has been kept small, then the shielding may not be good anymore and disruptions are taking over ...

The use of a buffer gives more freedom in the choice of cables

The "normal" buffer is the first link in the signal chain behind the guitar. In general, it is executed as a floor effect and there is still the cable from the guitar to the buffer. The buffer has a positive effect only on cables that are behind him. He can not make a bad cable in the entrance any better!
(Missing heights due to the RC combination of volume potentiometer and cable capacity, could possibly be compensated by means of EQ in the buffer, but the shifted resonance is hardly correctable again ...)
So the problem of the first cable remains:
The "typical" sound of a particular guitar or a specific pickup is probably with a connected medium-length cable and medium capacitance. As an average length, one could assume 3m and as capacitance the mentioned 100pF per meter, thus altogether approx. 300pF. Shorter cables or less capacitance make the guitar sound "brighter", not necessarily an advantage for distorted playing, longer cables or higher capacitances make the guitar "darker", perhaps less so for "clean" playing.
At least from 6m cable length you should expect with clearly audible effects, with (electrically) very good cables maybe only a little longer length, at (electrically) worse perhaps even with less length.
Try an old spiral cable of (undressed) 2m length. Here, the capacity easily reaches more than 1000pF, which is very clearly audible.
In general, one should try out the completely unaffected sound of his instrument by connecting it with a short patch cable directly to an amplifier and playing as clean as possible. If you then switch to a longer cable, the influence is usually heard clearly.
Incidental note:
How strong the cable affects the sound also depends on the type of pickup.
The higher the inductance of the pickup (the higher the output power), the stronger the effect seems. The resonance is already quite low in frequency anyway, and a further shift downwards makes the guitar seem completely lifeless.
Roughly speaking, the influence of the cable on humbuckers and P90-type pickups (high inductance, high output power) is greater than with single coils (smaller inductance, less output power). For certain single coils, the resonance without a connected cable is so high that only a cable of medium length or capacity makes the instrument perfect and helps it to the desired character.

CONCLUSION: Actually, I have to look for a specific cable that suits the pickup of my guitar and possibly also my preferred playing style (rather clean or rather distorted ...).

- Do I use the volume pot of the guitar or is it always at 10?
- I would like to keep a rather high-pitched tone at a low-pitched volume or like it
a slight drop in altitude better?
- Does my guitar have a "treble bleed" capacitor?
- Do I always play more distorted or rather clean?

The answers to these questions depend on which cable is right for me. Maybe I can not find that all my wishes optimally fulfilled.
But can I possibly eliminate the influence of the first cable that connects my guitar to the "saving" buffer or perhaps directly to the amp?
Yes, the buffer just has to sit directly in or on the guitar. That's also the idea of "active" pickups. The buffer is simply built into the guitar, the pickup itself does not "see" the cable connected behind the buffer, so it can not be affected by resonance and resonance.
If my guitar has no active pickups and I do not want to install any later (possibly connected with woodwork to make room for the battery or their exchange), then there is the solution with an external buffer, directly on the guitar or the guitar strap is to attach.
An example of this is the
AG1.0 / The PURR .

What sets a good buffer apart from a less good one?

The buffer should have a high input resistance in order not to unnecessarily damp the resonance of the pickup. Of course, the smallest possible input capacitance is desired at the same time, but even with an unfortunate design of the buffer input, the capacitance of the connected cable is usually one order of magnitude higher.
At the output, the buffer should provide the signal as low impedance as possible and provide sufficient power to drive larger connected cable capacitance without loss.
Often, buffers (or amplifier circuits in general ...) have a fundamental problem with too large capacitances at the output (very long cables ...). This is due to the often used circuit technology, which is based on a very high gain, restrained by an internal negative feedback (operational amplifier). With large capacitances at the output it can lead to instabilities, the buffer or the amplifier starts to oscillate. On the other hand, the ideal buffer should be immune using appropriate circuitry.
The buffer should also be able to process large input signals distortion-free. Depending on the circuitry used, the internal (battery) or external power supply of nominal 9V is not sufficient. Then the supply must be increased internally, which requires additional circuitry.
Overvoltages, either caused by a false connection or especially by discharges of static electricity (ESD), endanger electronics in general, but especially when field-effect transistors (FETs) are involved. The buffer should have corresponding protection circuits in input and output.
If the buffer can also be supplied by a battery then of course the lowest possible power consumption has to be taken into account, with only external power supply this point is not that important. In general, the power consumption of an analog buffer is often negligible compared to other, perhaps digitally working effect devices (or to my maybe 100W strong tube amp, with which I can heat a room ...).
Of advantage may be a way of gain setting, so that the buffer as a booster or as an attenuator (often desirable value for humbucker pickups with very high output power ...) can be used.
Some buffers additionally offer a possibility of influencing the sound by means of treble, center and / or bass controls or switches. Of course, at least here you can not necessarily speak of a "neutral" buffer, possibly there is at least one switch or pot position, which sets the buffer to "neutral". Ultimately, the buffer is just another link in a possibly long signal chain. The negative influence of a cable is avoided, an absolutely "neutral" processing of the signal is perhaps not necessarily important, as long as the sound is right.
A "true bypass" in the off state is an advantage, so the device can always be in the signal path, even if its services are not desired.

What is the function of a buffer with a transformer in the output?
For devices with mains supply, their ground potential (housing) is usually connected to the protective earth of the socket for safety reasons. Also, devices (such as effects) that are powered from an external multiple power supply may be grounded as a consequence.
In the audio cabling of these devices, for example by means of jack cables, the ground connections of the devices are now once again linked via the cable shield.
This may result in large-scale loops in which interference voltages can occur ("ground loops"). Especially with mains-powered devices, this effect occurs when two audio-connected devices, e.g. be fed from two different sockets.
In order to avoid interference by such ground loops, a separation of the audio grounds must be made (the protective earth ground must not be easily separated for safety reasons ...). Responsible for this is a transformer, which transmits the signal of the respective device ground-free to the output. Thus, it is of practical use to have a universal device, such as a buffer, with which one can sever ground connections in case of cases. The buffer would then not be used as the first device in the signal chain behind the guitar, but rather as the last device in front of the network-supplied (and connected to the protective conductor ...) amplifier.

Does it matter whether the buffer is built using operational amplifiers or discrete transistors or FETs?
Not really !
The operational amplifier has the advantage that it is often not necessary to internally increase the 9V supply voltage to achieve high dynamics. There are operational amplifiers available that can provide practically the entire supply voltage range as signal voltage available (rail-to-rail operational amplifier). And with 9V output voltage, you can also buffers powerful humbucker pickups.
To be able to do without an increase in the supply voltage, has the decisive advantage that the necessary circuits can very easily produce interference due to their mode of operation.
With discrete components (single transistors or FETs) it is much more difficult to map a large dynamic range if only a very limited operating voltage (like the 9V ...) is available. An internal increase in the operating voltage is useful, but the interference generated thereby must be kept away from the output signal with increased effort.

Often one hears the opinion that the discrete circuit sounds "musical".
In practice, this usually means that it generates more distortion than the operational amplifier circuit. So, a discrete buffer may not be as "neutral" as one created with OPs, but it can sound more musical because the distortions it creates enrich the sound. Also a transformer in the output of the buffer can make the signal sound more "musical" due to the distortions it generates.
Conclusion: If I want a possible "neutral" buffer with great dynamics, then I am pleased about operational amplifier in the device. If I accept or even prefer a "musical" coloring, the discrete buffer may be the better choice (... can I still call the device a buffer, or is it an effect?)

Who does not need a buffer?
If I connect my guitar directly to my amp with a 3m cable, then I certainly do not need a buffer in between. The only reason to use a buffer would be if, for example, I wanted to amplify my guitar signal in front of the amp to get into the distortiion range more easily, or vice versa, my signal has to weaken somewhat, because I definitely want to play the amp "clean" and my guitar gives off too much power for it.
Here could a "buffer" with boost and / or attenuation function perform well, for example a
AG2.0 or a AG2.1.
When using longer cables, I should definitely try out how the guitar sounds with a short cable compared to the long cable. Do I need the long cable, but rather wants the sound of the short cable, then a buffer directly to the guitar would be a solution (
AG1.0/The PURR).
Some guitars already have a buffer on board. Either as a separate unit or in the form of "active" pickups. Powered usually by one or even two 9V batteries, they already fix most cable problems, eliminating the need for an external buffer.
Or at least I bypass the problem of the cable in direct connection with the pickup by using a radio link between guitar and amplifier / effects.
But that is another story...