Most Transparent Buffer?

[upspoon12]

Hey everyone,


Just wondering what the most transparent small buffer circuit out there is. I'm thinking IC as opposed to a fet, but then again, what do i know?!
After searching around and looking at many many many different circuits, which ones do you guys like?

what are your thoughts?

[Mark Hammer]

It will be the one with a big-enough input impedance whose power supply and consequent voltage swing drastically exceeds any possible input signal.

That's a pretty big list.

[ashcat_lt]

It's really hard to beat a basic non-inverting opamp buffer most times.  Like Mark said, pick a big enough power supply and set the impedance on either end with resistors.  Spend whatever you want on the opamp.

[R.G.]

Also, be honest with yourself about what you mean by "transparent". That word has been overused until it's almost meaningless.

Practically any opamp buffer is dramatically more "transparent" in the sense that the output is just like the input, than any simple JFET buffer. Every one-JFET, or one-MOSFET, or one-bipolar buffer will introduce some small distortion. Some of these are pleasant-sounding distortions to some people, but they are a diversion from following the input. Opamp buffers follow their inputs with very little distortion from the input signal. Below 0.1% distortion is common, and many get a lot below that.

If you mean "I can still tell it's my guitar even if it's pleasantly distorted" by "transparent" as many guitarists do, that's a whole other ball of wax.

[upspoon12]

My definition of transparent is to color the tone as little as possible and not distort.  To maintain as much of thr original signal just amplified without adding color.  It seems like opamp is the way to go for me.  When I'm designing this buffer,  does it matter if I put the filter before thr impedence/bias resistors or vice versa?

[Seljer]

Its a buffer. They're all like only 5 components. You can solder them together without a PCB in under 10 minutes in whichever construction method you favor.

Spend an hour building a bunch of them and do blind testing A vs B testing if you think theres going to be that much of an audible difference 


The only time I've had a buffer not be "transparent", is when it runs out of headroom and clips (curse you high output pickups!). Or when driving low impedance loads (i.e. not a guitar amp), if the output impedance is low enough some of them do get loaded down and you may lose 1 or 2 dB ....but still way less than what you would lose the buffer.

[Mark Hammer]

My definition of transparent is to color the tone as little as possible and not distort.  To maintain as much of thr original signal just amplified without adding color.  It seems like opamp is the way to go for me.  When I'm designing this buffer,  does it matter if I put the filter before thr impedence/bias resistors or vice versa?

Any decent buffer will colour your tone....relative to what your guitar might sound like plugged straight into the amp.

I often recommend the following experiment.  Plug your guitar into your amp with the longest cable you have.  Set the amp for the cleanest tone and flattest frequency response you can.  Strum away.  Now plug your guitar into the amp using the shortest physical cable you have that will still allow you to play your guitar without tearing anything.  The difference in tone you hear is the effect of cable capacitance and sub-optimal impedance matching.

The buffer hasn't added any treble, but it wil retain as much treble as your pickups can offer.  That treble may be more than you are accustomed to hearing.  It's not "coloration" in the same sense as adding harmonic distortion, but it IS a tonal change.

[GibsonGM]

My definition of transparent is to color the tone as little as possible and not distort.  To maintain as much of thr original signal just amplified without adding color.  It seems like opamp is the way to go for me.  When I'm designing this buffer,  does it matter if I put the filter before thr impedence/bias resistors or vice versa?

Any decent buffer will colour your tone....relative to what your guitar might sound like plugged straight into the amp.

I often recommend the following experiment.  Plug your guitar into your amp with the longest cable you have.  Set the amp for the cleanest tone and flattest frequency response you can.  Strum away.  Now plug your guitar into the amp using the shortest physical cable you have that will still allow you to play your guitar without tearing anything.  The difference in tone you hear is the effect of cable capacitance and sub-optimal impedance matching.

The buffer hasn't added any treble, but it wil retain as much treble as your pickups can offer.  That treble may be more than you are accustomed to hearing.  It's not "coloration" in the same sense as adding harmonic distortion, but it IS a tonal change.

EXACTLY.   For instance, you could try making something with 'the ultimate in input impedance - 100 megs!!".  But for some instruments (I'm thinking piezo pickups here, especially), that would make your output so BRIGHT sounding as to cut off peoples' heads in the audience!  And it was always there!   Just that all those capacitances dumped it, and what we were left with sounded good.

We're constantly enjoying and loving the NON-transparent things that happen all around our instruments, all the time.

Seljer said "build a few, see which one you like!" and that seems like great advice!     You'll build 2 or 3, and one of them will stand out as good and transparent to YOU - and you'll swear that's the only buffer worth anything, he he!     

[R.G.]

When I'm designing this buffer,  does it matter if I put the filter before thr impedence/bias resistors or vice versa?

Filter?? Filter is not equal to transparent. What did I miss?

[roseblood11]

I often use this simple opamp buffer, with this horrible layout (which fits vertically into the small side of a 1590B:


It's got a high input impedance and uses a quality opamp and a voltage doubler. I wouldn't say it's perfect, but nobody needs more than that for a guitar...

I don't have a schematic, but it's easy to draw one from the layout...

[upspoon12]

Everything you guys are saying is certainly making a lot of sense. That's awesome.  Perhaps in my infancy in pedal making and electronic engineering I don't have the full vocabulary of terms and understanding.  But I'm glad to have the words of intelligent people to guide my way.  Thanks everyone!  Everything is dually noted.  +1 understanding

[upspoon12]

The little red square with thr circle on on the board represents a cut in the trace?

[Jdansti]

Yes. You can use a drill bit and just twist it with your fingers.

[Gus]

As posted before what does transparent mean?

A buffer is a gain of one circuit a gain stage/boost is more than a gain of one circuit
The buffer part is easy a gain of one circuit at 9VDC made from an opamp or a jfet(s) or a BJT(s) does not have much distortion if built correctly

The input resistance matters, output resistances matters but often not as much, most circuit you will find on the web have a low enough output resistance.
What cable will be use between the guitar and buffer?  The cable capacitance matters
What input resistance will you want?
What passive guitar pickups will be used?

The input resistance interacts and can have a resonance peak with the LRCs of the guitar and cable between the input of the buffer.  This can change the tone and as a extra bonus the tone can change with volume control setting of the guitar
SO there is no such thing as a transparent buffer
the input resistance will change the interaction with the guitar

Search the site for "buffer
I posted a simple single pickup guitar sim in the thread below.  Different pickups have different inductance and resistance etc.
http://www.diystompboxes.com/smfforum/index.php?topic=107397.0

[upspoon12]

the cables i use are cables that i make myself. Being canare GS-6 Cable and Neutrik Connectors. i have a selection of lengths 20' and under.

I have a selection of guitars i use, strats, tele, vintage les paul. the resistances of the pickups in said guitars range from 5.65k in some of the single coil pickups to 16.5 humbucker. and an assortment in between.

for the input resistance, i want something that will allow through a nice amount of highs, but not so much sizzle that it will take your head off. I think i will have to breadboard it up and play with different resistances in the impedance setting resistors to find the sweet spot and i realize that each guitar will have a different sweet spot.

I am wondering if it is necessary at all to filter out the inaudible frequencies. Part of me thinks maybe, to free up headroom at the top, but i feel like the more correct answer will be it is unnecessary.

[MrStab]

I am wondering if it is necessary at all to filter out the inaudible frequencies.

only the troublesome ones, ie. radio. a 33k resistor in series followed by a 100pF cap to ground at the very start of the signal path (before input cap) is one way. afaik. but with a unity-gain buffer, you'd probably only need that if something before it was unprotected.

[Lurco]

My definition of transparent is to color the tone as little as possible and not distort.  To maintain as much of thr original signal just amplified without adding color.  It seems like opamp is the way to go for me.  When I'm designing this buffer,  does it matter if I put the filter before thr impedence/bias resistors or vice versa?

It`s not a simple buffer anymore, if it amplifies.

[R.G.]

I have a selection of guitars i use, strats, tele, vintage les paul. the resistances of the pickups in said guitars range from 5.65k in some of the single coil pickups to 16.5 humbucker. and an assortment in between.

for the input resistance, i want something that will allow through a nice amount of highs, but not so much sizzle that it will take your head off. I think i will have to breadboard it up and play with different resistances in the impedance setting resistors to find the sweet spot and i realize that each guitar will have a different sweet spot.

Good realization! You have mutually conflicting needs.

Here is something to ponder: the input impedance of the buffer is not the only place in the signal chain that you can control how much high frequency is apparent from each guitar. First, the humbuckers naturally have less highs to start with because of their larger inductance and self capacitance. The problem with them is getting enough highs. The single coils will put out more treble, but even a single coil cuts off somewhere around 7-8kHz. So for the guitars exclusive of any effects that generate more high frequencies, the output typically doesn't even go into the top octave of audio.

Since you use many guitars, you might want to tune the guitars to match the setup instead of tuning one setup to match many guitars. A small capacitor on the jack of the guitar, or a resistor even, would cut the highs by a trimmable amount. Just a thought. At least this is possible without sacrificing your perception of the right amount of highs.

I am wondering if it is necessary at all to filter out the inaudible frequencies. Part of me thinks maybe, to free up headroom at the top, but i feel like the more correct answer will be it is unnecessary.

It isn't necessary.

[karbomusic]

A small capacitor on the jack of the guitar, or a resistor even, would cut the highs by a trimmable amount. Just a thought. At least this is possible without sacrificing your perception of the right amount of highs.

While being both facetious and serious, I use the tone knob for that; it's built in already. For the OP, I also think that most any decent opamp style buffer would do the trick as RG mentioned before. I want to be gentle when I say, simply having a decent buffer should take care of the signal well enough because by then the differences between guitars exist because they are different guitars, allow them to have their respective personalities, and exploit that instead of masking it.

[upspoon12]

Here is something to ponder: the input impedance of the buffer is not the only place in the signal chain that you can control how much high frequency is apparent from each guitar. First, the humbuckers naturally have less highs to start with because of their larger inductance and self capacitance. The problem with them is getting enough highs. The single coils will put out more treble, but even a single coil cuts off somewhere around 7-8kHz. So for the guitars exclusive of any effects that generate more high frequencies, the output typically doesn't even go into the top octave of audio.

Interesting.
This all certainly makes sense to me. While i am building this for myself, i also will be building one for my business partner who is a session guitar player. I wonder if i could make the Impedence setting resistors switchable  and find say 3 different resistors with 3 different values? Try and find a happy median between the three for different guitars?
while for mine i would likely just set one but for him he is like any true session guitar player. Tone is his god, and anything that prevents the proper tone from coming through is unacceptable. And while both he and i know that you couldn't possibly find a resistor value for each guitar in the same circuit, do you think 3 interswitchable could work? I picture it like an old DS-1 i have that i have different clipping diodes in that are switchable via toggle.

I am wondering if it is necessary at all to filter out the inaudible frequencies. Part of me thinks maybe, to free up headroom at the top, but i feel like the more correct answer will be it is unnecessary.

It isn't necessary.


thought so - thanks!

A small capacitor on the jack of the guitar, or a resistor even, would cut the highs by a trimmable amount. Just a thought. At least this is possible without sacrificing your perception of the right amount of highs.

While being both facetious and serious, I use the tone knob for that; it's built in already. For the OP, I also think that most any decent opamp style buffer would do the trick as RG mentioned before. I want to be gentle when I say, simply having a decent buffer should take care of the signal well enough because by then the differences between guitars exist because they are different guitars, allow them to have their respective personalities, and exploit that instead of masking it.

Certainly i want to maintain the charactoristics of each guitar, otherwise there would be no use in having several different kinds of guitars.

After much reading and taking your opinions into consideration i definitely thinkg that this non inverting opamp as close to unity gain as possible with potentially 3 switchable value impedances is the direction i want to head.

This buffer is to be placed inside a 5 Channel Looper Pedal, I was also considering making it switchable to have it in the begining of the chain or end of the chain.