Which buffer do you like?

[midwayfair]

Op amp is best if you need headroom, high input impedance, and bandwidth because you can just use a better op amp if you need to. A TL072 isn't the best buffer in the world, but with guitar you aren't likely to notice any shortcomings. Current consumption is higher than a single transistor. Output impedance is so low you have to add a resistor to the output.

A BJT is good if you don't need sky-high input impedance but you do want lots of headroom and can sacrifice a tiny bit of bandwidth (in part from cable capacitance and lower input impedance). You can get 1M input impedance from a BJT which ought to be enough for any bass player to be happy. Current consumption is very low. I strongly disagree with CultureJam that it's the worst of the lot.

A JFET is good if you need very high input impedance (for our purposes effectively infinite) and bandwidth but don't need headroom. A JFET can be set up with gate biasing for more headroom but you are still contending with a device-level problem. Current consumption is very low.

A MOSFET is a truly lousy source follower buffer (try it and see), but if you can live with an output impedance of, say, 1K (I think that's about as low as I'd go on the drain and source resistors), and don't need to conserve current (i.e. you aren't using a battery), you could use it as a unity gain inverter and get the extremely high input impedance of a JFET, excellent bandwidth, and tons of headroom.

Hybrids: There are lots of two-transistor buffer arrangements, but that might be more complicated than you need.

Positive feedback: You can use a bit of positive feedback to make up for low input impedance. You usually get less noise and can correct the bandwidth loss from the cable capacitance... sometimes. IIRC the Cornish buffer does this.

If it were me doing something for a bass, I'd just use a BJT with a 1M at the base tied to a 22K/22K voltage divider and a 10K emitter resistor, and run it on 18V (two batteries if it's going inside the bass) if I managed to distort it and just live with the imperfect nature of such a simple device. Basically the second thing on Jack Orman's basic buffers list.

[ShredderNemo]

There really is no such thing as "the Klon buffer". The buffer in that circuit came straight off the op amp datasheet in the "example applications" section.

Might as well call at the National Semiconductor or Texas Instruments buffer. 

Is there a good location to view example applications of op-amps? It seems like a surprising number of big-name manufacturers used an example design in this fashion.

[culturejam]

I strongly disagree with CultureJam that it's the worst of the lot.

I should have qualified my statement a bit. I don't personally think BJTs are lousy buffers. But the general perception, at least from the many articles and forum posts I've read, is that they aren't as "good" because you get a little less than unity gain and the input impedance is not has high as an op amp or JFET. Do these things matter in most designs? Probably not.

I actually use them all the time in my designs. And it's probably just psycho-somatic, but I swear BJT buffers are not quite as "bright" as op amp buffers or JFET. I've not scoped anything or done any waveform analysis, so I might be full of shit on this.

[midwayfair]

but I swear BJT buffers are not quite as "bright" as op amp buffers or JFET. I've not scoped anything or done any waveform analysis, so I might be full of shit on this.

They aren't. The input impedance isn't as high no matter how big you make the base resistor. It's a device limitation. Some BJT-input op amps are marginally better than a single device (and some are far worse), but there you can run into problems if you make the bias resistor too large anyway.

[jubal81]

but I swear BJT buffers are not quite as "bright" as op amp buffers or JFET. I've not scoped anything or done any waveform analysis, so I might be full of shit on this.

They aren't. The input impedance isn't as high no matter how big you make the base resistor. It's a device limitation. Some BJT-input op amps are marginally better than a single device (and some are far worse), but there you can run into problems if you make the bias resistor too large anyway.

Having great success using these in distortion circuits in places other than the input stage. Few parts, outstanding headroom and the slight low-pass is actually a bonus. Very nice putting one of these after the tone stack and before the volume pot, for example.


Jon's on point. For standalone, can't really beat opamps. I like using the second in parallel for lower noise and more power. Even better would be with a split supply ...

[jimilee]

I need to learn how to take schematic to vero I think. Googles, here I come!


Sent from my iPhone using Tapatalk

[jubal81]

I need to learn how to take schematic to vero I think. Googles, here I come!


Sent from my iPhone using Tapatalk

I'll lay out a PCB for you. Share it on OSH.


You want a switch? Toggle? Footswitch?

[jimilee]

Foot switch would be great! Thank you sir!


Sent from my iPhone using Tapatalk

[jubal81]

Foot switch would be great! Thank you sir!


Sent from my iPhone using Tapatalk

3PDT or DPDT/Optofet?

[jimilee]

3pdt would work


Sent from my iPhone using Tapatalk

[jubal81]

3pdt would work


Sent from my iPhone using Tapatalk

How's this look?

[jimilee]

That's really sharp, thank you.


Sent from my iPhone using Tapatalk

[m-Kresol]

How's this look?

the buffer will be always in the circuit and you can switch something behind it on/off if I'm not mistaken. Was this the purpose of the application here? your schematic is 100% correct (although imho C6 and R9 are overkill), but I'm not sure that's what Jimi intended. I thought the switch was to toggle the buffer on and off.

I have the regular opamp buffer used in the klon in my oshpark projects (useable as 2 buffers), altough without a toggle, if that helps.

[jubal81]

How's this look?

the buffer will be always in the circuit and you can switch something behind it on/off if I'm not mistaken. Was this the purpose of the application here? your schematic is 100% correct (although imho C6 and R9 are overkill), but I'm not sure that's what Jimi intended. I thought the switch was to toggle the buffer on and off.

I have the regular opamp buffer used in the klon in my oshpark projects (useable as 2 buffers), altough without a toggle, if that helps.

Yeah, misunderstood what Jimi wanted. Going to redo it as a standalone buffer pedal with the buffer on a footswitch.

[JC103]

There really is no such thing as "the Klon buffer". The buffer in that circuit came straight off the op amp datasheet in the "example applications" section.

Might as well call at the National Semiconductor or Texas Instruments buffer. 

I've heard that both the Cornish and "opamp-in-the-klon" buffers are straight up Text Book buffers. Would love to find the old British data sheet or electronics mag that Cornish sourced-nicked-borrowed his design from.