Suggestions for buffer/booster circuit

[egasimus]

I'm looking for recommendations about what opamp-based guitar buffer-booster circuit to build. I'm going to make one for personal use, and, even though few things are simpler than an opamp buffer, I'd like mine to have some tricks up its sleeve.

Ideas:
I'll definitely slap a LM3914 LED bar graph in there. More info on interfacing it with the buffer?
"Treble boost" and "bass boost" switches, perhaps?
"High" and "low" inputs in case I plug something with an onboard preamp into it.

I haven't ruled out the possibility of designing my own, but I still feel quite insecure about it. Any pointers would be helpful.

[Gus]

You are posting about a circuit with  EQ, a bar graph and a buffer.   That reads like a guitar/bass/etc  preamp.

Google Ask etc for opamp EQ and 3914  bar graph app notes for a start then guitar preamp schematics.

[egasimus]

Well, I guess it is, although I want to keep it simple, something that could be built in one sitting. I guess I'll have a go at designing my own and post the schematics here for discussion.

[anchovie]

I'll definitely slap a LM3914 LED bar graph in there. More info on interfacing it with the buffer?

http://www.diystompboxes.com/smfforum/index.php?topic=91629.0

"Treble boost" and "bass boost" switches, perhaps?

For perceived treble boost, switch in a smaller input cap. For "bass boost", switch in a low pass filter at the end.

"High" and "low" inputs in case I plug something with an onboard preamp into it.

Why worry? It's no different to having a buffered pedal (e.g. tuner) first in your chain with other stuff after it.

[egasimus]

Thanks, the Glass Blower was what I had in mind, but I forgot the name so I couldn't find it. But I'd prefer to have some actual treble and bass boost, so I'm currently studying active filters... about time I did, too

[egasimus]

I understand that my posts might seem a bit random, but I'm researching as we speak. My newest find: ROG Mr. EQ. Probably would do a good enough job as a buffer, too. I can only source parts locally, though, so, if I can't find an unbuffered 4049, can I replace those inverters with ordinary inverting opamp stages?

[merlinb]

if I can't find an unbuffered 4049, can I replace those inverters with ordinary inverting opamp stages?

Not directly no.

There are lots of schematics for treble/bass EQs online though. If you want a one-knob active tone then a tilt control is a nice compromise (basically an active bigmuff tone control):
http://gilmore2.chem.northwestern.edu/images/eq_tilt.gif

From this page:
http://gilmore2.chem.northwestern.edu/projects/equal_prj.htm

[Tony Forestiere]

http://gilmore2.chem.northwestern.edu/projects/equal_prj.htm

That was a great read. Thanks for posting it.

[egasimus]

Ok, this is my first attempt at actually designing a schematic from scratch. Just a run-of-the-mill buffer, which I based on the TL074. So amidoinitrite?


Questions:
1. Input and output cap values OK?
2. Should there be a DC removal filter between two opamp stages if they're both biased to Vb?
3. I'm not sure if I tied the correct legs of the pots together - I've never actually used any log pots. Can I use some sort of linear pot + tapering resistor combination in this scenario?

[Gurner]

Those 1M resistors on your -ve inputs are excessive.....go with lower values (better still, if you simply want unity gain, go with a non inverting voltage follower...higher input impedance & you'll save a couple of resistors)

No, there's no cap is needed between your opamp stages (since you asked)

You don't need those resistors to your opamp +ve pins.

[R.G.]

Ok, this is my first attempt at actually designing a schematic from scratch. Just a run-of-the-mill buffer, which I based on the TL074. So amidoinitrite?

That does result in a buffer with a 500K input impedance and variable gain. There are some subtleties that you missed; however, there is no reason for you to have run into them before now. These are all the nit-picky details of opamps and non-ideal-ities of using them.

1. The (-) input of an opamp sits at virtual ground. So the input impedance of an inverting stage is equal to the input resistance at low frequencies. You are using a 1M series resistor for this resistor, and a 1M pulldown, so that gets you to the half-meg low frequency input impedance. You then invert it again, and add a variable gain. Still OK.
2. Every ohm of resistance you add to your input adds thermal noise of its own. So 1M in series with the input and a 1M feedback cap will be substantially noisier than smaller values.
3. This is an AC-only amplifier, so DC balance isn't that big a deal, but the bias current of the opamp will cause current through the feedback resistor and biasing resistor on the + side to generate an offset voltage. With FET input opamps, this isn't a big deal, but it gets that way quickly with BJT input opamps. And it can get very bad if you happen to pick an opamp which has a low input impedance. The NE5532, otherwise a great low noise audio amp, only has an input impedance of about 100k on the + or - input, so using that in a 1M/1M inverting amp will cause problems.
4. 1M and up resistors have problems with contamination of the PCB around them and parasitic capacitances changing the frequency response you thought you had.

It's one of those truisms that it's hard to get high input impedance on an inverting opamp input. There are tricks for how to do it, some of them posted at Geofex; but in general, of you don't have to make your input opamp stage inverting, getting high input impedance is easier with a non-inverting stage. Of course, noninverting stages bring their own set of problems, but low input impedance is not necessarily one of them.

This leads down to the question: did you have a reason to go inverting for your first stage? There are situations which need that, but if you can go noninverting, you can get both high input impedance and lower noise (a bit, at least).

A second question is whether you need the gain to go all the way to zero, or if you can accept unity gain as the minimum. If you can, you can use one stage for your buffer, and save the second stage. As you've shown it, the second stage has a gain from zero (with the feedback pot at zero ohms) to unity.

1. Input and output cap values OK?

F = 1/(2*pi*R*C)

2. Should there be a DC removal filter between two opamp stages if they're both biased to Vb?

It depends. Theoretically, no. Practically, in most cases with near-unity gain, still no. With problems with bias-current offsets caused by high value resistors, maybe; especially if you use BJT input opamps. With JFET or MOSFET input opamps, the opamp DC accuracy is worse to start with, but they have almost no input bias current so the bias current offsets don't add up and get you. Net - it depends on which opamp and resistors you use.

3. I'm not sure if I tied the correct legs of the pots together

Gain of an inverting stage is G ~= Rf/Ri. Ri is fixed, 10K. Rf is variable between 0 and 10K. Gain varies from 0 to 1. The correct pins to connect are the ones which make the resistance of the pot be 10K when the shaft is turned fully clockwise.

I've never actually used any log pots. Can I use some sort of linear pot + tapering resistor combination in this scenario?

Yes, but do the math to figure out what degree of tapering you want, and which way.

[egasimus]

Thanks. This made a lot of things clear to me, and, of course showed me how much I don't actually know. Was it Socrates who first observed this phenomenon?



So, yeah I'm not really sure what I'm doing here. Also left finding out which pin corresponds to the clockwise lug and the related tapering resistor calculations for later.

[egasimus]

Bump. This isn't how I bias a non-inverting stage, is it? More like 100k resistors from Vb to the non-inverting input.

[merlinb]

I thought it was active tone controls you were trying to design? There are plenty of clean boosters out there to copy, but you seem to be trying to reinvent the wheel?

[PRR]

> Bump

[egasimus]

but you seem to be trying to reinvent the wheel?

Well, yeah, kinda. I have no formal electronics education, and I think I've learned way too little for the three years I've had this as my hobby. So I want to be sure that I know what I'm doing, and not just copying stuff...

PRR, thanks - this'll help.