Simple (FET) preamp schematic needed that goes from unity gain to about +4dB

[Speeddemon]

Hey scaesic,
so basically a lineair 25k pot would work good there?

Now for the total circuit, is all I need this:
-input cap (47n or 100n?)
-voltage divider resistors (2x 22k?)
-R1 100k
-TL071
-R2 100k
-25k pot
-output decoupling electrolytic 10uF?

did I forget anything?

[scaesic]

actually, if you have it in non-inverting mode you could drop one resistor out as the eq for that is 1+R2/R1,    R2 = 25k pot, R1 = 200k

http://www.radio-electronics.com/info/circuits/opamp_non_inverting/op-amp_non_inv_with_capinput.gif

[Speeddemon]

But where should that resistor from the + (pin 3 on a TL071, right?) go?

Also, were my cap-values correct?

[scaesic]

dunno if they're correct, go check a simple op-amp distortion pedal.

[lovekraft0]

After realizing how confusing my previous post was, I breadboarded this up for giggles this afternoon, using what I had on hand  - the likely issues are headroom and those nasty electrolytics in the signal path, but it appears to work fine, and sounds pretty transparent  .


Feel free to mod as you like - R5 should probably be a 20K trimmer, but the 10K worked fine on my board (biased the drain to about 5.5 volts - close enough for proof of concept). As always, YMMV.

[Steben]

Are those electrolytics necessary?
With 1M inpout impedance I would say not (even 0.0022 MKT would be enough), but maybe with the feedback loop...?

[brett]

Hi.
That's a cool schematic.
Simpler again would be to drop the 27k and 33k resistors, and make R4 = 12k.
Then it's a buffer (gain~1.05 ) with C4 switched OUT, and a booster with several dB of gain with C4 switched IN.  Making C4 smaller (e.g. 1uF) reduces the gain to somewhere near your desired 4dB, but it might also introduce a small amount of treble boost (probably inaudible and/or desirable).
And for maximum transparency, the input cap (0.1uF) and output cap (0.22uF) should be film (MKT) or greencap types.
cheers

[d95err]

the likely issues are headroom and those nasty electrolytics in the signal path

You can put a small cap (e.g. 0.1uF) in paralell with the electrolyte to ensure a flat frequency response with no effect from the electrolyte impedance. This trick is used e.g. in Jack Orman's Super Buffer: http://www.muzique.com/lab/superbuff.htm

[Speeddemon]

Hmm, I might wanna stick with the op-amp preamp then, might be a tad simpler.

Btw, is the GGG IC Buffer useable as a template to make such a preamp?
http://www.generalguitargadgets.com/pdf/ggg_ic_buffer.pdf

If so, what would needed to be change?
Would the 2 2M2's need to be replaced with 22k for the voltage division, and between pin 6 and 2 the 100k + 25k potmeter?
Where would the 100k (R1) go from that negative feedback diagram calculator?

[Speeddemon]

actually, if you have it in non-inverting mode you could drop one resistor out as the eq for that is 1+R2/R1,    R2 = 25k pot, R1 = 200k

http://www.radio-electronics.com/info/circuits/opamp_non_inverting/op-amp_non_inv_with_capinput.gif

What is R3 here, a de-popping resistor (1M or something that size)?


EDIT:

Would this be correct:

[d95err]

Speeddemon:
There seems to be several problems with your layout.

First, the input cap goes to ground and never gets anywhere close to the opamp. Probably just a simple mistake.



Second, the schematic you found at radio-electronics assumes that the opamp is fed with bipolar power, e.g. +/- 9V. To use singular (e.g +9V only) with an opamp, you need to "trick" the opamp into thinking it is seeing a bipolar supply. This is done by creating a voltage of about half the supply (e.g. 4.5V for a 9V supply). This is called a bias voltage. The bias voltage is added to the signal, thus putting the signal halfway between +9V and 0V. I.e. when the signal is completely quiet, it will actually be a DC voltage of 4.5 volts.

Look around at various FX pedals using opamps.  Check out distortions, delays, choruses, mixers etc. Look at how they use the bias voltage and try to figure out how to do that for your circuits. If you come up with something, it's probably best to post the schematic here first to let the community review it before you start designing a PCB.

(I'll see if I can do some kind of sketch for it tomorrow, but right now I'm too tired)

[lovekraft0]

Are those electrolytics necessary?

No, they're what I had on hand - the coupling caps could both probably be as low as .22, or even .1uF, so long as you're not using this for bass. I'd recommend the large cap on C4, unless you want shelving bass cut, in which case, tailor it to suit ya. With C1 and C3 changed to .22uF film caps, response should be flat (+/-3dB) down to about 26Hz.

[Speeddemon]

Yeah, I saw that mistake now about the input cap being grounded!

About the 4.5V for the op-amp, that's what I had the 2 22k resistors for; the voltage dividers; 1 from 9V to signal and 1 from signal to ground.

So, apart from the input cap grounding, is the rest correct?

[d95err]

Yeah, I saw that mistake now about the input cap being grounded!

About the 4.5V for the op-amp, that's what I had the 2 22k resistors for; the voltage dividers; 1 from 9V to signal and 1 from signal to ground.

So, apart from the input cap grounding, is the rest correct?

The 200k resistor in the feedback loop should go to the 4.5V connection, not to ground. However, you can't connect it to the current voltage divider, since that is also the input. The buffer can use the voltage divider on the input since it doesn't have any other parts that need to connect to the bias voltage. On a gain stage, you need a voltage divider that is separate from the input and then connect from that to the input with another resistor. You usually put a filter cap on the 9V and another on the 4.5V as well. Check the opamp schematics on the bottom of the AMZ Lab Notebook booster article for a schematic of the gain stage. Check out any opamp circuit for how to use bias.

For example, the Rat: http://www.generalguitargadgets.com/diagrams/rodentsc.gif