Guitar LED circuit sucking tone

[T-Flex]

Hello,
I have an acrylic fat strat I frankensteined together for shows and added a small, op-amp based circuit that contols 4 LEDs behind the pick guard.  Trouble is, this circuit is cutting off some of the high end of the signal.  I cant just throw in a bypass switch because I want the tone to be corrected when the circuit is active too. Raising the input impedence destroys the desired LED effect. Would a FET based input buffer help or perhaps I need to rearrange the circuit layout?
Thanks

Circuit is set up like this:

Pickup_________________________Output
                        |
                        |____OpAmp____
                                                   |
                                                 Gnd

Pic:

[slacker]

A buffer should do the job, from pickups into buffer output of buffer to opamp and output, that would probably stop the opamp circuit causing tonesuck.

If you posted your circuit someone can probably figure out a way to stop it causing the problem though 

Looks very cool by the way.

[Hides-His-Eyes]

Yeah, just insert a buffer before anything happens! You've already got the power in there.

Is the op-amp in an inverting or non-inverting config?

[T-Flex]

Its non-inverting, just a 386 for attack sensitivity.  I just built a wah that features j201 based buffers, do you think something like that would suffice? I have all the parts available.
Thanks.

[slacker]

The 386 has quite a low input impedance, about 50k I think, so a buffer in front is probably the easiest solution.
A simple J201 buffer will do the job fine, there's some good examples here http://www.muzique.com/lab/buffers.htm.

[T-Flex]

Excellent.  Thanks for the pointers.

[Hides-His-Eyes]

Its non-inverting, just a 386 for attack sensitivity.  I just built a wah that features j201 based buffers, do you think something like that would suffice? I have all the parts available.
Thanks.

I agree with slacker, a Jfet buffer will be fine.

Just FYI an LM386 is not an op-amp. It's a special function amplifier chip that just happens to connect in a convinient op-amp style way.

[MrStab]

not to dig up two dead threads in one day or anything... but i was wondering if the buffer required to make this work would definitely need to be placed before the signal splits, or if it could be placed after (so just before the LED circuit)? i think it's already been implied twice that it needs to be before anything happens, but if so, is there no contingency to keep the audio signal independent besides a separate battery for the buffer (or fast switch-flicking)?

gonna just make a simple TL072 buffer, hoping it'll be as transparent as possible.

cheers!

[slacker]

I'm not sure what you mean by keeping the audio signal independant.
If you want you can put the buffer after you've split the signal. The input of the buffer will then be in parallel with what ever you plug the guitar into but if you make it have a high input impedance this won't noticably affect the sound.

[MrStab]

ideally i wouldn't want the signal coming out of the guitar to pass through the buffer at all, i guess what i'm asking is can that be done?

i know this prolly demonstrates a poor understanding of impedance, but what better way to learn than flashy things! i can afford some leeway in terms of screw-ups as it's not my primary gigging guitar.

[Gurner]

Imagine a Y cable** ...one leg of the Y goes to your guitar amp as normal, the other leg goes to a buffer.....

                           _______ To guitar amplifier as normal.
pickup signal-------|
                          ---------  to a 'buffer' circuit feeding the LED board.

Signal does not pass through the buffer.

Buffer (feeding your LED circuit) does not load the signal.

Job done.



(**it's probably better described  'T off' ...as i  just when you 'T off' a pipe from say a main water pipe)

[MrStab]

dude, you summed that up perfectly! thanks a lot, i was unsure as my braindead intuition made me think "hmm, thing fix sound. thing go right at start just in case."

out of curiosity, would it even work if you did pass it through the buffer first and then split? or is it just the same-o?

parts should arrive on saturday or monday hopefully. my concerns now will probably be any possible interference from the LED's being in close proximity to the (unpotted) pickups and routing everything. if tone-shaping transparency isn't an issue anymore, i can build the laziest POS buffer imaginable. lol

[slacker]

It works either way, one advantage of splitting the signal before the buffer is your guitar will still work if the battery powering the buffer dies, if you took the signal from after the buffer it would only work if it had power.

[deadastronaut]

interesting...

@G:  so thats why i had a little , but noticable interference when using the 'Y' with the 386 on the diy sustainer then?..

[MrStab]

good to know, Iain - so turns out it's advantageous in another way altogether to not screw with the tone.

it was your thread where i first heard about using the LM386 for this purpose, Rob. talking of which, i've tried combining a similar schematic (modified? i "credited" your name for it either way!) i found with a simple buffer from Muzique.com. reckon it'll work? there are probably some glaring mistakes there, feel free to mock!

[Gurner]

interesting...

@G:  so thats why i had a little , but noticable interference when using the 'Y' with the 386 on the diy sustainer then?..

Hard for me to say (cos I don't know your exact circuit situation), but for any sustainer circuit essentially what you are doing is that same as outlined in my last post - you're essentially T'ing off the guitar signal to a sustainer circuit, which has a buffer at the front so as not to colour the original guitar signal. An LM386's input impedance is too low for a typical guitar pickup - it needs a buffer in front of it ...else your tone-age will be full of suckage.

my concerns now will probably be any possible interference from the LED's being in close proximity to the (unpotted) pickups and routing everything. if tone-shaping transparency isn't an issue anymore, i can build the laziest POS buffer imaginable. lol

If your pups are single coil & your LEDS are too close, then yes, you'll get interference. Humbuckers are more foregiving.

Re the circuit you intend using....I think 4 LEDs in series as shown might be a stretch (possibly *just get away with it with RED leds, but it won't driver blue, green, white LEDs etc)

[MrStab]

If your pups are single coil & your LEDS are too close, then yes, you'll get interference. Humbuckers are more foregiving.

Re the circuit you intend using....I think 4 LEDs in series as shown might be a stretch (possibly *just get away with it with RED leds, but it won't driver blue, green, white LEDs etc)

i'm a humbucker guy so i have that on my side. duly noted about the quantity, i'm guessing the only solution would be a trade-off between resistance and brightness? prolly not much room for reflections in the pickup slots, but maybe i can find way to spread out the light of a lesser amount of LEDs

[deadastronaut]

@G: cheers man, i'll try that when i blow the dust off it.. .

[MrStab]

lack of current resistor(s) and unrealistic amount of series LEDs aside, anyone see any fatal flaws in the layout i posted? ( http://i.imgur.com/9xLOL3y.jpg ). i just built it and am trying to establish whether it's the layout or the build that's wrong. not sure if the original voltages from either "half" of the circuit would apply here, if the whole thing's merged. would be cool to have this whole thing on one piece of strip/vero.

it's hooked up to a 1/4" plug to test post-guitar, atm, although i doubt that'd prevent it from working at all. i did overheat an LM386, but shoved a 220uF output cap in there and replaced it, and it seems okay now, so that doesn't seem to be it. just doesnt do anything! lol

[Gurner]

This is one of those circuits that's a kludge at best!

Ok, with no AC signal present, the LM386 is going to have half the power supply on its output pin. Therefore for 9V there's going to be 4.5V sitting on the output pin.

From what I can gather, the trick with this circuit, will be to make sure you use just enough LEDs so that they are just visibly'off' with no AC signal present....and that's where the problem starts creeping in.

RED LED fwd voltage is about 2V .... so for the 4.5V present on the LM386 output pin two LEDs would be too little, three would be too many

Blue LED fwd voltage is about 3.3V  .....1 LED would be to little, two would be too many.

What I'd be inclined to do (just as a starter & experiment), is remove the output cap, & connect two blue leds (or any other led with a fwd voltage greater than 2.5V) ....if you've connected everything right, they should be off. whack your guitar & they should light up.

What I'd probably do first though is connect a speaker *with* the output cap in situ, just to make sure you han hear someting on the output & to provie the circuit is working.