Hey, unfortunately I don't have a scope at hands.
The changes are: 47p instead of 100p, the two resistors and caps in the 
feedback path going to gnd are basically just adjusted to the pot 
change, so same freq. cuts here.(no more electrolytic caps) And I also 
change the 1nF right in front of the opamp with a 3.3nF one to prevent 
the oscillation that was going on there.(figured that out in one of my 
earlier builds, that are quite the same and just work fine)

show your real and layout schematic.

what pins did you use on the 4558?

Don't really have a schmematic. But as the "same" circuit worked before 
and I didn't spot any easy mistakes I guess it must be something that 
can not be seen by looking at my build. Maybe someone has a idea what 
could cause such problems and then I can focus on that part. But of 
course I can give you an updated schematic and a picture of my real 
build, as I maybe have overlooked some easy mistakes. The green wire is 
the input, the top yellow one is Gain 1, the other Gain2&3. Red +9V, 
Black GND. I triple checked all the wires going to the pots and else, 
that shouldn't be the mistake.

PS: I disconnected the zener, but it's still the same.

2-3-6, is that the pinout you used with the 4558?

Oh sorry, forgot to change that in the schematic. Of course 6 becomes 1 
here.

Also realized the volume pot on my guitar creates some cracking and 
noise by turning it up an down, sometimes even the LEDs light up.

what about pin7?

where is your virtual ground comming from?

only solution i see to fix this build:
re-draw a complete new schematic by hand from your actual build.
then check what's wrong.

V+ going to pin 8, GND to pin 4. Here's a layout of the power supply. 
+4.5V comes in trough a 1M resistor after the first 22n cap. Got about 
9V at pin 8 and 2.x at pin 3. That seems to be ok. But is it normal to 
have 5V at pin 2? It must be caused by the opamp as I have 0V there when 
it's not in the socket. Pin 5 also has a connection to virtual ground 
and 6 and 7 are connected so the second opamp acts like a buffer.

There are 5V at input pin :o . I guess that could be the problem, but 
also only when the IC is not in its socket.

i say it again:
re-draw a complete new schematic by hand from your actual build.
then check what's wrong.

i'm out until then, it's a waste of time otherwise.

Sorry 5v at input when IC IS in its socket, otherwise 0V

R4 versus your pot allows an amplification of up to 250000. I doubt that 
your OP-amp can follow that.

Regarding the "bassy" sound:

The 47nF capacitor has 340Ω at 10kHz and 34000Ω at 100Hz. So this 
capacitor weakens the high frequencies. If your pot is at 0Ω, the 
capacitor has no effect because it is completely bypassed by the pot. 
But on the other hand when your pot is 250kΩ, then only a very small 
part of the signal hoes through the pot, most part of the signal goes 
through the capacitor so its frequency-dependant resistant become 
noticable.

> up to 250000

Sorry, too many zeroes. It is up to 2500, this should be supported by 
any OP-amp.

Lothar M. wrote:
> With open input? Without a guitar connected?
> Then you have some wiring problems or a defective C1. At least...

Yes without a guitar connected.

Lothar M. wrote:
> As long as the 4558 is working as it should, you have the very same
> voltage at pin 2 and pin 3. Thats the job of the 4558: to keep the
> difference between them both zero.

Pin 3 has the 2.5V through bias, Pin 2 still gets 5V but I don't know 
where it comes from, must be from inside the IC cause it's only there 
when it's connected. Already changed the IC but same result.
Another weird thing: when i measure voltage between gnd and pin 3 the 
gain pot works like a dimmer for the LEDs in the upper range. They're 
brightest when it's maxed.

stefanus wrote:
> Regarding the "bassy" sound:
>
> The 47nF capacitor has 340Ω at 10kHz and 34000Ω at 100Hz. So this
> capacitor weakens the high frequencies. If your pot is at 0Ω, the
> capacitor has no effect because it is completely bypassed by the pot.
> But on the other hand when your pot is 250kΩ, then only a very small
> part of the signal hoes through the pot, most part of the signal goes
> through the capacitor so its frequency-dependant resistant become
> noticable.

It's a 47pf one, it's only there to cut off the useless very high freq. 
and prevent the circuit from oscillation.

Luca wrote:
> Pin 3 has the 2.5V through bias, Pin 2 still gets 5V but I don't know
> where it comes from

As long as the Opamp doesn't show the voltage of the VCC/2 rail on all 
of its  inputs and outputs without signal, the opamp is not working 
correctly. Exchange it with a known good working part. Before inserting 
the new chip make sure theres VCC/2 on the non-inverting input. If thats 
the case, a good opamp will then show this voltage on all its inputs and 
outputs.

matzetronics wrote:
> As long as the Opamp doesn't show the voltage of the VCC/2 rail on all
> of its  inputs and outputs without signal, the opamp is not working
> correctly. Exchange it with a known good working part. Before inserting
> the new chip make sure theres VCC/2 on the non-inverting input. If thats
> the case, a good opamp will then show this voltage on all its inputs and
> outputs.

I put in another IC of one of my working builds, still the same.

I don't get it, the only thing connected to the input that will let DC 
pass is the 1M resistor to ground

Found it!!!!

A extreme tiny piece of a strand connected the In rail and the one 
beneath it(which connects between the 2k7 resistor and the 1u cap).
A massive thanks to all of you for helping me find the problem. So I 
knew where to look at. You guys are great!