Help with Op Amps... need help from the smart people :)

[Laguna]

Hi Guys, I've been trying to do a distortion... but long story short, i couldn't get the amplification to work.

Right now, i have this circuit that's basically my best copy of a non-inverting amplifier. The most basic one.



So, i have a RC4559D (or was it P?). I searched and found the pinout. For what i was able to read, this is completely standard on dual op amps:



So, i built it:





Of course... it didn't work.

Just to explain:
1) On the left is the input. The yellow cable goes to pin 3. (In+)
2) The blue cable is +. Connected to pin 8. (Vcc+)
3) Pin 4 is connected to ground (the other yellow cable) (Vcc-)
4) Pin 2 (in-) is connected to ground through a 1k resistor
5) Pin 2 (in-) is also connected to pin 1 (out) through a 100kB pot. (red cable goes left, black cable goes back). I only connected the 2 right-most pins of the pot.
6) The input is on the left, the 2 black cables. Ground to -, input to to pin 3 through the yellow cable
7) The output is on the right, the 2 black cables. Ground to -, output through the red cable ==> then the black cable to pin 1.

Now.... It doesn't work...

I tried another battery too

Now, to measure. I forgot the exact measurements BUT between pins 2 and 3 i get like 18mV but on pin 1 i get 8.5V () is this normal??

You can't see it on the pictures, but the Ic has the dot on the bottom left. (not in the middle left as the picture says).

Is it because i connected ground to -Vcc? i have a battery, it can't do -9v!

Oh, i also tried adding a 68k res between input and ground. I found it online. It didn't do anything that i could see.

This is for people smarter than me, i know
I really appreciate the help. It's not that i'm trying to make a specific pedal right now, i'm just trying to learn.

Thank you!

[antonis]

Although I can't preclude myself to "smart" people, let's try to find out what's the problem(s) with your DC amp..!! 

Pin 2 (inverting input) seems to be happy as it is (although it should be happier with a capacitor in series with R2 & GND, to roll-off DC gain to unity - cap is considered open circuit to DC so no DC can flow through R2 to GND, hence no DC amplification..)

Now, pin 3 (non-inverting input) needs a voltage to refer on (any voltage within power supply margin is wellcome..)

So, you could just place a resistor from pin 3 to GND and bias it to 0 Volts..!!
(beware of the wink icon 'cause is kind of sneaky..)

If you try the above zero bias, you'll result in asymmetrical distortion(*) (only positive signal waveform will be amplified 'cause single supply op-amp cant go lower than zero volts..)
(*) It could be interesting, depending on your personal taste of sound.. 

If you want a more "faithful" amplification, you have to form a resistive voltage divider, between Vcc & GND, of equal value resistors and connect pin 3 to their junction..
(you now bias pin 3 to half the power supply so your output can equally swing above and below it..)

Of course, you need an input and an output capacitor to "isolate" any DC coming to or leaving from op-amp..


P.S.1
In & Out Cap values together with voltage divider resistor values will be discused in a following post after a short reading about High & Low pass filters and that %&^%#@#  thing called Impedance..!!! 

P.S.2
About your measurements (in brief): pin 1 (OUT) is hitting the positive rail and pin2 <-> pin3 show you amp's input offset voltage..

[Laguna]

Antonis, thank you very much for the reply!

I did try a 22k res between input and ground and it didn't change anything. (and input is connected to pin 3, so this is what you were saying, right? or am I missing something?)

"About your measurements (in brief): pin 1 (OUT) is hitting the positive rail and pin2 <-> pin3 show you amp's input offset voltage.." ==> errrrr, sorry, what does that mean? how do i fix it? hahahaha

I understand what you're saying about the asymmetrical distortion. I didn't know it was going to do that. After the basic one works (at least gives me any sound :S), I'll ask the smart people how to do the biasing correctly. For now, I just want to do the basic and get some sound so I can understand what did I do wrong and learn from it

[antonis]

First of all, check your breadboard wiring..

Then form a divider with 2 x 10k to 22k resistors and connect their junction (node) to pin 3 via a 100k to 470k resistor..
(a 10μF to 22μF cap from their junction to GND shouldn't be considered as lap of luxury..)

Place a 100nF to 22nF cap in series with input BEFORE the Voltage reference resistor & pin 3 junction..

Place a 10μF cap between R2 & Ground (negative cap leg to GND) or you'll also amplify any DC present by the amplification factor of 1 + R1/R2..
(for a max gain set of 101, you only need 84mV DC to make OUT hit the possitive rail..)


Forget anything about op-amp "nasty" habits (like input voltage & current offsets, output hitting rails, etc) 

[GibsonGM]

Take a look at "Single supply non-inverting amplifier" Laguna...the pic shows what Antonis is explaining.   
R3 and R4 in the pic are setting the bias voltage (using 2 10K resistors here is fine).   You see the cap to ground as "C3", to block amplification of the DC as he also said.

Look up the schematic for something like Tube Screamer or Big Muff Pi, and you'll see that generally the bias network is done as a subcircuit with an output called "Vref", or reference voltage - but sometimes it IS shown as these 2 resistors at the input, behind the input cap (which blocks the DC used to operate the opamp or other active device from other portions of the circuit but allows your AC guitar signal to pass).   More often with transistor circuits, in my experience.

https://www.electronics-notes.com/articles/analogue_circuits/operational-amplifier-op-amp/non-inverting-amplifier.php

[Laguna]

Guys, this is GREAT!

I'll do it tonight and hopefully i can get some output hahahaha.
I wanted a big gain because the whole idea was to clip to ground for distortion and then do a voltage divider for the volume. But that's when it works!

Another thing, any suggestions on the C1, C2 and C3 caps that i'm adding?
The video says to calculate C2 with the load, but that's the Pot, so it varies. I can try to play with the values.

Thank you!!!!!!!!
I'll tell you how it went tomorrow

[GibsonGM]

Using the "Single supply non-inverting amplifier" from the link I posted, Laguna, you could try:

.1u or .22u for C1, 1u for C2, 10u for C3.  Common values.   

These are values to mess around with!!!   C1 will form a high-pass filter with the opamp's input impedance, so other values will do other things..bigger cap here lets more bass in.

C2 is the output cap...another chance to trim bass (works with vol. pot, but also the next stage of amplification...an effect, or your amp!  Vol pot looks small compared to a following high input impedance...).

Same for C3 (look at the feedback loop of a Tube Screamer to see how they use this to trim unwanted frequencies using the NEGATIVE feedback  properties of the opamp).

So mess around with these values!  Tube screamer has .047u in C3!!

[Laguna]

Guys!!! It worked!!!!!!!!! I'm so happy!!!

Pics (Before you say anything, please ignore that one of the diodes is connected to the positive rail, i fixed it but i didn't take the pictures again):





After the gain i added 2 diodes for clipping and a voltage divider for volume.

These are 1N4001 diodes because they are the only ones i had available.
I saw that 101x wasn't enough to clip a lot with those diodes, so i changed the res to 330. It did the trick, although there's a lot of noise.

I bought  1N4148 and 1N914 but coming on friday. I don't know exactly what would be the difference for the clipping TBH, but I'll try them. I'm hoping they'll clip to less voltage so i don't have to amplify that much and reduce the noise, but I'm not an expert ^_^

Thank you guys! this was extremely fun and i did learn a lot.

[GibsonGM]

Great to hear, Laguna !     Much of the noise is probably due to being on breadboard and not inside a shielded enclosure....lots of it goes away when you 'house' it.  There are methods to tame a lot of it.  But some will always be present in a high-gain circuit, as distortions are. 

1N4001 diodes are fine...the 4148/914's (same part actually) are small signal vs. power diode like the 1N4XX series.   If you want to clip sooner, try some germanium ones!   1N34A, 1N60....they clip at like .35V vs. .5 to .75 for 914's.   That is the "Vf", or forward voltage of the diode (the voltage it turns on at).   Germanium clips sooner. 

[antonis]

I'm hoping they'll clip to less voltage so i don't have to amplify that much and reduce the noise

As Sir Mike already told you, "noise" can be reduced in by several tricks..

One of them is, of course, less amplification (gain) - but.. 
(always exists a "but" - sometimes more than one..) 

Clipping voltage isn't the major term of distortion - which is the (gain depended) waveform curve slope..
(the higher the gain  AND the lesser the clipping threashold  => the "harder" the clipped waveform - almost square wave..)

[Laguna]

sounds good, i'll try germanium transistors as i'm buying a couple more things now in smallbear.

Another thing i noticed, it can behave like a distortion or like a fuzz depending on the capacitor values. (I'm sure you knew, but I didn't ^_^)