I don't think I understand op-amp powering/negative voltage

[petirkelieby]

Hi,
noob problem here: I have made a few 9v builds with transistors and one with a single op amp (LM386) and had no issues with powering or my signal, but I have a couple builds with dual op amps (TL072, LM358P) that I have not managed to get a signal through.

For example, one pedal i've been trying to build on my breadboard so I can tweak it and begin to understand what does what is this bitcrusher (images included), but I've not managed to get a signal through although I though I'd placed all components in the right place. I think the problem lies in me getting confused about where ground is etc. for the circuit and what negative voltage is and how that might affect where ground is. I've watched a few long, in-depth videos on op-amp on youtube and grasp most of it but still not sure...

Like, is it implicit I connect pins 4 and 8 to V+ and V-?
Should I ground my outputs to common ground or to V-?
How would I include an LED?
What mathematical equations should I know?
Probably quite a dumb question but I'm still learning! Any help on this would be greatly appreciated.

[pokus]

Hey welcome to the forum!

With opamps it's important to know that they are biased with typically half of the supply voltage.
So everywhere Vb coming to your circuit, it should be connected to the Vb point of your power supply circuit (where the voltage divider out of R10 and R11, and C5 meet)

On the breadboard your ground is v-

[pokus]

Looking at your picture of the breadboard, it seems like the opamps aren't powered at all. Like you mentioned pin 4 to v-, pin 8 to v+. The other connections on the left opamp should be ok as far as I can see. What is that green wire doing going nowhere?

Also I can't see any source for V+

[Fancy Lime]

Welcome petirkelieby,

and let me start with stating that some kind of first name to address each other instead of the nickname by is always appreciated here. No pressure, though.

Let me go on with saying: There is no ground, young grasshopper.

Ok, that does not help you at all, I assume. Let me explain: What we usually call "ground" is actually just a reference voltage that is stable and does not change as the signal changes. At least it should be stable in almost all commonly used circuits for them to work as expected. This reference voltage CAN be equal to the actual ground potential but it does not have to. For anything connected to 110V or 230V mains power, it is often a good idea to make "real ground" the reference voltage. For anything you can power with a battery, there is no connection to "real ground" and therefore we have to make ourselves a "virtual ground", aka the reference voltage, which we, for simplicity's sake and because we're a lazy bunch, just call "ground". We're not doing this to confuse newbies, I promise.

Now, if you use a wall wart power supply and want to feed more than one circuit with it, you need some kind of reference voltage between them for everything to work. Since we only have to wires connecting the effects (tip and sleeve of the 1/4" connector) and for connecting the wall wart to the effects, the common reference needs to be either the positive or negative supply voltage. Nowadays it is overwhelmingly the negative one but with older effects, especially 60's fuzzes, it can be positive. To make matters worse, opamps and many other bits of circuitry need an additional reference voltage between the two supply rails. In guitar effects and the like, this is usually considered "V 1/2" or (in case of the most common +9V supply and negative=ground) +4.5V. But in devices with a so called symmetrical supply (typically studio rack units and other "professional" audio equipment), this "middle reference voltage" is actually the common reference of all devices, which then need to be connected with a 3-wire cable (XLR connectors or the like). The power supply for these devices also have three wires: Positive supply, negative supply and ground. This particular instance of ground can again be "real ground" and connected to the mains ground or it can be a "virtual ground" (meaning: just another reference voltage). This is also why such devices often feature a "ground lift" switch, which means you can disconnect some of the ground connections to avoid hum problems with AC voltages between the "different grounds". Confused yet?

In the schematic you posted, positive supply is marked "+9V", which implicitly means that the ground symbol (the three horizontal bars forming a triangle pointing down) is connected to the negative wire of the power supply. the resistors R10 and R11 form a voltage divider, which provides the virtual ground, which is called "Vb" in this schematic but goes by many different names in general. You can calculate the Voltage of Vb as:

+9V * R10/(R10+R11) = +4.5V

Capacitor C5 is there to stabilize this reference voltage against AC influences from anything that is connected to the Vb reference point. Opamps X1, X3 and X4 are all referenced to this 4.5V point Vb. X2 is not (although it may look like it at first) because there is a capacitor (C2) in between but that does not matter because X2 can take its reference voltage (aka "bias") from the output of X1, which also sits at 4.5V because that opamp is correctly biased to Vb. C2 is there to form a "sample and hold" circuit together with transistor Q1.

Alright, so much for general theory about what "ground" even means. Or doesn't. Now to your questions:

Like, is it implicit I connect pins 4 and 8 to V+ and V-?

Yes, the power pins of any opamp (or rather opamp package, which may contain 2 or 4 actual opamps) need to be correctly connected to the positive and negative supply. This is often not explicitly printed in schematics and is also not shown in the schematic you posted. Still have to do it, though. See the manufacturers datasheets when in doubt, which pin goes where. Or google "xxx pinout", where xxx stands for the opamps name.

Should I ground my outputs to common ground or to V-?

The schematic tells you that the input and output referencing resistors R1 and R3 go to "common ground", which is identical the negative supply here. They are not connected to "virtual ground" Vb, which you seem to think is called "common ground". These are two different things and not knowing that has caused me a lot of headaches when I started. The "arrow down" symbol with the three lines is always common ground. It can at the same time be V- (as in your case) or the reference voltage at half supply (not the case here but seen in many schematics, especially reference implementations from manufacturers datasheets and in many electrical engineering books, where they do things properly and not as haphazardly as we in the guitar effect tinkering crowd).

How would I include an LED?

A million different ways. Depends on what you want the LED to do.

What mathematical equations should I know?

All of them. Sorry, could not resist. There is no fixed set of equations to learn. You'll automatically learn the ones you need along the way. That's what this forum is for.

Hope that helps a bit.

Cheers,
Andy

[GibsonGM]

Just a note: LM386 isn't an opamp, it's a dedicated audio amplifer, and can't be used in place of an opamp in anything!   

Welcome to the forum, you're in good hands!

[ElectricDruid]

Sounds to me like you're getting confused between "bipolar power supplies" and "single supplies".

Most op-amp circuits (outside of pedal-world) use bipolar supplies, where there is both a positive and negative voltage with a ground in the middle (so three power wires). You'll see stuff like +/-12V or +/-15V commonly. If you've been watching "how to" op-amp videos, many of them assume a bipolar supply (as do most websites and textbooks) without really telling you that's what they're doing.

In pedal land, we use a simple single supply, with ground/0V and +9V (or maybe +12V if we're lucky). No negative power to be seen. So how do we power an op-amp? This is where what Andy said about "there is no ground" comes in. In fact, "all voltages are relative, man". If we use a voltage divider made of a couple of resistors to give us a midpoint 4.5V voltage level, we now have 0V, 4.5V and 9V. But we can equally well call it -4.5V, 0V, and +4.5V. And there's our bipolar supply for our op-amp!!

That's the basics of it. We make a midpoint voltage, and then we use it as if it were ground (a "virtual ground" in the lingo) and the other two power wires become the negative and positive supply.

[GibsonGM]

We can play with AC signals...we can 'give them a DC offset', which is what Tom is describing.   Without the 2 power supplies, you can't cross zero and amplify (or whatever you're doing) an AC signal; you'd rectify your signal - half would be cut off.   We are usually stuck with our 9V battery, and ground.    If you 'jack up' the sine wave to 1/2  your supply voltage, 4.5V here, then you CAN work on the 'lower part' of the wave...and because capacitors block DC voltage, at the output, the AC signal goes back to having its normal positive and negative parts - the DC is blocked from going to the next stage!! 

[petirkelieby]

Thank you all for your responses, it has definitely been helpful and I am making some headway! I'll no doubt post any problems I run into with an updated pic of the breadboard should I need to.

Looking at your picture of the breadboard, it seems like the opamps aren't powered at all. Like you mentioned pin 4 to v-, pin 8 to v+. The other connections on the left opamp should be ok as far as I can see. What is that green wire doing going nowhere?

Also I can't see any source for V+

V+ is at the top of the board, I had 3 different circuits in different boards aligned next to each other; the green cable going to nowhere is the input cable, going to one of those different circuits.

[merlinb]

is it implicit I connect pins 4 and 8 to V+ and V-?

Yes!

[antonis]

Next post should deal with "why my J201 doesn't sample and/or hold..??" 

[petirkelieby]

Next post should deal with "why my J201 doesn't sample and/or hold..??" 

Ahaha can you see my problems unfolding before me?

I've got a 'working' circuit now, but the output is barely audible and it's self oscillating like crazy (i switched one of the 100k resistors to a 100k pot), have i done anything obviously wrong? It won't let me include a video of it but here is a current picture... thanks again everybody!

[antonis]

Can't estimate if it oscillates due to R5/R7 or R9..

[petirkelieby]

Can't estimate if it oscillates due to R5/R7 or R9..

R7 is the resistor i changed to the 100k pot; it does effect the oscillation along with the 1m pot but only oscillates in certain places