Operational Amplifier Not Working as Expected

[Smauel]

EDIT: sorry for the wall of text, I explain the issue more succinctly two posts down

Hello
I'm a relative beginner to working with electronics, but am very enthusiastic to learn. I also recently picked up the bass guitar. I figured it would be a fun project to make a distortion pedal consisting of an amplifier and some diodes to clip the signal of the instrument. After googling other peoples designs to see how they've done it, I found some common designs and copied them, but after plugging everything into a breadboard, it didn't work properly (It sort of worked for a little bit, among various horrible squealing noises, and getting no signal at all, but upon attempting to troubleshoot the issues I only managed to make it stop working entirely). I took it all apart and started again, less ambitiously with the goal to create a basic amplification circuit using an op-amp.
Unfortunately, I managed to cook two opamps and have nearly done the same to my last one another two times. I am currently using an NE5532P.
As a power supply I am using an old wall plug fixed power supply rated at 9V and 510mA that is from some unidentified old device (Perhaps notable is that one day the power supply was supplying 14V and the next it was supplying 15V. I have no clue what that could mean, especially considering its only supposed to be 9V).
I have a voltage divider using two 10kOhm resistors. Ground goes to the negative supply of the opamp and the high voltage goes to the positive supply of the opamp.
The divided voltage goes to non-inverting input via a 1MegOhm resistor, and the input audio signal from the instrument is also plugged in there.
The noninverting input is connected via a 1kOhm resistor to ground, and a 100kOhm gain controlling potentiometer to the output.
However, the moment I plug in the power, the opamp draws way too much current and gets extremely hot, no matter what the gain resistor is set to. I put a 470ohm resistor in series with the power supply to limit the current draw to allow me to probe the circuit with my multimeter before anything is destroyed. I found that there is an 80mV potential difference between the inverting and noninverting inputs. I don't know whether thats because im starving the opamp for power by limiting the current or another issue with my circuit.
This should be a simple circuit, but I just can't figure out whats wrong with it.
I know this is a terribly vague question, but does anyone have any troubleshooting tips, or know what might be causing the particular issues I'm running into?

[WattsThat]

Lesson one: Using words to describe a circuit is an exercise in futility.

Draw a schematic and post. Until then, there is very little to discuss.

[Smauel]

I'm sorry, I really don't know what I was thinking, that was rather diabolical of me.

To reiterate the important facts more succinctly: I need to amplify a signal using an opamp.
I'm using an NE5532P opamp, and an old 9V power supply thats outputting a lot more than 9V for some reason.
Upon powering the circuit, the opamp immediately heats up, regardless of what the potentiometer is set to, and if I didn't unplug it, it would fry the opamp.
Additionally, while powered, (if I limit the current from the supply with a 470 ohm resistor, which also promptly heats up to be uncomfortable to touch), there is an 80mV difference between the inverting and noninverting inputs, even when the potentiometer is set to minimum resistance.

[xavier60]

Very unusual to have problems like that with opamps.
Wrong pinout maybe? The  NE5532P is a dual opamp

[Etesla]

Here's some answers to your questions.

The 9V power supply is likely providing more than 9V because it is unregulated. It should output around 9V when it is loaded at it's rated current.

The reason your circuit doesn't work is because it is not the correct circuit. The circuit that I think you are going for looks like the one I attached. You basically had the input going to the wrong spot.

It's easy to see why your version of a single supply opAmp circuit doesn't work.
With no input signal, you would expect the voltage on the + input to be at 4.5V if the opAmp was working as expected.
This means that the output would have to make a voltage that changes the - input to 4.5V through the voltage divider made of your potentiometer and 1k resistor. Say that the potentiometer is set to 10 k ohms. This means that the output would have to make a voltage of (10+1)/1*4.5 = 49.5V to get the - input to be 4.5V. That is obviously impossible, since the opAmp's output can really only go up to maybe 8 volts with a 9V supply. So what happens now? Well, the opAmp goes into saturation and does weird things. The inputs are no longer high impedance, the output voltage will be 'railed', and you may blow up your opAmp depending on what you have hooked up to it.

Try the circuit I attached and let us know how it works out.

[Smauel]

Thanks, that explanation makes a lot of sense. Unfortunately, it didn't solve the problem.
I built the circuit. I got an extremely noisy signal out before plugging in the input or power. Plugging in the power and input, the input signal was not amplified (the same noisy signal was output), and the opamp started overheating immediately again.
Would the circuit I attached be a better way to power the opamp, by avoiding having to bias the input? Would that cause any problems for a 9v wall plug power supply or is that a fine solution?
Are there any other reasons this could be happening? It's also strange that even when the potentiometer is set to zero, on both circuits the opamp overheated even though I would think the gain should be at its minimum, though I may be mistaken.
I'm still very confused about whats going on, and I really want to figure this out now.

[Etesla]

I do not think it would be smart to try to do a split power supply in that way. Uneven current draw by the opAmp from the negative or positive supply rail will cause issues, and the maximum current out is then limited by the 10k resistors. If you are familiar with the 555 timer, a better way to get a negative supply rail would be to throw together a quick inverting charge pump circuit, but the whole point of the circuit I attached in my previous post is to work with a single supply. Would you mind taking a picture of your breadboard so that we can troubleshoot possible wiring errors?

[John B]

The op amp should not be able to heat itself up catastrophically unless you had large +/- power rails and were sourcing or sinking significant current on the outputs. Since the output is AC coupled, the current may be high frequency AC content. Do you have the means to check the output signal and see if it's oscillating?

Also, where did you source the op amps from? These wouldn't be ali express/ebay ones? I would be assuming that they have internally shorted.

[Fenton Bresler]

OP, your original circuit should work with the simple addition of a capacitor between the 1K resistor and ground.

[Vovk_Z]

There must be a 1-100 uF electrolyte or 0.1 uF ceramic cap across the opamp power lines (somewere close to pins). But if opamp is bought somewere in Aliexpress then it can be just broken.

[Smauel]

I took a picture of the breadboard and highlighted where everything is going. I can't see any mistakes myself, but do take a look at it and see if you can catch something I missed. I don't know if the values of the capacitors aren't high enough, but I wouldn't have thought that would make the difference.

If you mean an oscilloscope, I don't have means to check for oscillation. The only piece of test equipment I have is a relatively (but not dirt) cheap multimeter.

As for the opamps being cheap Aliexpress/ebay cheap opamps... Unfortunately I can't be sure; I got them from "guitarpedalparts.com" but I don't know where that website sources their opamps from. Though, if they have internally shorted, I think that would be measurable by a multimeter? All three opamps, when I first started using them had several megaohms resistance from Vcc+ to Vcc-. Is that not high enough? The two opamps that I killed now have 40 Ohms and 80 Ohms, while the NE5532, which I've been more careful with, has 3 or 4 megaohms resistance across the power rails. Also, where is a good place to get good deals on reliable parts like opamps?
I do have a pair of old broken speakers that I took apart, which have two opamps inside. I'm not sure if they're working, (because I have no idea what caused the speakers to stop working, or even when they were last working), but might it be worth desoldering the opamps from the board and trying to use those instead? It's got a BA4560 and a TDA2007A, the latter of which is attached to a fairly beefy heatsink.

I tried adding a 10uF electrolytic capacitor across the power lines near the opamp pins but that had no effect on the overheating problem.

[WattsThat]

The power supply connections are backwards. The center indented pin is the positive, the two outer straight, flat connections are the negative and switched contact.

[magic]

This one doesn't look like typical fakes. The dot is similar to legit parts from Texas Instruments. Markings on TI chips are laser engraved, rather thin and hardly visible at certain angles.

That thing is probably oscillating. Add 100nF ceramic or film capacitor between V+ and V-, shorten the feedback path somehow.

Also, I'm not sure if the pot is even connected correctly. The wires go to its ends, but one of them should go to the central pin unless the cental pin is shorted to one of the ends.

[Smauel]

WattsThat, it appears you are correct; I googled how to connect that power connector several times and each time found that what I had was correct, despite what my intuition was telling me, so I trusted it, but it was wrong all along. Thats an embarassing reason for it not to work.

[WattsThat]

Time to invest in a multimeter. Even a $5 harbor freight DMM is better than nothing for sussing out issues like that connector, there is no substitute for knowing what a voltage or a resistance actually is.

A no name inexpensive meter is okay for low voltage work (<50 volts) and resistance checking but don’t use them on mains voltage, they don’t meet any safety standards. If you can afford better, by all means buy a name brand meter, even Fluke has some inexpensive models, they’re an investment like any good tool.

I bought my first meter some fifty years ago, a Triplett 630. Still have it. Still works. Bought it from a neighbor that was an EE with GE, cost me $20 which was about 5 lawns I had to mow. I remember the cost because it seemed like a lot of money at the time but it helped put me on my life's course.

[Vovk_Z]

WattsThat, it appears you are correct; I googled how to connect that power connector several times

- there is no need to google that, just check with multimeter where is "+" and "-".

[StillTrying]

[SilverSolder]

WattsThat, it appears you are correct; I googled how to connect that power connector several times and each time found that what I had was correct, despite what my intuition was telling me, so I trusted it, but it was wrong all along. Thats an embarassing reason for it not to work.

It is intelligent to learn from your mistakes.  It is ingenious to learn from other people's mistakes!