No matter what I do my cmoy amp won't work proprorly

I've made several cmoy amps on a breadboard without any problems at all, but it refuses to work when I solder it to a prototyping board. The amp as it is now has a constant hissing noise with difference electrical sounds seemingly coming and going at random, but the sounds do change in intensity if I touch or move the amp. The constant hissing noise increases and decreases in intensity if I move my hand closer or further away from the amp (I didn't think that was possible). When it comes to the actual music I'm trying to amplify this also varies alot, something not beeing there at all or it can be heard quite well (allthough the sound quality is pretty bad), but it is usually very weak in the background of all the other noises.

I've made two completely independant amps with about the same problems in both of them. I got a new temperature controlled soldering iron yesterday and with a seperate container for flux, resin cored solder and a wet sponge my solderings are pretty good so there shouldn't be any bad connections. I have also gone over the solderings several times to make sure everything is as it should be. The measures I have taken is to completely bypass the input cap, remove the volume pot and changed the batteries (running it off two 9V batteries) with no differences.

This is the schematic I have followed. The power supply caps are 330uF (not 220uF) and the amps I have tried are Op275 and NE5532R (same result).

https://tangentsoft.net/audio/cmoy-tutorial/misc/cmoy-tangent-sch.pdf

It looks like the amplifier(s) are oscillating. It could be at a very high frequency. Have you kept the wire running to the input pin well away from the output pin?, Running the wire to the volume control in screened wire might help. You could try to reduce the high frequency gain by putting a capacitor of 470pF across R4.
Frank

I removed the input all together, I figured I needed the varying buzzing noise to be gone before I could think about getting it to amplify an input. It allmost seemed like putting a 470pF disc in parallel with R4 lowered the buzzing abit, but it is still there and touching sertain components will magnify it. I also see now that touching ground makes the amp completely silent. I don't just want to build this amp, I also wan't to understand why these problems occur. I am also building on an amp using LM1875's which also seem to have some of the same problems but if I touch ground on the 1875 amp, the buzzing increases in intensity. I've been told here that this might be a "grounding issue" or a loop ground, but all the connections to ground goes to the exact same place and there are not several paths to ground (this accounts for both the cmoy and the 1875). How can building these amps on a breadboard go without any problems what so ever, but as soon as I am aiming for a decent and permanent designs with propper connections, it fails completely?

I want to open this thread again. A lot has happened since I last made this thread. I've taken a couple of classes at uni and I've got a better overall understanding now. I've mastered the circuit I linked in the opening post, but the problem is now with my new bass-boost function. Here's my circuit:

https://www.circuitlab.com/circuit/8ad2cs/cmoy/

I can get it to work, but the problem occurs when I wan't higher gain. I've been selling these and someone with a headset with an impedance of 300Ohm wanted one so I figured I needed a pretty high gain. Because of R3 and R4 causes a voltage divider I only get half the input voltage as I otherwise would have. The thing is that I can't increase R4! The original schematic gives a value of 100k, but my amp refuses to work with that value. The odd thing is that I can't see whats going on. The bass gets all bad (sounds like its oscillating perhaps) and this is constant regardless of output power. The strange thing is that it plays a low frequency sine wave just fine though so I can't see anything when scoping it. I've also tried to scope it when there's nothing connected to the input and I see no oscillations. If I increase the gain by increasing R5 or decreasing R6 I would have to further lower the value R4 to get it to work again thus making a higher gain useless There's also an issue with my virtual ground shifting off zero when an output is connected, the higher the value of R4, the more it shifts. The opposite is true for the output impedance, the higher the impedance of the headset, the higher the value of R4 can be. This should work out if its only used with a 300Ohm headset, but I wan't the possibility to connect other headsets as well.

What causes this and what can I do to fix it?

C3 is too low, try 1MFD. Both R4 and the R6 (1K) should roughly equal in value to balancer out the input currents,try X10 the 1k and the 10K feedback resistor it might just work. Even better make the - input the "live" one, Then instead of having the tone controlling cap at the input put one 1/10 of the value across the feed abck resistor (now 100K) - you get a much bigger bang for your buck that way!
Frank

Thanks for the reply I have now gone over to a single rail supply with an output cap instead. I was previously afraid of the distortion caused by such an electrolytic cap on the output, but I guess it's nothing to worry about, this isn't exactly high-end equipment There was several reason I put the bass-boost function at the input instead of in the feedback path and all of them was because of ignorance The way I did it I would had to have a make before break switch to avoid a loud pop when the bass-boost was switched on. I was also unsure of how to calculate the break frequencies in the feedback path and by putting them at the input instead I mistakenly thought I could connect the C4's of both channels together and have a single wire connecting them to ground (thus removing the need for a 2P switch). I see now that all of this was silly and I'm now listening to a fully functioning amp with the bass-boost in the feedback path and no virtual ground.

That doesn't help that fact that I etched 10x of the previous circuit that was linked And I still wan't to understand what's going on. Does this has to do with offset voltage and input currents? I can see that C3 is too low though, it has a break frequency of like 140hz which is unacceptable I still don't understand what's happening to the circuit though, why isn't it working? Is it the amp outputting a constant DC that's throwing the ground off it's center thus making it more prone to clipping? That doesn't explain how the sound can be equally bad at all levels though.

the op amps output voltage should be 1/2 way between the Vcc + and -. This means that the + and - inputs should be very accurately at the same DC level. What load are you using?, OPamps can't drive much current at their output, perhaps 20 mA or so, hence the 32 ohm load in the figure.
Frank

The load is a headset (I'm testing with a 32Ohm headset aswell as a 62Ohm), the Cmoy is a headset amplifier. The op-amp should output 1/2 supply, but it isn't because the virtual ground shifts when a load is connected. I decided to abandon the bass-boost on these. I removed the input resistor, increased the input cap to 470nF, but I still need R4 to be 10k, 100k would still make it sound crappy. I tried increasing R5 and R6 by x10, but that didn't make a difference. It all seems good though (it works on my two headsets), but giving that different headsets can yield different results I still don't trust my current design.