Guitar Headphone Amplifier

[Circlotron]

Most guitar amps have an input impedance of ~1M.  Lower and it can roll off the highs (not that electric guitars have much in the way of high frequency content).

I read somewhere that a low input impedance can damp the string vibration somewhat as well, causing the sound to die away quicker. I suppose it's possible in theory at least.

[tech_builder]

It’s been a while since I last posted as I haven’t had much time to work on this project, but I managed to build up the op-amp based ground rail and do a few simple tests. One thing that I noticed is that my output swing is much more limited. Previously I could get over 3.5Vpk-pk output at 1kHz.



 I can get about 1.75Vpk-pk output (1kHz sine) at most before distortion, but the shape of the distortion is strange. It happens on the lower half of the waveform by sneaking in and cutting it off.



The upper half starts distorting when the output reaches about 2.5V.



I’m not too sure what would be causing this other than the op-amp may not be able to put out enough current to maintain ground? The point at which distortion starts is frequency dependant. At 200Hz I can’t get more than 1Vpk-pk output before distortion begins,



 but at 10kHz I can get more than 3.5Vpk-pk.

[Audioguru]

I think your fake ground is causing the output distortion. If you make it with a single supply then you should not have the distortion. Have you measured the battery voltage when the amplifier is distorting?
But you also might have the pins of the transistors mixed up or have some polarized capacitors connected backwards.

Since the problem occurs only at low frequencies then maybe a capacitor (or a few) values might be much too low.

[tech_builder]

I did some more testing, specifically with the bypass capacitors on the power rails. I gradually increased the size of the bypass caps till they were 100uF. This size allowed an output voltage swing of about 4Vpk-pk at 200Hz. It sounds pretty good with headphones as well. Not burning with tone that'll knock your socks off, but pretty good!



I noticed that the crossover distortion has increased, possibly because I can now drive it to larger amplitudes on the output. Although at 10kHz and using 2.2uF caps I didn’t notice any crossover distortion at a similar output (see my previous post).

[Audioguru]

Guess what? The voltage from your battery is falling flat on its face unless it has a pretty big capacitor parallel with it to help it.
I repeat, "Have you measured the battery voltage when the amplifier is distorting?"

[Zero999]

I noticed that the crossover distortion has increased, possibly because I can now drive it to larger amplitudes on the output. Although at 10kHz and using 2.2uF caps I didn’t notice any crossover distortion at a similar output (see my previous post).

Try increasing the values of R8 and R9 slightly. It will increase the quiescent current slightly but should reduce the cross-over distortion.

[Circlotron]

If OP1 doesn't have enough grunt to drive the speaker directly but needs a buffer, how do you expect OP4 to be strong enough to hold the "grounded" end of the speaker steady? Same current flows in both.

[Zero999]

If OP1 doesn't have enough grunt to drive the speaker directly but needs a buffer, how do you expect OP4 to be strong enough to hold the "grounded" end of the speaker steady? Same current flows in both.

Indeed. There's an error in the schematic I missed. The speaker's return should be connected to the negative rail, not the output of the op-amp.

[tech_builder]

I connected the ground of the headphones to the negative rail and replaced the bypass caps back to 2.2uF and I didn’t see any distortion. Everything seems to work great! Thank you so much for the help. This design wouldn’t be anything without this forum! So I think I am going to call a design freeze here and work on designing the PCB and enclosure. I can still make small modifications if I need to, but if I don’t call it soon it’s never going to be made.
Audioguru, I was testing it with a power supply (HP 6235A). I did measure the ground voltage on the scope and it was stable. I haven't made the same tests with a battery.
Attached is the schematic I will be working from. I will be using KiCAD for the PCB layout.

Thanks!

[tech_builder]

I found some time to get everything together and start learning KiCAD. Here is the schematic I have put together for the headphone amplifier. I added a guitar input connector and a headphone output, which I had to make. I managed to find a mono-jack but I couldn’t find a stereo jack. I searched for it to no avail online. It seems that the library is now auto managed through GIThub or something? It was a little confusing.

Another thing I added was a break in the negative power supply rail that cuts the power when the guitar is not plugged in, so that if it is left on accidentally it won’t drain the battery. This required me to reference the guitar input to the negative rail instead of ground. I might add one in series to the headphone as well, any thoughts on if this implementation is horrible? Any better ideas?

I will now begin the next step which is part footprint association, which hopefully won’t take as long as the schematic to do.

EDIT: Schematic updated to fix errors in op-amp orientation. Updated file dated Aug 7, 2017

[Audioguru]

R12 connects to ground then the output signal overloads the output of the voltage splitting opamp. Maybe R12 should connect to the negative terminal of the battery instead?

[Zero999]

I found some time to get everything together and start learning KiCAD. Here is the schematic I have put together for the headphone amplifier. I added a guitar input connector and a headphone output, which I had to make. I managed to find a mono-jack but I couldn’t find a stereo jack. I searched for it to no avail online. It seems that the library is now auto managed through GIThub or something? It was a little confusing.

Another thing I added was a break in the negative power supply rail that cuts the power when the guitar is not plugged in, so that if it is left on accidentally it won’t drain the battery. This required me to reference the guitar input to the negative rail instead of ground. I might add one in series to the headphone as well, any thoughts on if this implementation is horrible? Any better ideas?

I will now begin the next step which is part footprint association, which hopefully won’t take as long as the schematic to do.

Disconnecting the power supply when the headphones or guitar is disconnected is a good idea.

[tech_builder]

That took longer than I thought. Anyways, I have finished laying out the circuit board. Not the prettiest thing, but I want to have something to try out. I decided to fill the top layer with the +4.5V net and the bottom layer with the -4.5V net. Not too sure if there are any concerns with doing that. It made it a lot easier to connect everything.

In general I found KiCAD to be a pretty good program to use. The only thing I wish is that I could get the traces to stick to follow along other traces when drawing them, which is how some of the other programs I’ve used have worked. I really like the 3D rendering of the board. Super easy to use!