Wow, I must say... I've seen a lot of schems in my time, and hand-drawn one are rarely that clear and readable!
I have only glanced at the schematic, which I may look at more closely later on, so right now I'm just going to blurt out some thoughts.
Firstly: Just checked wiki about the 5th gen nano. Uses a TFT (lcd) display, LED backlit. TFT's tend to be rather quiet when compared to the new OLED displays - which have to multiplex much higher current since they are powering the LED's as pixel, not just liquid crystal segments. So, I'm thinking the display itself is probably fine
Seems you have tried a LOT of standard methods to get a 'clean' supply, but I am still unsure of your
basic setup. I mean, you have a nice schematic there, but is your amp powered separately form the iPOD? Does it charge the iPOD? I see a max712 in there, with some batteries, which I'm guessing is for portability (the 712 being a cool on board charger for those batteries) ... is this what you meant by 'putting it on charge'? Or did you mean.... you plug your iPOD into the amp, whilst the iPOD itself is charging?
Apologies for all the questions
About switched mode power supplies. Generally, they are very noisy, switching high peak currents through PCB traces/wires, and inductors and so radiating RF/EM all over the place. There are fixed frequency PWM SMPS, which have advantages, as they don't vary their frequency to maintain efficiency, you can easily filter out the noise. I wouldn't start using one of those until you've got rid of your noise, because it will probably just 'add' to the problems, making it harder to narrow it down. So firstly, I guess trying to spot whats going on.
Noise is either conductive, going though the power lines, or transmissive, magnetic fields. As you've got some heavy filtering there, it seems if it is conductive, it's going to require an larger LC filter, 100uh, depending on the value of the capacitor looks a bit low to me.
It could be transmissive: that is, when your ipod is on charge, even though the noise from the on board SMPS charger (in the ipod) is well filtered from the audio out, the connections are picking up the noise via EM, and transmitting it through your amp, where of course its amplified. That means that the noise is getting into your signal path before your amp! However, this would be noticeable if you were to use headphones as well.
Maybe I'm over complicating things, but it could be noise from a beat frequency from two noise sources. Say, when your ipod charges, it kicks out say 240kHz. Sure your amp will heavily attenuate that, but what if your power supply for your amp produces a heavy line at 245kHz? They mix to produce 5kHz, which you can't filter out, as its slap bang in the audio band. - I@m out of my depth here though.
Ok, so, a coupe of questions which might help myself, and others narrow things down (as in things *I* would try).
1) Charging the iPOD with it playing through headphones. Leaving your amp out of it. Still noise?
2) iPOD going to your amp, both running off batteries. Any nose?
3) iPOD -> mains powered, amp.
4) Using a very cheap, off-the-shelve wall wart instead of your power supply. And running your amp off that. Charging your ipod at the same time and playing tunes. Sure, the wall wart WILL be noisier... but it won't be exactly the same noise (frequency, level etc..) as your power supply. If it sounds clean you know your filter circuits are doing jstu fine, and its the frequency of the noise thats giving your grief.
Again,I could be totally off here, perhaps I@m slow but I'm still unsure exactly when the noise rears its head.