First of all, I will say that I don't at this point have any images or schematics. I will work on uploading some tomorrow but for now I just have to describe this circuit. Sorry
The amplifier is a transistor power amp with an output stage like a Gilmore Dynamic amp...four parallel pairs of 2SC1815 and 2SA1015, each with a 10 ohm emitter resistor. For the prototype, I hand-matched eight 10-ohm resistors on each amplifier board. The output transistor bank is biased on by a totem pole of 680 ohm resistors and 1N4001 diodes.
The input stage is an op amp that directly drives the output transistors through the totem pole (at the junction of the 2 diodes). It uses a 5.6K, 1K resistor, and 100 uF capacitor for negative feedback taken from the output of the power stage. The input uses a 100 uF capacitor with two 1M resistors for the input network. The volume attenuator is a 50k stereo pot.
The amp is designed with op amp rolling in mind. The prototype is kind of wasteful, as each amplifier channel is built on separate boards. I wired the amps for dual op amps so the other side of a given chip is wasted as its pins are all tied to ground. At first I was using TLC272s which sounded pretty good. Right now I have LM4558s in it, which sound good at the expense of a slightly higher output offset voltage. With the 272s, it was nearly zero.
I made the mistake of trying the amp with a 24v supply (although I designed it for that). I first heard some uncomfortable clicking in the phones, then saw a thump against the underside of the chassis/tin. I opened it to find the TLC272s blown apart. I don't know if the power supply was defective but the chips should have been able to withstand +/- 12v.
Again, sorry for the lack of tangible data. I'll work on getting some tomorrow. Here's a picture of one of the blown TLC272s. I'll provide more data tomorrow.