I have a close friend that is a really great engineer and has worked up several DIY headphone amplifier designs. He was in contact with the Rocket Scientist guy who started the flurry of activity around low cost and high performance…yes I know he ruffled a lot of feathers and this is NOT to start anything like that up…
My friend AGDR took a look at what the desktop version might have been and came up with a DIY design which I built a few years back. Its a SS amp that really works very well…been using it for years every day…
Its the red one in the middle. I never sent the faceplate out for engraving…so this is a rare amp and only 6-7 were ever built…
Those on the front are RCA outputs from a simple preamp circuit…to go to a AVR or another amp.
The green leds are the +/- power rails and the led thats off is a clipping indicator, gain is 1x, 2x, 4x and 8X. Never use the 4 or 8 x positions.
· Up tp +/-16Vdc power rails with adjustable regulators for up to a 11Vpeak swing. Useful for high impedance headphones.
· Lower noise voltage regulators, LT1963A and LT3015. Probably won’t make any audible difference, though.
· Twice the output current capability and power dissipation - 280mA per channel. Useful for low impedance and low sensitivity headphones.
· 4 NJM4556A chips to handle the current and +/-16Vdc dissipation, two per channel. Uses the SIP 8 pin inline version, NJM4556AL.
· NJM2068 replaced with OPA627, which is now in a feedback loop with the NJM4556 chips to null out DC offset and reduce distortion even further. DC output offset voltage should be around 0.3mV = 300uV per channel.
· Has input RCA jacks and output ¼” Neutrik jack in addition to better (Switchcraft) 3.5mm jacks.
· Bass boost circuit – switchable on/off.
· Rotary gain switch with 4 gain settings.
· Relay-based no-thump circuit that waits 2 seconds to switch in the headphones and then drops them out quickly on power switch-off.
· Should have even lower background noise than the O2 headphone amp at high gain settings. 4 layer PCB with full middle ground plane.
· Volume pot is on the input now rather than the middle of the circuit, so it can attenuate “hot” sources as much as needed. Still no pot turning noise.
· Coupling cap is on the input, 4x as large to work with the 10k pot vs. 40.2k resistor in the O2, to block all incoming DC from the source.
Some of this changed during the build discover process, but for the most part it has what Rocket Scientist mentioned before he dis-appeared.
No batteries, both 1/4" and 3/5’’ head outs, more power, lower noise etc…it has even driven a set of Klipsch high efficiency speakers to fairly high levels! Updated to better MTBF parts, like the low cost push buttons and switches in the original O2.
Here is a link that shows an ODA that was sent to a fellow to do some objective measurements and for grins he attached the amp to a set of loudspeakers to demonstrate the power output of the ODA…
The ODA amp has 3x the current sourcing capability of the original O2 and measures better…it will drive almost any headphone to its limits with the exception of electrostatics…
