I've been doing some googling and haven't quite found an answer - but wasn't there meant to be a ODA (desktop amp), whatever happened to this concept?
.
The designer of the O2 headphone amp never did post his ODA before he dissapeared last year. Earlier this year I whipped up a DIY version of an ODA, based on the original O2 design. Same overall design as the O2 just with each section upgraded to take advantage of the additional space and power supply in a desktop design. See this post:
http://www.head-fi.org/t/616331/o2-amp-odac/1635#post_9748634
Hey my appologies to Currawong! I hadn't followed up on this thread in awhile and never saw his post #1677 until now, when I went back to find the post number of my old post. I'll follow up with him to find out the best way to post the files here. I haven't tried posting those yet because I want to do a couple of small updates (then fab and test the boards), but got sidetracked on another O2-related DIY project, an O2 output upgrade PCB. This one slides right into the top slot on O2's standard B2-080 case, above the O2 PC board, and plugs in place of the two O2 NJM4556A chips.
The O2 booster and upgrade board does a few things. It replaces the O2's two NJM4556A output chips by a lower distortion (by datasheet numbers, NOT dScope tested) and lower output DC offset OPA140+LME49600 pair. I have measured the DC output offset. 20uv vs. 3mV (3000uV) for a standard O2 with the NJM4556A chips, a 93% reduction. The O2's designer noted in his posts that the output distortion of the NJM4556A output chips swamp that of the NJM2068 gain chip, making it pointless to replace the NJM2068 with something even lower distortion. The upgrade board solves that problem. The board also includes a few optional O2 modifications I've posted elsewhere over the years including: output anti-thump resistors; power manageement latch circuit to prevent that oscillation noise when the batts get low; and an optional 1/4" Neutrik jack that fits upside down (but the 1/4" jack does require the taller B3-080 case).
The OPA140 is a DC precision op amp with good AC datasheet parameters (around 8x better than the O2 designer's posted dScope measured numbers with the NJM4556A) while the LME49600 is the well known 250mA audio output buffer. Note the O2's power supply in the O2 won't allow continuous (sine wave) output past around 200mA but it certainly can burst up there in a "music power" sense where peaks in music are 3x to 4x the average. The OPA140 has extremely low current draw, just 2mA, and the LME49600 is run in low bandwidth mode (still 110mHz which is way above what is needed for audio). So the upgrade chips pull only slightly more current than the original NJM4556A, making it still battery friendly. The OPA140 can be replaced with the OPA827, OPA627, OPA1641, LME49710, LME49990 but all with differing effects on DC output offset and current draw. See the build instructions.
Both boards are 100% DIY. The files contain Gerber files you can send out to Seeed Studio or other 4-layer fab house and make your own boards. If Currawong will allow I may make single PC boards available at cost if anyone is interested if there is some way to do that and stay within forum rules. The fab house has a 5 board minimum which is a bit of a DIY hurdle. The ODA board has a lot of surface mount parts and has parts on both sides of the board, although I kept the minimum part size on either board 1206 SMD (larger size) to make them as DIY friendly as possible. The O2 upgrade board is all surface mount but on just one side. So both are essentially solder skills V2.0 vs. the O2.
Best to wait a little bit on the ODA since I am doing a few small updates on the design and want to fab/test it first. I'm slowly tinkering around a whatever a V2.0 might be, but that wouldn't be until later next year if at all.
(image missing)