The Objective2 (O2) Headphone Amp DIY Project

[qusp]

if you only have a fairly modest need for the cells like i did (I bought 18 x ANR26650m1 cells), not sure what the limit is without becoming an oem, i bought them directly from A123. Its not the cheapest price, but guaranteed legit, well packed and the packing is legal, so easy to hot-knife it to reuse yourself for storage, or shipping. its come in handy for flying with all the regulations that come with lithium cells now.

as a bonus, that way you have a direct line for any questions you might have and if things escalate you have a point of contact. there are a couple other options without resorting to ebay, which i would never do. they also sponsor a radio control battery brand, A123 racing funnily enough and their batteries use the m1 cell, you can break it down. there are a couple of US RC retailers that sell the single cells under that model name, just perhaps steer away from the asian RC stores like hobby king, some will be legit as manufacturing goes on there, but you just dont know, although the price is good and i did get some of the smaller capacity cells there, but ive also heard of fakes.

one last one, not sure if its still like this, but black and decker, or ryobi use the m1 in their cordless drills and other electric power tools and you can remove the casing to get access to them,

if you are keen on getting some from A123 direct let me know, i'll dig out the link to the oem page where you buy them, its not obviously placed. you'll need to have a company name though to place the order. actually i just had another look, looks like you can still do it, but definitely only as a lead up to full production through a company name (same process really). you can still by the a123 racing packs though and good hobby stores have them. i got mine about 12 months ago now. looks like they have just released an SLA replacement, like car battery size nanophosphate at 12v and 4.6wh. a couple of those for a power amp would be sweet.

[RocketScientist]

Thanks for all the info qusp. I'm especially interested in SLA replacements but not for audio. But I don't want to get too far off topic here

If anyone has any other questions, comments, criticisms, suggestions, etc. for the O2 please share them? I certainly get things wrong, or may fail to realize there are better ways to doing some things. Ultimately my goal is to raise the bar for how much performance you can get at a nearly Cmoy price. So anything that improves the performance and/or appeal while retaining the bargain price should be on the table.

[qusp]

no problem, good luck!! well worth the effort, i dont forsee having to replace my batteries until there is something better and the extra cells have actually turned out really handy for quickly knocking up a power supply for testing something in situ without changing the layout or wiring.

too large for a regular pocket size portable amp though; my battery takes up about 50% of the unit and adds a good chunk of weight, the 4 cell pack turned out ~25 x 110 x 65mm

[agdr]

Another long term reliability suggestion - DIP8 heat sinks for U3 and U4. I've got a bunch here from Arrow. Mouser and Digikey don't have stock, but Jameco sells them too.

They are cheap and work well. I'm using a couple on the DIP8 BUF634s in my NG98 amp.

Data Sheet for 580100b00000g

Heatsink at Jameco

Heatsink at Arrow

[idjoel2000]

are you planning to launch the pcb/kit anytime soon, rocketscientist?

[RocketScientist]

Thanks agdr. I'll try to figure out if there's room. If so I'll add them as an option. I'm not sure if you read my paragraph on dissipation in the Circuit Description section of last article? I did the math and made several measurements of power-in vs power-out. I also measured case temps. For anything but sinewave testing it's a non-issue unless perhaps the amp was operated in an extremely hot ambient environment, on AC power, into a worst case load at clipping, etc.

@idjoel2000 I won't be offering anything myself as this is entirely a non-commercial project for me. I'm hoping someone else will have the bare PCB for sale around the end of the month (ready to ship) or the first week of September. Kits would likely a come a bit later based on demand. Pre-assembled boards are another possibility. Please see the first two sections of the O2 Details article for more.

[agdr]

Actually I hadn't read that section but did last night. Well that is certainly good, the relatively low power dissipation.

On the zeners/resistor g-s clamps, I know it looks kind of cheesy and ads some lead inductance, but maybe they could fit if put vertically, on end? Just a thought.

[paulb]

Originally Posted by RocketScientist

Hmmm... way to go Google! I've had random problems with Google Docs which is sort of amazing when you consider they want companies to run their entire businesses on Google Docs and Google Apps.

But, that said, I just checked and it works for me. Here's the link:

https://docs.google.com/viewer?a=v&p...iOTc4&hl=en_US

I get the same message. I'm using Firefox, and I'm not from China.
Any chance of you posting the schematic here?

[RocketScientist]

The Google problems are just weird. I just tried the link above and it works fine. And it's working fine for most everyone else.

In the article I've also added larger versions of the schematic, board drawing, and a screenshot of the main part of the BOM. Just click on the pictures in the article and a new window will open with the full size image. They're nowhere near as sharp as the PDF, but they're legible.

@agdr - I did think about vertical, and it can be a useful last resort in some cases. But, in this case, I'm sure others would complain or criticize the O2 for having vertical components. Vertical axial parts are also a bit less robust and DIY friendly (they can short out against each other or other parts if not properly bent/installed, the polarity is less obvious for diodes, etc.). If you're really bothered by the Vgs issue and plan to build your own O2 you can certainly add your own gate protection.

At this point I'm trying to avoid significant changes to the PCB unless there's a *really* good reason. Otherwise I'm going to have to do yet another prototype spin of the board before releasing it, build another amp, all the testing, etc. And that's going to add a lot of work, more expense, and a few weeks of delay. To try and make room for the gate protection means pushing several other components around, including some fairly critical power supply and grounds tracks.

[RocketScientist]

Originally Posted by agdr

Another long term reliability suggestion - DIP8 heat sinks for U3 and U4. I've got a bunch here from Arrow. Mouser and Digikey don't have stock, but Jameco sells them too.

I like the idea, but unfortunately the heatsinks won't fit on the current design and I doubt I could make room no matter what. But if I do a desktop spin of this board without the batteries, I'll make sure there's room. If nothing else they will help the amp better survive sustained sine wave testing.

They're also available from Newark/Element 14 and they also carries the Box Enclosures cases. So someone could order the heatsinks with the enclosure and save paying $8 shipping for $1 worth of heatsinks.

My only slight concern is I assume the metal bit on the bottom is in contact with the PCB? The anodizing is an insulator, and production PCBs will have solder mask, but are there any sharp edges that facedownward? If so they might over time cut through the thin anodizing and solder mask and could short against traces routed under the chip. It's difficult to do proper power rail routing and bypassing of DIP8 chips, and get all the signals in and out, without using that space for traces.