O2: High Performance Portable/Home Amp

[NwAvGuy]

Originally Posted by b0ck3n

AMB is what the Head-Fi administration calls a "real scientist". "We're in trouble when real scientists don't want to join the discussion" - indeed, they don't return once their erroneous claim have been refuted.

The mantra of most real scientists and engineers I know is "show me the data". And it's kind of amazing how little credible data is presented both for, and against, headphone gear. When I take a strong position I always try to back it up (or certainly will upon request) with hard science, math, and real measurements on professional equipment. But that's far too rare.

Products, designs, and even people, become "cult favorites" with little to no credible objective data to support putting them on a pedestal. It would be like Ferrari claiming they've produced the fastest car in the world but offering zero verifiable test results or even any numbers to support that claim. That just doesn't happen. They know the magazines will get their hands on one and test it on a track. And they know they'll look really foolish if it falls far short of their claims.

I'm not afraid of filling the role of the car magazines but for headphone gear. I think the more claims that are independently verified the better. And if some come out looking foolish, it's not my fault they made false claims for their products.

[CrowdAds]

[khaos974]

I don't think it's locked, I still see the "reply" case.

What about his claim?

The NJM2608 opamp is not rail-to-rail. It could only swing to about 2V above the negative rail and about 1V below the positive rail.

You forgot to mention it in your reply.

[Soaa]

Originally Posted by khaos974

I don't think it's locked, I still see the "reply" case.

Oh, they reopened it. I swear it was locked earlier. More evidence of administration acting on a whim?

[NwAvGuy]

Originally Posted by khaos974

I don't think it's locked, I still see the "reply" case.

What about his claim?

I was trying to keep an already long reply full of confusing numbers more concise. It does not swing to the rails but neither does the "rail-to-rail" AD8397 in the Mini3. In this case the NJM2068 gets within 2.6 volts total (divided between the two rails) versus the 3 volts he's claimed. So 18 volts - 0.2 - 0.2 (diode drops) - 2.6 = 15 V p-p = 5.3 V RMS measured. If you drop the battery voltage to 8.4 volts see my post above.

The datasheet only specifies the drop with a 30 volt supply which is very different than a ~18 volt supply. In reality AMB's 3 volts is close. But he's skewed all the numbers in a way that makes the final result look much worse than it really is.

As for asymmetrical clipping, the proof is in the graph published above. THD takes all forms of clipping, and any sort of other non-linearity, on both halves of the waveform into account. 1% THD is the industry accepted method. AMB hasn't even published THD+N vs output graphs for any of his designs--or their 1% THD+N output levels into proper loads. He claims to just "eyeball" clipping on his scope which is highly subjective and notoriously inaccurate.

Others will be verifying the O2 on an Audio Precision system. I'm extremely confident they will get numbers far closer to the ones I've published than the ones AMB has creatively invented.

[razorblader]

Originally Posted by NwAvGuy

Ouch! Well if the minimum output impedance is 10 ohms, the Arrow officially rates a solid FAIL for any headphones under 80 ohms. So all you Grado, AKG, Denon, Shure, Ultimate Ears, Westone, HiFiMan, Etymotic, etc. owners can forget it unless you want messed up frequency response and/or under damped bass. See Headphone & Amp Impedance for why.

Just for clarification, would this also be true if the added impedance occurs after the output stage? Kind of like an impedance adapter? Just curious.

[NwAvGuy]

Originally Posted by razorblader

Just for clarification, would this also be true if the added impedance occurs after the output stage? Kind of like an impedance adapter? Just curious.

In this case (and most cases) the added output impedance does occur after the output stage. The only issue if it's enclosed in the feedback loop or not. Feedback tries to correct for the drop across the added resistance which has the effect of lowering the output impedance, but as the Mini3 demonstrates, it comes at a significant "price" (worse performance, marginal stability).

Most amps, like the FiiO E9 and many others, simply toss resistors in series with each channel outside the feedback loop (the same as if you did it outside the amp). And because the headphone impedance varies with frequency, the amount of drop across the resistor varies with frequency. That means the output voltage (and power) vary with frequency rendering your nice accurate high-end headphones, much less accurate.

And if the output impedance is more than about 1/8 the headphone impedance it can degrade the damping of the bass as well. The bass becomes less controlled, more "boomy", "flabby", etc. Both issues are explained in my Impedance article. You might also want to see Wikipedia Damping Factor.

[Pinkys.Brain]

Originally Posted by Soaa

Oh, they reopened it. I swear it was locked earlier. More evidence of administration acting on a whim?

It's getting more ridiculous by the minute over there on head-fi. Now it's almost all members vs. admins regarding the O2-thread because of the admin' questionable fiats when handling the thread. The last maybe 4 pages everyone is questioning their behaviour.
Even the OP distances himself from his own thread, as he cannot understand the admins anymore:

Originally Posted by Willikan

As much as it pains me to troll my own thread, reading through it leaves me with a bad taste in my mouth.

I think it is likely that I am going to ask for this thread to be closed. Steve Eddy put it perfectly when he highlighted how unfair it was to have a good design quite simply held up for ridicule. Deleting links that allow NwAvGuy to defend his own design is simply wrong. I did not start this thread so someone I respect can be held down and kicked with no chance to defend himself.

EDIT: Perhaps I was too hasty, seems to have calmed down a bit.

[MrSlim]

For non DIY readers interested in trying the O2, I have posted a build service offer using boards from DIYaudio group buy in the Classified Forum.

[Mid-Fi]

Two interesting posts about the O2 on other sites.

Over at head-fi, user deadlylover rigged up a point-to-point version of the O2 and compared it to his friend's beta22 using LCD-2 as the reference cans. The differences, he says, were small. The non-blind comparison is here. In the past, deadlylover was somewhat critical of the O2.

At diyaudio, user MrSlim is offering a build service for the O2.

[NwAvGuy]

Thanks for the update Pinkys.brain. In my opinion, anyone there interested in the O2, who's not already active here on ABI, should be encouraged to join the thread here. This is a much higher quality discussion.

The ABI admins have politely requested we keep this thread reasonably on topic and not turn it into a gripe-fest about other forums. So I'm not suggesting others come here to vent their frustrations. But if they're genuinely interested in the O2, or even have valid criticisms of the O2, ABI is a much better forum. The "signal to noise ratio" is vastly better and there's almost no commercial bias here on behalf of the admins/mods.

That's interesting news on both counts Mid-Fi! MrSlim did contact me but I wasn't aware he started offering the builds, and more significantly, has priced them quite reasonably! That's well under the cost of a Mini3 build.

And DeadlyLover seems to mostly enjoy building lots of stuff and seeing how it sounds. I have not measured the beta22 but it's a far more expensive design and most, in sighted listening, are going to expect it to sound better so that's what they'll hear. In light of that, finding the differences "small" is impressive. His actual quote:

"any differences I heard were small, and may very well have been my imagination."

I'm really looking forward to some blind tests against some of the "reference class" amps!

[Pinkys.Brain]

Here's the full post of Deadlylover so that you're in the loop:

Originally Posted by Deadlylover

I just spent about 3 hours at a friend's place ping ponging back and forth in a level matched comparison between the Obj2 and beta22, using the LCD-2. I was unable to discern any large differences in sound between either amps, so any differences I heard were small, and may very well have been my imagination.

There was practically no difference tonality wise, but I felt that the beta22 sounded a little less congested, that's pretty much the best I can put it. Since I kind of tunnel vision the vocals when listening to music, the main difference for me was the vocal separation with the rest of the track.

Anyway, I don't claim to have a large amount of experience with the LCD-2, and I realise some differences may only show up after an extended audition, so that's why I left the Obj2 with my friend, so I can forward his impressions eventually. I'm no audiophile, I've never heard any real live music, I have tin ears and I don't particularly care how things sound as I am not nearly as passionate about music as most of you guys. I also loathe dynamics, so I was really trying to pick between the lesser of two evils here, so please take my impressions with a grain of sugar, spice and everything nice.

On the other hand, I may very well be a sucker for subjective bias, as the comparison was only volume matched, and not done blind. The beta22 was actually balanced, so it only had a 4pin xlr out, and I had to 'unbalance' it for this comparison. It was a pain having to switch the LCD-2's cable each time I went back and forth, which is why we couldn't be bothered to do a proper blind test. I would have liked to build a milosz style switcheroo box, but I was just too lazy.

Do keep in mind that my Objective2 was not done on the proper PCB, and I very well may have ballsed up the amp, so again, don't take my impressions too seriously.

p.s.: While direct links to this thread (and such kind) are forbidden on head-fi, I think the interested will find the way here anyway.

Update: Now whole posts of the thread starter Willikan got deleted, so even the least bit of criticism of the administrators isn't allowed anymore...sad, sad, sad
And a post by me in which I critize that got deleted in under 5 minutes...wow

[razorblader]

Originally Posted by NwAvGuy

In this case (and most cases) the added output impedance does occur after the output stage. The only issue if it's enclosed in the feedback loop or not. Feedback tries to correct for the drop across the added resistance which has the effect of lowering the output impedance, but as the Mini3 demonstrates, it comes at a significant "price" (worse performance, marginal stability).

Most amps, like the FiiO E9 and many others, simply toss resistors in series with each channel outside the feedback loop (the same as if you did it outside the amp). And because the headphone impedance varies with frequency, the amount of drop across the resistor varies with frequency. That means the output voltage (and power) vary with frequency rendering your nice accurate high-end headphones, much less accurate.

And if the output impedance is more than about 1/8 the headphone impedance it can degrade the damping of the bass as well. The bass becomes less controlled, more "boomy", "flabby", etc. Both issues are explained in my Impedance article. You might also want to see Wikipedia Damping Factor.

Thanks for the clarification. Looking forward to building my own O2.

[SpaceTimeMorph]

Just to reiterate... thanks NwAvGuy for this design. Based on DeadlyLover's impressions, it seems that this amp, even subjectively, is the real deal. I mean, when a fully assembled B22 can cost ~$1500-$1800 and the O2 can be had for <$100 there should be more than "small differences." I can't wait to get my hands on a non-point to point flavor of the amp. This is truly awesome.

[Rapide_23]

I'm not so much worried about soldering up the board as I am about case-work. I haven't been able to keep up with all the posts but is there a "standardized" case in the works for purchase?

[NwAvGuy]

Originally Posted by Rapide_23

I'm not so much worried about soldering up the board as I am about case-work. I haven't been able to keep up with all the posts but is there a "standardized" case in the works for purchase?

Yes there is, see:

O2 Enclosure Options

I'm working on an "O2 Summary" that will make it much easier find things like the section linked above.

But, to answer your question, the case is around $11 and a pre-made front panel is $16 - $22 or so. The board just slips right in with no modifications required.

[Rapide_23]

Exactly what I wanted to hear

[upstateguy]

I have a turn on thump question:

Is it better to plug in the headphones first and then turn the amp on, or is it better to turn on the amp and then plug in the headphones?

[NwAvGuy]

Originally Posted by upstateguy

I have a turn on thump question:

Is it better to plug in the headphones first and then turn the amp on, or is it better to turn on the amp and then plug in the headphones?

I would just leave the headphones plugged in and not worry about it. If the turn on "click" is bothersome just don't put your headphones on/in until after the amp is on. If it's AC powered you can also just leave it on 24/7. It's not worth the wear and tear on the jack and the headphone plug/cable to always plug/unplug the headphones. And the turn off thump is even less audible.

If you need an amp for only desktop use, and want total silence, you might want to wait for the new project as it won't have any turn on/off noises.

[WalkGood]

Originally Posted by NwAvGuy

... If the turn on "click" is bothersome just don't put your headphones on/in until after the amp is on. ...

I'm sure it's implied but I'd also keep the volume POTs on zero too. BTW I have a portable amp that I have to do it this way too, but I mainly wait till after power up and the thump is minimized completely.

[Satellite_6]

Meh. I always have my headphones unplugged when I turn my amps on and off. Mine are both MADE IN CHINA tho. heh.