[adydula]

Yes there is a O2 Booster Board :
 
The final circuit uses the OPA140 op-amp (FET input, DC precision, low THD+N) looped around an LME49600 audio current buffer on each channel to replace the O2 Headphone Amp's original NJM4556A output chips. Many other op amps would also work, such as the OPA827 and OPA627 with varying effects - see the full op amp rolling write-up in the build instructions. The board includes several optional "upgrade" modifications for the O2 headphone amplifier including an "anti turnoff thump" resistor, mosfet "on" indicator LEDs, and a 1/4" jack. The board produces a 93% reduction in the O2 amp's DC output offset voltage to around 20uV from the O2's standard 3mV. Output current capability is increased from 120mA per channel to 200mA for "music power" (playing music rather than sine wave testing). The O2's power supply has limitations that would prevent continuous current draw above 200mA per channel.
 
There are (3) versions of the pluggable PC board....check  it out at DIY audio....
 
.http://www.diyaudio.com/forums/headphone-systems/244473-o2-headamp-output-booster-pcb.html
 
Alex

 

[adydula]

In regards to the headn hi fi desktop kit etc...the agdr version is a lot more than just adding mechanical jacks or better switches. Its a totally new upgrade which involves lots of changes to "improve" on the O2...but cost is not one of them!! It is more expensive to build etc.....
 
Here are the highlights for the agdr ODA DIY:
 

1. · Up tp +/-16Vdc power rails with adjustable regulators for up to a 11Vpeak swing. Useful for high impedance headphones.
2. · Lower noise voltage regulators, LT1963A and LT3015. Probably won’t make any audible difference, though.
3. · Twice the output current capability and power dissipation - 280mA per channel. Useful for low impedance and low sensitivity headphones.
4. · 4 NJM4556A chips to handle the current and +/-16Vdc dissipation, two per channel. Uses the SIP 8 pin inline version, NJM4556AL.
5. · 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.
6. · Has input RCA jacks and output ¼” Neutrik jack in addition to better (Switchcraft) 3.5mm jacks.
7. · Bass boost circuit – switchable on/off.
8. · Rotary gain switch with 4 gain settings.
9. · Relay-based no-thump circuit that waits 2 seconds to switch in the headphones and then drops them out quickly on power switch-off.
10. · Should have even lower background noise than the O2 headphone amp at high gain settings. 4 layer PCB with full middle ground plane.
11. · 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.
12. · 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.

 
It is extremely versatile and there are many options when building that you can add, change or leave out, like gain settings, preamp usesage that is very versatile, bass boost if desired, etc....
 
It has leds that show the power rails working, a clipping led, power off switch in the rear....etc
 
Alex

 

[eimis]

Did somebody actually measure it or is it just a theoretical improvement?

 

[MrMateoHead]

 
The only question about a 3.5mm jack failure is when, not if.  You might be able to blame it on the "low-cost" selection, but I don't think so.  There is simply a limitation of geometry when it comes to 3.5mm jacks.  The contact area is super small, so any small potential for corrosion has a huge impact on connectivity.  Similarly, the spring strengths in the size of the metal required fatigues rather easily, also causing intermittent contact problems.  It's the nature of the beast in 3.5mm connectors.  You are best to keep things plugged in as continuously as you can. 
 
Just an FYI, but the physical geometry critical point is well exceeded with 1/4" connectors.  Equipment stored up and un-used for decades often makes perfect electrical contact with 1/4" connectors.

Thanks for the reply. The problem is, I typically leave things hooked up and virtually never disconnect - and yet experience this issue. I've only owned the O2/ODAC for perhaps a year or so. I assume that one day I will jack in and that will be it - one or two channels out, static, etc. I happen to love the amp so it is disappointing to hear that, apparently, this is "the nature" of 3.5mm jacks. Certainly I have many other products that use them without these symptoms.

 

[Xenophon]

Agdr should imo put some distance between his new build and the odac/o2. It might have been inspired by them but at least to me it's clear that his is a much higher specced product sold at a commensurately higher price. Only drawback is that I don't think it is something that just about anyone could build, been following your construction of it on diyaudio and this is most def. Not for novices. When it becomes available in hopefully a couple of weeks I'll seriously consider it for my office but first I'll build a power supply for it.

 

[adydula]

MrMateoHead....
 
Not to worry, its an easy replacement if it ever comes to that.....
 
Alex

 

[adydula]

eimis,
 
take a look at the ODA thread at DIYAudio.
 
Some stuff has been measured its a 40+ page thread, very informative, you will get a good idea on how it came to become what it is today...
 
I will have mine here probably late this week or early the next and I intend to do some critical listening etc and report back.
 
Alex

 

[SilverEars]

Ok, I have a question regarding the sonic difference I am hearing with my smartphone headphone out and ODAC/O2 out with iems.  These are iems and they are very sensitive with low impedance.  
 
What in hell make them sound punchier, dynamic(and transparent?)  with ODAC/O2?  Is it feeding more current than what my smartphone is feeding it?  Which doesn't make sense, but I thought I throw that out there.
 
I was going to throw this on the science forum, but since we got the Objective geeks here, I put it here.

 

[adydula]

Do you know what the output impedance of your phone is?
 
Alex

 

[r010159]

I understand you have made changes to the O2 to come up with this design, correct? Considering this, how faithful are you to the original design of the O2? Or is this a total redesign of the O2? Will any objective measurement be taken to verify performance?

 

[SilverEars]

  Do you know what the output impedance of your phone is?
 
Alex

1.2 ohms.  

 

[jring]

  1.2 ohms.  

 
What phone is this - the value is amazingly good for a phone if it's true and should be ok for typical IEMs with 16 ohms (headphone impedance divided by 8 or 10 is the usual rue of thumb).
 
If the output impedance is higher than 1/8 or 1/10 of the headphone impedance the frequency response will usually be quite non-linear. This is because headphones are not a purely resistive load and thus the actual impedanc depends on the frequency of the signal.
 
Joachim

 

[Smugsie]

I have this ODAC/O2 Combo
http://www.jdslabs.com/products/48/o2-odac-combo/

The FiiO X3 has a Coaxial Out option (3.5mm male to Coaxial female connector). Is there a way to plug in the Coaxial Out to the Odac?
 
 

 

[BenF]

  I have this ODAC/O2 Combo
http://www.jdslabs.com/products/48/o2-odac-combo/

The FiiO X3 has a Coaxial Out option (3.5mm male to Coaxial female connector). Is there a way to plug in the Coaxial Out to the Odac?
 
 
 

No

 

[adydula]

Hello,
 
On the question pertaining to the ODA and measurements. I chatted with AGDR and he replied with some information:
 
"I don't have a dScope (what NwAvGuy used) or the equivalent Audio Precision tester. I did get a QA400 distortion analyzer, essentially a sound card set up in a box for measurements, and posted those. To the limit of the QA400 (which is about 10dB noisier than an AP) the O2 and ODA showed the same noise floor and distortion. So, at least via the QA400, it is no worse than the O2, at least in terms of background noise and THD, and likely better from the data sheet measurements. But no, we'll never know for certain until someone out there with a dScope or AP does the tests.

I do believe the noise floor on the ODA is lower though, since I can hear it myself with the gain turned up on both. With a dScope or AP with that extra 10dB of range, I would expect to see the ODA edging out the O2 on noise floor. And from a circuit standpoint I would expect the same, given the lower noise power supply, lower noise pot and ground return resistor in the middle, and lower noise LME49990 chips in the gain stage. Having more NJM4556ALs in parallel also increases S/N since signal adds linearly while noise adds via RMS.

Distortion *should* be slightly lower, via the datasheets numbers of the LME49990 vs. the NJM2068 chips. But interesting thought experiment there. If you were to take an O2, add the booster board, and then replace the NJM2068 with a dual LME49990 adapter, that would probably match, or even possibly beat, the ODA in distortion. But not noise due to the O2's 10K pot. No way to get rid of that - can't fit in the pile of 4.7uF coupling caps needed.

Power though I can defininitely measure. Just a load resistor and scope. But it is really the distortion and other measurements at a high power output that matter.

The main tangible benefits of the ODA are lower background noise, something I can hear myself between the O2 and ODA on high gain, and the ability to drive higher current loads. Plus the clipping indicator and the pile of additional jacks and input select switch the O2 doesn't have. But even there the O2 can be modified to have more in/out jacks if folks use the taller B3-080 case."
 
I might add that there are these comments about things that should be better than the O2 and have been measured:
 

1. · Twice the output current capability and power dissipation - 280mA per channel. Useful for low impedance and low sensitivity headphones.

       Lower DC offset voltage.
 
Alex