O2 AMP + ODAC - Page 73

 

So do you have any preference between 1x and 2.5x gain? I personally prefer 1x gain for the transparency. Higher gain could possible feel more "powerful" but unity gain delivers pure clarity. To my ears, of course.

I can't speak for others, but unity and 3x gain (what I have mine set to) sound the same. The only difference is how far I have to turn the pot to reach desired volume. As long as you're not clipping the input it sounds great. Of course there's some hiss with iems on 3x gain, but I have to have the pot at 2 o'clock+ and that would seriously hurt my ears

My ears and my head are telling me the unity gain setting is sooo much better....but I know thats not totally accurate.

 

I do notice that with unity gain you have to crank the volume knob much farther to get things to liven up so shall we say...

 

I dont think there is any reason for me to used anything but unity gain with my setup...

 

I run from a destop pc, usb to the ODAC etc....mostly listen with batteries until I have to charge them up,

 

I would say that a pair of LCD2's and the O2 and ODAC is pretty much up there with the best....it would be hard for me to imagine anything being an orde of magnitude better than this setup...

 

And this is not even considering the cost....which we all know it this case is the deal of the century for head amps and dacs....

 

Alex

Pardon my ignorance, but I've heard it being used here for a while. What is unity gain?

 

1x gain.

 

I wouldn't say sooo much better... But after testing two different units and different gain configurations, I must say the difference is noticable if you listen carefully. It's really subtle, not something that all the people have to care about but it's there. IMHO.

 

The Beyer at 600 ohm would cause any inherent gain issues with the O2 to manifest themselves most easily.  From what little I know about it, the design took a departure from accepted headphone amplifier design and placed the volume pot inbetween the gain stage and output buffer.  That means that any input signal beyond a certain voltage will be clipped.  A high-impedance load such as the 600 ohm Beyers will require much more gain than other headphones.  When that gain is dialed-in, if the source is not below a certain voltage, the O2 will clip.

 

AMB made these same arguments to the designer.  It reduces the applicability of the design in the real world, perhaps in favor of better testing results.  The volume pot after the gain stage may be less noisy than otherwise.  Plus, 600 ohm headphones are not everyday, run-of-the-mill headphones.  So it may have been a design risk worth taking.

 

Personally, I find it interesting that some are commenting that lower gain settings are in favor of better sound, but this is exactly what one might expect with that kind of design decision (locating the volume pot in-between amplifier stages).  Clipping is not an immediate, "OMG - it sounds bad!" result in every case. Instead, you may have peaks and highs that somehow seem stilted, reduced, distorted, per se, or a sense that something is not quite right with the settings.

 

There's no reason an amp can't handle a strong source, unless the attenuation was placed after the gain stage.  Just my personal opinion again, but it seems somewhat ironic to find that a source from a designer may exceed an amp's gain capability that comes from the same designer.

 

Power is not the question.  This is often misunderstood when looking at gain.  A high-gain setting on an amp, or a strong source may seem like there's a lot of power present - simply because things get loud at very low volume-knob-travel settings.  It doesn't mean there's a lot of power at all - things may start clipping at just a little bit further up on the volume knob.  You could easily have ear-splitting volumes at 9 o'clock, but clipping could occur at 9:30 or 10.

 

I wouldn't call the design "defective," either.  The volume pot was designed to be put in the place in the circuit where it resides.  It was a conscious decision, perhaps interesting in its deviation from accepted practice.  However, there are consequences for that decision.  An advantage is that there is generally less noise when placing the volume in the interstage instead of at the signal input.  The disadvantage has already been mentioned.  One can design to meet measurements to the exclusion of all else, or one can design a product for the best usability. The two goals may not have the same results.  In this case, if a source is a bit stronger or a pair of headphones needs higher gain, the O2 may have trouble without attenuating the signal at some point in the circuit other than the volume pot.

 

 

Again, gain does not mean power.  Just because 10 o'clock on the volume knob is ear-splitting, doesn't mean the amp is not clipping at 10:30 - or even somewhat clipping at 10 o'clock.  You may simply not notice the increased distortion.  Your acknowledgement that unity gain is superior in sound quality may be a confirmation of this.

 

As for the USB related connections with the ODAC, I was surprised to see the schematic when it was finally released.  There is no power capacitor anywhere on the PCB.  A typical DIY DAC might have 1000uf to 1500uf onboard the PCB somewhere.  There is only a single SMD 22uf capacitor on the 5V USB buss in the ODAC.  It's no wonder it's super-sensitive to USB power fluctuations, laptops, etc., and that a powered-USB hub works wonders.  Yes, I would consider this a design-flaw.

 

Same-same as above.  The lower gain perhaps sounds better because of the minute distortions occurring from clipping at higher gain settings.

 

Just some guesses and observations - I'm not claiming to be an absolute expert.  It just seemed as if you guys needed some possible explanations or elaborations.

Edited by tomb - 3/10/13 at 8:38pm

 

To me this is an interesting point as well.  The designer claims to have tried the ODAC on all sorts of USB sources and had no issues on all of them, yet in the field there are a lot of people who talk about the issues they've had.  Maybe the designer's laptop and other USB sources are too high quality compared to what's out there?  Is that really the explanation we're going with?

 

As you say, most DIY audio DACs have some significantly larger bulk capacitor somewhere, but the ODAC does not.  Maybe my ignorance in analog electronics is showing, but my interpretation is that the circuit was reasonably optimized.  I mean, a larger cap is worse in some respects than a smaller one (and not just cost and size).  It should be noted the ODAC's a non-DIY design intended to be only SMT, but it's not like it couldn't have larger capacitance somewhere if it were desired or warranted.  There's plenty of spare PCB space, after all.  The power consumption is pretty low, anyhow.  There probably weren't four PCB revisions and supposedly "more than 100 variations of components" tested for no reason, and the 111 dB dynamic range (on at least some USB port...?) using a chip capable of 112 dB didn't happen by accident.  

 

Has anybody been able to eek out worse than claimed performance out of the ODAC (say by using noisy USB power) and confirmed it?  I'm curious what difference would show up if somebody could capture that on an audio analyzer or scope.

I think we have to keep in mind that it isn't a commercial design, but rather aimed at the diyaudio crowd who are electronics people.

In any case, all that anyone needs to know is that one should use low gain to avoid clipping issues unless the volume doesn't go high enough. In neither case will clipping occur. It's such a simple rule that it would justify doing this for a commercial design.

Remember this was an experiment in a low cost amp and DAC to prove a point.

 

It was an objective experiment.

 

The design goals and their arguments have been posted and discussed in great detail on the designers site and other DIY sites etc...

 

It was not designed to work with 100% of the cans out there. (the amp part that is)..

 

 

It is not perfect, but for what it is, it met the design goals and is a very good amp and dac solution that just happens to compete with much pricier

amps and dacs.

 

Many of us have used the amp and dac with little or no issues.

 

I have been in the PC industry in development since PC1 days.

 

My 2 ODACS have been used with laptops, desktops, DIY systems, etc...I mean a lot of different brands and designs.

 

Never once have I had any USB issues with enumeration, setup, whining noises, static, yada, yada, yada...it just works.

 

The ODAC cant control the power fed to it.

 

If your pc's USB implementation is crap...or you have a zillion things plugged in to the ports, or have it near sources of interference well...

 

I for one use my stuff in a clean environment, I dont place my stuff next to the microwave oven, or have it near AC line cords...etc

 

Remember with a little common sense for $99 for a ODAC, and $99 for an amp...you can have world class performance, totally transparent....

 

and the designer has stated its 'not perfect'....but for me its very, very close....for 98% of the cans out there...

 

and now back to the music...

 

All the best Alex

 

(Note...one of you exerimenters.....tack a cap to the odac and see it it 'fixes' any of your issues with USB?"

Let us know the results?)..

 

I have an Acer laptop, nothing spectacular in audiophile terms... Haven't experienced any problems with powering ODAC and audible degradation of sound. I guess playbacking in hibernate mode helps (Jplay Mini player) a bit.

 

However, as I need to purchase an USB hub, I'll get one with dedicated power supply, switch on/off button for each USB port, overload protection etc. I will try to compare whether the performance of ODAC is better or not with the hub. I'll post my opinion here.

 

In the above quotation the followig caught my eagle eye:

 

 

 

 

For comparison, please see this http://www.head-fi.org/t/603728/o2-vs-totl#post_8272835   

(One can search the 'O2 vs. TOTL' thread for the 'authority' word.)

 

Not being an audio authority at all, what does  'authority' mean in the audio context?

 

JJ

Edited by JakeJack_2008 - 3/11/13 at 9:44am

Thanks for the link to this other thread....

 

Authority to me in the context of the O2 and the stuff I source from and listen to ......means that there isnt anything that I cant hear or experience well with my cans connected to the o2amp and ODAC combo....

 

The headphones I use to listen with are easily driven to volumes that are very uncomfortable to listen to and would most likely cause hearing loss / damage etc....

 

Any amp that can drive cans to ear-splitting levels mean that to me...with authority in terms of enough power to excite them well.

 

But not all amps can do this "well"....ie transparent, noiseless etc..

 

Alex

Edited by adydula - 3/11/13 at 5:37pm

So I tried to compare quality of playback using my new USB hub (50USD) with a power supply, having only one of 4 USB ports switched on and having two HQ usb cables with ferrite for connections:

 

1) Laptop --- USB ferrite cable --- ODAC

2) Laptop --- USB ferrite cable ---- USB hub ---- USB ferrite cable --- ODAC

 

At first, I compared both setups using Kmplayer and two differents songs... I don't use Kmplayer to play music at all, it was just for fun. There was absolutely no difference. The only thing I realised is how much worse music sounds with Kmplayer in comparison to Jplay Mini in hibernate mode - really really bad quality.

 

After that, I compared both setups using Jplay Mini in hibernate mode. With the first song, I couldn't hear any difference. With the second one, I felt that both setups sound virtually the same but I very slightly prefered the first setup (laptop - cable - ODAC). It could very well be placebo but I thought the second song sounded a little bit more distorted. But the difference was extremely small - like 0.05% (subjectively, of course).

 

OUTCOME (IMHO):

 - don't bother with special audiophile USB hubs unless you have some super noisy environment around you

 - get USB ferrite cables (they are very cheap) instead if you want to feel "safe"

 - digital cables do not matter at all unless you have some really bad ones or use very long connections

 - the more direct connection, the better (try to avoid everything redundant in your audio chain, on both software and hardware side)

 

The end of story.

Edited by RustA - 3/11/13 at 12:35pm

You ask good questions and I'm not sure I have the answers.  About the only reason the design wouldn't have a power cap is if the current draw was already high.  Then there might be some concern with the inrush current needed to charge the cap.  That could conceivably cause the DAC to get knocked off, too.  From what you say in terms of power consumption, this is not the case, though.

 

IMHO, a good-sized power cap is cheap insurance for any power problems that could occur on a USB-powered DAC.  The USB-powered DACs that have them are too numerous to mention.  It should be no different than if there were much larger caps than normal in the PC's power supply - or in a separate USB hub.  AFAIK, it would have no effect whatsoever on the DAC's audio performance, but rather increase the usability factor by minimizing power issues. 

 

As for the SN/dynamic range, that's probably a function of the board layout and judicious selection of parts - particularly with regard to the smaller SMD sizes that result in much smaller trace lengths and less interference.