[r010159]

For craps and giggles I decided to hook my iPad up to the DAC part of the O2 + ODAC. I have never heard my iPad this good! I am looking forward to the C5D that I have just ordered. I will be quite a sight with the Denny's restaurant crowd whom I visit every morning, with full-sized headgear and all.
 
Bob Graham

 

[JacobLee89]

I'm currently waiting for the output booster mod kit for the 02, by AGDR. The thread is in the modding section of the forum ((under the thread name: O2 headamp output booster & modification PCB)) . As far as I know, the mod eliminates the turn off thump entirely, alongside lowering distortion and a whole list of changes that would be easier explained by just directing you to the document.
 
Currently I am using my O2 + ODAC with my HD650s, which has the general consensus that it (the headphones) works better with tube amps. Some have claimed that with the O2 mod by agdr, the amp sounds much less "digital". This may or may not be beneficial to the HD650's "less than neutral" sound signature. Unfortunately, the mod requires some direct tampering with the original PCB meaning I may not be able to directly compare the O2 with the modded variation, unless of course I get another one.
 
For others with the HD650's and this objective stack, I'll be sure to comment back on this thread to say how well this combination works with the (still awaiting) mod.

 

[Xenophon]

Looks like an interesting mod for sure and at least theoretically it should improve some aspects of the design.  Only, I don't feel it's still a good deal if you have to first get the original, then purchase the kit and install it.  We're then reaching a price point then where I believe better overall quality to be available.
 
But I'm curious to hear your impressions (BTW:  coupling a HD-650 to a tube amp...only for serious warmth lovers imho).

 

[agdr]

The booster board will also increase the O2's slew rate.

   I added some information about that to the build instructions at the Google Drive link yesterday after the good discussion in this thread made me realize that was another benefit of the board I had forgotten about.
 
The O2's orginal NJM4556A's have a slew rate of 3V/uS while the NJM2068 gain chip is 6V/uS from the datasheets.  The NIM4556A would be the limiting factor end-to-end in the O2 amp then, giving an O2 amp slew rate of around the 3V/uS.
 
The OPA140 on the booster board has a slew rate of 20V/uS, while the optional OPA827 slews at 28V/uS.  They loop around LME49600s that has a slew of 2000V/uS, but in that circuit arrangement the op amp is the limiting factor. 
 
So with the default OPA140s the booster board slews at around 20V/uS, replacing the 3V/uS NJM4556As, which in turn lets that NIM2068 chip's 6V/uS become the new limiting factor.  Long story short the end-to-end slew rate of the O2 amp should double from around 3V/uS to 6V/uS with the booster board.
 
To go a step further the O2's NJM2068 gain chip can be replaced with an LME49720 which has similar low noise specs, slightly better distortion specs, and slew rate of 20V/uS.  The O2 designer has a rather favorable writeup about the chip (vs. the NJM2068) in the "op amp measurements" section of his blog.  So then you get 20V/uS slew end-to-end with the LME49720 slewing at 20V/uS, as does the OPA140+KME49600 combination on the booster board.
 
The O2's designer had a good point about slew rate in his blog.  He notes real world audio sources don't slew much faster than 0.6V/uS and the signal processing chain at the studios has similar limitations.  So he doubled that to around 1.2V/uS just to be safe, if I'm remembering correctly, then noted the NJM4556A's 3V/uS is still about twice that making it more than adequate in his view.  He also noted the only downside to more slew is the potential for oscillations from a badly designed circuit.  I've measured the booster out to 200mHz so that isn't a problem here.
 
On the other hand there is that link to a good writeup on slew that was posted a few posts ago.  That author makes similar remarks to the O2 designer about real world source slew limitations, but then goes on to say something like "that is for 20kHz but well designed amplifiers are usually designed to be flat out to 100kHz, requiring a slew of 10V/uS" (paraphrased) and goes into some supporting detail. So there are certainly other interesting opinions about how much slew is needed and why.

 

[r010159]

  The booster board will also increase the O2's slew rate.

   I added some information about that to the build instructions at the Google Drive link yesterday after the good discussion in this thread made me realize that was another benefit of the board I had forgotten about.
 
The O2's orginal NJM4556A's have a slew rate of 3V/uS while the NJM2068 gain chip is 6V/uS from the datasheets.  The NIM4556A would be the limiting factor end-to-end in the O2 amp then, giving an O2 amp slew rate of around the 3V/uS.
 
[snip]
 
The O2's designer had a good point about slew rate in his blog.  He notes real world audio sources don't slew much faster than 0.6V/uS and the signal processing chain at the studios has similar limitations.  So he doubled that to around 1.2V/uS just to be safe, if I'm remembering correctly, then noted the NJM4556A's 3V/uS is still about twice that making it more than adequate in his view.  He also noted the only downside to more slew is the potential for oscillations from a badly designed circuit.  I've measured the booster out to 200mHz so that isn't a problem here.
 
On the other hand there is that link to a good writeup on slew that was posted a few posts ago.  That author makes similar remarks to the O2 designer about real world source slew limitations, but then goes on to say something like "that is for 20kHz but well designed amplifiers are usually designed to be flat out to 100kHz, requiring a slew of 10V/uS" (paraphrased) and goes into some supporting detail. So there are certainly other interesting opinions about how much slew is needed and why.

 
Can you give me a link to the O2 designers blog, where he discusses his design decisions and the reasons behind them? I am sure I will find this interesting to read.
 
Thanks!
 
Bob Graham

 

[miceblue]

Nope. It's illegal to post links to websites of previously banned members of Head-Fi.

You can just Google "nwavguy o2" though and it should be within the first two listed websites. XD

 

[r010159]

OK. Sorry. I was not thinking.
 
Bob Graham

 

[Xenophon]

   
Can you give me a link to the O2 designers blog, where he discusses his design decisions and the reasons behind them? I am sure I will find this interesting to read.
 
Thanks!
 
Bob Graham

You'll have found it by now.  It actually IS an interesting read and he undoubtably knows what he's talking about but be aware that it only presents his viewpoint of course.

 

[agdr]

Yep, he managed to get himself banned. If you find it do a search for "slew" and "slew rate" in the topics around July and August of 2011.  There is more than one chunk of stuff.

 

[Xenophon]

For a really good explanation on slew rate, check this out (math challenged persons should take a deep breath before clicking though)  http://www.onmyphd.com/?p=slew.rate

 

[r010159]

That article on slew rate reminds me too much of the electrical engineering classes I took in a previous lifetime. Still, an interesting read on the basics. The article demonstrates to me how much I have forgotten.
 
Looking for past articles on slew rate by the designer of the O2 (the one whose name cannot be spoken), I found a treasure-trove of information spanning several threads. Much of the discussions involve the technical aspects of amps. I can see a long night ahead of me.
 
Bob Graham

 

[JacobLee89]

" To put this into laymans terms and by my attempt to steer this into basic audio talk, is the slew rate useful for lowering distortion throughout the frequency range or does it lower distortion in the extremes of the frequency ranges? Of course I am referring to the frequency ranges that is audible to the human ear."
 
Was what I was wondering until I read somewhere that Slew Rate is associated with the "speed" of the amplifier and affects transients in music such as a hard hitting drum beat. But I shall leave that question there in case it has any validity.
 
Much thanks in advance.

 

[Xenophon]

I don't know if I understand you correctly but slew rate requirements for a component to deliver distortion free output at a fixed, maximum voltage are indeed frequency dependent is the short answer.  
 
If you want to calculate it and for a more complete explanation, see here:  http://hyperphysics.phy-astr.gsu.edu/hbase/electronic/a741p3.html
 
If you want the background for the formula used, check the link I posted previously, it's all there.

 

[JacobLee89]

  I don't know if I understand you correctly but slew rate requirements for a component to deliver distortion free output at a fixed, maximum voltage are indeed frequency dependent is the short answer.  
 
If you want to calculate it and for a more complete explanation, see here:  http://hyperphysics.phy-astr.gsu.edu/hbase/electronic/a741p3.html
 
If you want the background for the formula used, check the link I posted previously, it's all there.

Thanks for the clarification!
 
It does seem that the improved slew rate is beyond the margin of my hearing ability, but that of course is only part of the many improvements.

 

[RAZRr1275]

Are there any pretty cheap solutions for the ODACs power supply issues?