O2 Measurements & Review

Discussion in 'Headphone Amplifier Measurements' started by tomchr, Jan 26, 2017.

  1. ultrabike

    ultrabike Measurbator - Admin

    Staff Member Pyrate MZR
    Joined:
    Sep 25, 2015
    Likes Received:
    8,960
    Trophy Points:
    113
    Location:
    Irvine CA
    The clipping is strange. The maximum voltage swing might be dependent on the supplied voltage. The "standard" O2 comes with a 2.5x gain in "low" mode. But if the power brick does not swing 15VAC then things may indeed clip. I will try battery also. Some things come with 12VAC bricks and if that is the case I recommend getting a 15VAC one.

    It seems you had an underpowered O2.

    That said, not everyone is going to like the same stuff. Many don't like it. Such is life.

    As far as negative feedback and opamps, I don't see the use of those in a negative light. In fact I like negative feedback and opamps. But that's just me.
     
  2. BenjaminBore

    BenjaminBore Friend

    Pyrate
    Joined:
    May 23, 2016
    Likes Received:
    2,842
    Trophy Points:
    93
    Location:
    London, UK
    That’s cool, my nonsense wasn’t directed at anyone specifically.

    I found a photo of what I think is the same adapter that reads: “AC16V 700mA 11.2VA”. Either way wasn’t the O2 known to have a design flaw that lead to early clipping?
     
    Last edited: Dec 2, 2018
  3. ultrabike

    ultrabike Measurbator - Admin

    Staff Member Pyrate MZR
    Joined:
    Sep 25, 2015
    Likes Received:
    8,960
    Trophy Points:
    113
    Location:
    Irvine CA
    That was the claim.

    The deal was that the pot is installed between the gain stage and the current buffers. Many folks install the pot prior to the gain stage. That is not a design flaw, that is a design choice.

    I would do similar because doing so avoids amplifying the noise from the pot (a relatively high impedance). It also may help in having more control with the DC block capacitors, and reduce noise when rotating the volume knob.

    The design will not clip for line level of up to 2.65 Vrms in low gain w a 15VAC thingy. But it might with a 11.2VAC thingy.
     
  4. ultrabike

    ultrabike Measurbator - Admin

    Staff Member Pyrate MZR
    Joined:
    Sep 25, 2015
    Likes Received:
    8,960
    Trophy Points:
    113
    Location:
    Irvine CA
    Wait. I dunno about AC16V, but you said 11.2VAC? That might be the problem. 16VAC should be fine.

    My AC brick reads...

    OUTPUT: 15VAC 500mA 7.5W

    Go well below 14VAC and things may indeed break apart.
     
  5. BenjaminBore

    BenjaminBore Friend

    Pyrate
    Joined:
    May 23, 2016
    Likes Received:
    2,842
    Trophy Points:
    93
    Location:
    London, UK
    This looks identical to my old one: https://sg.carousell.com/p/epiphany-acoustics-ehp-o2-objective-2-nwa-headphone-amp-44836979/

    I am ill equiped to comment on the technical aspects of it’s design. But I can say with absolute certainty the model I had clipped when it should not have, in a way that nothing else I’ve used has, and in a repeatable fashion.

    Indeed with a lower line level going in, mains powered, the gain set low it was okay. But in others configurations it would clip anywhere from just an ocassional pip of a clip, to constantly if fed 2vrms. Hooked up to the Modi Multibit it would sound through headphones like one never ending violent fart, and reducing gain only reduced the level violence.

    It’s possible the early clipping was caused for a different reason to that speculated. If I still had it I’d offer to send it to you, alas it’s home is now that of some fellow in his dotage happy as Larry listening to his AKGs. Which I learned over some 12 after sale ebay messages...
     
    Last edited: Dec 2, 2018
  6. ultrabike

    ultrabike Measurbator - Admin

    Staff Member Pyrate MZR
    Joined:
    Sep 25, 2015
    Likes Received:
    8,960
    Trophy Points:
    113
    Location:
    Irvine CA
    Indeed. Your O2 clipped because the brickwall adapter supplied 11.2VAC instead of the 14VAC minimum, and that AFAIK is massively under the O2 requirements.

    Ephiphany Acoustics fucked up.

    Get one of these:
    http://www.schiit.com/products/wall-warts

    I would go for the 16VAC 500mA option with the appropriate voltage/plug selection depending on country.
     
  7. Ardacer

    Ardacer Friend

    Pyrate Contributor
    Joined:
    Oct 15, 2017
    Likes Received:
    986
    Trophy Points:
    93
    Location:
    Croatia
    My O2 came with underpowered PSU as well, and made weird clicks. So I made a psu for it myself. It's been fine since.
     
  8. Poleepkwa

    Poleepkwa Friend

    Pyrate BWC
    Joined:
    Oct 1, 2015
    Likes Received:
    1,557
    Trophy Points:
    93
    Location:
    Finland
    @BenjaminBore . My Epiphany O2 back in the day did the same in regards to clipping and generally lame sound compared to a JDSlab O2. That was more punchy sounding too. I think there was design differences too when looking at the PCB boards. Oh well...
     
  9. monacelli

    monacelli Friend

    Pyrate
    Joined:
    Jul 23, 2017
    Likes Received:
    1,649
    Trophy Points:
    93
    Location:
    The Land of Enchantment
    I think we’re misreading his post about the specs. It sounds like the Epiphany wall-wart is 16Vac and 11.2VA (Volt-Amps, or W if you like). A wall-wart transformer with an 11.2 Vac output would be supremely weird.

    Per the above: “AC16V 700mA 11.2VA”

    Confirmed by: 16*0.7 = 11.2 VA
     
  10. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,778
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Perhaps not design flaw, but bad design decision to gain what? A few db in lower noise floor. Too much of an assumption that people buying the O2 would only use portable low voltage out DACs. Compounded by the fact that the O2 really isn't all that portable. Bad engineering to make assumptions like this, especially for consumer gear.
     
  11. BenjaminBore

    BenjaminBore Friend

    Pyrate
    Joined:
    May 23, 2016
    Likes Received:
    2,842
    Trophy Points:
    93
    Location:
    London, UK
    @ultrabike Be that as it may mains power still provided more oomph and less clipping than the dual 8.4v PP3s, and battery mode shouldn’t have exhibited this behaviour either. Just tracked down an old email. After a short time Epiphany Acoustics (now defunct) set the gain to 1x and 5x on them, mine included. With that and if @monacelli is correct it really should have been less of an issue rather than more. @Poleepkwa could be on to something. Maybe they were naughty and many never got a real O2.

    Found an old photo I took:
    [​IMG]

    An aside about those PP3s. Apparently you have to stick to using 8.4v as the O2 expects to see certain voltage levels to manage the smart charging circuit. So anyone with ideas like I had to juice it up with a pair of 9.6v might not want to be doing that. Same goes for 7.2v. Also, if I’m not mistaken, the hp out would occasionaly audibly misbehave if run without any batteries when mains powered and require a power cycle.
     
    Last edited: Dec 2, 2018
  12. tomchr

    tomchr MOT - Neurochrome

    Joined:
    Oct 28, 2016
    Likes Received:
    513
    Trophy Points:
    93
    Location:
    Calgary, Canada
    Home Page:
    Precisely. With the gain stage after the volume pot, you amplify the noise of the pot. In a low-noise amp, the volume pot is generally the dominant noise source, so having the pot after the gain stage makes perfect sense, assuming the gain stage can be bypassed or set to unity gain. Without the bypass/unity gain option, you won't be able to use the amp with sources that provide significant output voltage. That would be a design flaw in my opinion.
    I generally design circuits such that both pro sources (up to 8 V RMS) and consumer sources (down to 900 mV RMS) can be used.

    Tom
     
  13. tomchr

    tomchr MOT - Neurochrome

    Joined:
    Oct 28, 2016
    Likes Received:
    513
    Trophy Points:
    93
    Location:
    Calgary, Canada
    Home Page:
    Hi-Z won't strain the amplifier much. 32 Ω or 50 Ω are more telling tests. The THD often degrades under heavier load.

    For the max power spec, I generally use 0.1 % THD+N as the "onset of clipping" threshold. Some manufacturers use 1% or even 10% for this.

    Tom
     
  14. Azimuth

    Azimuth FKA rtaylor76, Friend

    Pyrate
    Joined:
    Nov 9, 2015
    Likes Received:
    6,840
    Trophy Points:
    113
    Location:
    Nashville, TN
    Home Page:
    Subjectively, to me, I found best gains were at 1x (unity) and 2.5, even when used with HD650's. It gave more potentiometr resolution in the bottom end of the pot and gets it out of the channel imbalance section. Not much difference between those two gain settings, but I used 1x most of the time and 2.5 for quieter tracks like light jazz or classical.

    Also, subjective impressions of the O2 can vary widely with different op amps. I found the stock chips bland and strident and at times kind of dry. It is all in my O2 op amp swapping thread:

    https://www.superbestaudiofriends.org/index.php?threads/o2-op-amp-rolling.2109/

    Eventually I moved on, but I think the op amp selection should be more selective than the lowest measured noise possible.

    Also on that thread you will find that if you want a good and well designed op amp driven project, Pete Millet's Butte amp fits that category. I dove head first into a Dynalo, lol.
     
  15. ultrabike

    ultrabike Measurbator - Admin

    Staff Member Pyrate MZR
    Joined:
    Sep 25, 2015
    Likes Received:
    8,960
    Trophy Points:
    113
    Location:
    Irvine CA
    AFAIK VAC is RMS.

    The Epiphany wall-wart is 11.2VAC (RMS), 16 V peak. The O2 needs 14 to 20 VAC (RMS). So the Epiphany wall-wart seems inadequate.

    Yup. It is a design decision to get a lower noise floor.

    However, one can use DACs with line out up to 2.65 Vrms on low gain, which should cover many non-low voltage DACs. Specially considering the amp is single ended. All Schiit DACs put out about 2.0 Vrms single ended.

    The O2 is indeed not really all that portable if having a phone in mind. It is also not USB powered, which would constraint performance. As such, it is more of a small desktop amp. A Magni equivalent.

    Having a unity gain option (aka no-amp all current-buffer) would be nice. There is an O2 option with 1x gain (low) and 3.3x gain (high) from Massdrop.

    I picked the standard 2.5x and 6.5x gain. Using it in low gain w/o issues so far. Haven't seen many 8 Vrms DACs. I don't have one such DAC anyhows.
     
    Last edited: Dec 2, 2018
  16. ultrabike

    ultrabike Measurbator - Admin

    Staff Member Pyrate MZR
    Joined:
    Sep 25, 2015
    Likes Received:
    8,960
    Trophy Points:
    113
    Location:
    Irvine CA
    BTW guys, if you don't care about the O2, that is fine. But if you have one, and have problems with it, maybe some of the posts here can help.

    I will also try to make some further measurements with loads, and then move on to other stuff I have here and there.

    Also, don't be concern to ask or participate in this thread, regardless of how you feel about this amp.
     
    Last edited: Dec 2, 2018
  17. monacelli

    monacelli Friend

    Pyrate
    Joined:
    Jul 23, 2017
    Likes Received:
    1,649
    Trophy Points:
    93
    Location:
    The Land of Enchantment
    What I'm saying is that the 11.2 VA figure is not a voltage specification at all. It's an apparent power specification in Volt-Amperes. If the power handled by the transformer is entirely real, this is equivalent to 11.2 Watts. Both of the other two figures, specifying the voltage and current, are in RMS. There's nothing wrong with the transformer, and it's actually got a higher current rating than the Schitt offering.

    16 Vac (RMS) * 700 mA (RMS) = 11.2 VA

    That's a continuous rating, and none of those are peak figures. I don't care about the O2, but the analysis about the transformer seems incorrect to me.
     
  18. ultrabike

    ultrabike Measurbator - Admin

    Staff Member Pyrate MZR
    Joined:
    Sep 25, 2015
    Likes Received:
    8,960
    Trophy Points:
    113
    Location:
    Irvine CA
    I understand what you are saying. Maybe the amp was defective then. The O2 should not clip with 2 Vrms line inputs on "low gain" of 2.5x as far as I can tell and given measurements, assuming a 16VAC/700mA amp.
     
  19. Azimuth

    Azimuth FKA rtaylor76, Friend

    Pyrate
    Joined:
    Nov 9, 2015
    Likes Received:
    6,840
    Trophy Points:
    113
    Location:
    Nashville, TN
    Home Page:
    I'd personally like to see some other measurements with some other op amps, but that is just me. I can probably supply a few to send you.
     
  20. tomchr

    tomchr MOT - Neurochrome

    Joined:
    Oct 28, 2016
    Likes Received:
    513
    Trophy Points:
    93
    Location:
    Calgary, Canada
    Home Page:
    If the O2 is working, you should have no trouble getting over 2 V RMS into even a 20 Ω load. Mine measured nearly 5 V RMS before clipping into a light load (300 Ω). I ran it on the included batteries.

    Tom
     

Share This Page