+ Reply to Thread
Results 1 to 8 of 8

Thread: Wisdom needed on amp design and theory...

  1. #1
    Join Date: May 2008

    Location: Well below the Mason-Dixon line

    Posts: 370

    Default Wisdom needed on amp design and theory...

    There has been a long-standing and often-repeated conventional wisdom over on Head-fi, the audiophile headphone forum, that all dedicated headphone amps are almost automatically superior to the headphone sections of all receivers and integrated amps, because sometime in the 80s, manufacturers stopped using the main amps, stepped down through a nest of resistors, to power headphone sections and went, instead, to cheap op-amp based headphone sections that were, more often than not, stuck in as an afterthought.

    There has been some dissension in the ranks lately as a few newbies have bought headphone amps and reported back that they, in fact, didn't sound better than the headphone jacks on their regular hifi equipment and they felt they had wasted their money. Recently a member did the research to test the conventional wisdom. He wrote to Denon, Yamaha, NAD, Cambridge Audio, Onkyo, Marantz, Harmon Kardon, and Pioneer. He has heard back from all but Onkyo. Every one of them say their headphone sections are powered by the main amps, stepped down through a nest of resistors.

    Ooops. Conventional wisdom = myth.

    Now, of course, the true believers are flooding into the mythology gap to find other reasons why their investments in $400 headphone amps yielded dramatically different results than simply plugging into their $500 integrated amp. The argument now seems to be that headphones and speakers have very different requirements because they present very different loads, and that an amp designed for speakers isn't appropriate for cans.

    That's completely possible, I suppose. But it could also be that the difference between a great-sounding headphone amp and an average-sounding receiver is much more mundane -- the elegance and simplicity of design; the quality of components and execution.

    Can anyone help me here? Does a 300 ohm headphone need a fundamentally different amp design than an 8 ohm set of speakers? If the theoretical quality was equal, would stepping a speaker amp's output down through a nest of resistors cause those headphones to sound and behave differently than they would when being driven by a headphone amp that only puts out milliwatts in the first place?

    I'd really like to know, because I listen through cans a lot and a headphone amp is one of the items on my list of wants. But if I can get similar results from the headphone jack of a Yaqin or Glow, or even a NAD, it sure would be a more effective use of the money.

    Tim

  2. #2
    Join Date: Jun 2008

    Location: London

    Posts: 228
    I'm Rob.

    Default

    Interesting subject.

    I think much of the difference people hear between headphone amps/stages is down to output impedance. The effects of varying output impedance can easily be shown and heard and it varies hugely between stages.
    A phone that sounds forthright and direct on one stage can sound recessed and soft on another because the output impedance is altering the transfer function in a non-linear way - certain frequencies are boosted while others are reduced.

    Keith Howard did some testing on this recently in Hi-Fi News and the results were quite dramatic with some phones.

    When looking at headphones, as with loudspeakers it isn't always the headline average impedance figure that is of primary importance but the trends in the impedance curve. A phone that is say 8 ohms at 100hz might reach 30 ohms at 1khz and unless used with an amp having low output impedance it will sound quite different.
    Last edited by RobHolt; 26-06-2008 at 18:03.

  3. #3
    Join Date: May 2008

    Location: Well below the Mason-Dixon line

    Posts: 370

    Default

    Quote Originally Posted by RobHolt View Post
    Interesting subject.

    I think much of the difference people hear between headphone amps/stages is down to output impedance. The effects of varying output impedance can easily be shown and heard and it varies hugely between stages.
    A phone that sounds forthright and direct on one stage can sound recessed and soft on another because the output impedance is altering the transfer function in a non-linear way - certain frequencies are boosted while others are reduced.

    Keith Howard did some testing on this recently in Hi-Fi News and the results were quite dramatic with some phones.

    When looking at headphones, as with loudspeakers it isn't always the headline average impedance figure that is of primary importance but the trends in the impedance curve. A phone that is say 8 ohms at 100hz might reach 30 ohms at 1khz and unless used with an amp having low output impedance it will sound quite different.
    OK. To the extent that I understand it, this is no different with speakers. Your amp probably has an output of 8 ohms at the speaker terminals, and while your speakers are rated at 8 ohms average, the actualy impedance curve can vary pretty wildly, presenting a pretty challenging load to the amp. This is certainly the case with my favorite phones, high-ohm Senns. While they are rated at 300 ohms the impedence curve for HD600s, for example, looks like this:



    So what design criteria/specs would one look for to handle such a load? Do I simply look for a headphone amp or head stage in an integrated that is up near the top of the impedence curve? How might that effect the sound if I plugged in a very low impedance set of cans?

    Tim
    Last edited by tfarney; 26-06-2008 at 18:47.

  4. #4
    Join Date: Jun 2008

    Location: London

    Posts: 228
    I'm Rob.

    Default

    That's a high impedance headphone and it will be less affected by the output impedance of typical amps. Pretty much anything half decent should drive it with very little change to the phone's basic tonality.
    !0:1 is a good rule of thumb so given that the headphone doesn't drop below 300 ohms I'd ensure that the driving amp is <30 ohms output impedance.

  5. #5
    Join Date: Jan 2008

    Location: Central England

    Posts: 2,932

    Default

    Hi Rob.

    Welcome to the Art of Sound.

    I always though headphone amps were basically for folks without a headphone socket on their amp first and foremost. If they sounded better after that it was a bonus.

    I have a Rega Ear headphone amp from 2001 (the one that accompanied the Clamshell range) for sale that I no longer need. It used to drive Sennheiser HD600 headphones, now sold, rather well.

  6. #6
    Join Date: Jun 2008

    Location: London

    Posts: 228
    I'm Rob.

    Default

    Quote Originally Posted by Steve Toy View Post
    Hi Rob.

    Welcome to the Art of Sound.

    I always though headphone amps were basically for folks without a headphone socket on their amp first and foremost. If they sounded better after that it was a bonus.

    I have a Rega Ear headphone amp from 2001 (the one that accompanied the Clamshell range) for sale that I no longer need. It used to drive Sennheiser HD600 headphones, now sold, rather well.
    Thanks Steve.

    The Rega ear is a nice stage and would drive just about any 'phone well.

    Some integrated amps (and pre amps) include a headphone socket but these vary in quality for one main reason. Often they are simply a feed taken from the main speaker line and attenuated. These often have high output impedance and as a result give varying performance depending on the 'phones used. Some however drive the headphone socket via a dedicated circuit and those are usually better.

  7. #7
    Join Date: May 2008

    Location: Well below the Mason-Dixon line

    Posts: 370

    Default

    Quote Originally Posted by RobHolt View Post
    Thanks Steve.

    The Rega ear is a nice stage and would drive just about any 'phone well.

    Some integrated amps (and pre amps) include a headphone socket but these vary in quality for one main reason. Often they are simply a feed taken from the main speaker line and attenuated. These often have high output impedance and as a result give varying performance depending on the 'phones used. Some however drive the headphone socket via a dedicated circuit and those are usually better.
    Interesting. The story over on Head-fi has been the opposite, that most modern receivers and integrated amps have dedicated headphone circuits that are based on op amps and have little power behind them and are, therefore, inferior to both dedicated headphone amps and older integrateds/receivers with the main speaker amps stepped down through resistors...particularly with "hard to drive" phones like high-ohm Senns.

    I'm not saying that's accurate. Head-fi is full of contradictions, mythology and hyperbole. I'd really like to find out for myself what to look for. I'm a frequent headphone listener and I'd love to be able to upgrade my speaker and headphone amp with a single purchase. I would assume that in tube amps, a separate circuit powered by op amps would not be the solution, but the main amps stepped down. Is that a fair assumption? And is it also a fair assumption that if I look for headphone sections with out puts at or around 30 ohms, that if I like the sound of the amp, I'll like it through phones?

    ON EDIT: Rob, a lot of amps list a wide range (32 ohms to 600 ohms is not uncommon) for headphone output impedance. I assume they're talking about headphone impedences the amp is appropriate to drive?

    Tim
    Last edited by tfarney; 27-06-2008 at 13:15.

  8. #8
    Join Date: Jan 2008

    Location: Wrexham, North Wales, UK

    Posts: 110,012
    I'm AudioAl'sArbiterForPISHANTO.

    Default

    Hi Rob,

    I echo Steve's comments - welcome to the forum

    Please continue to offer your thoughts and opinions on all aspects of hi-fi. When reading your posts on PFM I often agree with much of what you write, although perhaps not quite so much when it comes to the subject of cables

    Can you tempt Markus to venture over, too? I enjoy his Teutonic ramblings

    Marco.
    Main System

    Turntable: Heavily-modified Technics SL-1210MK5G [Mike New bearing/ETP platter/Paul Hynes SR7 PSU & reg mods]. Funk Firm APM Achromat/Nagaoka GL-601 Crystal Record Weight/Isonoe feet & boots/Ortofon RS-212D/Denon DL-103GL in Denon PCL-300 headshell with Funk Firm Houdini/Kondo SL-115 pure-silver cartridge leads.

    Paul Hynes MC head amp/SR5 PSU. Also modded Lentek head amp/Denon AU-310 SUT.

    Other Cartridges: Nippon Columbia (NOS 1987) Denon DL-103. USA-made Shure SC35C with NOS stylus. Goldring G820 with NOS stylus. Shure M55E with NOS stylus.

    CD Player: Audiocom-modified Sony X-777ES/DAS-R1 DAC.

    Tape Deck: Tandberg TCD 310, fully restored and recalibrated as new, by RDE, plus upgraded with heads from the TCD-420a. Also with matching TM4 Norway microphones.

    Preamps: Heavily-modified Croft Charisma-X. LDR Stereo Coffee. Power Amps: Tube Distinctions Copper Amp fitted with Tungsol KT-150s. Quad 306.

    Cables & Sundries: Mark Grant HDX1 interconnects and digital coaxial cable, plus Mark Grant 6mm UP-LCOFC Van Damme speaker cable. MCRU 'Ultimate' mains leads. Lehmann clone headphone amp with vintage Koss PRO-4AAA headphones.

    Tube Distinctions digital noise filter. VPI HW16.5 record cleaning machine.

    Speakers: Tannoy 15MGs in Lockwood cabinets with modified crossovers. 1967 Celestion Ditton 15.


    Protect your HUMAN RIGHTS and REFUSE ANY *MANDATORY* VACCINE FOR COVID-19!

    Also **SAY NO** to unjust 'vaccine passports' or certificates, which are totally incompatible with a FREE society!!!


+ Reply to Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •