Amp topologies and their effects on sound

Discussion in 'General Audio Discussion' started by thegunner100, Nov 15, 2016.

  1. thegunner100

    thegunner100 Hentai Master Chief

    Pyrate
    Joined:
    Sep 25, 2015
    Likes Received:
    2,461
    Trophy Points:
    113
    Location:
    NYC
    Ever since the the big LA meet and my experience with the Sonett 2 and First Watt F3 over the past 2 weeks, I've been wondering about amp topologies. It's not the obvious solid state vs tube differences that are easily generalized. There are some questions I want to ask of the amp designers and owners who are experienced with various topologies. I think this was discussed to an extent back on Changstar but it's hard to dig up old content.

    1. Are there distinctive sound qualities that one can attribute to a certain topology?
    I'm thinking along the lines of OTL, Transformer coupled, push-pull, etc for tubes; JFETs and opamps for solid state. DNA likes to use a SET transformer coupled design with zero feedback, EC also uses transformers with the exception of the ZD and BW, and ECP likes to experiment with crazy designs.

    2. Similarly to question 1 but more specifically, I find that simple circuits and designs (Sonett 2 and FW F3) produce a very resolving but pleasing sound to listen to. What is it about those simple designs that make them sound good?
     
  2. murray

    murray Friend

    Pyrate
    Joined:
    Sep 29, 2015
    Likes Received:
    478
    Trophy Points:
    63
    Location:
    N.Z.
    Significant differences can be heard between low/no feedback and high feedback topologies. The Pass SS designs are low feedback. Op-amps, by their nature, are high feedback. Feedback is applied to reduce many types of Harmonic distortion and hence make the circuit measure better. However, feedback may have an undesired effect on transient signals.
    I built a Pass Zen amp years ago. It is a single stage SS class-A amp with very little feedback. It had a very pleasant sound, but I'm pretty sure it would have measured poorly.
     
  3. bazelio

    bazelio Friend

    Pyrate
    Joined:
    Nov 16, 2015
    Likes Received:
    3,417
    Trophy Points:
    113
    You might add Class A, Class AB, etc operation to your list. Theoretically we should be able to find Class A OTL for headphone usage, since power can be appreciably low? Anyhow, I'm not going to participate in the thread and hope not to see gross generalizations. So just beware of probable shitposts here. I'd like to see inputs from @dsavitsk @tomchr @Donald North @Marvey et. al. here though.
     
  4. thegunner100

    thegunner100 Hentai Master Chief

    Pyrate
    Joined:
    Sep 25, 2015
    Likes Received:
    2,461
    Trophy Points:
    113
    Location:
    NYC
    Yeah, I didnt know all of the different amp topologies so I just listed the ones that popped up first in my head. No doubt there will be shitposts but id love to hear some both technical and anecdotal responses.
     
  5. thegunner100

    thegunner100 Hentai Master Chief

    Pyrate
    Joined:
    Sep 25, 2015
    Likes Received:
    2,461
    Trophy Points:
    113
    Location:
    NYC
    @Marvey I noticed that you mentioned SET amps in the Audio 101 and the 3 speakers shootout thread. Could you expand a little bit more about SET amps?
     
  6. Pyruvate

    Pyruvate Friend

    Pyrate Contributor
    Joined:
    Sep 29, 2015
    Likes Received:
    1,425
    Trophy Points:
    93
    Location:
    Here
    @Marvey wrote this in regards to the BH SEX and driving fullrange speakers on another thread and I decided to repost this here since I feel that it is relevant to this discussion.

    What is it about the parafeed topology that saves cost? I'm curious because a lot of the "higher end" bottlehead amplifiers use parafeed which Doc really hypes about. Also, I recall that parafeed is also found in the Torpedo III and the highly acclaimed ECP L2 headphone amp.
     
    Last edited: Nov 18, 2016
  7. murray

    murray Friend

    Pyrate
    Joined:
    Sep 29, 2015
    Likes Received:
    478
    Trophy Points:
    63
    Location:
    N.Z.
    Interesting that you mention Bottlehead's S.E.X. Amplifier. They have only just announced the new version 3 S.E.X. I was keen to get one (ver 2) but then they discontinued it. So this is good news for me.

    I answer to your question, I don't claim to know the full answer, but I would presume that the parafeed output transformers are much cheaper as they don't need to take the full DC bias. On the other hand, the parafeed circuit requires a sizable choke and a high quality high voltage parafeed capacitor, so I'm not sure if this cancels that savings out. There are other quoted advantages for parafeed that improve sound quality, so hence my interest.

    Edit: Sorry, I hadn't realised that @Marvey had already posted about the new S.E.X. v3 amplifier release. I have only just now backtracked to that post (curiosity), and seen it is in another thread that I had not seen before.
     
    Last edited: Nov 18, 2016
  8. murray

    murray Friend

    Pyrate
    Joined:
    Sep 29, 2015
    Likes Received:
    478
    Trophy Points:
    63
    Location:
    N.Z.
    I am planning for my 2017 project to be the S.E.X. (ignoring all the half finished projects lying around!). I am equipped to do measurements, so might have something to post next year. I am keen to compare it to my 27 year old Tim De Paravicini designed Musical Fidelity B1 (class AB integrated).
     
  9. dsavitsk

    dsavitsk Friend

    Pyrate
    Joined:
    Oct 2, 2015
    Likes Received:
    1,616
    Trophy Points:
    93
    Home Page:
    There is no correct answer here. In general, single ended amps as opposed to push-pull amps were developed to be cheaper, not better. Single ended really only became seen as being better with their revival in the late 80's.

    Within SE amps, parafeed offers the advantage of a smaller OPT which does not carry DC meaning that it is both cheaper and of higher quality. But single feed has the advantage of being able to swing above the PS. A single feed can get colse to 25% efficiency. In order to really take advantage of the parafeed advantage you need to load the output tube with a current source, but that means you need to effectively double the voltage and double heat for the same output power, meaning 12% efficiency. The solution is to use a choke, but then you have the same issue as with single feed which is a big piece of iron that needs to handle DC. But you add to that weird circuit resonances that need to be handled. Both single feed and parafeed can be done well - the devil is in the details.

    My view is that parafeed makes more sense in small signal circuits - like the L2 where the gross inefficiency is less of an issue. (The "parafeed" in the T3 is a totally different thing and part of a different discussion.) And the real way to get the advantage of a smaller higher quality output transformer while not sacrificing efficiency is with push pull. Push pull has a bad reputation, but really only because of class B and ultralinear and high feedback circuits. But class A differential triode push pull is about as good as it gets.
     
  10. bazelio

    bazelio Friend

    Pyrate
    Joined:
    Nov 16, 2015
    Likes Received:
    3,417
    Trophy Points:
    113
    In the case of Cinemag, do their cores have limitations that you've not seen in Lundahl in terms of saturation? I get the impression Cinemag favors capacitor coupled in order to avoid gapping with steel laminations which would increase THD and hurt bandwidth. But I don't know why, unless for some reason they don't have access to the same core materials that others are using.
     
  11. sphinxvc

    sphinxvc Gear Master (retired)

    Staff Member Pyrate BWC
    Joined:
    Sep 28, 2015
    Likes Received:
    3,320
    Trophy Points:
    113
    Really good article, @dsavitsk, thanks for sharing.
     
  12. batriq

    batriq Probably has made you smarter

    Pyrate
    Joined:
    Oct 31, 2015
    Likes Received:
    1,041
    Trophy Points:
    93
    I wouldn't use the SEX to power speakers if I didn't have to. I have the SEX 2.1 and I tried it with efficient speakers, but it sounded veiled. The comparison though was with a 211 SE amp (with big iron), so perhaps slightly unfair. At the time, iirc, PJ mentioned it's hard to compete with a DHT. For headphones though, the SEX is very powerful.
     
  13. murray

    murray Friend

    Pyrate
    Joined:
    Sep 29, 2015
    Likes Received:
    478
    Trophy Points:
    63
    Location:
    N.Z.
    My reason for choosing it is that I have no basis for comparison of Triode SE amplifiers, and so wish to have a base line to understand its pros and cons. The closest I have got is building a Single-Ended Pass Zen MosFET amplifier.
    1. Preferably minimal cost & minimal time, otherwise need to save up all year.
    2. Best to have a kit of parts so I can hope to finish it in finite time.
    3. Prefer to have something that suits headphones and speakers: more things to try.
    4. I typically listen to speakers at way less than 1 Watt max, so 2 Watts max is OK.

    If enough of you guys (first @Marvey and now you) convince me that it is a bad choice I will reconsider. Obviously over at the Bottlehead forum they must have all drunk the Kool Aid.
     
  14. Pyruvate

    Pyruvate Friend

    Pyrate Contributor
    Joined:
    Sep 29, 2015
    Likes Received:
    1,425
    Trophy Points:
    93
    Location:
    Here
    [​IMG]

    Haha Kool Aid making these speakers singgg. :confused:
     
  15. ultrabike

    ultrabike Measurbator - Admin

    Staff Member Pyrate MZR
    Joined:
    Sep 25, 2015
    Likes Received:
    8,960
    Trophy Points:
    113
    Location:
    Irvine CA
    Dunno if it's Kool Aid, bit if it is, it's good looking Kool Aid.
     
  16. dsavitsk

    dsavitsk Friend

    Pyrate
    Joined:
    Oct 2, 2015
    Likes Received:
    1,616
    Trophy Points:
    93
    Home Page:
    A big topic that can't possibly be covered in a single post. I don't think Lundahl has access to different materials, I think they just have different philosophies on how to design and build transformers. Lundahl is very unusual in how they do things, Cinemag is much more traditional.

    Adding a gap actually tends to linearize transformers at the cost of lower inductance. Indeed, you generally want to shoot for the lowest inductance you can as more inductance means more turns which means more copper losses, more capacitance, more distortion, etc. It is why a crappy 600 Ohm transformers sounds better than a great 10K Ohm in an application where both will work. Sometimes it makes sense to interleave, however, in order to get the inductance you need on a smaller core. A very fun thing to play with is the Intact ACV as you have the ability to restack the lams in different patterns. It is a great way to start to explore some of the tradeoffs.

    Really, transformer design involves evaluating tradeoffs between numerous variables. Talk to 3 transformer winders and you'll get 6 opinions on how to do it. There are certain applications where I like Lundahl and feel they are as good as it gets. But they can be quite costly and they are not always cost effective. Indeed, there are other places I prefer Cinemag. I have also used plenty of Electra-Print, Onetics, Magnequest, etc. parts and they all make great transformers.
     
  17. batriq

    batriq Probably has made you smarter

    Pyrate
    Joined:
    Oct 31, 2015
    Likes Received:
    1,041
    Trophy Points:
    93
    These are all good reasons to get the BH SEX. Again, my comparison was very unfair since the 211 amp, if I were to estimate its cost, is probably 10x the price of the SEX. But if your primary use case is speakers, then I would recommend going for the Stereomour kit assuming the funds allow it. BH almost always have a 10-15% Black Friday sale. In my case, I ended up using the SEX solely for headphones, and I slightly regretted not getting the Stereomour instead.
     
  18. batriq

    batriq Probably has made you smarter

    Pyrate
    Joined:
    Oct 31, 2015
    Likes Received:
    1,041
    Trophy Points:
    93
    These are pretty awesome speakers!
     
  19. Cspirou

    Cspirou They call me Sparky

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    8,200
    Trophy Points:
    113
    Location:
    Northwest France
    Might want to add pentodes vs triodes, ultra linear vs triode, voltage amp vs transconductance.
     
  20. Maxx134

    Maxx134 Dunning–Kruger effect poster boy

    Pyrate
    Joined:
    Oct 27, 2015
    Likes Received:
    818
    Trophy Points:
    93
    Location:
    NYC
    I have heard a small bottle head sex amp and although it was nice,
    I did not like the topology of using huge output capacitors.

    At the time I owned woo wa7 which I felt sounded superior.

    One thing I have consistently noted about most tube amps in general was the ability to play better at low levels,
    And the soundstage always being superior in many different aspects from holography to soundstage ambient cues...

    Just recently I found this link explaining some aspects of tube superiority over solid state:

    http://kenrockwell.com/audio/why-tubes-sound-better.htm

    It's a bit long winded but very insightful.
    :)
     

Share This Page