 
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
|
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
64.13.14.96
In Reply to: RE: Distortion and Audibility...replay. posted by morricab on June 18, 2008 at 15:42:23
Morning!
Your post has been rattling 'round in my head all night. Guess I'm trying to sort out how much you blame global feedback and how much you blame 'highly non-linear devices' as the culprit for not enjoying solid state amplifiers as much as SET's. In another of your posts you mention that rather than enjoying the distortion of tubes that it's more the case that it's a tolerable price to pay for their low-level linearity. (My words, I'm too lazy to do a good clip and paste).
I'm curious if you have listened to, for lack of a better description, SET amplifiers where the 'T' stands for transistor rather than tube? Having never heard an SET, I'm really trying to learn vicariously why people like what. Since too many people that should know better prefer them, they can't be dismissed out of hand.
But what is the dominant factor?
-Ultra clean performance at low levels.
-"Built-in" peak compression.
-Mild signal 'enhancement'.
-Moderate damping factor.
-Natural sounding clipping.
-Low EMI susciptibility.
-No global feedback.
-Easy grid drive for preceding stages.
-The warm nostalgic flow of the filament.
-Watching the blue glow on the envelope.
-Being X-Rayed by the secondary emissions.
Now I don't expect you to have all the answers, but with your experience I'd settle for informed opinion.
Here's why I'm interested: I'm old enough that I come from the tube era, and I'm loath to go back for various and sundry reasons. I believe that there are no perfect devices, designs or systems and the best are those that have the highest strength/weakness ratio with respect to the user's requirements. Not only that but I don't regard transistors or op-amps as nasty little distortion generators and do believe in feedback. So there! Gotta get my biases clear. However I should clarify that it's the concept of feedback that I believe in. Whether it's a force for good or ill depends...
That being said, I find myself getting more and more interested in messing about with amplifiers which doesn't make much sense as I'm happy with the ones I have. Corrupted by AA? Probably. I'm starting to get the niggly notion that I might be missing something. It's also a matter of intellectual curiousity, Andy_c sucked me in with his wretched Spice models and now I keep noodling over how good of an amplifier can be built without any global feedback. Just because. I know many others have slid down this treacherous slope so I may be doomed to tinker.
Looking at all this from another angle, my main amplifier probably is similar to the Threshold that you had. It sounds good to me and measures as a serious, wide-bandwidth voltage source. But you have moved on because other designs sound more like "live music". In what manner?
Sorry for the rambling post, advice welcome.
Rick
Follow Ups:
"I'm curious if you have listened to, for lack of a better description, SET amplifiers where the 'T' stands for transistor rather than tube? Having never heard an SET, I'm really trying to learn vicariously why people like what. Since too many people that should know better prefer them, they can't be dismissed out of hand"
Yes I have heard some of the Pass Labs SET (T for transistor) amps and while I think they were very good in some ways (I am talking about the older Aleph series and not then newer First Watt stuff, which I would love to hear!) I think they were also compromised by using too much negative feedback.
Now my current amps are using single ended (and single devices even!) transistors for the input and driver stage and then feeding a custom designed tube for the output (KR Audio makes their own tube designs including the manufacturing as well as the output transformers.). So there is no negative feedback in the circuit so I am getting in these amps a dosage of JFET and MOSFET Class A "signature" (although this will be a minor component compared to the output stage I think...but I could be wrong). Now FETs run Class A with no negative feedback behave roughly quadratic so one would expect only even and low order harmonics to be present (based on the modeling of such devices by Boyk and Sussman). In other words largely inaudible up to perhaps several % with real music.
So I am not necessarily against transistors per se, I am against unnecessary complexity of circuits that don't lead to greater linearity of the inherent circuit (mismatching, thermal tracking issues etc. etc.). Negative feedback is used to clean these circuits up because without it they would either be completely unstable or have quite high distortion.
You have to be careful with transformer coupled amplifiers (be it tube or transistor). If the output transformer saturates in the bass (and nearly all do from the measurements I have seen whether PP or SET) then you will get highly distorted bass (audibly so). One of the main strengths traditionally granted to SS amps is their bass tightness. This is generally attributed to the high damping factor. However; I know of at least two amps with low damping factor that have superb control over any kind of bass arrangement. The one thing these amps have in common is lack of distortion in the bass, meaning the output transformers are not saturating. This coupled with an article I read in Audioholics about damping factor, which basically shows that damping factors down to 5 are not significantly less controlled than essentially infinite damping factor.
Read this article:
http://www.audioholics.com/education/amplifier-technology/damping-factor-effects-on-system-response
For SETs this is the biggest weakness because they put the biggest demand on the output transformer. It takes a hell of an output transformer not to saturate to avoid the sloppy, slow and weak sounding bass that many SETs suffer from. But its not the circuit itself that is the problem as much as it is this necessary evil coupling device in between.
For an example of what I mean by transformer saturation look here:
http://www.soundstagenetwork.com/measurements/amplifiers/esoteric_a100/
Look at chart 3.
or here:
http://www.soundstagenetwork.com/measurements/amplifiers/hyperion_ht88/
again Chart 3.
and a SET for good measure:
http://www.soundstagemagazine.com/measurements/song_sa300_mb/
Chart 3
Here is a very good sounding SET:
http://www.soundstagemagazine.com/measurements/wyetech_topaz_572b/
Chart 3. Notice that up to about 10 watts the distortion rise is arguably below audibility at these very low frequencies.
And here is one of the champions of tube sound from Lamm:
http://www.soundstagemagazine.com/measurements/lamm_ml11/
Now look at Chart 3. Notice how flat the distortion is even right down to 20hz and also up to 20khz!! Superb and it sounds that way too.
http://www.soundstage.com/revequip/lamm_ml11.htm
"http://www.soundstagemagazine.com/measurements/manley_snapper/"
here you can see the saturation hitting hard at around 30 watts. Below that though not too bad. HF distortion doesn't look so great though.
For a final comparison of tube amps here is a good OTL:
http://www.soundstagemagazine.com/measurements/atmasphere_ma1_mkii2/
Notice also the flat distortion vs. frequency response. The Lamm is just as good even WITH an output transformer...it says a lot about that output transformer (it must cost a fortune).
"http://www.soundstagemagazine.com/measurements/monarchy_se160/"
Nice sounding hybrid amp with flat distortion at both low and high frequencies. Has a SET like distortion vs. power and harmonic structure most likely due to the fact it has zero negative feedback and only active stages in total
http://www.soundstagemagazine.com/measurements/lamm_m12_reference/
Excellent sounding hybrid. Notice modest rise in HF due to a bit of negative feedback in the cicruit. Notice also the harmonic content.
Now look at what most SS amps do:
http://www.soundstagemagazine.com/measurements/redgum_rgm175/
look at chart 3 and Chart 5. See the dramatic rise in high frequency distortion? See the picket fence of harmonics? This IMO destroys natural sound.
http://www.soundstagemagazine.com/measurements/psaudio_hca2/
Here is a Class D mess of an amp (I know I owned it once). Great for bass but sterile sounding. Note all the odd high order harmonics at high levels.
"http://www.soundstagemagazine.com/measurements/threshold_s5000e/"
Notice the worst problem in Chart 3 with HF distortion is at the lowest power! Bye Bye soundstage.
http://www.soundstagemagazine.com/measurements/simaudio_moon_w6/
Nearly Ten fold increase in distortion from 1Khz to 10kHz.
http://www.soundstagemagazine.com/measurements/anthem_statement_p2/
Like the others.
http://www.soundstagenetwork.com/measurements/amplifiers/conrad_johnson_premier_350/
This is just to illustrate that there are some fundamentally different behavior going on with these amps and tube amps (transformer coupled).
"Ultra clean performance at low levels"
For good amps yes.
"-"Built-in" peak compression"
For triode output stages apparently there is a natrual compression that offsets the increase in distortion. Increasing distortion is perceived as loudness and the compressor is working in the opposite direction (ie. making the sound relatively softer compared to the input voltage).
"Mild signal 'enhancement'"
This I don't believe at all.
"Moderate damping factor"
Not the culprit for bass response it is widely believed to be
"Natural sounding clipping"
See above regarding signal compression
"Low EMI susciptibility"
Don't know about this
"No global feedback"
Amp is not susceptible to back EMF distortion nor does it have the effect of biasing in favor (and generating) high order harmonics. IMO, very important in what is seen in the amps differences above. Note that in the case where HF distortion rises rapidly it is almost always accompanied with a sudden drop in damping factor. The negative feedback is no longer performing as it did at lower frequencies and I think that you are beginning to see (on the graph) and hear the inherent non-linearity of the underlying circuitry. This HF distortion is IMO a main culprit of SS sound.
"Easy grid drive for preceding stages"
The driver stage can make a lot of distortion in a tube amp if not done well.
"The warm nostalgic flow of the filament"
Rubbish. My Own amps you basically can't even see the filament glowing unless you look in the tiny slot.
"Watching the blue glow on the envelope"
Same as above
"Being X-Rayed by the secondary emissions"
??
" keep noodling over how good of an amplifier can be built without any global feedback"
Very well in fact.
"But you have moved on because other designs sound more like "live music". In what manner?
"
Mainly two areas, tonally and dynamically. Other amps I have had since were simply tonally more natural sounding. Also, the Threshold (and many other SS amps I have tried) sound, for lack of a better word, constipated as if the sound would burst out if you just turn it up more but it never does let loose like you expect from the live experience (unamplified that is). My current amps, even at low levels, give this alive sensation of dynamic shift that many of my previous amps did not. Maybe they were overdamped in a way.
Also, the highs are simply NOT natural with most amps and now I see from measurements a possible reason why. This I have learned negatively affects soundstage because the distortion makes the highs more prominent (remember distortion increases the perception of loudness) and pushes them more forward. High frequency attenuation is one cue for distance (the further away you are the more attenuated the highs). If the highs are perceptually pushed forward the net result is a foreshortening of the soundstage as things seem closer.
I want to point out that these comments are my thinking to date. As I observe and read more I continually refine what my position is. In other words, if someone can demonstrate to me a ultra high power, high feedback multi transistor amp that sounds as good as the best tube and hybrid amps I have heard then I will be all ears as to the explanation of why. I will also pore over the measured data with great interest. So far, I have heard the best SS amps and been left underwhelmed, although the Edge NL Reference ones were pretty good as was the darTZeel, both despite mediocre measurements but they are both doing some unusual stuff (the darTZeel has no global feedback and only six transistors per channel for example).
The amps I have found to sound the best with regard to the live concerts I have attended and recordings I have made are the ones like the Lamm ML1.1 or the Atamasphere OTL or Monarchy Audio hybrid or my own KR Audio hybrid, which measures quite similar to the Lamm.
Thanks for the detailed response and links. I was especially thinking about Pass amps when I asked the SETransitor question so I'm glad that you had.
I appreciate the transformer comments, I aim to avoid them for all the reasons you mentioned. I was frustrated as a kid because I couldn't afford decent plate to voice coil transformers. Never again.
Now I'm reflecting upon you constipation comments. I know just what you are talking about, and it was with a shock that I realized that it's one thing I never experience with my current systems but have with others in the past. I always suspected the speakers and think one of the reasons that I like acoustic suspension woofers is that they don't seem to require a minimum loudness to "turn on". I'll keep my ears open if I mess with Amps. Sounding good at all levels may be some sort of telling test.
Rick
One other thing about low level...when you are listening to wide dynamic range music, like orchestral music you often have very loud and very soft sounds being produced at the same time. Often the loud sounds will mask the soft sounds in a system that does not do low levels very well. The abillity to hear the soft sounds within an envelope of loud sounds is something that the amp designer Allen Wright refers to as downward dynamic range (DDR). It is related to low level resolution and microdynamic capability of a piece of electronics (or speaker).
Again, now that you point it out I know just what you are talking about. And those low level signals really provide a lot of the aural clues that make the difference between sounding live and canned.
I suspect this is where resonance, or stored energy whatever folks call it in speakers makes a big difference. I like film tweeters and I think one of the reasons is that they are light enough that the they have a pretty good impedance match to the air so it provides the damping. I've heard the same thing with horns, but often accompanied by other drawbacks.
Electronics wise I suspect the relevant factors are very low intermod when driving a speaker load. Any non-linearity in the whole system tends to mess up the "micro-dynamics", that has become all too apparent as I have grown older and the dreaded "cocktail party" effect has settled in. You know the one I mean, not the one where you have too many and make an ass out of yourself. That one has stayed rather constant...
Rick
"Sounding good at all levels may be some sort of telling test"
It is and it is a sure way to tell when a system is lacking in real dynamic range (from an audibility standpoint). The lower the level a system can play and stay consistent the better its lower end of dynamic range. Also, this means that its low level resolution is superior when listening at a normal volume level and you want to hear subtle details. If these drop out then the realism is dimenished.
Hi Rick,
What would be nice, is if you could pick a few of your favorite sounding cds and visit Earl's place in MI (he is hospitable to such requests) to hear the soundwaves/soundfield created by his op-amp/NFB/RS wires, etc. Summa setup. Same for SL.
Then, visit Morricab (yes, I know he is safely encapsulated abroad) or some other audiophile SET/no NFB/magic wires setup and hear the soundwaves there.
When you were done laughing, perhaps then you might reconsider the path of SETs, non-NFB, etc and whether to further pursue this, if for no other reason than amusement.
cheers,
AJ
Funny thing, in Audiophile circles if you can hear things others cannot, you are revered, in society, you are sent for treatment.
I've already got equipment with plenty of feedback and op amps and it both measures and sounds excellent. I'm not unhappy, just curious. It has both a certain technological allure and who knows, maybe it would sound even excellenter...
I don't have any "magic" wires in my system, but one of the best things I ever did was change all the interconnects to ones with very low capacitance and dielectric absorption. Perhaps that suggests that at least my line driver stages could stand some improvement. It also taught me that if a bunch of people say something is so that it may be worth looking into even if it seems implausible. And it actually seems plausible to me that a single ended class A amplifier could be better, especially at low levels, than a class B scheme.
Rick
I don't have any "magic" wires in my system, but one of the best things I ever did was change all the interconnects to ones with very low capacitance and dielectric absorption. Perhaps that suggests that at least my line driver stages could stand some improvement. It also taught me that if a bunch of people say something is so that it may be worth looking into even if it seems implausible.
I don't recall "a bunch of people" saying anything like that. Where and when? I do recall there usually being one or two people in audiophile circles having a slight comprehension of physics and perhaps measuring and understanding why high capacitance in cables wouldn't be good.
"A bunch of people" say lots of things, like cryo treating everything in sight, including photos of one's self, improves sound quality. Any plans for looking into that seemingly implausible something? Do you doubt that they are actually "hearing" these "improvements"?
And it actually seems plausible to me that a single ended class A amplifier could be better, especially at low levels, than a class B scheme.
Ya think? You know, this isn't the first time amp clipping/fidelity has been discussed here (link)
(I'll try to find the link for you to Earl's low level linearity test method (yes, unlike a chemist, he has one) over on Diyaudio.)
cheers,
AJ
Funny thing, in Audiophile circles if you can hear things others cannot, you are revered, in society, you are sent for treatment.
Friends and dealers, that sort of bunch of peoples. And it was about twenty years ago. But since I knew better I refused to even seriously look into it for years. Years in which I could likely have had better sound had I not been so arrogant and bull-headed.
You bring up an interesting point about the Cryo. processing. I actually haven't given it much thought and it's not something that I can do on the cheap in the privacy of my own home. But yes, obviously people do hear a difference so it's something to keep in mind. Just anecdotally I'm not convinced that it's always for the good. Have you tried it? Naturally I'm thinking of it's application to metallic conductors, not photos. That may make a difference also, but I suspect the mechanisms are different...
Rick
Do you know anyone with Summa speakers? I don't.
Do you know anyone with Summa speakers? I don't.
I know someone with mediocre Acoustat speakers. Powered by gutless 30 watt KR Marshalls (or is that Fender?). I laugh at them every day.
cheers,
AJ
Funny thing, in Audiophile circles if you can hear things others cannot, you are revered, in society, you are sent for treatment.
Thanks for confirming your Trollosity. My guess is that you are not familiar with either product. And you confirmed that you know no one with Summa speakers.
Funny thing is AJ, that people don't laugh when they leave my place...they often try to buy the same components.
"...and now I keep noodling over how good of an amplifier can be built without any global feedback"
The Ayre MX-R has no global feedback and has surprisingly low distortion. Its measurements are linked below.
"The Ayre MX-R has no global feedback "
If you read the comment at the end Atkinson says low overall loop feedback so apparently it has some global feedback.
Still, it is supposed to sound really good...I would like to audition it.
"If you read the comment at the end Atkinson says low overall loop feedback so apparently it has some global feedback."
I suspect that's an error on John Atkinson's part. Over at DIYaudio, Charles has been a staunch advocate of zero global feedback for years and has stated on a number of occasions that his power amps have no global feedback. See for example here . Charles is to global feedback as Todd is to ASRC :-).
Maybe but look at the THD vs. power curve
http://stereophile.com/solidpoweramps/407ayre/index4.html
There is definitely some feedback in the amp but perhaps it is only local feedback.
"There is definitely some feedback in the amp but perhaps it is only local feedback."
That's a given. Without local feedback (AKA emitter/source/cathode degeneration a la Boyk and Sussman), the circuit gain cannot be controlled accurately because it will be strongly dependent on device parameters. Emitter/source/cathode followers all have 100% local feedback, and just about every BJT power amp uses an emitter follower output stage.
If there is some manufacturer claiming no local feedback in their designs, they are probably stretching the truth at best. Charles insists on not calling local feedback "feedback", but instead calls it "degeneration". People familiar with that concept understand what he's saying, so it's not misleading in that case.
I was not referring to degeneration but I know what you mean. I am thinking that perhaps there is a loop wrapped around either the output stage or perhaps the input and driver stages but not from input to output. I know that Simaudio was touting that they didn't use any negative feedback in the output stage. DarTZeel uses only smaller loops as well.
Obviously, every SET amp that has a cathode follower as a driver stage is using degenerative feedback in that stage at the very least.
Nice, thanks Andy,
Clearly the answer is "a plenty good one".
Rick
You would think that CH would proudly display on his website the DBT results where everyone (or at least someone) heard a difference and even consistently preferred his amp over say, a Halcro.
But then that would indicate that the no NFB was somehow audible, wouldn't it.
Now wasn't that the POTY amp that also blew up?
I bet that was audible.
cheers,
AJ
Funny thing, in Audiophile circles if you can hear things others cannot, you are revered, in society, you are sent for treatment.
| FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
||||||||||||||
This post is made possible by the generous support of people like you and our sponsors:
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2025, Audio Asylum, all rights reserved ]
