Amp Design/what goes into building a great amp

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I have been an audiophile all my life--please don't hate me. I know most of the reviews in the audio rags and have heard most of their systems. I have gone to CES for 20 straight years. But I have come over to the other side--DIY. Why? Because mainstream audio is a millionaires club bereft of new ideas for the large part. I don't hear much progress of late.

I wanted to be more than a consumer and a critic.

So I read Doug Self's Book and Randy Sloan's book on amp design at the same time.

I think Randy Sloan's book is a blatant rip off of Self's book, it even borrows the same organization and adopts virtually the same opinions across the board.

That being said, I thought both books were fantastic and should be read together because randy being a teacher explains so much that Self expects the reader to have the background to understand and he is more practically oriented. I still love Selfs writing style but I can't say I understand everything he says.

But what I find appaling is that both of them are anti-subjectivists to an extreme and believe that science alone will lead to a good sounding amps. They leave no room for art and discounts the importance of careful zen like listening in amp design.

They act like you can just wipp this stuff together and achieve audio nirvana.

John Curl stated in an interview in some audio rag that he designs the circuits and leave to others more gifted to the parts selection and the execution. He appears to give equal importance to all 3 aspects.

I don't know any great sounding system that uses any of the products designed by pioneering designers of Self, Leach, Pass. Charlie Hanson of Ayre, and others. Don't get me wrong. I worship these guys but I don't think they deliver on the final product. I think they ignore 2 and 3 above.

My group of firends believe that parts selection plays a much larger role then believed by the anti-subjectivists. And by that I don't mean whether a 2sc1214 should be used rather than a 2sc1540, and I don't think we are able to measure this stuff yet. We scour the junk yards for old parts that are reputed to sound great.

I am talking about basic parts here. Transformers, chokes, resistors, capacitors, rectifiers, IEC jacks and even wire. I am partial to cloth covered wire. I listen to parts and I hate doing it it. But I know the importance of parts selection.


I am curious what all of you wonderful minds out there think.



c

e iothers to the par seletiotsame

arunderstand that xlp cir

Roa
 
What goes into a great amp ?? Initially , a few screwups (the best teacher). :)

I am talking about basic parts here. Transformers, chokes, resistors, capacitors, rectifiers, IEC jacks and even wire. I am partial to cloth covered wire. I listen to parts and I hate doing it it. But I know the importance of parts selection.

Some would disagree.. (had that told to me today - parts don't matter). Power components (trafo , main caps etc.) have a noticable effect on sound :eek: . Individual semiconductors less , if the circuit is designed to make the most out of them.

OS
 
I am am not a subjectivist, and I believe design, parts and layout (including heat sinks, and cases) all make a difference. However, the better the design, the less the parts and layout matter, but they still matter. (I would guess Self and Sloan agree with these points) Self does talk about device parameters like beta and Cob where they make an obvious difference (to go beyound that would probably fill another book). There are good transistors (tubes,caps, tranies etc) and bad transistors but I dont think a certain transistor has a type of "sound". It depends on the circuit and layout and in the case of output devices (probably the most important) interaction with the speaker/feedback network.

I believe you will be able to measure the difference between a real good amp and a really really good amp, but were not quite there yet, people are working on it. Read some of the posts on transient distortions. (some of the subjectivists are not helping by insisting that measurements are useless, and many "hi-fi" manufacturers listen to this nonsense, and spend more on marketing than R&D) This measuring is complex and goes way beyond THD and noise (which are often quoted with very little detail ie one number when they should be a bunch of curves).Would you trust a transitor that specs a beta of 100 without a temprature or current atached to the beta? yet this how most amps are sold!

My bigest problem with the subjectivists is exactly that, its subjective. What does "it sounds good" mean? It can mean different things to different people(and even the same people on different days). And you cant prove anything. Its a great cop out for amp designers too. (There a $250K tube amp that has 1% distortion and a 10db boost at 100hz, but "it sounds great") and it just might... for some people, but should I be trying to design and amp with those specs? I would be laughed out of this forum.

Amp design is still engineering, and last time I checked engineering is still mostly science, at least when you get away from "hi-fi" audio. Otherwise go out and buy yourself a $600 power cord and marvel at how good it sounds.

Sorry for venting but as an old audiophile who has recently come back to it, I am shocked at the Voodoo that has become prevailent in this field, even to people that are educated electrical engineers (my cousin for one). Maybee this is why people like Self stear clear of such topics (you just cant win), and thank God for him and his like!
 
Topologie sounds, i can even know how an amplifier will sound just watching the

schematic...and the analisis are not connected to parts or values, if not too much wrong biased and too much wrong coupled..if not too much errors into the feedback and compensation... i can forecast how the unit will play... in brazil all them sounds the same, as all them uses the Leach amplifier schematic..ahahahahah!

I am not a special mind..i am someone that has already learned a lot and concluded that i have one more ammount alike the one i have learned to learn into the future days. i am just a hard worker that has built thousands units and have listened all them and compared a lot.

The ones thinks they are special mind are the arrogant ones... so wrong they are that i could see that most of them are alike empty bags... walking calculating machines without no consistence.

But speaker changes sounds even more than amplifier topologie.

I do not believe in magic parts, speaker stands, special audio cable and special speaker cable.... and i have tested some.

The more complicated the schematic the worse is the sound is what i could perceive..but there are exceptions when complication is not made inside the audio patch.... surrouding sub circuits are

not a always a bad idea.... sometimes helps changing entirelly an amplifier character, creating a sub topologie.. a variation that changes sonics.

During some productive weeks i use to build 5 amplifiers... the average is 2 units each week and sometimes i do 8 units into a single week..this was done during 48 years long... i have 57.

Even if i was very stupid, building all those units i would learn something... and for course, as a natural consequence, i did.

I was superviser into Television Broadcasting Stations..but someone easy to find into the electronics laboratory...that was my place... were i used to be found, always building amplifiers and using the TV Station laboratory equipment....my own office was usually empty, or i was together enginneers from the operational area or into the laboratory.

Magazines, sometimes, has paid reviews.

regards,

Carlos
 
What I think is that both subjectivists and objectivists keep this maddening hobby going.
The former seek to achieve the 'perfect sound' and since this is ill defined the pursuit is never ending. The latter set themselves the task of satisfying a bunch of parameters, say various distortions which is practically endless.

One thing which does amaze me is how many different topologies can exist or whether they are actually only variations of the same theme?

The trouble is the goal of hifi sound reproduction is no longer well defined. Am I right?
 
Professor smith said:
The trouble is the goal of hifi sound reproduction is no longer well defined. Am I right?

It's a question that doesn't have a yes/no answer. You set your own goal. You can make it as woolly or as specific as you like. It's a personal thing, like many other goals in life.
 
Hi , professor .. Bigun

By P. smith - One thing which does amaze me is how many different topologies can exist or whether they are actually only variations of the same theme?

That only depends on how you define "the theme". Broadly , we seek to convert the DC of the power supply into a complex AC waveform. The power supply does not know or care what amp it is driving , so.. if it is "supersized" , an unregulated PS can be eliminated as a source of "bad sound".
The only thing left is the amp , dumping gobs of electrons into those poor speaker voice coils. Everyone has their favorite topology , I DON'T. Some are better suited for different jobs. I don't think my only blameless sounds good for listening , But as a subwoofer driver,excellent. They are variations of the same theme , they amplify.


The trouble is the goal of hifi sound reproduction is no longer well defined. Am I right?
That has different ,but well defined answers for different groups.

-- OEM's (walmart , chinese hi-fi ) :xeye: have profit as first priority with the goal of acceptable , reliable sound at the greatest (best) cost /perceived value ratio. No dual mono (better CCS in the chip) , 6-8 A bridge and 6-8000 uf X 2. Just enough so it does not blow up for a year :hot: and will match the custom speakers that go with it.
(20-40$ materials = $200-300 finished system)

-- High end - here we flirt with the subjective. More detail in the design. We see separate PS's (dual mono) , GOLD , special topologies (class A and separate output stage/ input stage supplies). In General , industrial grade components with Aesthetic considerations (fancy case). Another factor is the group these products are marketed to seem to be frivolous and subjective anyway, so a lot is spent on reviews and hype (advertising , pretty pictures). "show them pretty pictures , have someone famous tell them it is BETTER , and they will buy it"
($200 - 500 materials = $3k - 8k)

-- DIY'ers - Here we can get the best - or the worst. Hopefully, if we carefully observe the first two groups , we can better them both. Some will just consider the amp in isolation (it specs out better , so it MUST sound better). Most DIY'ers will end up with better than OEM equipment simply because this was the motive to DIY in first place. :D Others will produce at least "flat" or "sterile" sounding amps , even dangerous ones :hot: :hot: , which hopefully will trigger an increased learning curve (fix it now and make sure it never happens again .. BOY)
OS
 
ostripper said:
-- High end - [...] Another factor is the group these products are marketed to seem to be frivolous and subjective anyway
OS

It would be nice to think so sometimes, but most likely there are plenty of customers who value good well designed and well marketed goods, who appreciate their quality without having a clue how they work and can readily afford it. I'd like to be able to be a bit more frivolous myself more often ;)
 
fglabach said:
I have been an audiophile all my life--please don't hate me.

Roa


(':D')


cbdb said:
I am am not a subjectivist, and I believe design, parts and layout (including heat sinks, and cases) all make a difference. However, the better the design, the less the parts and layout matter, but they still matter. (I would guess Self and Sloan agree with these points) Self does talk about device parameters like beta and Cob where they make an obvious difference (to go beyound that would probably fill another book). There are good transistors (tubes,caps, tranies etc) and bad transistors but I dont think a certain transistor has a type of "sound". It depends on the circuit and layout and in the case of output devices (probably the most important) interaction with the speaker/feedback network.

I believe you will be able to measure the difference between a real good amp and a really really good amp, but were not quite there yet, people are working on it. Read some of the posts on transient distortions. (some of the subjectivists are not helping by insisting that measurements are useless, and many "hi-fi" manufacturers listen to this nonsense, and spend more on marketing than R&D) This measuring is complex and goes way beyond THD and noise (which are often quoted with very little detail ie one number when they should be a bunch of curves).Would you trust a transitor that specs a beta of 100 without a temprature or current atached to the beta? yet this how most amps are sold!

My bigest problem with the subjectivists is exactly that, its subjective. What does "it sounds good" mean? It can mean different things to different people(and even the same people on different days). And you cant prove anything. Its a great cop out for amp designers too. (There a $250K tube amp that has 1% distortion and a 10db boost at 100hz, but "it sounds great") and it just might... for some people, but should I be trying to design and amp with those specs? I would be laughed out of this forum.

Amp design is still engineering, and last time I checked engineering is still mostly science, at least when you get away from "hi-fi" audio. Otherwise go out and buy yourself a $600 power cord and marvel at how good it sounds.

Sorry for venting but as an old audiophile who has recently come back to it, I am shocked at the Voodoo that has become prevailent in this field, even to people that are educated electrical engineers (my cousin for one). Maybee this is why people like Self stear clear of such topics (you just cant win), and thank God for him and his like!

+2


Professor smith said:

What I think is that both subjectivists and objectivists keep this maddening hobby going.
The former seek to achieve the 'perfect sound' and since this is ill defined the pursuit is never ending. The latter set themselves the task of satisfying a bunch of parameters, say various distortions which is practically endless.

One thing which does amaze me is how many different topologies can exist or whether they are actually only variations of the same theme?

The trouble is the goal of hifi sound reproduction is no longer well defined. Am I right?

Hi fi and it's related components will always be subjectively driven , regardless of the science .......I mean look at the OP original question , bet there will be no consensus ;)
 
many "hi-fi" manufacturers listen to this nonsense, and spend more on marketing than R&D

Please don't forget that many (most?) people are not techies and do not understand tech-stuff. Otherwise everybody could build his own amps ;)

As you say yourself, THD does not tell the whole story so what would a bunch of figures improve for the customer? And if it did, you need to know the whole picture, like that 1% THD at full power is still less than what a typical speaker produces at this power.*

As for the manufacturer marketing is certainly cheaper than R&D, especially in a field where real inventions are rare (audio technology has about 100 years of history!).

That's not to say that I wouldn't like to see more plots in their documentation :D

Have fun, Hannes

* ever seen THD-plots for speakers? I know of a single manufacturer that specifies THD for this stuff (JBL pro equipment).
 
cbdb said:
Otherwise go out and buy yourself a $600 power cord and marvel at how good it sounds.
I've got to remember that one. It reminds me of an ad that I saw a while ago for high-priced gold-plated toslink cables.

As you all know, gold is a good plating material for electronics because it does not tarnish and therefore consistently makes a good connection -- electrical connection, that is.

Toslink cables are digital optical cables and make no electrical connection at all. Thus, gold doesn't help at all. One can't even argue that the gold looks good, because it's plugged into the back out of sight. But there this company was selling them.
 
You're not referring to the $600 power cord, I hope.

$600 power cord's are for the poor. $6K and you really have something.
http://www.dagogo.com/StageIIIZyklop.html
An externally hosted image should be here but it was not working when we last tested it.


Wow , handmade in the USA by greedy capitalists. :D
Seriously , the benefits to detail and soundstage that are reported would be better realized with a 1KVA industrial toroid and 100k uF of capacitance ( only $300).
OS
 
You have been reading the wrong books.

Try 'Zen and the Art of Motorcycle Maintenance' by Robert M. Pirsig, or 'The Teachings of Don Juan - A Yaqui Way of Knowledge' by Carlos Castaneda or even 'The Integration of the Personality' by C.G.Jung.

You are suffering from a fragmented world view. Aspects of your post are mutually incompatible.

w

Do not adjust your set. Normal service will resume as soon as possible.
 
It is difficult to separate the insults from the recommendations.
Most importantly, audio design is a bit, both objective and subjective. That is what makes it interesting. For example, you can make some amp or preamp that MEASURES very, very well, yet doesn't sound as good as you would presume that it should. Why? How can this be so? Yet, another amp or preamp, might sound better, yet its specs, while somewhat different, should be inaudible by every objective standard of what the ear can hear.
IF and when you have been subjected to such an experience, especially if your own design failed to satisfy critical ears, that you've grown to trust, then you go back to the 'drawing board' and hopefully use approaches that don't necessarily follow standard engineering, but are more akin to 'Zen and the Art of Motorcycle Maintenance'. If they work: Everybody loves you, thanks you, including audio reviewers, and this is progress.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.