AKSA amps

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
And Hugh says that this emitter resistance value is critical and cannot be replaced

For other values without losses.

I just cannot tell the value Hugh selected... for sure it is a commercial normal value, not special order made by Hugh...value can be found everywere.

There are a lot of small things that were founded during reseaches.

You know...very small increase here in the emitter resistances... another small increase int he feedback line.. other increase in the stop resistances...other increase in this, that and those....those things altogether are unbeatable.

Hugh is very rich related those details.... and they result special, as i could not beat his amplifier without use the same sub-circuits and parts he sent me.

My Dx amplifier turns much better using Hugh subcircuits and parts...but, of course, will be another Aksa Nirvana plus 55.....and not more Dx amplifier, not more a simple and cheap amplifier.

I have perceived bass worst using emitter resistances.... worst in level, ..but quality is another thing, very subjective and very hard to describe.

The first idea i had was that i was impressed because i have enormous respect by Hugh...also i know that i am not so experienced as he is...because he spent less years than i spent searching for perfect sonics...but he is much more organized into colecting informs and select them to tests and comparisons....i was very lost and confused doing that.

I made A to B testings.... the Aksa eated my Dx amplifier without too much problems.... loose in bass (quantity) but winned in treble, voices and sound stage...had lower distortions and overall sonics were very, very, very different.

I strongly recomend Aksa...go to listen one...and them return to tell me if i am wrong.

My dream was that Hugh, when finished the Lifeforce design, would say to me:

- "Well Carlos, this one is superated by Lifeforce and then i decided to give you all the old Nirvana board i have, as people will not order than anymore"

Ahahahah...not only he did not made that, but also he asked the Nirvana boards back as a pre condition to sell the Lifeforce...he do not want to see Nirvanas free to speculations.

regards,

Carlos
 
Richie,

I mount 0R47 emitter resistors on the outputs UNDER the pcb. They are mounted a couple of mm below the pcb and enjoy good air circulation. This is, of course, rather important at high power.

If you do the math on the optimum bias for an AB, you find the emitter resistor should drop 26mV, which is a significant figure related directly to the Bolzmann constant.

If you use a 0.22R resistor, this assumes a bias of 120mA. If you use 0.47R, the bias is 55mA. Since mosfets require at least 100mA of quiescent to place them in a reasonably linear portion of the parabolic transfer function, and I choose to use bipolars, I go for the lower bias.

You can see that to achieve high efficiency, important in this Class D switching world with low quiescent dissipation, the larger emitter resistor is preferable, BUT it costs rail efficiency, which is not so good.

OTOH, the larger resistor promotes quiescent stability, improves current sharing between parallel pairs AND linearises the output device, no bad thing either. I go for a quiescent variation of no more than 3% between complementary pairs, since I match carefully. Since we are designing for maximum sound quality, linearity is crucial and too important to pass up, so I choose 0R47. I have had no problems since the domestic situation, unlike pro-audio, is not too concerned if we drop a few watts from max power. A nice stiff power supply avoids the usual sag of the rails anyway, and largely makes up for this loss of rail efficiency.

Cheers,

Hugh
 
AKSA said:


OTOH, the larger resistor promotes quiescent stability, improves current sharing between parallel pairs AND linearises the output device, no bad thing either. I go for a quiescent variation of no more than 3% between complementary pairs, since I match carefully. Since we are designing for maximum sound quality, linearity is crucial and too important to pass up, so I choose 0R47. Cheers,

Hugh

Hugh, not understanding your comment that a larger resistor improves linearity. This is contrary to what I've read. A lower value supposedly reduces higher order harmonics of crossover distortion, and lessens gm-doubling aberations. Also, a few weeks ago, Curl stated, if I recall correctly, that higher quiescent current in the output transsistors reduced higher order harmonics (I presumed that the subject distortion was due to the transistor, per se, as opposed to crossover distortion.)
 
Hi Pooge,

I've found this gives very good sound. I can't argue too much about this, because such debates become interminable and people soon loose interest in the finer academic points, but a larger emitter resistor constitutes degeneration, reduces transconductance, and to my understanding always improves linearity by dwarfing the non-linear b/e junction with ohmic resistance.

There is some controversy on this point, with proponents invoking higher math to prove a point of some conjecture.....:clown:

gm doubling is a function of both transistors being on at the same time during crossover, and its effect is to rob very low level detail, the bane of Class AB. It can be reduced by better controlling the output of the voltage amplifier.

Cheers,

Hugh
 
Greg and Hugh,

I have not personally found any value to be the best. I have not personally tried different values and listened to them yet (or read any other subjective reports), although I just ordered new values last night to try just that. As I opted for going lower, Hugh's comment about going higher gave me a tweak.

I've been studying the situation lately just to decide what to use (and order), and opted for .22 ohm based on going as low as practical without going too low to compromize output sharing.

I was just struck by Hugh's comment that raising the value linearizes the output. Didn't know if that was a subjective or objective observation, and it struck my curiousity.

Hugh responded by stating that larger gives more degeneration. This may be true if standing currents were the same, but it is the voltage across the resistor that is fed back, and this is made virtually the same by adjustment in bias current.

As to "dwarfing the non-linear b/e junction", the intrinsic emitter resistance is an inverse function of Ic, so if the bias current is doubled, the intrinsic emitter resistance is halved. So along with a halving of emitter resistance from say, .47 to .22, the relative proportion of added resistance to intrinsic would appear to be the same, with a lower intrinsic resistance to modulate in the case of .22 ohms with higher bias.

As to gm-doubling being reduced by better controlling the output of the voltage amplifier, I can not follow this with any available knowledge. It is my understanding that lower Re reduces this.

Not holding Hugh to the fire for a rock solid explanation of cause of why he thinks the larger resistor sounds better. It's just that I can't tell when Hugh makes a statement about such things as linearity, whether it is subjective or objective, because he wears both hats (a good thing). But I never know what hat he is wearing when he makes such a statement.

If there is another explanation of why larger resistance might sound better, I am all ears, because I am happier when my right brain gets along with my left.
 
Those debates finishes normally very bad.... we have some bricks that use to enter

behaving very unkindly.

It is a pitty, but, i have learned a lot reading those debates.

I have colected many informs during those things, you have gave me those secrets for free during conversations and because of negotiations we have made about some Aksa 55......but i could perceive that some forum readers, having attention, could receive nice informs about your circuit.

During the massacre they use to make.... some clever guys had insigths and clear nice informs that pop out during the hot argumentation.

We have to leave the experience to learn things... theories and conclusions about are not enougth in audio sonics.

My amplifier had emitter resistances removed, clearly i had more "punch" in the bass...so... to reduce parts count, to make it more simple and more cheap, the emitter resistances do not exist anymore.

But finishing the unit, i have apreciated the sonics, but comparing with mine Aksas, my units loose into the comparison....so.... i conclude that researches made by Hugh, related many things, including this emitter resistances, altogether, made the difference.

There are a lot of things that do not make any sense... and sometimes those non sense things are very near us... around us....a nice man can married a very bad and crazy woman.... that's love... non sense too.

Input differential, in every theories, works better using costant current sources and line voltage regulators... some schematics using sophisticaded methods to drain charges can be used too...using Fets transistors and many other interesting parts...... the problem is that theories do not produce sound.... the amplifier with simple differential sound even better.... very strange those things...and only the A to B comparative testing, using the same circuit with both aproaches can show you how strange those things are.

You have to use Blind testings, because our "beliefs" will influence the result...not knowing what is playing (ask some help) you will be fair...cannot detect, conclude, discover, what circuit is playing...decision must be only because of sonics.

Levels must be the same...same power, same speaker, same music, same effects, same place...dark place...cannot see nothing..confortable place and repeat tests to see if they make sense....... also change letters of reference.... invert signals... try to confuse every clever brain not to discover what is A ...and what is B.

Also do not loose time searching because or curiosity...only Aksa owners knows those circuits, subcircuits and values...and they will not publish so gratefull they are related Hugh.... well...Hugh proved his amplifier is the best for customers...they are happy because Hugh gave them the chance to use the best possible...they perceived that, as they have other equipments to compare...so..they do not talk about....only personally, and with another one that has Aksa for sure.

That P61 have not the subcircuits... was a very early circuit and the critical parts were not published those early days.

regards,

Carlos
 
Those debates finishes normally very bad.... we have some bricks that use to enter

behaving very unkindly.

It is a pitty, but, i have learned a lot reading those debates.

I have colected many informs during those things, Hugh gave me those secrets, for free, during conversations and during some negotiations we have made about some Aksa 55......but i could perceive that some forum readers, having attention, could receive nice informs about his circuit.

During the massacre they use to make.... some clever guys had insigths and clear nice informs that pop out during the hot argumentation.

We have to leave the experience to learn things... theories and conclusions about are not enougth in audio sonics.

My amplifier had emitter resistances removed, clearly i had more "punch" in the bass...so... to reduce parts count, to make it more simple and more cheap, the emitter resistances do not exist anymore.

But finishing the unit, i have apreciated the sonics, but comparing with mine Aksas, mine units loose into the comparison....so.... i conclude that researches made by Hugh, related many things, including this emitter resistances, altogether, made the difference.

There are a lot of things that do not make any sense... and sometimes those non sense things are very near us... around us....a nice man can married a very bad and crazy woman.... that's love... non sense too.

Input differential, in every theories, works better using costant current sources and line voltage regulators... some schematics using sophisticaded methods to drain charges can be used too...using Fets transistors and many other interesting parts...... the problem is that theories do not produce sound.... the amplifier with simple differential sound even better.... very strange those things...and only the A to B comparative testing, using the same circuit with both approaches can show you how strange those things are.

You have to use Blind testings, because our "beliefs" will influence the result...not knowing what is playing (ask some help) you will be fair...cannot detect, conclude, discover, what circuit is playing...decision must be only because of sonics.

Levels must be the same...same power, same speaker, same music, same effects, same place...dark place...cannot see nothing..confortable place and repeat tests to see if they make sense....... also change letters of reference.... invert signals... try to confuse every clever brain not to discover what is A ...and what is B.

regards,

Carlos
 
Well excuse me!

Hugh made a statement that linearity is improved with larger resistance. I guess I just assumed that was an objective statement with something objective to support it other than subjective listening.

Carlos, while you may throw away objective theory when something that doesn't measure as well sounds better, I try to find out why. For example, in your differential constant current source example, items such as noise, problems with rf interference, and perhaps this new memory distortion theory or other unknown issue may come into play to make the sound better or worse than any advantage a current source can provide. I don't make any apologies for trying to understand why something sounds better or worse.

As I don't care to get into a debate about tastes in sound, I guess this topic ends here.
 
Dear Pooge,

Let me put my case with some accuracy. If someone makes a suggestion that such and such is subjectively superior, and then declines to mention any objective reason for it, then that is his choice. Have you considered why that person makes this choice?

Let me elucidate, I regret having to spell this out because this !@#$ forum aggravates me seriously:

#1 The reason given is usually conjecture in areas where the experts disagree. This is just such an area.

#2 Conjecture leads to arguments, insults, !@#$ competitions and eventually serious brain damage.

#3 There is no obligation to so explain for anyone here - this is NOT an academic discussion and reputations are all unproven. This applies to all comment, subjective and objective.

#4 I was asked. I gave an answer. If you do not find it satisfactory, I politely point out that the onus is on you to research it and find why to your satisfaction. Experience is not transferable, but it is easily replicated. You do not have to tell anyone; no one here has placed obligations on you, Sir.

#5 This is my livelihood, and while I may choose to reveal a few things, I would be a fool to reveal all, particularly if I perceive commercial advantage in playing my cards close to my chest.

#6 Previous questions on difficult theory attract incisive, at times deeply offensive comment. Witness the dreadful time Graham Maynard has had because he dares to be a prophet in his own time. To what end, I would ask, particularly as all anyone here wishes to do is make a better mousetrap.

#7 While the electrical aspects of amplification are now well understood, the sound/brain interface and just how music is processed, pivotal to a fully objective understanding of why things sound as they do, is not yet fully understood. Some inkling of this lies with distortion spectrum, and even memory distortion, which are presently flavour of the month, but I put it to you that if high end were automobiles we would be just moving into automatic transmissions and disk brakes.

#8 This is my weekend, I may have other things to do......

If you are not happy with this response, and wish to contact me and discuss this in private without burdening others, please do so via my forum or email. While I'm somewhat hostile to the forum, I am otherewise affable and quite prepared to talk about it. Previous comments from me and others here have attracted nothing but derision; I'm not a masochist, and have no wish to 'prove' myself a fool by offering something which can be argued either way - there is simply nothing in it for me and my days of altruism are past.

My point remains; MY subjective analysis has revealed highish emitter resistors are a better option. If this doesn't float your boat, then so be it. It's not important in a world riven by famine, disease and injustice!

Cheers,

Hugh
 
Yes pooge, trow away may be the correct name.

Happened because i have concluded, listening many precious equipment that were very good into specifications.

Perfect waveforms.... nice small distortion..... and bad sonics.

Than i concluded that those things, objective quality and subjective quality do not march together.

Was simple.... and a very simple conclusion.

After you realise this, you do not bother, anymore, with specifications, as this do not make any sense to our ears/perception.

Objective things, values and measurements serves as common language to communications between certain level of experts.

They can conclude what is better, what is worst.... into meterings ...and they do this inside that "closed club"...but when you put the equipment outside the club, result can be extremelly disappointing.

There are exceptions, good meterings do not means bad sonics..there are many valuable exceptions....but more an exception than a rule i could find listening for very long time, thousands of equipments.

Objectivity is very nice..... a very charming world of numbers that allow you to compare, to compete, to increase, to develop....but real world, outside...humans listening...people ears.... tastes...sonics..this is entirelly subjective.

A matter for Human perception, connected to psychologie.... less enginnering....more humanity.

I am, with your permission, unsubscribing me from this thread, as i am sniffing problems.... i want to continue happy... and i have no problem to let others be happy with their beliefs.

regards,

Carlos
 
Carlos,


I couldnt have said it better, music is about the human spirit and emotive qualities that it contains and cannot be wholly represented by numbers and measurements as the final word. For example, the blameless measures well but as I have learned through listening it offers better measurements than it does music in my honest opinion, I couldnt live with it happily, atleast without earplugs.



Colin
 
We are all subjective - some more than others!

;)

We are all subjective - some more than others!

Whether we like it or not, we are all subjects.
In the meanining we are our selves.
Your Self, My Self
and then we have the sometimes controversial Doug Self :D

When I went to school I learned ( in my mother tongue lessons )
that a statement, sentense have usually some components:

subjekt - predikat - objekt

Example:
I talk to you.
I = subject .... he who does
Talk = predikat .... what he does= verb
You = objekt .... the thing, remote person, that the doing is applied upon

Humans have mostly strong Self. In Greek called EGO.
This is in order to survive - survival of the strong, the fitest.

So, whether I like it or not, I should be Subjective
.. or become slave, used, defeated and ultimately die out, be killed off.


Now, the difference is if you can or want to be anything but Subjective :confused:
It is very much your ability to keep a distance to your self,
be able to see things form other's view, empaty,
trying to be in-partial, even in the wildest of argumentation with your self invloved.

I have this ability, to a certain degree, I think.
Try to get a 'bird's eye overview' of the situation I am in.
See thing from the other side.

It is about let go of your prestige, for a moment.
People with good selfestime, has easier to relax
and not to 100% try to 'be winner' and 'maintaine diy audio status'.





The more,
someone will keep on Sticking to His Own Self Subjective opinion
and not recognize, try to understand and respect other opinion,
the less self confidence it shows. At least in my eyes.

Sometimes, even reminds about small children,
in that age, when they start to discover there own EGO
and always say OPPOSITE to parents:
-------------------------------------------------------------

Mother:
- Billy, you should do this and that!
Billy:
- No!
Mother:
You know you should, Billy.
Billy (screaming):
- No.
- I do not want to.

-------------------------------------------------------------

At a certain age, child will prcatice in the art of having an ego
and trying to be a factor to count with.
( In english, using my Swedish-English dictionary:
- Assert One Self
This is natural process for a young human.

When it becomes a bit patetic, is when adults, grown up
get into too much use of making this.
This is what I mean when thinking of children in a SandBox
when I see most highly educated engineers get into fighting about details.

( we have right now one great guy in sinbin, by a silly and too hot argue )

----------------------

:cool:
We have to try to keep in mind:
- We are here with a common hobby, interest.
- We are by this all On The same side.
- diyAudio is just a tiny bit of life - not Life
... if you think it is Your Life

I have to tell you, in all friendliness:
... get your self, your ego - a rest, a day off or two
... get a life


Regards
lineup
.....................................................................................................
- a very subjective person, trying to get an overview, prerspective
.. and most of all trying to see himself, in the view of other eyes
.. Can I stand the sight of me .. or not, this is the question :)
 
AKSA said:
Hi Pooge,

I've found this gives very good sound. I can't argue too much about this, because such debates become interminable and people soon loose interest in the finer academic points, but a larger emitter resistor constitutes degeneration, reduces transconductance, and to my understanding always improves linearity by dwarfing the non-linear b/e junction with ohmic resistance.

There is some controversy on this point, with proponents invoking higher math to prove a point of some conjecture.....:clown:

gm doubling is a function of both transistors being on at the same time during crossover, and its effect is to rob very low level detail, the bane of Class AB. It can be reduced by better controlling the output of the voltage amplifier.

Cheers,

Hugh


pooge said:
I SIMPLY asked why you stated that larger resistors are more linear. YOU made the statement. If you don't want to answer my inquiry objectively, that's up to you. Sorry if I cannot yet tell which hat you wear when you make such statements.
One way to find out if this is true or not is to simulate it. Hugh is right here and it's basic knowledge but this emitter resistor has also a max practical value.

pooge, you can ask nicely and Hugh is not obligated to educate you.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.