Alpha Nirvana 39w 8ohm Class A Amp

Jump to Latest
thompsontechs

thompsontechs

Member
Joined 2019
Luckily, I did mono builds and the board wasn't hard to remove. I've been looking at the two resistors in question and have two ideas. The first id extracting, which can be a real pain. I will try that first, and if it's to much of a pain, I will just cut it out and solder the new ones on from below. Not ideal, but workable.


I have to do something, just knowing they are not ideal would fester, lol
 
xrk971

xrk971

Founder of XSA-Labs
Joined 2012
etsy.com/shop/XRKAudio
Paid Member
thompsontechs said:
Do we need to stay away from Metal Film in those positions? I ask because there isn't much available out there in the 1%. I guess, I will have to do some matching if metal is out.
Click to expand...

If you have quality metal thin film 1% like Dale CMF55’s or Takman’s or even Vishay’s - you are fine. Don’t change it to CF.

We are talking about a tiny difference. If you have carbon comp or metal oxide then maybe consider swapping.
 
thompsontechs

thompsontechs

Member
Joined 2019
xrk971 said:
If you have quality metal thin film 1% like Dale CMF55’s or Takman’s or even Vishay’s - you are fine. Don’t change it to CF.

We are talking about a tiny difference. If you have carbon comp or metal oxide then maybe consider swapping.
Click to expand...


Thanks X, ya they are what was on the Mouser cart and that is Carbon Comp 1/2 Watt. Once in a while I push a system pretty hard and I just feel better knowing that I have the what the designer intended and not just, good enough. Maybe we should edit the cart though, for future users.


It's a pain, but even if you had caught the carbon deal in the BOM I still would have to go back in anyway having the cart .5W instead of 1W. I appreciate the time it took to do all the cart work... it's up to the person that orders it to double check.



So it's on me....


JT
 
AKSA

AKSA

Member
Joined 2001
aksaonline.com
I first John Darko in 2008, IIRC. We got along very well; he is a very tall Brit, teaching ESL at that time, and I'm a very tall Ozzie, trained to teach ESL as well. We shared a deep love for music and audio (in that order!) and he then put my original NAKSA in media lights. I was very grateful, and wrote a long analysis of my design philosophy.
Here it is:
“My approach is essentially informed by musical considerations. The human is unique in that he can make sense of cacophonous polytones – music by another name. But the inter-relationships must abide by certain rules to make mellifluous music – and chief amongst them is the rule that even orders are acceptable, where odd orders, particularly the fifth harmonic and beyond, are definitely not.”

“Psychoacoustic studies reveal that to humans the fifth harmonic, sometimes called H5, is musically discordant and unappealing. H2, the second harmonic so redolent with valve circuits, adds warmth and engagement; H3 adds a certain sharpness, and H4 adds ‘body’. H7, H9 and H11 are quite unattractive, sometimes described as ‘machine’ tones, since they do not sound sympathetic to music and in fact contribute to listener fatigue. Crossover distortion on push pull amplifiers is often made up of these components.”

“Even order harmonics are produced by distortions to one half of the waveform – but NOT to the other half. This type of distortion is sometimes called asymmetrical, because it manifests on one half of the waveform, but not the other. Most amplifier designers strive to have the amp clip at the same amplitudes of the two waveform halves; this is electrically efficient, and gives a pleasing symmetry.”

“However, there is a trap here. The human animal is primordially drawn to symmetry, after all, we have two arms, two ears, two eyes, two legs, and two hands. We are irresistibly attracted to symmetry in art, in our visual appraisal of others, in mathematics and in science. Much of our architecture is symmetrical, and even automobiles are often marketed on the basis of symmetry, so there is clearly something to it.”

“The trap is that symmetry of waveform distortion means that any distortions will be odd order. Be mindful that distortion is inevitable; it is a fundamental part of the amplification process, however small, so we must design to accommodate this distortion. Corruptions of the upper half cycle duplicated in the lower half cycle will manifest as odd order harmonic distortion, and this sounds very bad to our ears. In fact, there is a pronounced asymmetrical quality to our hearing; the transfer function of air is different in compression to rarefaction, and this means that everything we hear is slightly corrupted by the very medium through which we hear it! It transpires that the human ear also deals asymmetrically with compression and rarefaction; and as a result all incoming sound waves are augmented with H2 as they enter the ear canal. Estimates of this phenomenon vary, but some experts put the figure in excess of 30%. This is a shocking revelation to those who insist on vanishingly low THD, since, along with huge distortions added by the speakers, the amplifier issues are, on the face of it, dwarfed by other phenomena occurring in the human hearing system.”

“This all has repercussions for audio systems, particularly digital sources and power amplifiers. It means that if odd order harmonics, even at very low levels, are delivered to the speakers then the human ear will respond to them, perhaps not directly, but certainly at a subliminal level. This may cause us to cry out in sonic rapture, or quietly leave the room after only ten minutes.”

“At Aspen we believe that these considerations are pivotal to audio amplification, and our many careful tests over the years confirm that an amplifier with higher even harmonics than odd harmonics is more musical. This has informed all our recent designs, with the NAKSA series following this philosophy to the letter. We fervently believe we are on the right track, have developed a series of metrics to assist with the design cycle, and would ask that you listen carefully to the new NAKSAs and hear for yourself just how musical is this novel approach.”
Click to expand...

I was always fascinated with X's measurements using carbon resistors v. metal film resistors in the feedback networks of the Alpha. He demonstrated conclusively that metal films bring up the third harmonic over the second; the carbons do the opposite.
Best sounding amps have a monotonic decrease of successive harmonics, and the more linear, the better. This is from Jean Hiraga in the 60s - nothing is new and demonstrates how the marketing speak drives the equation rather than the truth.

Cheers,

Hugh
 
Last edited:
B

bucks bunny

Member
Joined 2020
taking this Quote
"but some experts put the figure in excess of 30%. This is a shocking revelation to those who insist on vanishingly low THD, since, along with huge distortions added by the speakers, the amplifier issues are, on the face of it, dwarfed by other phenomena occurring in the human hearing system.”
I consider any discussions of amp THD signature that amount to a total THD below 0.00xxx% as pretty irrelevant. This includes the actual hype about post-filter feedback as well. If there are any serious reports about audibility of THD levels that low I would be very interested in link(s).
 
S

spind

Member
Joined 2003
Paid Member
X--weren't those measurements done with SMD resistors on a preamp? Is it safe to generalize from that to a general 'rule' for everything? Has the experiment been redone on amps with through hole resistors? Would be interesting to look at. (though most generalizations seem to be that carbon comp aren't great).
 
adason

adason

music soothes the savage beast
Joined 2004
edsaudiopages.blogspot.com
Paid Member
AKSA said:
I first John Darko in 2008, IIRC. We got along very well; he is a very tall Brit, teaching ESL at that time, and I'm a very tall Ozzie, trained to teach ESL as well. We shared a deep love for music and audio (in that order!) and he then put my original NAKSA in media lights. I was very grateful, and wrote a long analysis of my design philosophy.
Here it is:


I was always fascinated with X's measurements using carbon resistors v. metal film resistors in the feedback networks of the Alpha. He demonstrated conclusively that metal films bring up the third harmonic over the second; the carbons do the opposite.
Best sounding amps have a monotonic decrease of successive harmonics, and the more linear, the better. This is from Jean Hiraga in the 60s - nothing is new and demonstrates how the marketing speak drives the equation rather than the truth.

Cheers,

Hugh
Click to expand...

Hugh, nice post. But how do you actually bring speakers into equation? A particular amp may sound great with certain speakers, but will fall short driving another. Some speakers are so notoriously hard to drive, only a subset of amps will do. Other speakers present easy load and completely different group of amps will shine. Should not the sound of an amp be judged alone, but rather a result of a combo...an amp and the speaker?
 
adason

adason

music soothes the savage beast
Joined 2004
edsaudiopages.blogspot.com
Paid Member
Just to illustrate, I once used to have Bruce Rozenblit's otl amp, 2watts per side, but in reality more like 1.6 watts. Its no longer offered on his site, this was a while ago. That poor little thing was not able to drive almost any speakers, except the most efficient ones with high impedance. So I made mtm with two betsy's and selenium/jbl horn tweeter, wired for 16 ohm. With estimated efficiency of 96dB, all it took was half a watt to be loud. (woofers were separate classAB). What I want to say is that judging that amp with 4 ohm speakers with 86dB/w/m would be a terrible mistake. Amp had heavenly sound with right load. I ended up using it as headphone amp for a while, but later realized that burning 1000 watts of electricity to run headphones is barbaric and I sold the amp.
 
Last edited by a moderator:
AKSA

AKSA

Member
Joined 2001
aksaonline.com
Hi Adason,

Good question. I have thought long on this too.

Distortion from speakers arises from mechanical issues (breakup of cones, inertia of cone+voice coil mass, compliance at spider, rubber surrounds, whizzer cones; electrical parameters of the crossover/coils; restrictions on voltage and current in the intrinsic design of the voice coil and electromechanical system, and finally, the response of the driving amplifier from the back EMF, which impresses itself on the output stage at a considerable period later, and causes horrific fb signals sent back to the amps input stage.

These complications make it possible to offer hundreds of different speaker and amp systems, as we know, and we need to choose carefully for matching.

SS amps are generally more versatile than tube amps, and need different speakers, viz the ES57 v. ES63 Quad systems. Now to address amps, my particular interest.......

One of the greatest issues for amps is the back emf which plays havoc with the feedback system. All you can do is either make the fb system very fast, OR, as I do with my commercial amps, reduce the global fb to no more than 33dB. If you listen to a zero global fb amp (cf. Ayre, Pass, etc), you hear the depth of image most tube amps deliver. This alone makes people fall in love with them, but with no global fb you have serious deficiencies, such as damping factor, slam, and high level resolution. There are exceptions, of course, but not many. Effectively you are forced to choose lots of unhappy compromises to meet the customer expectation, and this has given us a diversity of products in the audio world and a lot of dubious marketing and a lot of prejudice.

I don't expect this answers your question, but at least it defines the major issues!!

Hugh
 
Last edited:
Francois G

Francois G

Member
Joined 2004
Paid Member
xrk971 said:
If the BOM specified carbon composition that would be an error on my part for not catching it. It should be carbon film or metal thin film. And only at the feedback and shunt resistor positions. The carbon film has a slightly higher level of overall THD but gives a greater ratio of second harmonic vs third harmonic compared to metal thin film. Metal thin film are the safest bet and sound fine. Don’t use metal oxide thick film under any circumstance for the feedback resistor. These are sometimes the 1w or 2w “pink colored” matte finish higher power resistors. They add higher third harmonic distortion vs either carbon film or metal thin film. This was discovered in the Aksa Lender preamp thread. In SMT, metal oxide thick film is the most common cheapest surface mount resistor.
Click to expand...

X, could you please post the promised correction/clarifications on #1 in the thread and rule out further misunderstandings? Lot has been said but lacks precision. R124, 126, R133, R135?
 
Last edited:
adason

adason

music soothes the savage beast
Joined 2004
edsaudiopages.blogspot.com
Paid Member
Hugh,

I understand what you are saying. Too much negative feedback already has bad rep.
So we should be focusing on buffers, no feedback, just current.

Long time ago i remember visiting web page dealing with amp, where there were several output pairs, but the pair driving speakers was not part of the feedback. Can't find that page anymore.

Is output transformer in tube amps reducing EMF? McIntosh used to make ss amps with output transformer, but i never owned or heared one, so no idea if it has any benefit.
 
You must log in or register to reply here.
Top Bottom