[jules]

Every now and then the 55W N+ Aksa is compared with the 100W N+ Aksa. It seems to be agreed that the 100W gives more "slam" but the 55W wins on pure sonics (I haven't seen this described exactly. Does this mean for example that it excels on midrange voice for example? ). The touring 100W N+ recently fell into the middle of a nest of tube enthusiasts and survived with a great deal of honest and generally positive approval.

As a newbie I noticed that the "dark siders" are of necessity using v high efficiency speakers [above 95db] and trading off "slam" and performance in music from the sixties onward for clarity and excellence with vocals and acoustic instrumentation (if I can summarize it that briefly).

Am I right in thinking that high efficiency speakers require lighter cones/coils etc, softer neoprene/foam surrounds and in general less weight?

What I am getting at here is, would it be better to tailor the 100W and 55W amps more specifically so that the 55W is combined with a 95db + speaker system and aimed closer to the "tube" sound market while the 100W is more suited to music needing outright power?

When Hugh designed the Aksa 55 it was intended to get as close to the sonic warmth of tubes as could be achieved with a non tube system [please excuse me putting words in your mouth Hugh] and a constant thread runs through all reviews of this amp to say that he has got closer to this ideal than any other SS amp.

Much attention goes into the use of a tube pre-amp (such as the GK-1) to add the final touch to the sound of the Aksa 55 but should we be looking more closely at the other end of the system and finding a more specifically matched set of speakers for these amps?


Jules

[AKSA]

Jules,

Characteristically insightful post about loudspeakers and the general philosophy of the amp.  Many thanks!

Yes, by and large you are right.  Interestingly you have placed the design of speakers center stage!

High efficiency speakers almost all have total cone and voice coil mass well below 20 grams, very stiff suspensions for strong mechanical damping, and very high BIL factors (product of voice coil length, current and magnetic field strength - all leading to force factor), and tiny, often less than 1 millimetre, excursions.  This gives high acoustic output for low electrical input.

These sorts of speakers have trouble achieving a low resonance and deep, high impact bass;  but they are extremely sensitive.  You can't have it all!  Significantly, their very strong mechanical damping puts no demand on a low damping factor from the amplifier.  A tube amp with a DF of only two or three can sound pretty good with these drivers.

A strong disadvantage of these speakers is the full range requirement.  Typically, you might have a 7" or 8" cone, large enough for bass down to 40Hz, but too large for 4KHz upwards.  So a small whizzer cone is attached right at the voice coil join to pick up and transfer the rapidly moving voice coil for these very high frequencies.  The problem is that the rest of the cone, made usually of very light, strong paper, starts to 'break up', and these resonances can be most objectionable, and often around 2-4KHz.  At even higher frequencies, the whizzer cone, unsupported at its outer perimeter, also tends to break up, and this too creates objectionable resonances.  Sometimes driver manufacturers even recommend electrical notch filters to outright remove these resonant peaks from the musical material;  I need not stress that this is probably not the right answer!

It so happens that tube amplifiers have high source impedance, and act as current sources, not voltage sources like SS amplifiers.  The global feedback loop on a SS amplifier is accommodated by most common <92db/watt/meter drivers in the market;  they are designed for high damping factor, and demonstrably for voltage feedback.  This is in sharp contrast to the full range speakers used on low power SET amplifiers, which are VERY different, and designed for low damping factor.

If you use the AKSA with a full range high efficiency speaker, the combination of high electrical and high mechanical damping factor will remove most of the bass from the sound and emphasize the high frequencies.  In short, overshoot is negligible.  And the huge power of the SS amp at low frequencies will drive enormous currents through the full range speaker which will stress the suspension unduly, but the overdamping will ruin bass decay and give an 'off' sound.  Lowthers sound great on SETs of 8 watt, but lousy on AKSAs.  In many ways, I see the comparison as similar to a high speed European gasoline engine versus a slow speed US V8;  the transmission designs are very different.

Hope this is of some explanation;  this is a tricky area to fully understand and I don't pretend to be a speaker expert!

Cheers,

Hugh

[jules]

Many thanks for the explanation Hugh. It's always been one of the outstanding features of this site that you (and others involved with Aksa design) are prepared to give full and reasoned responses to all manner of questions. Maybe this is why much of the discussion here transcends the more usual two or three sentance exchanges.

It's true that I've put speakers centre stage here just as it's true that for the most part this forum concentrates on the Aksa amps and pre-amp. At the same time Aksonics do exist and maybe they should receive more attention.

One of the reasons I raised this issue was that I have a set of three way speakers [about 25 years old] that are now on their last legs. They were designed for an amp of about 55W and used a 10" woofer [with a passive bass radiator] along with a dome midrange and dome tweeter. When I had the 10" bass repaired recently the bloke who did it commented on the fact that it was very high quality construction but used a relatively light coil/cone/magnet compared to modern practice. Unfortunately I can't find out the efficiency rating of this system but good listening levels were easily achievable without having to twist the knob too far and room evacuation levels were quite achievable at the top of the scale.

Now an Aksa 55 is clearly much more powerful than most of tube amp systems so there's no need to go to the extreme end of the efficiency spectrum for power reasons but I'm still left wondering if the use of lighter, more efficient speakers might compliment the "tube" warmth of this amp and that this form of differentiation from the 100W [as against a purely room size approach] might be a good thing.

So, to take the bull by the horns. It seems a little wrong to me that both the 100W and 55W amps are served by the same set of Aksonic speakers. Surely this is a compromise for one or both amps. Would it be of advantage to design separate Aksonics for each amp with an eye to defining distinctly different roles for each amp?

Jules

ps My amp case is close to finished Hugh but I'm resisting the temptation to do the show and tell until I get there.

[gonefishin]

Hi Jules,

    It's hard for me to tell from your post if you talking strictly about HE single driver speakers...or HE speaker systems in general.  I'm guessing single driver only.


    What are your thoughts on using one of the AKSA 25n+ amps for higher efficiency speakers?  

   I know I've been converted over to the "dark side" for some time now.  But I'm not over here because of what the tube amps don't do well (bass, bass impact and highs on many SET's).  I'm here because of what they do do well (mids, detail without edge and dimensionality with a pulse).
 
  I've always tried to keep an open mind (ahhh...make that ear) about audio gear.  Stating that simply because something had a tube in it, didn't mean it's going to sound good.  Just the same...Because something was SS didn't mean it would sound bad either.  But I've just been underwhelmed with a lot of SS amps out there.  This is what led me to find AKSA.

   I also love the fact that the AKSA 25, 55 and 100 are voiced  similar to each other.  An AKSA amp is an AKSA amp...take it or leave it.

  What I've decided to do with my system is bi-amp the mid/treble (106db) with a 25n+.  After which...if I like the sound I'm going to build a 100n+ for my low efficiency (95db) woofers.

   Has anyone tried a 25n+ on higher E speakers?


  take care,

 dan

[SamL]

From my limited experience, high efficient speaker tent to sound good even when playing at lower volume. It is able to provide the attack and highlight and making  music interesting at low volume level. A lower efficient speaker tent to need to be "crank up" to sound good and exciting. It is hard to move away from high efficient speaker once you have listen to one.
Old speaker can sound quite good actually. I went to a friend place and his old 3 way have beautiful sound. Its 100hz to 250hz reproduction put my newly build 2 way to shame.  

Have fun,
Sam

[DSK]

Jules,
I have listened to the stock, Nirvana and Nirvana Plus versions of the 55w and 100w AKSA amps on numerous occasions. While it is true that the 55w bettered the 100w and the 55N bettered the 100N, you would indeed need very golden ears to differentiate the 55N+ from the 100N+. It took Hugh much longer to "crack it" with the N+ kit for the 100w, but "crack it" he did.

My speakers (89db) probably didn't require the 100w, but because the 55N+ and 100N+ sounded identical, I went for the 100N+ for the extra headroom. Had the 55N+ sounded even marginally better, that is the route I would have taken.

So, for your current contemplations and musings, you need to consider the power issue and price difference, but not the sonics.

Note: the above comparisons are only valid when the two amps are built with identical internal wire, connectors, etc etc.

I have not heard the 25w or 40w variants of the AKSA. But according to those who have, they do indeed sound better again. So, if that is all the power you need, it would seem to be the best option.

[jules]

So many answers!!

Firstly, I am in danger of falling on my face very quickly when it comes to technical analysis of these issues but where I'm coming from here is a sort of holistic look at Aksa amp/speaker systems in an effort to find out how it is best to enhance the unique amp sonics with speakers that further those qualities. I'm not trying to play the 100 off against the 55 nd my points about high [or perhaps that should be mid-high] efficiency speakers aren't so much about the efficiency so much as the different response qualities of speakers with lower inertia.

So ....

Dan; I can't comment on any experience with a 25W Aksa but it does seem that the number of people who bi and tri amp their systems gives a great deal of support to this approach to system improvement. I have to say that your suggestion that a 95db speaker is inefficient places you firmly on the dark side .

Saml; Your observations about listenability and High efficiency tend to support the direction of what I am asking. One of the points frequently made about Aksa amps is that it is possible to listen to them comfortably for long periods and that has to be a fantastic comment on the way they have achieved thier design goal.

DSK: I take your point [and Dan's] that the 100 and the 55 are similarly voiced but what I'm really trying to get into here is if it is not so much related to power as the question of getting closer to a 'tube" sound via speaker selection.

Jules

[gonefishin]

Hi guys...


  ok...I STILL haven't heard the recently finished AKSA25N+ on my main speakers.  I've been running the amp in on some 89db bookshelf speakers that I have at anywhere from low to moderate levels.  I've currently got over 150 hours on them.  

   Last night I took the amp over to my brothers house where we're working on some modified Altec speakers using a much larger (and stronger) 10cuft vented cabinet, upgraded crossover parts and a Fostex t90 tweeter.  Overall...these would be placed in the not a single driver catagory...but of decent efficiency.  He currently uses an Anthem integrated amp and a Rega Planet CDp.

   Even tho we were using the small Creek passive (yuk) for volume control, I was still quite impressed with the performance of the AKSA amp.  When listening to other SS amps I've always had a feeling of the sound being flat.  Nope, I'm not talking about the frequency response...I'm talking about the life of the amp.  Simply flat and no emotion, weight, depth or breath in the music (told you I come from the dark side ).  So far I've been very let down by solid state amplifiers of many makes and price ranges.  

  Enter the AKSA on high efficiency speakers>>>  This thing had me taken back a bit.  Sure, I've read the stellar reviews from customers around the world...and even a very low number of people who were less than happy with the performance of the AKSA amp with their particular speaker.  I've even read of all the "tubies" that gave this amp the nod of approval.  But I still worried if this amplifier would meet my expectations (or maybe hopes is more accurate).

  While running thru all kinds of music this amp was a genuine surprise to listen to.  It has control and attack that you may expect from SS but without that harshness that makes you cringe.  I wouldn't describe the amp as being smooth, but it certainly isn't harsh or cold and sterile (like many of the big boy SS amps).  This things breaths a controlled quiet, dynamic life into every part of the music it touches.  

    The highs in this amp were a delight to listen to...nice sweet and delicate.  Perhaps not with as much delicacy and "air" as some SET amps, but still very nice.  What they did right was nice pinpoint imaging and to have a quiet background.  This let you hear the decay of each and every strike, splash or wack of a cymbal or triangle.  But the control and tone of each strike seemed to be more accurate and clear than other tube or SS amps that I've listened to.  On a few cymbal crashes the amp reminded me a bit of the sonic signature of the Jule-Electra Splashhhh.

   The mids were again clear, quiet and detailed without edge.  The varying sound from trumpet to trumpet was more noticeable on this amp than on any other SS or tube amp I've listened to no matter what tube or design.  Going along with the trumpets above...the amp maintained this excellent tone and clarity thru the high dynamic range it plays as well.  While many SS amps will play loud...I've always thought that the "way" they play loud was a bit annoying.  The AKSA25N+ gave me dynamics in much the same way decent horns do... unrestricted dynamics from the single instrument in question.  It just sounds as though it's free to do what it will.
 
  Voices were also nice on this SS amp.  They had the breath and life that a nice tube amp could provide.  Clean clear and dimensional.  (hmmm...just thinking here...I'm going to have to try some Gospel music with this amp)

   Bass was right in line with the rest of the amp.  I've heard descriptions that the AKSA was a bit weak in this area but I found the opposite to be true.  The control, dynamics and clarity while maintaining tone was astounding.  From the pluck of a stand-up bass to pedals of a organ...impact of percussion.  This amp just gives you more of the same from the highs to the lows.  Clean, clear, detailed, quiet with great tone and life presented in an SS amp that could actually give you a stage that has width and depth...pinpoint imaging on top.  (could you tell I have to get going...our daughter just woke up from her nap).  

  In short...I've ordered a second AKSA100N+ to power my woofers and my brother had me order him an AKSA55N+ as well.


   well...I suppose that may even make for a short review as well...but there's soooo much more to say

  thanks,
 dan

[AKSA]

Thanks Dan,

Very nice comments;  much appreciated.  

I have checked my email;  two more comin' up......!

Dan, you touch on harshness, the top end to make you cringe, the emotional communication.  These are probably the most significant aspects of subjective sonic performance.  Perhaps this is the time to talk of these things from a design viewpoint.     WARNING:  No maths here, all theory and speculation, no proof, lots of conjecture, possible smoke up the !@#$, but the honest to goodness thoughts of what is happening here.  Flame me at will, I enjoy words, love a challenge, and have an answer for most of it..........

I have only two CROs, three DMMs, a couple of function generators, a variac, and LOTS of parts in my junkbox.  But significantly, I have five pairs of speakers!  I do not have an Audio Precision 1, the best distortion analyser in the business, or even an HP339A.  This is because these things cost heaps, and I never for one moment believed that the measure of ampwlifier sonics was the distortion measurement, particularly THD.  Too many amps measure well and sound dreadful to truly believe in this specification, so here goes:

1.  Biggest problems with PP Class AB SS amps are crossover artefacts, decay of notes, and 'flatness' (some might say 'lifelessness') of tone.
2.  Respectively, these problems are addressed with scrupulous design attention to the output stage, the feedback network, and the lag compensation.
3.  The recording process strips much of the 'emotion' from most performances.  This is because of bad miking and directional effects, which remove a portion of the higher harmonics because they are off-axis.  At the least the spectral distribution of the harmonics is skewed towards the fundamental.  It is the care with miking and mixing in particular which distinguishes good from mediocre recordings, though increasingly software is being used at the digital level to 'retouch' the recording.  Much of the emotion is contained in the H2 and H3 overtones;  reduce or even strip them away and the recording will sound lifeless.
4.  Tube amps add H2 and H3, particularly SETs.  Lots of it, up to 3%.  This partially restores the orginal harmonic distribution of the original performance, and certainly it's a big improvement.  Now, it should be possible to engineer the SS PP Class AB amp to do the same by careful attention to the voltage amplifier and input stage.  It does not need to be much;  0.05% is probably sufficient, as long as it is solely H2 and H3.  Because of the exponential transfer function of the bipolar transistor this tends to imply that this reconstitution should be done passively, to avoid creating higher order artefacts which sound very bad because they are musically dissonant.  On the electrical analog of a musical waveform this directs us towards asymmetrical distortion, best done with single ended, passive circuitry.

This last observation is actually heresy, because it utterly contradicts the 'straight wire with gain' notions on which audio engineering was founded.  A great many audio engineers have spent their careers striving for lowest measureable distortion, so this represents a life-changing contradiction.  No wonder this sort of talk generates animosity!!  However, no-one ever claimed that tube amps were of the 'straight wire with gain' genre;  the subjectives are conveniently ignored.  In fact, they are often buried under layers of feverish marketing, giving us one of the greatest blind spots in audio technology.  The world talks red but drinks white;  it promotes front drive, but prefers rear drive;  it shouts 'straight wire with gain' from the rooftops, but, notwithstanding, the cognoscenti know that the better single ended triodes are sonically (that is to say, subjectively) the best amplifiers in the marketplace.

Knit this cognitive dissonance together, steadfastly ignore the audio marketing machine, ponder and tinker for a few years, and you have the AKSA........

Cheers,

Hugh

[Felipe]

Hi guys,

I'd like to throw a question about speaker combos here, as it was referred earlier in the post that are "some" designs that do NOT couple with Aksa.

Hugh said that Full-Range Louthers/Fostex designs sound terrible with Aksa - wich comes to me as a shock beacuse i was thinking of building a pair of Voigt Pipes!  
If the problem is the high-damping factor, this leads to bad combinations of AKSA with any kind of full-range being that Voigt Pipes,TQWT,TL,TQWP TL, Horn or even Bass-Reflex based Full Range types i assume      Hugh  

So, what is the perfect loudspeaker for AKSA ? 2-way ? 3-way? Bass-Reflex ? Sealed ? Electrostatic ? Infinite Bafle ? 86dB ? 95dB ?

[jules]

Felipe and forum,

I usually have to read posts from Hugh, Tinker and others such as Occam at least 5 times before I start to get the point and this was the case with Hugh's first reply to my post but as I see it he has been quite specific about the type of high efficiency drivers that are unsuitable ie low coil/cone mass <20gm., low excursion <1mm,  and stiff suspension. There are other details but in short these are drivers specifically tailored to SET applications and I would presume they might also have upper limits on power input that might make them unsuited to Aksas [though this is a guess].

I don't see that this rules out all forms of high efficiency drivers [although I would like to know if I'm wrong]. Ribbon tweeters run from 95db to over 100 db efficiency and appear to be suited to ss amps. Along with this [and it's apparently preferable to have reasonably closely matched efficiencies between the drivers in a system], a mid-bass driver such as the PHL 6.5" has a coil/cone weight of 12gm,  a  max excursion of +/- 8.5mm with an efficiency of 93db and a power rating of 100W.  There are other drivers with similar specs and it appears as though they should be fine with an AKSA

Perhaps my original post wasn't clear but carefully selected higher efficiency drivers seem to make sense as a match for Aksa 55s.

jules

[AKSA]

Thank you Jules, very logical reasoning, and quite correct!

Felipe,

The AKSA sounds fine with any speaker NOT specifically designed for single ended triodes.  These normally are 98dB/watt/metre or even higher, and many are horns.  The AKSA sounds fine with direct radiating speakers of any efficiency, actually, but I don't recommend it with horns.

The reasons for this fill a large volume.  If you'd like to research it further, I commend you to reference the work of Dr Bruce Edgar, a scientist who worked in Defense Industries in California and who started a business a few years ago selling the now legendary Edgarhorn.  I heard he'd been rather ill in recent times, but a couple of seminal articles are available here:

http://www.enjoythemusic.com/magazine/manufacture/whyhorns.htm
http://www.positive-feedback.com/Issue4/edgarinterview.htm

This man is possibly the world's foremost expert on horn loudspeakers, and comes from a pedigreed scientific background.

I have heard AKSAs with open baffles, the Linkwitz Phoenix system (Ben's system, sensational!!), reflex designs, Voigt pipes (direct radiating of course), sealed boxes, and sixth order bandpasses.  All sounded pretty good to me - perhaps because modern drivers are designed to work best with voltage sources, which have low source impedances like SS amplifiers.

Hope this helps,

Cheers,

Hugh

[gonefishin]

Hi Hugh,

  I've read and heard arguments of why SS may not get along with high E speakers...and why tubes may get along well with high E speakers.  

  One of the arguments that I've heard is regard to distortion in tube and SS amps.  I've heard it argued that the reason why SS sounds bad or harsh with H.E speakers, is because the distortion levels are at their highest in SS amps when almost no power is being drawn from them.  Being much of the distortion is of the "undesirable" type...SS doesn't get along well with H.E.  While tube amps distortion levels are at their best when there's a low draw of power.

    One of the nice attributes about some single ended designs is their portrayal of detail within music with good separation, imaging and a certain "blackness" in the background (which I think leads to the better detail portrayal, in spite of the distortion ratings).  

  It just seems as if traditional amplifier distortion ratings don't describe what's happening in the music reproduction of a system...or give you any hint if the amplifier will give you good detail retrieval or not.  

   A SET amp can give you (or me) the portrayal of a cleaner...less distorted sound.  No matter what the distortion ratings of the amp are.  While there may be some sonic differences between your AKSA amps and various SET amps, there are some similarities.  One is in the way a clean, clear precise scape is drawn with a seemingly quiet separated background.  

    Your amps certainly don't seem to be designed (or voiced) to sound like other SS amps.  I've heard some bad examples of SS amplification with high E horns...your amps aren't one of them.

  One effect that surprises me about your AKSA25n+ is how well it maintains it's voice and attributes (clean, clear, good imaging, dynamics) at both high and low volumes...doesn't seem to matter.  I found this to be true on the lower E speakers that I was burning the amp in on...and two sets of higher E speakers (some modified Altecs and my DIY EdgarHorns).  

   I know my ignorance plays a large part in my confusion on this subject.  But your amp doesn't portray music in the same way as many other SS amps.  I know it doesn't have the same offensive sound of some other SS/HE speaker combinations that I've heard.  Should it...according to the measurements?

   On my brothers high E modified Altecs...the amp really performed nicely.  From top to bottom...very consistent, detailed, clear, precise, quiet, controlled...great tone (nice job on the voicing!) and yes...musical.  However, I currently have some concerns so far on the AKSA combo on my DIY edgarhorns.  But I'll reserve final judgment until I get the woofers powered by the AKSA100N+.  The AKSA 25n+ amp (driving my mids/tweets) seems like it may just be running away from (dynamically) the amp that's currently powering my woofers.  So I don't think it's fair to pass judgment until the second 100Watt amp is finished.  But this doesn't seem like anything similar that I've heard with other (harsh) SS/horn combo's.  


   please...feel free to be as frank as you feel necessary.

  thanks,


 dan

[Martin]

I was looking into building a set of fostex back loaded horns using the Fostex FE207E according to the plans found on Madisound's drivers page.
However these drivers have a 96dB efficency rating. I guess I'll start looking for a different project. These are of pretty good size...probably not what the wife had in mind
 As usual...the good folks here helping us make the most of our audio dollars.
Thanks
Martin