[gonefishin]

Hi again (I'm not gone yet )


   I have read some threads in the archives, but I still have some more questions.


   

A lower power version of this amplifier can be made which offers wonderful sonic qualities. In this case, the power supply rails are reduced from +/-36V to +/-24V, resulting in an amplifier of 25W rms power output into 8R which is highly suited to efficient speakers of the type often used with SE tube amplifiers or active crossover systems. This variant is available at the same price on request.

   Can you describe some of the "wonderful sonic qualities"?


   When selecting the transformer for the lower powered version of the Aksa55, What specs should I look for?  (  ex)"two 160VA transformers and 4 x 4,700uF for the 55W stereo AKSA, and two 300VA transformers and 8 x 4,700uF for the 100W AKSA. You may use other transformer sizes and capacitor values, but in our experience the sonics will not be optimised.")

   Has anyone tried varying +/- 24V transformers to find what suits this amplifier the best? What transformer and capacitor values work out the best for sonics to be..."optimised"?


   Is the Nirvana Plus Upgrade version of this lower power Aksa 55 (25)available or ready?


   What's the "stock" gain in this amp?

     What's the "stock" gain of the Aksa 100 amp?



  thanks again, (more questions to follow )

   dan

[gonefishin]

Are the components identical between the Aksa55 and the AKSa 25?  


  If so, I suppose the Nirvana Plus Upgrade would be available for the lowered version also...dunno tho.


   Does anyone have an idea of the minimum height the case need to be for the AKSA 25?


  thanks again,
   dan

[JohnR]

Hiya, the 55 fits nicely in a 2u case, which is 3.5" high. The 25 would be the same, the heatsinks will prevent you from getting much shorter than that.

[gonefishin]

ahhh....thanks.  I was hoping it would fit in a 3" case...perhaps I'll have to modify something on the case...or look at a new sink...or???  


  thanks,
  dan

[JohnR]

3" might work, depending -- the heatsink is 3" tall (actually 75 mm)

[AKSA]

Gonefishin,

I'm always exasperated when people ask how an AKSA sounds.

How does one communicate the appearance of a nice home to one who has never been to see it?

The AKSA is extremely transparent.  It has very little sound of its own.  It comes down to source and speakers, and enables you to hear more of the music.  Bass is strong, midrange is neutral and uncolored, top end is very clear.  Hard to go beyond this.

Gain is 38.2, or 31.6dB, both for 55W and for 100W.  Both share a topology and many components.  Output pairs on the 100W are tightly matched both for Vbe (to 1mV) and beta (to 3%).   Nirvana upgrades are very similar for both.

At this stage there is a proven Nirvana Plus upgrade, which is quite pricey but which adds more transparency.  The 100W Nirvana Plus is still under development, however.

The supplied heatsinks with these amplifiers are all 300mm long and 75mm (2.95") tall.  A 3" tall case would be fine for the lower power versions of the AKSA;  heat is never a problem and so the case height can match the heatsink;  this is not recommended however for the 100W AKSA which needs some flueing at the base of the sink.

Hope this answers your queries, if you have more, let's have 'em.....

Cheers,

Hugh

[gonefishin]

Thanks Hugh


    Above, when I was asking about the sound...I was actually referring to the sound differences between the regular 36Vr AKSA55 and the 24Vr AKSA55(25).

   You made mention on your website that there are some "wonderful sonic qualities".  I was wondering, since I don't need the extra power...could I have some improvements to gain with the lower power AKSA 55(25).

   Can the gain be changed some, while keeping the integrity and stability of the amp?  

   I'm looking into using the AKSA55(25) in a bi-amping situation...and while I may end up getting a AKSA 100N+ also...that may be some time away.  If I could adjust the gain easily (and while maintaining stability) that would be all the much better.


    Do you have any advice on transformer selection for the +/-24V 25Watt version?

 I have heard the advantages in transformer selection for DIY tube amps...but, are there differences in the transformers that I need for the Aksa55 (25), like there are in tube PS and OPT transformers?

   Does the quality of the Torrid make a difference in sound quality?

   Are there things that you DO want to look for when selecting a transformer for your amps...and is there anything you want to be sure to avoid when selecting a transformer?


  thanks (again)
   dan

[AKSA]

Hi Dan,

Oh dear - quite a few questions!!

You made mention on your website that there are some "wonderful sonic qualities". I was wondering, since I don't need the extra power...could I have some improvements to gain with the lower power AKSA 55(25).

Yes, more refinement, slightly deeper sound stage, better top end.  These are the chief benefits.

Can the gain be changed some, while keeping the integrity and stability of the amp?

Yes, within prescribed limits.  You can drop gain from the present 38 to 26 without appreciable change to the sonics.

I'm looking into using the AKSA55(25) in a bi-amping situation...and while I may end up getting a AKSA 100N+ also...that may be some time away. If I could adjust the gain easily (and while maintaining stability) that would be all the much better.

It is no problem, and neither is stability, which is enhanced a little with lower voltage rails.

Do you have any advice on transformer selection for the +/-24V 25Watt version?

The transformers, of which there are two, one for each channel, should be rated at approximately three times the power rating of the amp. For a 25W AKSA, this would be 24V rails, with an AC secondary of 18-0-18V.  Rating should be 80VA, a commonly available transformer.

I have heard the advantages in transformer selection for DIY tube amps...but, are there differences in the transformers that I need for the Aksa55 (25), like there are in tube PS and OPT transformers?

This is a vexed question, but yes, I believe there are differences.  I believe the double C core transformers are the best, in fact, but cannot say why precisely.    Toroids have a frequency response to beyond 200KHz, which makes them quite susceptible to mains born noise, and they are also very leery of any DC on the mains, which is actually quite common these days, typically to 200mV or so.  EIs and C Cores have a less stressed iron core which is less susceptible to saturation and thus clipping, and their frequency response is very poor over about 10KHz.  I suspect these differences are audible since the transformer is sorely taxed in an audio amp, feeding as it does a very highly capacitive load in the filter caps.

Does the quality of the Torrid make a difference in sound quality?

I cannot say categorically, but I believe it does, and generally I look for a good regulation.  Sadly, most toroids in 80VA have no better than 10% regulation, so my first reaction would be to move to EIs, which you may be able to find surplus at low cost.   Toroids still sound wonderful on the AKSA, but I've reason to believe Double C cores sound better, although the evidence is there that bedding in can take weeks.

Hope this is helpful,

Cheers,

Hugh

[gonefishin]

Hi Dan,

Hope this is helpful,

Cheers,

Hugh

   Very helpful...thanks!

[Occam]

Gone,

Just some info to pass along... I bought a couple of 100va 36V ct transformers to try on my 55. Haven't gotten around to it, but its a low cost way of trying out your 55 as a 25 @$10/ea for an EI transformer.

http://www.skycraftsurplus.com/cgi-bin/ePages.storefront/1535664471/Product/View/32227-630493

The MCI Transformer site bears out these specifications, but sadly, don't give the regulation %. Iff'n I had a power rheostat I could calculate the regulation, but alas...

[AKSA]

Hi Paul,

Thanks for the tip on these excellent transformers!!

There's a way to determine trafo regulation - quite simple, too!

Do you have a 10W 47R resistor?  Or something you could make up to around this figure?  If so, here's how:

1.  Wire up the transformer to the mains;  active neutral to power.  Insulate, and keep pinkies away.
2.  Using a decent DMM, switch to AC volts and measure the output, from both of the windings.  Secure the leads onto the windings handsfree for a continuous, rms measurement.  If it's 36Vac CT, then one side to the other should be around 36Vac.  Let's say voltage is 38.48 Vac (guestimate!!).
3.  Now connect the 47R resistor briefly across the windings, and measure the voltage.  Naturally, under load it will drop, but not by much.  Let's say it drops to 38.12 Vac, a drop of 0.33 Vac.
4.  Now, the math.  The drop here has been (38.48 - 38.12) = 0.33Vac for a current draw of 38.12/47 = 0.811A.  We need to know the current output of the transformer at full rated output;  let's assume it's 3A, which would make it (3 x 36) watts (108VA) rating.  It is an estimate, but it's valid;  if drop is 0.33Vac for 0.81A, then for 3A it will be approximately proportionate, namely 1.22Vac.  The only way to verify this, however, would be to load it down with a 12.4 ohm 100W resistor, and this might be impractical.
5.  We now divide the extrapolated 1.22Vac drop at full power by the no load voltage, 38.48, and express it as a percentage.  In this case this is 3.2%, which is a very good regulation indeed.

Note here that all voltages and currents are express as rms;  this ensures that energy levels rank equally with DC voltages of the same value.

Cheers,

Hugh

[Occam]

Hugh,

I hadn't thought of doing a linear extrapolation. Good idea. I'll hit the Rat Shaque tomorrow and buy the highest rated resistors I can find (hopefully as large as 1 watt), and paralell them. Sorta funny, I had a bunch of appropriate big sucker power resistors till  last week, when I sent them of to a friend

[gonefishin]

Occam...thanks for the link...That site will cerrtainly get a bookmark!



   Does anyone else have some links to some good Double C core trannys?  (in the US or overseas is fine)


   thanks!!!!

[gonefishin]

Hugh, I'm having troubles finding information on the Double C-core transformers.  I'm having trouble finding DIY'ers opinions on various double c-core transformers used in a SS application like your amps.

   If your able, I'd like to know what transformer company you have used.  (thanks)

   Also, does there seem to be any problems with the (2)C-Core transformers picking up noise?  Would shielding be recommended(or mandatory)?  


   thanks again!
   dan

[AKSA]

Dan,

Very hard to find.  I had some 250VA Double C Cores wound professionally here in Melbourne a couple years back, and sold a few on to constructors building 100W AKSAs.  They reported a different 'sound' to the stock 300VA toroids I normally recommend, all other things being equal.

At present a good friend is doing a transformer shoot out here with double Cs v. toroids using a higher voltage 55W AKSA Nirvana Plus.  I'll appraise you and others of the outcome when known;  could be another couple of weeks.

I've combed the US web market too and found almost nothing in double C core.  It's rare.  However, a toroid has the very real disadvantage of high frequency response, permitting hash and other mains peturbations to pass through readily to the DC circuit.  EIs and Double C Cores don't do this anywhere near as much, so there might be some advantages for top end purity and absence from hash.

Cheers,

Hugh

[PSP]

Hi Hugh,
If we suspect that the possible impact different transformers on amplifier sonics may be due to the frequency response of different transformers (and that torroids may pass very high frequency noise components from the mains), etc. ... it seems to me that a much simpler and more direct approach would be to experiment with ferrite clamps on the leads from the power supply caps to the amp boards (I know that you can't just throw ferrites at a problem, but we could do some focused experiments on the subject and report results back).  We could also try some 0.1uf line rated caps across the line (like the Jon Risch line filter, without the inductors, see   http://www.geocities.com/jonrisch/surge.htm, perhaps placed across the IEC inlet).

It seems like this approach would be more of a surgical, focused attack on the problem.  We could do this kind of experiment without a major re-build of the amp; in particular, the ferrite clamps could be put on in five minutes so the sonic impact could be quite apparent right away.  Thoughts?  

Peter

[andyr]

Hi Hugh,
...We could do this kind of experiment without a major re-build of the amp; in particular, the ferrite clamps could be put on in five minutes so the sonic impact could be quite apparent right away. Thoughts? ...

Good idea, Peter,

I have some ferrite clamps so I'm happy to experiment, however, I think we should define the test which should be carried out - so we can all report our results against a common base.  I dunno ... first try 1 ferrite, then 2, then 3, then 4??

I suspect, however, that there is more to using different types of transformers than what ferrite clamp experiments will show up.

Regards,

Andy

[AKSA]

I would suspect Andy is quite right;  namely that transformers determine more sonic characteristics than a ferrite clamp.

Further, ferrite clamps are available in a huge range, all different and all presumably with different effects.  This could be a very large experiment, and to narrow the field some physics would need to be done.

There is also the variation in electrical and magnetic properties of different transformers, and the inevitable interaction with the clamp.

I won't commit to R&D on this;  if anyone else is interested, please let us know your results.

I've got an ocean of fish to fry at present!

Cheers,

Hugh

[PSP]

My understanding of how a transformer functions in an amplifier power supply is certainly too simple, being limited to the view that a transformer's function is to supply the right voltage to the rectifier diodes such that after rectification the voltage across power supply caps is as specified by the amp designer.  I also understand that a transformer and power supply caps must be "fast" so as not to limit amplifier dynamics.  That is all I know:  if there is more to transformer functionality in power amps than that, I am clueless about it.

If filtering noise from the mains is the problem and we wish to do this without limiting amplifier dynamics, it seems to me the place to attack is to filter any high frequency noise on the DC supply to the amps--or, perhaps better, between the transformers and the power supply.  There is a huge variety of ferrites, but these have specifications (so it's not a totally random search) and the AKSA group is pretty smart.  

It's also possible that this is a wacko idea.  I am wide open to a good technical explanation of just why it's wacko.  I'm sure that I'll learn something.

That said, I will be looking at this "someday" (after I get my GK-1 built, the 55N "plussed", and (I think) build a pair of Jens' Equilibrium speakers.

Peter

[SamL]

Can't remember where I read it but a few of those who replace their CD/DVD player's power caps recommend Jensen 4-pole electrolytic cap. It is said that the cap reduce the noise level significantly and is better than BlackGate.
http://www.referenceaudiomods.com/Merchant2/merchant.mvc?Screen=CTGY&Store_Code=RAM&Category_Code=JE4PA

Too bad the US$ has gone up, otherwise I'll get 4x 10000uf/63v for my aksa100n

Sam