Alpha Nirvana 39w 8ohm Class A Amp

I think slight asymmetries in the circuit are result of a lot of fine tweaking to ensure that the dominant harmonic distortion is not odd (3, 5, 7, etc) ordered. Perfectly balanced will usually result in lower overall THD at the expense of elimination of even order distortion. This would result in a harsher and more fatiguing sound. It’s very subjective of course, but these are what I refer to earlier as the “Aksa-approved” harmonic distortion profile.

It doesn’t get any simpler than 6 actives for this category of amp does it?

Why SE Class A and not PP? Again, SE Class A tends to have a more natural sounding harmonic distortion profile (dominant H2) vs PP which tends to have dominant H3.

X,

Exactly, perfectly balanced H2/H3 to my ears is pretty hideous!This is the salient point in audio for most people, whether they realize it or not.

As usual, thank you for all of your hard work!

Cheers,

Greg
 
Member
Joined 2014
Paid Member
.....I am thinking of keeping it close to the original Alpha 20 PCB board footprint and UMS compatibility. Maybe add some Molex connectors to allow for flying lead mounting of the MOSFETs.

Hi X,

Yes, please include the option of remote mounting with using the vertical mount Molex Minifit’s. Probably include a set of the P-mos/N-mos mini mounting pcb’s with onboard snubber also.

Thanks
 
Founder of XSA-Labs
Joined 2012
Paid Member
Some predictions of the harmonic distortion profile in LTSpice at various power levels.

Code:
Here is 8Vpp or 1W into 8ohms:
Harmonic	Frequency	 Fourier 	Normalized	 Phase  	Normalized
 Number 	  [Hz]   	Component	 Component	[degree]	Phase [deg]
    1   	1.000e+03	4.001e+00	1.000e+00	   -0.87°	    0.00°
    2   	2.000e+03	1.848e-04	4.619e-05	 -117.68°	 -116.81°
    3   	3.000e+03	1.053e-05	2.631e-06	   58.71°	   59.58°
    4   	4.000e+03	2.313e-07	5.781e-08	  104.21°	  105.08°
    5   	5.000e+03	7.156e-07	1.789e-07	 -175.07°	 -174.20°
    6   	6.000e+03	5.836e-07	1.459e-07	  178.69°	  179.56°
    7   	7.000e+03	4.936e-07	1.234e-07	  178.79°	  179.66°
    8   	8.000e+03	4.319e-07	1.079e-07	  179.01°	  179.88°
    9   	9.000e+03	3.839e-07	9.596e-08	  179.12°	  179.99°
   10   	1.000e+04	3.455e-07	8.636e-08	  179.21°	  180.08°
   11   	1.100e+04	3.141e-07	7.850e-08	  179.28°	  180.15°
   12   	1.200e+04	2.879e-07	7.196e-08	  179.34°	  180.21°
   13   	1.300e+04	2.658e-07	6.642e-08	  179.39°	  180.26°
   14   	1.400e+04	2.468e-07	6.167e-08	  179.43°	  180.30°
   15   	1.500e+04	2.303e-07	5.756e-08	  179.47°	  180.34°
Total Harmonic Distortion: 0.004626%(0.004808%)

Here is 27.2Vpp or 12W into 8ohms:
Harmonic	Frequency	 Fourier 	Normalized	 Phase  	Normalized
 Number 	  [Hz]   	Component	 Component	[degree]	Phase [deg]
    1   	1.000e+03	1.387e+01	1.000e+00	   -0.87°	    0.00°
    2   	2.000e+03	2.483e-03	1.791e-04	 -121.70°	 -120.83°
    3   	3.000e+03	6.206e-04	4.475e-05	   59.08°	   59.96°
    4   	4.000e+03	1.880e-04	1.356e-05	    9.49°	   10.36°
    5   	5.000e+03	7.426e-05	5.355e-06	 -104.64°	 -103.77°
    6   	6.000e+03	2.763e-05	1.992e-06	  177.05°	  177.92°
    7   	7.000e+03	9.412e-06	6.786e-07	   89.53°	   90.40°
    8   	8.000e+03	1.830e-06	1.319e-07	  -14.32°	  -13.45°
    9   	9.000e+03	1.956e-06	1.411e-07	 -144.36°	 -143.49°
   10   	1.000e+04	1.602e-06	1.155e-07	  176.37°	  177.25°
   11   	1.100e+04	1.070e-06	7.715e-08	  172.05°	  172.93°
   12   	1.200e+04	9.599e-07	6.922e-08	 -179.95°	 -179.08°
   13   	1.300e+04	9.312e-07	6.714e-08	 -179.74°	 -178.86°
   14   	1.400e+04	8.648e-07	6.236e-08	  179.34°	  180.22°
   15   	1.500e+04	8.025e-07	5.786e-08	  179.38°	  180.25°
Total Harmonic Distortion: 0.018517%(0.018564%)

Here is 50Vpp (just before clipping) or 39W into 8ohms:
Harmonic	Frequency	 Fourier 	Normalized	 Phase  	Normalized
 Number 	  [Hz]   	Component	 Component	[degree]	Phase [deg]
    1   	1.000e+03	2.496e+01	1.000e+00	   -0.89°	    0.00°
    2   	2.000e+03	1.405e-02	5.630e-04	 -158.12°	 -157.23°
    3   	3.000e+03	1.391e-02	5.574e-04	   56.44°	   57.33°
    4   	4.000e+03	8.589e-03	3.441e-04	  -21.73°	  -20.84°
    5   	5.000e+03	6.813e-03	2.729e-04	 -135.45°	 -134.56°
    6   	6.000e+03	3.939e-03	1.578e-04	  118.95°	  119.84°
    7   	7.000e+03	3.691e-03	1.479e-04	    3.52°	    4.41°
    8   	8.000e+03	3.049e-03	1.222e-04	 -101.34°	 -100.46°
    9   	9.000e+03	3.149e-03	1.261e-04	  169.39°	  170.28°
   10   	1.000e+04	2.769e-03	1.109e-04	   84.89°	   85.78°
   11   	1.100e+04	2.897e-03	1.161e-04	    5.17°	    6.06°
   12   	1.200e+04	2.717e-03	1.088e-04	  -77.65°	  -76.76°
   13   	1.300e+04	2.811e-03	1.126e-04	 -163.48°	 -162.59°
   14   	1.400e+04	2.513e-03	1.007e-04	  107.42°	  108.31°
   15   	1.500e+04	2.402e-03	9.623e-05	   16.13°	   17.02°
Total Harmonic Distortion: 0.098380%(0.099638%)

I have added an RFI filter to the input and this dropped the gain a bit to 27.0dB. I am using models for the KSA992, and also replaced Q4 with a KSA1381 (or TTA004) to provide room for a heatsink in case it gets too hot.

The FFT for 8Vpp into 8ohms looks like this:
attachment.php


And at 27.2Vpp into 8ohms (12W), the FFT looks like this:
attachment.php


I think Hugh may be making some last minute tweaks to the circuit, so I won't post any updated schematics until then.
 

Attachments

  • Nirvana-8Vpp-8ohms-FFT.jpg
    Nirvana-8Vpp-8ohms-FFT.jpg
    115.8 KB · Views: 4,883
  • Nirvana-27.2Vpp-8ohms-FFT.jpg
    Nirvana-27.2Vpp-8ohms-FFT.jpg
    119.7 KB · Views: 4,860
Disabled Account
Joined 2015
With Tina , it is possible to start the distortion measurement after an adjusted delay because the beginning of the burst there is the transient function . To know the measurement is the steady state distortion the phase shifts should be 180° for the pair harmonics and on phase for odd ones . Your measurements are wrong , at 13Vp or 10W , Tina after 10ms delay measures 0.012% lower than yours , look at post 16.
 
Last edited:
Disabled Account
Joined 2015
R7 plays important roll in trade off ,speed vs distortion . I adjust it for optimum speed to get 0.0055% Dtot. There is still margin in NFB as the sensitivity is high, but I will leave it to the spec of the designer. Tina doesn't have the bc327/40 model , with, it will still be better.
 

Attachments

  • ALPHA NIRVANA4.JPG
    ALPHA NIRVANA4.JPG
    299.1 KB · Views: 1,672
Member
Joined 2010
Paid Member
Still confusing.
To my knowledge, Hugh has used "Nirvana" to identify upgrade kits for his commercial products since the original AKSA 55. This is nothing new in the commercial AKSA scheme of naming things, at least.

The "Audio Nirvana" company just seem to market their own tube amps and a class D amp, their own speaker drivers, box plans and a powered subwoofer box, direct from St Louis. I wouldn't expect to see much confusion arising on this forum, at least.
 
Disabled Account
Joined 2015

Attachments

  • ALPHA NIRVANA,4OHM.JPG
    ALPHA NIRVANA,4OHM.JPG
    319 KB · Views: 1,659
Alpha Nirvana, not Audio Nirvana.

SE Class A can be higher efficiency if there a reactive “push-pull” provided by either an active CCS, as is the case here, or for example, by an inductor like in the MoFo.

I'm no expert, but my guess is that with global negative feedback the output is 'voltage output push pull' - if you want reactive inductor or current source behaviour you need to remove the global negative feedback. This amp uses a CCS yes, in a SRPP type of topology (it's been used before as an alternative to the Aleph current source) but the speaker will 'see' an amplifier with low output impedance dominated by the behaviour of the feedback loop.
 
All correct, Bigun........

One thing; the active CCS is concertina, the voltage across the two sources resistors will remain constant at 0.65V, but the two resistors will pass conjugate currents which will always sum to 3.4A. The two resistors are identical so that the see-saw action is completely symmetrical.

HD
 
Last edited: