Working on an Allison biased class A ~25W

[jkuetemann]

Seeing some examples of Allison type bias has piqued my interest in trying a class A design for a change. I submit my take on a class A design for a little peer review.

1kHz @ 4V p-p ~1W RMS into 8 Ohm
-------------------------------------------------------
Fourier components of V(vout)
DC component:0.0153455
Harmonic Frequency Fourier Normalized Phase Normalized
Number [Hz] Component Component [degree] Phase [deg]
1 1.000e+03 4.053e+00 1.000e+00 -1.20° 0.00°
2 2.000e+03 1.872e-07 4.619e-08 -76.48° -75.29°
3 3.000e+03 6.902e-09 1.703e-09 68.07° 69.26°
4 4.000e+03 2.661e-09 6.565e-10 98.87° 100.06°
5 5.000e+03 3.951e-09 9.748e-10 157.31° 158.50°
6 6.000e+03 2.744e-09 6.771e-10 -52.02° -50.83°
7 7.000e+03 9.757e-09 2.407e-09 -5.41° -4.21°
8 8.000e+03 4.728e-09 1.167e-09 -101.76° -100.56°
9 9.000e+03 3.267e-09 8.061e-10 -112.14° -110.94°
10 1.000e+04 4.823e-09 1.190e-09 99.33° 100.53°
Total Harmonic Distortion: 0.000005%

20kHz @ 4V p-p ~1W RMS into 8 Ohm
---------------------------------------------------------
Fourier components of V(vout)
DC component:0.0153463
Harmonic Frequency Fourier Normalized Phase Normalized
Number [Hz] Component Component [degree] Phase [deg]
1 2.000e+04 3.852e+00 1.000e+00 -23.34° 0.00°
2 4.000e+04 5.308e-07 1.378e-07 -36.11° -12.77°
3 6.000e+04 2.978e-07 7.731e-08 165.08° 188.42°
4 8.000e+04 3.808e-09 9.887e-10 64.60° 87.93°
5 1.000e+05 7.181e-09 1.865e-09 -119.72° -96.38°
6 1.200e+05 1.105e-08 2.868e-09 -100.93° -77.60°
7 1.400e+05 3.870e-09 1.005e-09 -60.81° -37.47°
8 1.600e+05 6.793e-09 1.764e-09 27.16° 50.50°
9 1.800e+05 4.809e-09 1.248e-09 12.28° 35.62°
10 2.000e+05 4.042e-09 1.050e-09 -9.91° 13.43°
Total Harmonic Distortion: 0.000016%

1kHz @ 20V p-p ~25W RMS into 8 Ohm
-----------------------------------------------------------
Fourier components of V(vout)
DC component:0.0153499
Harmonic Frequency Fourier Normalized Phase Normalized
Number [Hz] Component Component [degree] Phase [deg]
1 1.000e+03 2.048e+01 1.000e+00 -1.20° 0.00°
2 2.000e+03 4.860e-06 2.373e-07 -73.30° -72.10°
3 3.000e+03 1.875e-06 9.158e-08 74.74° 75.94°
4 4.000e+03 1.799e-07 8.784e-09 169.59° 170.78°
5 5.000e+03 5.029e-07 2.456e-08 -93.64° -92.44°
6 6.000e+03 6.045e-08 2.952e-09 25.94° 27.13°
7 7.000e+03 1.884e-07 9.201e-09 76.25° 77.45°
8 8.000e+03 4.227e-08 2.064e-09 -63.99° -62.79°
9 9.000e+03 5.039e-08 2.461e-09 -105.24° -104.05°
10 1.000e+04 1.169e-08 5.710e-10 30.50° 31.70°
Total Harmonic Distortion: 0.000026%

20kHz @20V p-p ~25W RMS into 8 Ohm
------------------------------------------------------------
Fourier components of V(vout)
DC component:0.0153706
Harmonic Frequency Fourier Normalized Phase Normalized
Number [Hz] Component Component [degree] Phase [deg]
1 2.000e+04 1.946e+01 1.000e+00 -23.34° 0.00°
2 4.000e+04 1.622e-05 8.334e-07 -36.81° -13.47°
3 6.000e+04 4.229e-05 2.173e-06 137.43° 160.77°
4 8.000e+04 1.541e-06 7.920e-08 119.91° 143.25°
5 1.000e+05 1.347e-05 6.920e-07 -140.32° -116.98°
6 1.200e+05 9.367e-07 4.813e-08 -17.61° 5.73°
7 1.400e+05 4.160e-06 2.137e-07 4.49° 27.83°
8 1.600e+05 4.596e-07 2.362e-08 137.38° 160.72°
9 1.800e+05 1.220e-06 6.270e-08 144.33° 167.67°
10 2.000e+05 1.200e-07 6.165e-09 -93.03° -69.70°
Total Harmonic Distortion: 0.000244%

[lineup]

Very Good circuit with error correction.
And runs in Class A. Gives a fantastic result.
How much bias you have for 25 Watt output?

[jkuetemann]

Thanks. The sims look good to me, but I'm no expert. Even if the real performance is a order of magnitude worse it would still be good. I have it set at close to 2.2A, mainly to accomodate impedance dips down to a little lass than 5 Ohms.

[Bigun]

Looks superb, on paper at least. I hope you build it and report back on the sound.

[jkuetemann]

I'm looking to construct this for sure. I'm no expert on layout however, though I've read a number of threads and have read books by Slone, Self and Cordell and hopefully can create a layout that works well. I will likely strip out an old NAD chassis and try to design into that. My collection of old transformers lacks a suitable 20-0-20, most are much higher voltage. That I will likely have to buy new.

I've been away from DIY projects for some time and would appreciate some suggestions on component selections with respect to good quality capacitors and resistors (not audio-fool stuff). Brands, product series and the like to aim me in the right direction. Also suggestions for Canadian suppliers is welcomed.

[lineup]

Originally Posted by jkuetemann

I have it set at close to 2.2A, mainly to accomodate impedance dips down to a little lass than 5 Ohms.

You need a big heatsink. Which you already know

[jkuetemann]

I have a few fairly large efficient ones pulled from old amps that I'll reuse, at least for trial purposes. For a real polished build I'd see about getting some new ones but repurposed ones are free and great for prototyping.

I just lack a suitable transformer at the moment. I looked at buying a 400VA 22-0-22 Plitron but the direct order minimum was $250, so after selecting another to fill the order +shipping +tax I was looking at close to $400. Ouch. I have a 22-0-22 that unfortunately is only 160VA and two identical old ones with single 20V secondaries, unknown VA rating (they aren't particularly large). The 160VA may be OK for a single channel to test the circuit, and even then not for continuous full power testing, but is way under sized for a stereo build.

I'll begin laying out a PCB shortly and post it for comments and refinements. This is the part where I'll need some insight as I go. Thanks for the encouraging comments so far, this will be an interesting foray into class A.