GainWire-NGNFB-classB-PowerAmp

The BJT buffer could be used in an IPS with CFA and get very, very low distortion, THD20k only 0.77 ppm output voltage at 80 Vpp.
Damir
 

Attachments

  • CFA-pre-high-gain-high-voltage-sch.jpg
    CFA-pre-high-gain-high-voltage-sch.jpg
    114.2 KB · Views: 690
Finally here is non GNFB amp, but with the CFA IPS, where the BJT buffer plays main role, with something lower distortion then pure non GNFB.
This amp could be configured as full GNFB and lower distortion more.
Damir
 

Attachments

  • ClassB-poeramp-IPS-CFA-dadobuff-TT-EC-OPS-sch.jpg
    ClassB-poeramp-IPS-CFA-dadobuff-TT-EC-OPS-sch.jpg
    172.6 KB · Views: 686
Here is a FFT comparison between the EC power amp with an IPS with no GNFB around it and an IPS with CFA around it. Quite similar harmonics distribution, but IPS CFA with lower distortion. Now which one to choose? I think that both can sound quite similar.
Damir
 

Attachments

  • FFT comparizon IPS-CFA to IPS-noNFB.jpg
    FFT comparizon IPS-CFA to IPS-noNFB.jpg
    277.3 KB · Views: 615
That should be working schematic of the 120 W/8 or 230 W/4
All could be seen from attached pictures.
Damir
 

Attachments

  • GainWire-ClassB- betterCCS-EC-TT-noGNFB-15V-DCservo-sch.jpg
    GainWire-ClassB- betterCCS-EC-TT-noGNFB-15V-DCservo-sch.jpg
    157.2 KB · Views: 1,034
  • GainWire-ClassB- betterCCS-EC-TT-noGNFB-15V-DCservo-gain.jpg
    GainWire-ClassB- betterCCS-EC-TT-noGNFB-15V-DCservo-gain.jpg
    252.1 KB · Views: 991
  • GainWire-ClassB- betterCCS-EC-TT-noGNFB-15V-DCservo-square.jpg
    GainWire-ClassB- betterCCS-EC-TT-noGNFB-15V-DCservo-square.jpg
    82.7 KB · Views: 822
  • GainWire-ClassB- betterCCS-EC-TT-noGNFB-15V-DCservo-clipping.jpg
    GainWire-ClassB- betterCCS-EC-TT-noGNFB-15V-DCservo-clipping.jpg
    94.2 KB · Views: 821
Disabled Account
Joined 2012
@ RNMarsh

Hi, i'm NOT being funny etc, but i thought that you used to design/build amplifiers commercially. So i'm just wondering why you don't design what you need, & then give us the benefit of your experiences ?

Regards

True as in Used to. Now semi-retired at age 72. Lots to tell, dont know where to start. But, Damir's designs are better than anything I did.


THx-RNMarsh
 
From the point of view of minimizing distortion, the optimal resistance of the resistor R3 is about 240 ohms. At the same time, the output impedance is also minimal, but since the loop gain at a frequency of 20 kHz has a phase shift of more than 60 degrees, the output impedance also has a phase shift. It is known from the distortion spectrum test that when the resistance R3 + -20 Ohm deviates from the nominal value, the level of the 2nd harmonic increases to no more than 0.0035%, and the 3rd harmonic increases to no more than 0.0025%. Measuring the output impedance using an 8 V peak voltage applied across an 8 ohm load to the output of the output stage gives the following results. See Figures. Negative output impedance in some cases can be useful for partial compensation of wire resistance.
ROWEN - schweizer high-end Audio
bible_d 1.pdf
 

Attachments

  • OPS_Santos-2013_1kHz_R3=240-Rout.png
    OPS_Santos-2013_1kHz_R3=240-Rout.png
    22.6 KB · Views: 81
  • OPS_Santos-2013_20kHz_R3=220-Rout.png
    OPS_Santos-2013_20kHz_R3=220-Rout.png
    21.6 KB · Views: 65
  • OPS_Santos-2013_20kHz_R3=260-Rout.png
    OPS_Santos-2013_20kHz_R3=260-Rout.png
    21.5 KB · Views: 76
  • OPS_Santos-2013_20kHz_R3=240-Rout.png
    OPS_Santos-2013_20kHz_R3=240-Rout.png
    22.8 KB · Views: 74
Class AB amplifiers are often subject to switching distortion.Let us measure switching distortion with three values ​​of R3.
As the test shows with the optimal value of R3, the crossover distortion does not exceed 0.001%, and increases to 0.005% at the extreme values ​​(220 and 260 ohms)

Let's compare the level of crossover distortion with a circuit specially designed to eliminate switching distortion. The circuit is designed according to the principle of Kazuaki Nakayama patent US4595883. The OPS also consists of a Darlington triplet with paired transistors and has a similar quiescent current.
 

Attachments

  • Class-i_NON-switching_SCH.png
    Class-i_NON-switching_SCH.png
    33.3 KB · Views: 152
  • OPS_Santos-2013_R3=260_cross-dist.png
    OPS_Santos-2013_R3=260_cross-dist.png
    31.1 KB · Views: 100
  • OPS_Santos-2013_R3=240_cross-dist.png
    OPS_Santos-2013_R3=240_cross-dist.png
    26.7 KB · Views: 89
  • OPS_Santos-2013_R3=220_cross-dist.png
    OPS_Santos-2013_R3=220_cross-dist.png
    30.5 KB · Views: 83
  • Class-i_NON-switching_cross-dist.png
    Class-i_NON-switching_cross-dist.png
    33.1 KB · Views: 151