100W Ultimate Fidelity Amplifier

Attachments

  • APEX AX11 stereo protect.JPG
    APEX AX11 stereo protect.JPG
    228.2 KB · Views: 2,304
A17 with PRO out pin

A17 audio amplifier is very similar with my project: MicroSOUND-100 (see attach below). However, I believe that A17 is not optimized. Convince yourself that, try make a comparison with my MicroSound 100 project - here's a question for you: in your A17 project, see VAS circuit, there is a 470pF capacitor in parallel with a resistor of 1Mohm. You're sure the other 1Mohm resistor from VAS does not need a capacitor too? Maybe not, if you're going to make a comparison with a Marantz MM7025 (and sure you did that). But are you sure really do not need? You did a rectangular signal test to see what happens if you mount a capacitor in parallel with that resistor? I don't believe you did, considering the result presented.
 

Attachments

  • MicroSOUND-100 v.2.2.5 - Schematic and PCB Audio Amplifier.pdf
    203.7 KB · Views: 919
  • MicroSOUND-100 v.2.2.5 - PCB 3D Top View.png
    MicroSOUND-100 v.2.2.5 - PCB 3D Top View.png
    403.1 KB · Views: 2,053
  • MicroSOUND-100 v.2.2.4 - PCB 3D Top View.png
    MicroSOUND-100 v.2.2.4 - PCB 3D Top View.png
    494.4 KB · Views: 1,966
A17 audio amplifier is very similar with my project: MicroSOUND-100 (see attach below). However, I believe that A17 is not optimized. Convince yourself that, try make a comparison with my MicroSound 100 project - here's a question for you: in your A17 project, see VAS circuit, there is a 470pF capacitor in parallel with a resistor of 1Mohm. You're sure the other 1Mohm resistor from VAS does not need a capacitor too? Maybe not, if you're going to make a comparison with a Marantz MM7025 (and sure you did that). But are you sure really do not need? You did a rectangular signal test to see what happens if you mount a capacitor in parallel with that resistor? I don't believe you did, considering the result presented.

VAS transistors colectors are shorted thru 10uF for HF and another 470pF will be in parallel (there will be 940pF), two 220pF can be on both 1M resistors in parallel if you want simetry.
Regards
 
Looks great Alex! Will you please post the foil as well?

Blessings, Terry
As always I'm helpful and all attached below :cool: It's 100 % tested
......:hohoho:

Regards,Alex
 

Attachments

  • PCB A33 APEX AMPLIFIER ALL GRAY.pdf
    216.1 KB · Views: 1,146
  • PCB A33 APEX AMPLIFIER BOTTOM MIRRORED.pdf
    100.1 KB · Views: 955
  • PCB A33 APEX AMPLIFIER TOP MIRRORED.pdf
    325.3 KB · Views: 1,053
  • pcb a33 apex pic.jpg
    pcb a33 apex pic.jpg
    1.1 MB · Views: 1,424
Last edited:
Hi Guys,

I started populating the A33 today. I'm using Alex's layout above. The first question I have is why there are over a dozen caps added to the circuit. Is there something wrong with Mile's design? I saw some of this with the A23 that olafk posted but his one has a least 16 caps added that are not on the schematic. Are you guys amplifier designers? Why all the changes to Mile's designs before they are even built?

Next question is there are two trimpots on the schematic. One is listed as 4k7 and the other 220R. Are those center values or are we supposed to find trimmers in those values?

Thanks, Terry
 
In schematics it's usually not necessary to specify exactly what caps to use in the power supply. Most DIYers can choose their own PSU or do a PSU project for however many watts the amp is.

On a PCB, there should be lytics as close as possible to the power BJTs, Alex's layout is smart in this regard. If anything this is what should be specified on the schematic, to place reservoirs very close to the outputs.
 
In schematics it's usually not necessary to specify exactly what caps to use in the power supply. Most DIYers can choose their own PSU or do a PSU project for however many watts the amp is.

On a PCB, there should be lytics as close as possible to the power BJTs, Alex's layout is smart in this regard. If anything this is what should be specified on the schematic, to place reservoirs very close to the outputs.

I'm not asking about the power supply. I'm asking why there is 3,000uf added to the each rail and then three additional 100uf and two 100n on each rail. Is Alex trying to make up for a weak power supply? Does this design really need all that extra filtering, and if so why? I'm just trying to learn something here.

Thanks, Terry