L20.5 AP SYS Test data My design

Also, have you compared the sound with a linear against the SMPS? If so, did you notice any differences?
Hello,
In my opinion, a conventional power supply is not worth it. Neither in terms of sound nor about the structural and financial effort
Not to mention the size and weight.

Best regards
 

Attachments

  • L20.5 Mono-Brückenamp.jpg
    L20.5 Mono-Brückenamp.jpg
    749.4 KB · Views: 85
  • L20.5 Mono-Brückenamp mit SMPS.jpg
    L20.5 Mono-Brückenamp mit SMPS.jpg
    899.3 KB · Views: 82
  • Like
Reactions: ziffel
Hello

try it here.

https://de.aliexpress.com/i/1005001983485605.html

But in my experience these Chinese-SMPS have not been available for over a year.

That's why I switched to the Hypex SMPS400A400 (for 8 ohm speakers) or the Hypex SMPS400A180 (for 4 ohm speakers). Unfortunately the Hypex SMPS are twice as expensive as the Chinese- SMPS.

The Hypex- SMPS work very well despite their small size.


Best regards

Michael
 

Attachments

  • SMPS-Vergleich.jpg
    SMPS-Vergleich.jpg
    651.2 KB · Views: 41
  • Like
Reactions: ziffel
For those who stuff their board with other transistors, please be aware LJM like to use NCC 5551 , which is used on my L20.5 board and has different pin out as the other 2n5551 type. NCC5551 pin ECB,2N5551 pin EBC.
The 2n5401 used on this board seems to be the normal one however.
 
Beware of the brand of D669 transistor used. I have changed the D669 to an original Hitachi D669 with similar HFE 130 to old one on board and it blew all power transistor and kill a 100R resistor. The type of D669 will change the bias current, in my case, the original Hitachi really up the bias current and burnt the 100R resistor after D669 and the rest of npn transistors. Make sure to change the bias resistor to higher value to more than 1K (the 1k resistor above 5551) , the one LJM used is actually 1050R and not 1k. Best is replace the 1K with a 2k pot and start around 1.5k .
 
Hi Michael,
i see a little difference between your board and the boards i got on my table a few days ago. The feedback capacitor on your board got 680 µf. On the other boards the value was 470 µf 16 Volt. I would take, to make something better, a non polar capacitor (green Muse) and a 100 nf Wima parallel (on the backside). 470uf is enough. I cant see the imprint on your board.
In the past the 100 nf parallel was on the board, not really necessarary, but better.
 

Attachments

  • Eingang.jpg
    Eingang.jpg
    375.1 KB · Views: 25
  • L20.5.jpg
    L20.5.jpg
    210.9 KB · Views: 27
Hello Peter

https://www.we-online.com/components/media/o127145v410 AppNotes_ANP071_Aluminium-Elektrolyticvs.AluminiumPolymerCapacitor_EN.pdf

https://www.we-online.com/components/media/o127146v410 AppNotes_ANP071_Aluminium-Elektrolyt-vs.Aluminium-Polymer-Kondensatoren_DE.pdf

Polymer elkos are the technical best product, I wanted to try it out.
Muse would be too big with 470/16 volts and diameter1m mm to fit on the board.

With 680µF I achieved the best sound results.

The most important thing in terms of sound is the power supply anyway.

Therefore, this double mono structure receives two Hypex SMPS400A180 again

Incidentally, by far the best end transistor is the on MJL4302/4281. The Kombi 2SA1943 / 2SC5200 is too imprecise and soft.

greeting

Michael
 
Last edited:
Hi Michael,

I know about the polymer capacitors. I will choose some for a phono stage. Panasonic has 470 uf 16 non polar with 12,5 mm diameter. If you are happy with the sound no need to change.
SMPS...i like Connex SMPS more, but i think the sound would be identical.

Greets
Peter