About possible Babelfish J interest

I have 50 unused hexfred diodes.
I've been meaning to try them.
I will also try the snubber but I will try the diodes first on one channel.
Cool Pico
Article from Linear Audio magazine show tests
and 3 $ full bridge with his proper snubber can bring superior performance
if compare to 25 $ bridge de luxe without.
Soft recovery diodes type with snubber is to radar :radar: with jig on the scope. Numero Uno Top 1
Psu electrolitics bypassed with small polypro MKP or MKC's is another thing too try :)
 
On second thoughts it's pretty bloody easy to install a snubber on the power supply.
What value do you guys recommend?
Maybe I'll do that first.

Is bit more complicated not " universal " caps and resistors values
but if you do you own Quasimodo jig and with little practice
you made stop ringing your psu toroid after 5 minutes of work.
CheapoModo: quick and dirty transformer snubber bellringer jig
Smooth operator
Simple, no-math transformer snubber using Quasimodo test-jig
Happy snubbing ;)
 

Attachments

  • 7670043D-D950-4D30-8AC2-D9FB3B09883C.JPG
    7670043D-D950-4D30-8AC2-D9FB3B09883C.JPG
    93.2 KB · Views: 402
  • 199A7E60-C54E-4999-A7BE-DB75B8DB7B0D.JPG
    199A7E60-C54E-4999-A7BE-DB75B8DB7B0D.JPG
    592.4 KB · Views: 413
  • 609BE28D-BD2E-4997-9E27-3EED894A19E6.JPG
    609BE28D-BD2E-4997-9E27-3EED894A19E6.JPG
    85.1 KB · Views: 399
Official Court Jester
Joined 2003
Paid Member
I'm getting ready to populate my boards. Is this the best guide at present?

489883d1435018162-about-possible-babelfish-j-interest-base-pcb-sch.jpg

to refresh your memory :

TP2 - set 10V across R11

then - with shorted inputs and no load on output .... TP3 is for Iq setting (desired Iq is measured as voltage drop across (any) mosfet source resistor), while TP1 is for output offset setting

ya know the drill - everything in temp equilibrium ..... put lid on , wait temperature to be stabilised , pull lid off - set in half** increments what you must set - put lid on , wait again

** half step to desired value
 
Founder of XSA-Labs
Joined 2012
Paid Member
I fired up Ch1, luckily no magic smoke. :) Set v across R11 at 10.0v (took quite a bit of turns). Set bias at 1.20 amps, output offset is still varying slightly as it warms up but no problem getting 0mV offset.

Output noise looks very promising at 0.1mV rms on Fluke 101. Could be used for a headphone amp at these low levels. Looks like I did my grounding topology correct too and no measurable hum.

Ripple on PSU rail under load is 26mV.

So far, an excellent start. Need to put flying leads on the other channel now for the IRFP150's.

So far, very promising...

attachment.php
 

Attachments

  • Babelfish-J-Startup-Ch1.jpg
    Babelfish-J-Startup-Ch1.jpg
    401.8 KB · Views: 474
Founder of XSA-Labs
Joined 2012
Paid Member
Both channels are working properly from bias, DC offset, and noise. However, there is no gain in the amp. 1.00vrms in = 1.00vrms out at speaker terminals. I am only driving +ve and GND of the balanced input. Do I need to do something special? All I can think of is that I set a feedback resistor somewhere off by factor of 10? So music is very weak when played on speakers.

attachment.php


Current is flowing through the MOSFETs as the amp is warming up. Still steady at 1.2amps. So not sure why I don't have music that has gain.

Is closed loop gain set by R20/R4 or R20/R2?
 

Attachments

  • Babelfish-J-Startup-Both-Channels.jpg
    Babelfish-J-Startup-Both-Channels.jpg
    468 KB · Views: 445
Last edited: