Build This MoFo!

If someone wanted to go all out on parts my guess is the Lundahl LL2733 filament current choke may work well. It has two coils which you can put in series or parallel. In series it gives 400mH at a recommended current of 1.7A saturating at 2.7A. Or in parallel it gives 100mH with a recommended current of 3.4A saturating at 5.4A. It's $100 from K&k audio (no affiliation). The LL1694 is also a contender. They claim to wind and gap them as meticulously as there audio transformers and IMO they look sexy if indeed a choke is capable of looking sexy.
 
LBHadju:

The Lundahl 2733 is the choke used in my MoFo monos. See post #669.

Steve

Apologies sledwards12375 I actually looked at the image, it's well built, but didn't realize it was the Lundahl. I should have looked closer as it's printed on the package.

Anchan if the inductance is insufficient, there will be insufficient magnetic field for it to swing negative getting worst at low frequencies. And swinging positive will also be difficult if it does not have sufficient impedance between the mosfet source and ground. I would need to do an LTSpice simulation to see what it looks like in practice.

Leve
 
Last edited:
question about the output cap:

in signe rail PP amp the output idle voltage is at 1/2 Vrail so the output cap is always porlarizedi in the same way, an electrolitic can be used

in the MoFo the output voltage is around 1V (very low resistance of the choke)
when it swings with the audio signal, will the polarisation if the output cap be reversed? do we need non polarized electrolitic?

Is-it better to reference the output (and maybe the input too) to the +V as done here
PFdesign.jpg


the Power Folllower
 
Official Court Jester
Joined 2003
Paid Member
input reference - pretty much always ref. to side where source of mosfet is ;

in fact , disregard that sentence ..... it really depends of particular case

and , it's wise to arrange that GND is on same side , just because gnd is usually cleanest rail in system

though .... where to reference load is another issue

looking at left side of pic you posted - "normal design" - input is flawless , speaker return can be moved from gnd to drain of mosfet for some benefits

looking at right side of your pic -"exclusive design" - only speaker is connected in best way ....... input side is very troubled ...... move generator marked as 1/2Vb to gnd side , and everything is suddenly OK

I'm surprised that we can't see anti-parallel resistors there ....... :devilr:
 

Attachments

  • best.jpg
    best.jpg
    32.9 KB · Views: 866
Last edited:
Fine, but PSRR will go down, not up in the case of MoFo. The drain of the MOSFET offers much higher resistance than the inductor. The inductor is not a constant current source.

I have run various choke loaded output stages with BUZ900P , RD16HHFN1 and Semisouth 085 . Subjectively I found the sound a lot brighter with the output connected across the MOSFET each time I tried . I preferred taking the output across the choke . Chokes used were the Thomas Mayer specified LL2733 700mH/175mH

You are right about PSRR . LCLC supplies are the way to go :)

cheers

316a