Single-stage class A 20 wRMS mosfet-amp: can you help me to develop it?

Jump to Latest
#41
Just looking to a pair of components that ensures minimum positive voltage for bias, as it ensures to swing more towards zero volt with the drain, so more clean watts can be obtained, but those are not available in TO247 formats.

I haven’t tried 4 Ohm loads, but for that you need for sure more current.

A more classic pair will be the first choice for tests.

The ratio of power dissipated by pmosfet and nmosfet depends also on the amount of feedback (more feedback means more watts to be dissipated by the pmosfet).

I’ve changed the opamp feedback because the first one was a drop-in of what I use in tube amps (driver, output tubes and output transformer involved), and here it can be way simpler.
 
Last edited:
  • Like
Reactions: dkfan9
#42
In sim, I'm seeing the opposite on dissipation: higher peak and average dissipation on the P-channel for R13: 39 vs. R13: 63.3 along with the higher distortion expected with reduced feedback. My schematic is updated to match yours but I did not change the 10k/100 values or the FETs/FET arrangement.
 
Last edited:
#44
I've played a bit with the simulator and got this:

1698706001084.png


Results are written above:
Zout = 3.46 Ohm so a DF of 2.31 (similar to alot of triode single ended amps)
1 Wrms at 0.55% THD with 3rd harmonic 27 dB below 2nd and 4th 16 dB below 3rd.
27 Wrms at 5% THD with 3rd harmonic still 1.5 dB below 2nd.

Are 16 mA enough to properly drive the gate of five IRFP240 ?

I have all six mosfet at around 9W of dissipation, which seems reasonable for TO220 enclosures.
Efficiency is very close to 50%.
 
#51
xrk971 said:
Here’s a tested and proven 20W Class A amp.
Click to expand...
Thanks, but it’s not what I’m looking for. I want to build a single gain stage amp, with local feedback only, with the harmonics and the damping factor of a Triode Single Ended amp, while chasing 50% efficiency.

Last circuit gets all points (better to raise the choke to 80 mH), I just need some help to define the choke and what is the standars practice for the 4.7 mF coupling cap (paralleling smaller film caps, etc…).
 
#52
If you like the last circuit (choke loaded cascode topology), check out the LuFo. Plenty of LU1014D’s available still and also a fully tested circuit. Drop voltage if you only need 20W.

https://www.diyaudio.com/community/threads/lufo-amp-39w-se-class-a-from-28v-rail.372679/

1698839868058.png


2.83vrms 10ohms:
1698839916801.png


1698839984828.png


It can use various input gain buffer stages with Yarra format input boards (7 different Yarra compatible ones available including Pass ACP+, WB18, Aksa Melbourne, Aksa Hakuin, XRK PCA, and HV OPA454 opamp even).

Again, tested and verified to work and can use microwave oven transformers (MOTs) for inexpensive 60mH 0.5ohm DCR.
 
  • Like
Reactions: zintolo
#53
@xrk971 thanks for the suggestion of the MOT, I wasn't aware of this possibility.
I would still like to go on with this project as it works on LTSpice.

Using TO247 and a proper heatsing I can probably use just two of them, each one dissipating around 25-30W maximum.
 
#57
Thanks, but I was talking about MOTs.
Alot of them on the other side of the pond don’t ship to Italy: too cheap to risk an international shipping.

I’ve tried the single IXIS IXTK88N30P in my circuit in LTSpice and it works fine with 1 Wrms at around 0.3% THD and 27 Wrms at 5% THD, always 2nd harmonic dominant.

I’m sure there would be better p-mosfets for the source follower at the bottom (lower Rdson, lower Crss), but I haven’t found that much through mouser.
 
You must log in or register to reply here.
Top Bottom