Last edited:
I'm slowly setting up my QuantAsylum 401 analyzer box, and learning how to measure...
Very soon I'm going to measure all my amps 🙂
The hour of truth is coming 🙂
Very soon I'm going to measure all my amps 🙂
The hour of truth is coming 🙂
You need a good notch filter and a generator with very low distortion. Or put spectraplus on your computer
All TO-126 transistors are lined up on the PCB on the purpose - there will be Alu heatsink going across both boards, and all these plastic transistors will be screwed to it - including temperature sensor BD139s.
MJEs and BDs will need mica isolators...
In reality only drivers need heatsinks (in the LatFet version no heatsinking was needed), but it won't hurt to have all of them cooled...
Скиньте файлик в лтс.Гляну,где причина
Give me the LTS spice file .I'll look at the error
I finished a purely discrete version .It is necessary to draw printed circuit boards
Give me the LTS spice file .I'll look at the error
I finished a purely discrete version .It is necessary to draw printed circuit boards
Last edited:
I'm basically fine with versions sim-ed and built so far.
Don't think there would be big difference in measurements/listening -
all these amps already have very impressive parameters.
If there is one thing I would like to improve - it's the slew rate.
So far - 15V/us measured - which matches the sim exactly.
But this I guess is limited by VAS configuration itself, not the input stage..
Just to compare I sim-ed square waves on discrete NMOS200 (blameless quasi hexfet that I built few years ago)
Power amp under development
which shows slew rate over 60V/us.....
15V/us is perfectly fine for an amp, but not impressive 🙂, perhaps more is possible?
Don't think there would be big difference in measurements/listening -
all these amps already have very impressive parameters.
If there is one thing I would like to improve - it's the slew rate.
So far - 15V/us measured - which matches the sim exactly.
But this I guess is limited by VAS configuration itself, not the input stage..
Just to compare I sim-ed square waves on discrete NMOS200 (blameless quasi hexfet that I built few years ago)
Power amp under development
which shows slew rate over 60V/us.....
15V/us is perfectly fine for an amp, but not impressive 🙂, perhaps more is possible?
Last edited:
The gain is too high .We need to reduce it.
You forget to put capacitors in the collector of current sources.Replaced the capacitor in the local OOS with inductance. It should work.
You forget to put capacitors in the collector of current sources.Replaced the capacitor in the local OOS with inductance. It should work.
I tried with these capacitors, they were making it even worse...
inductor in there? Not a resistor?
inductor in there? Not a resistor?
OK, it works fine with C10=330pF, C15=47pF, C16=68pF (referring to my asc file).
Not sure if this is optimal, will try to tune it up later on.
What kind of inductor it should be? Air coil? Or magnetic core?
Not sure if this is optimal, will try to tune it up later on.
What kind of inductor it should be? Air coil? Or magnetic core?
Total Harmonic Distortion: 0.000000%(0.000092%)
FFT also looks slightly better 🙂
>The gain is too high .We need to reduce it.
Now it works perfect even with my original gain (4k resistor in the input).
Wonder what slew rate will be. Will check later today..
FFT also looks slightly better 🙂
>The gain is too high .We need to reduce it.
Now it works perfect even with my original gain (4k resistor in the input).
Wonder what slew rate will be. Will check later today..
Last edited: