i want to use up my irfp140(i used the irfp240 becuse i couldnt find the model for the 140) so i designed this amplifier, the weirdest thing here is the second supply for the voltage gain stage, its here to compensate the common emitter gain stage losses and make the voltage swing a little bit bigger without adding more parts, also compared to my last se amplifier i used mosfets which are supprisingly not that awfull.
Try a bootstrapped collector load on Q4, and direct couple. That will let you drive the gate above the rail.
I am a little surprised by the linearity as it is essentially a bjt gain of 10 single transistor amp driven to 12v p-p. I used this in a RIAA amp but the output is far smaller.
i also wanted to make an weird hybrid feedback/bias thing using an pnp and conecting r11 to the output.
The gain stage is AC coupled by C2 so the supply here has no effect.
The problem is you have set the bias by setting the resistor values for the FET's to bring them just into conduction. A change of a volt or so on the 15 volt rail will massively change the bias current, a different FET of the even the same batch will be wildly different and the stability with temperature will be horrendous.
By removing one resistor and adding one NPN transistor you can fix all that 😉
The problem is you have set the bias by setting the resistor values for the FET's to bring them just into conduction. A change of a volt or so on the 15 volt rail will massively change the bias current, a different FET of the even the same batch will be wildly different and the stability with temperature will be horrendous.
By removing one resistor and adding one NPN transistor you can fix all that 😉
I would use the 15 vdc supply only. ONE psu for both channels.
Then I would remove C2, R5, R6 (and R7?-)
and R10 set between collector and basis of Q1, use a trimmer and set about 1/2 VDC (or about 10 V)
I suggest, recommend, to use TO-220 transes for the power stage.
To push the sound to the extreme, you can try many parts. Transistors, for example, sound as different as beer or wine taste;-) With a complex design, that would be pearls to pigs;-) I would try bjts - sound more bodied, more elaborate than MosFets usually.
C1 > 2 mF.
Don't forget to connect your speakers reverse. And use speakers without crossover components.
Like this.
Great amp, Your last one, maybe;-)
Then I would remove C2, R5, R6 (and R7?-)
and R10 set between collector and basis of Q1, use a trimmer and set about 1/2 VDC (or about 10 V)
I suggest, recommend, to use TO-220 transes for the power stage.
To push the sound to the extreme, you can try many parts. Transistors, for example, sound as different as beer or wine taste;-) With a complex design, that would be pearls to pigs;-) I would try bjts - sound more bodied, more elaborate than MosFets usually.
C1 > 2 mF.
Don't forget to connect your speakers reverse. And use speakers without crossover components.
Like this.
Great amp, Your last one, maybe;-)
Parts modulate the current, i.e. the sound. I would do without every unnecessary part.
I would also connect RIAA and DACs and cassette players and everything else to these 15 VDC;-)
I would also connect RIAA and DACs and cassette players and everything else to these 15 VDC;-)
No, No - Mooly’s suggestion requires an NPN. With the Vbe wrapped around the 0.33 ohm source resistor and collector to gate. That will force the lower FET current to be constant no matter what FET you throw in there or where the supply drifts. No, Cumbb. It’s not unnecessary - not in the real world. Regardless of what you do with the front ent it’s a good idea.
this is your original circuit. The supply changes from 12 to 30 volts. The change in FET current is a change of nearly 14000%
And this simple modification sees that change n current reduce to 6%
And this simple modification sees that change n current reduce to 6%
Simple answer is yes 🙂
You need to understand how a FET (or a transistor) behaves when biased as you have done. Good circuit design lets you change things like transistors or supply voltages dramatically with little or no change to performance.
You need to understand how a FET (or a transistor) behaves when biased as you have done. Good circuit design lets you change things like transistors or supply voltages dramatically with little or no change to performance.
For the current source, yes. But with the gain stage run off only 15V it will limit the amplitude of the signal so you don’t get the maximum possible power out of it. I suggested using a bootstrap, because it can generate more signal swing (not quite twice the supply). Without resorting to using a transformer.
now where should i bootstrap? from the amplifier output to the bias stage or from the gain stage output to the bias stage? or some other configurtation?
That’s not it. You’ve added a mild amount of negative feedback, that’s all. Boot strap goes to the collector load of Q1, and you need to direct couple into the gate. Once you do that, adding NFB from output to the base of Q1 might be a good idea, to bring the gain back down and linearize it.