We don't know what your models or sim files were to even suggest what your problem might be. You would need to simulate in LTspice and post the asc. file to enable a proper analysis but you could post an image of the Multisim screen display of the schematic and someone might be able to spot your problem if it's evident there.
Significant problem is that there is no thermal control for the bias of the output stage.
You can use a 680R+pot for the bias generator on a lateral mosfet, but not for hexfets.
I would suggest you use a small hexfet like a IRF710 to form the bias generator.
Then a few hours with LTSpice running through the simulations to be sure, then build and test it. This might give you more ideas about compensation too; I suspect it might more than 22pF caps.
HD
You can use a 680R+pot for the bias generator on a lateral mosfet, but not for hexfets.
I would suggest you use a small hexfet like a IRF710 to form the bias generator.
Then a few hours with LTSpice running through the simulations to be sure, then build and test it. This might give you more ideas about compensation too; I suspect it might more than 22pF caps.
HD
It looks like a very poor adaptation of a Lateral Mosfet amplifier circuit.
As shown bias is not enough by any means, you will have *horrible* distortion, and if you increase it to avoid crossover then amp will thermally overrun and burn.
And compensation was probably good for the original one, but this one may very well be unusably unstable.
In a nutshell, it is a mess.
Search for a "real" amplifier circuit, with proper background.
As shown bias is not enough by any means, you will have *horrible* distortion, and if you increase it to avoid crossover then amp will thermally overrun and burn.
And compensation was probably good for the original one, but this one may very well be unusably unstable.
In a nutshell, it is a mess.
Search for a "real" amplifier circuit, with proper background.
1. If you have simulated it, then you can probe the simulation to determine what is not right.
2. A very similar circuit is included in the latest (free) LTC spice down load, a "100Watt" amp.
3. Real problems with this amp include
-slew limit due to limited VAS current driving many gate capacitance
-VAS voltage limits prevent output from getting anywhere near the rails.
-differential VAS only complicates the circuit and does not provide any performance benefit.
2. A very similar circuit is included in the latest (free) LTC spice down load, a "100Watt" amp.
3. Real problems with this amp include
-slew limit due to limited VAS current driving many gate capacitance
-VAS voltage limits prevent output from getting anywhere near the rails.
-differential VAS only complicates the circuit and does not provide any performance benefit.
Thank you Fahey. Do you have a suggest? I want a good amplifier between 250 to 300w with mosfets
Mr astx creating some awesome amps. Look at his thread. Very well documented and also rebuild by some users:
Mos-Fet Amps threadl
Mos-Fet Amps threadl
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.