MOSFET Quasi Complimentary ???

[IIR]

Could someone please explain how a MOSFET quasi complementary output stage (all N-channel FETS) really works? Kind like the one off www.aussieamplifiers.com

Are the high-side FETs operating as source followers (for the postive side) while the low-side FETS acting like transconductance amps (for the negative side)? Please help!

 

[mikek]

...get a-hold of douglas self's 'audio power amplifier handbook'...third edition......
In brief, quasi-complementary MOSFET output stages are sub-optimal in nearly all respects......

 

[Kilowatt]

Unless of course you (for some reason or other) need to use devices that have no compliment.

Randy Slone's "High Power Audio Amplifier Construction Manual also explains how they work.

 

[Circlotron]

N-channels rule!

Look in any catalogue and you will find seriously big N-channel Hexfets and IGBT's. Big enough to need bolts to attach the conductors. In contrast, P-channel devices wimp out comparitively early on the buahahaha scale. Super duper sized amplifiers (not necessarily audio) may well need to use a pair of N-channel devices for this reason.

GP.

 

[UrSv]

The abundance of big N-channel FETs and lack of P-channel FETs could also be because the biggest ones are designed for switching which means no P-channel needed...hence no complimentary.

/UrSv

 

[Nelson Pass]

If you need an example of a quasi-comp Mosfet output
stage, you can check out the 60 watt Mosfet upgrade on
an HK Citation 12 at www.passdiy.com

Sounds tho, like D. Self doesn't like quasi-comp Mosfets

 

[carlmart]

D. Self doesn't like FET or MOSFET at all, complementary or not.


Carlos

 

[Kilowatt]

And Slone loves them.

 

[Nelson Pass]

I like to be agnostic about it, and I've made plenty of
money with both.

Bipolar are great if you are trying to get decent distortion
with a low bias AB amp. Home theater stuff is a good example.

Mosfets are the choice if you have high voltages and power,
or if you want simple Class A circuits. In general, they tend
to be easier to design with if you can afford lots of bias.

 

[hifidaddy]

Here is a quasi-comp Mosfet power amp

Hi,

I just browsed my MJ mag articles and found a kaneda Power amp from 1999, which has both N-type Mosfets in the output stage. Though he drives them rather symmetrically, in my opinion.

regards,
Hartmut

 

[haldor]

And Slone loves them.

Slone does state MOSFETs will have more distortion than bi-polar. Self is just a bit more retentive about distortion than Slone.

Phil

 

[Kilowatt]

He does, but he also states that the difference in distortion is pretty much insignificant for well-designed amps. This is a matter of opinion. He just likes them (especially for PA amps) because they don't blow up as easily.

 

[haldor]

Hi Kilowatt,

I agree, I built a pair of Slone's 250W MOSFET amps and am very happy with them. Self just has different priorities and going from 0.005% distortion to 0.001% is worth it to him. I'd rather have an amp that is harder to smoke.

Phil

 

[Nelson Pass]

Properly designed, the Mosfets are a lot more rugged
since they don't have second breakdown at high
voltages. Your typical 250 watt Bipolar becomes a
wimpy 50 watt device at 150 volts. Not so with the
Mosfet.

 

[nigelwright7557]

Properly designed, the Mosfets are a lot more rugged
since they don't have second breakdown at high
voltages. Your typical 250 watt Bipolar becomes a
wimpy 50 watt device at 150 volts. Not so with the
Mosfet.

MOSFET amps lose about 7 volts across the bias stage so unless you use a slightly higher voltage rail for the driver stage you lose a bit of power.

I designed a 450WRMS MOSFET power amp it has been very reliable for disco and guitar work.
I do tend to overdesign my amps with huge heatsinks and a pair of fans to keep it cool. It just never even gets warm.
I did find care has to be taken setting up the bias level and output DC offset voltage.