Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

MOSFET power amplifiers

Status
Not open for further replies.
Hi,
I've been struggling for the last couple of hours with crossover distortion with push pull followers (with mosfets). Why does this occur? What the heck is biasing the inputs?
Anyone got a comprehensive explanation / article. I am a student, so please don't be too advanced.

Thanks
TAM
 
Crossover distortion occurs when there is a threshold difference between when the top transistor stops conducting and the bottom transistor starts conducting during the amplification of an AC signal. The time it takes to cross that threshold difference puts a glitch in the waveform which is crossover distortion. It can occur with both MOSFET and BJT stages.

To avoid this, a bias network at the push-pull input generates sufficient voltage with no signal that both transistors are biased slightly on (called a Class AB stage), thus there is no dead spot when reproducing an AC signal (and thus no crossover distortion).

Here's a reference that should help further.
 
Thank you very much.
So when the inputs are biased, the MOSFET is slightly on, when an input signal is applied, the MOSFETs are already on, and will respond. Am I understanding this correctly?

Also, I understand distortion known as 'clipping', what measure can be taken against it?
 
Last edited:
Hi,

Yes that sounds right. With no signal both upper and lower MOSFET's are turned on a little bit, so when the signal goes through zero level there is no time when the MOSFET's are both off. Keeping them on slightly even with zero input means much less crossover distortion.
The trick is to design a bias scheme that will keep them both on slightly without too much current flow (for zero input). They are both in series across the power supply rails so too much current flow is not a good idea as it just heats up the MOSFET's.
 
Thank you very much.
So when the inputs are biased, the MOSFET is slightly on, when an input signal is applied, the MOSFETs are already on, and will respond. Am I understanding this correctly?

Yes, post your circuit so we can see what's going on (click on 'more options' and 'upload a file').

Here's a typical circuit from 'back in the day', the 1K pot is for adjusting the bias.

https://www.angelfire.com/sd/paulkemble/sound7g.html

Also, I understand distortion known as 'clipping', what measure can be taken against it?

Don't overdrive the amp, fitting a limiter before it is one solution - but it really depends what you're wanting to do with the amp, generally HiFi amps don't have any limiting, and only a fairly small number of PA amps do.
 
The way I understand it, this is what the, " Vgs threshold ", one a mosfet data sheet is used for. Correct?
 
Vgs threshold is a range of voltages, not a single voltage rating because each Mosfet is different. Therefore the bias voltage must be adjusted to match whatever Vgs thresholds your Mosfets have.
The current in a Mosfet at its Vgs threshold is probably too low to reduce crossover distortion.
 
Hi,
I've been struggling for the last couple of hours with crossover distortion with push pull followers (with mosfets). Why does this occur? What the heck is biasing the inputs?

This occurs because MOSFETs are hideous devices when used with the usual SEPP OTL topology that was originally designed for BJTs. The problem is that complimentary N-Channel/P-Channel pairs are a good deal less "complimentary" than are NPN/PNP complimentaries. The end result is nasty x-over behavior. When using MOSFETs, it's either N-Channel or P-Channel -- one or the other, but not both. That means that the SEPP topology is out. You can either use a Circlotron topology (complication: needs two floating power supplies) or an OPT. That's not done since no output xfmr is way cheaper than the cheapest OPT.

"Big Box" amps that use MOSFET finals are sonic abominations which the marketing dept. tried to sell by claiming the "MOSFET sound" was something new and uber-kewel when all it ever was was distortion. These thingies produce so much high frequency garbage they poof tweeters.

BJT finals sound much better, in the order of: Sziklai Pairs, Quasi-Complimnetary, Emitter Follower.

Anyone got a comprehensive explanation / article. I am a student, so please don't be too advanced.

Thanks
TAM

Here ya go: Power amplifier Design (Doug Self's site)
 
Hi,

When mosfet's first started to become popular back in the 1980's i had designed an amplifier using power mosfets and op amps as the mosfet controllers. The mosfet gate voltage changes with temperature so it's easier to NOT try to adjust the gate voltage to get the right quiescent current flow to avoid crossover. The trick is to design two voltage followers, one with an N channel and one with a P channel, the N channel putting out a tiny bit higher voltage than the P channel, then put them together with small resistors between them. The quiescent current then becomes a function of the two small resistors rather than the gate voltage. The N channel stage sources the output current and the P channel sinks the output current, so we get full wave operation similar to a good bipolar design except there is almost no crossover distortion.

For example, with a 0.1v offset in the N channel amp and 0.5 ohm output resistors, the quiescent current flow is 100ma:
Iq=Voffset/(R1+R2)

That virtually eliminates all crossover distortion.

We must also be aware that this is still a linear amplifier however, so there will still be high dissipation in the two transistors.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top