bias point of class-D amplifier

[23g]

Hi all, I notcied that Class-D is working in a switching mode. Like for class A we have 360 conduction angle so the bias point is around the middle.
But I saw a tutorial claims that Class-D has an 0 conduction angle. What does that mean?And how do we choose the bias point for Class-D? Thanks!!

 

[D.A.(Tony)Stewart]

Class D means a saturated PWM switching amplifier.

In theory Class-D amplifiers can reach 100% efficiency, as there is no period during a cycle where the voltage and current waveforms overlap as current is drawn only through the transistor that is on.

In practise whether BJT's or MOSFETS are used as complementary switches, there is a saturated On resistance and bias depends on devices selected. There are methods also to prevent overlap of both high side and low side driver being on at the same time and is called "dead time", often in the 1 us range.

 

[23g]

Thank you.
So does it mean that we prefer biasing class-D in class-A for less distortions?

 

[betwixt]

No, the term 'biasing' in class-A refers to setting the mid conduction point so the signal added to the bias can make it conduct more or less.

In class-D there is no static bias at all, it is digital so is all off or all on. The bias referred to in this situation is the offset from half output in the PWM waveform so the averaged output can be made equally smaller or larger. It's a time concept not a current consumption one.

Brian.

 

[D.A.(Tony)Stewart]

Class A depends on avoiding the region of saturation with around a 2V margin where linearity from current gain drops rapidly. Negative feedback improves this linearity and also lowers output impedance, but also lowers input impedance by the feedback ratio.

Class D depends on avoiding the linear region by switching between two drivers, one in full saturation with the other fully off.

The impedance of the load is frequency dependant along and is determined partially by the Class D switching rate.
A balance between reduced DC losses but rising AC losses from Eddy Currents can be selected with desirable characteristics of the electronic switch and magnetic core.