THAT＇S CLASS: Audio Amplifiers from A to T （3）

Published:2011/8/2 1:13:00 Author:Li xiao na From:SeekIC

By Harry Baggen

Classes 5 and T

Although the working principle of the Class D amplifier is already several decades old. it never managed to become truly established in hi-fi applications. This was primarily due to excessive distortion and a lack of good semiconductor devices (fast power FETs). In the meantime, several manufacturers have devised variations on this theme, and in many cases they have given them their own designations. For instance. Crown came up with the Class I amplifier. Sony developed its S-Master technology, and Tripath has devised its Class T amplifier. Unfortunately, the nice alphabetic sequence has been abandoned in favour of manufacturer-specific abbreviations. In its S-Master technology. Sony combined several techniques to make the Class D configuration suitable for domestic hi-fi applications. Here the process of convening the incoming signal into a corresponding pulse-width signal is called ’complementary pulse length modulation’ (C-PLM). Extensive atten-tion was given to suppressing jitter. This was accomplished by using an extremely accurate clock signal and a circuit called "clean data cycle’ that corrects the positioning of the output pulses if necessary (see Figure 6).



The method used to implement volume control is certainly an unusual feature of the Sony approach. In a normal Class D design, the full pulse waveform is always present at the output. with an amplitude of 50 to 100 volts peak-to-peak. Particularly with small output signals, it is very difficult to fully eliminate all residual components of the pulse waveform from the filtered signal. In the Sony design, the volume is regulated by adjusting the supply voltage for the output stage. This prevents any information from being lost at low signal levels. This technique has an effective range of 50 dB.

Another company. Tripath. has developed a technique that according to them combines the signal quality of Class A and AB amplifiers with high efficiency (around 80-90 %). This is done using a combination of analogue and digital circuitry; together with digital algorithms that modulate the input signal using a high-frequency switching waveform. The algorithms developed by Tripath are derived from adaptive and predictive algorithms already used in telecommunication systems. With the Tripaih amplifier, the majority of the analogue and digital circuitry is housed in a single IC. which may also include the output transistors (depending on the power). The block diagram of the amplifier is shown in Figure 7. The input signal is first buffered by an input stage. From there it passes to the Digital Power Processing block, which contains the signal processor, a digital conversion function, mute switching, overload protection and error detection. The output stage is driven via the qualification logic, and die loudspeaker is connected to a filter following the output stage. Thanks to its special algorithms, the processor in the Class T amplifier can shape the drive signal for the output stage to match die specific characteristics of the transistors used in that stage. This shaping takes into account non-ideal switching behaviour, mismatching of the complementary output transistors, dead-time distortion and the residual energy of the oscillator in the audio band.



The switching frequency of a Class T amplifier is continuously adapted to the magnitude of the input signal. At low input levels, the switching frequency is quite high (around 1.2 MHz). This has a beneficial effect on signal quality. The switching frequency gradually drops as the input level increases, in order to increase efficiency. The switching frequency ultimately reaches its lowest value (around 200 kHz) when the output is driven to maximum amplitude. Besides this, a form of noise shaping is applied to the peaks of the output signal in order to improve the signal waveform. As a result of all these measures, the Class T amplifier can deliver a sound impression that reminds listeners of audiophile analogue amplifiers.

The future

With the steady advance of digital audio, some form of digital output stage will ultimately be found in many consumer amplifiers. The reasons for this are higher efficiency, smaller size and lower manufacturing cost. It’s difficult to estimate whether this development will also come to prevail in the high-end realm. There are presently only a few high-quality digital amplifiers on the market. But if you’d like to try it for yourself, you can start by building the Clarity amplifier described in this issue.

Reprinted Url Of This Article: http://www.seekic.com/blog/project_solutions/2011/08/02/THAT＇S_CLASS__Audio_Amplifiers_from_A_to_T_（3）.html

Print this Page | Comments | Reading(319)