SIGNAL CHAIN BASICS #59: Saving power and minimizing heat with power supply volume control
(Editor’s note: Signal Chain Basics is an ongoing and popular series; click here for a complete, linked list of all installments.)
In open-loop Class D audio amplifiers, the audio level at the speaker terminals can be defined by Equation 1:
Audio Level =
(PWM modulation index) × (power supply voltage).
A look at some of the pulse-width modulation (PWM) devices on the market may lead to a section on power-supply volume control (PSVC). Some of you may already have tinkered with this function, while others may have blown straight through the datasheets, without realizing what a cool piece of technology PSVC can be. In home audio applications, most buyers are wowed by a “1000W home theater” or “600W subwoofer.”
However, in most home-TV settings, average listening levels are below 20W of power. Let’s look at a standard home-audio system to see how it can be tuned for higher efficiency, lower heat dissipation, and greater signal-to-noise ratio (SNR).
Figure 1 is a block diagram of a fairly common home audio system. Multiple amplifiers all run from a fixed, high-voltage power rail. The aim is to throw any signal from the audio source with enough headroom in the power supply to handle it. Now your system is ready to pump out 1000W at a moment’s notice!
Figure 1: A traditional home-audio system with a fixed power supply.
However, in the real world, most users don’t run their home audio systems with the volume knob maxed out at 11. Most users have the volume knob turned down to two or three. By doing so, you’ve immediately lowered your maximum potential output from your amplifier. This means that your maximum PWM modulation from a 50V power rail may only be 30 percent or so.
Many amplifiers with a zero signal continue to switch at a 50-percent duty cycle, which means you have 50V being switched on and off through the FET transistors in the H-bridge. FETs have a finite (although very small) on-resistance, therefore, the power is dissipated as heat (Equation 2):
P = V2/R
By reducing the power-supply voltage by half, the power dissipated by the power stage, when muted (but still switching at 50/50), is quartered. This essentially pushes a power stage from idling at 2W per bridge-tied load (BTL) channel. Think about six of these per system for a 5.1, or eight of these in a 7.1 home theater in a box (HTIB) in the power stage to 500 mW per BTL channel, This dissipation scales up and down with the VCC of the power stage.
This directly impacts the size of the heatsink required for your amplifiers and reduces the airflow requirements, too.
Impact on Class D amplifiers
Open-loop amplifiers:
In summary, power-supply modulation in open-loop amplifiers can achieve two different objectives:
- Power idle-current consumption
- Direct volume control.
If all you’re looking to achieve in the open-loop amplifier is lowered idle-current consumption, then this functionality requires a little more thought and planning. For instance, if power-supply modulation is in mid-playback, gain in the digital signal must be done to compensate for the lower power supply. Put simply, if you halve the power supply but expect the same average-voltage output at the speaker terminal, you’ll need to double the PWM duty cycle.
For example, in the TAS55xx PWM modulators, the digital gain compensation is integrated as part of the modulator. The gain required can be calculated for you, (For more details, see page 41 of the TAS5508C 8-Channel Digital Audio PWM Processor Data Manual,)
If all you need it power-supply volume control, then simply allow the I2S-to-PWM modulator work at gain G=1, and then modulate the power supply for different volume outputs.
If we now update our block diagram, we get:
Figure 2: A higher-efficiency home-audio system, using PSVC.
Closed-loop feedback amplifiers:
Closed-loop feedback amplifiers have their own personalities when it comes to power-supply modulation. A closed-loop amplifier maintains the same amount of gain on the signal (i.e., same output signal) whenever the power supply is greater than the output signal.
For instance, if the output voltage needs to be 30 V, then with a 36 VCC power supply, you’ll be smooth sailing. It is the same at 50 VCC.
However, at 28 VCC, the output becomes distorted. This means that it is not possible to change the output level by changing the power rail in a closed-loop amplifier system.
Nonetheless, all the advantages in lowering idle-power consumption still apply. Less power needs to be dissipated in the power stage at low levels. Additionally, idle noise is lowered, improving SNR at low levels. An important consideration here is that the PSVC module must not scale the data, as the gain compensation has been taken care of in the power stage, not the modulator,
•For more details on the closed loop amplifiers and a compatible Class G power supply, check out this reference design: www.ti.com/tas5630-ca.
•For more information on the I2S -> PWM Modulators, visit: www.ti.com/digitalaudio-ca.
Please join us next month when we will talk about tracking down spurs in high-speed data converters.
About the Author
Dafydd Roche is the home audio strategic marketing and systems engineer for the Audio Converter group at Texas Instruments. An avid musician in his spare time, Dafydd pours his passion and knowledge of audio and music-making into his work.
If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :
