www.ti.com

Power Management

1.5Power Management

1.5.1 Overview

In many applications there may be specific requirements to minimize power consumption for both power supply (and battery) and thermal considerations. There are two components to power consumption: active power and leakage power. Active power is the power consumed to perform work and, for digital CMOS circuits, scales roughly with clock frequency and the amount of computations being performed. Active power can be reduced by controlling the clocks in such a way as to either operate at a clock frequency just high enough to complete the required operation in the required time-line or to run at a high enough clock frequency until the work is complete and then drastically cut the clocks (that is, to bypass mode or clock gate) until additional work must be performed.

Leakage power is due to static current leakage and occurs regardless of the clock rate. Leakage, or standby power, is unavoidable while power is applied and scales roughly with the operating junction temperatures. Leakage power can only be avoided by removing power completely.

The DSP has several means of managing the power consumption, as detailed in the following sections. There is extensive use of automatic clock gating in the design as well as software-controlled module clock gating to not only reduce the clock tree power, but to also reduce module power by freezing its state while not operating. Clock management enables you to slow the clocks down on the chip in order to reduce switching power. Independent power domains allow you to shut down parts of the DSP to reduce static power consumption. When not being used, the internal memory of the DSP can also be placed in a low leakage power mode while preserving the memory contents. The operating voltage and drive strength of the I/O pins can also be reduced to decrease I/O power consumption.

Table 1-19summarizes all of the power management features included in the DSP.

Table 1-19. Power Management Features

Power Management Features

 

Description

 

 

 

 

Clock Management

 

 

 

PLL power-down

 

The system PLL can be powered-down when not in use to

 

 

reduce switching and bias power.

 

 

 

Peripheral clock idle

 

Peripheral clocks can be idled to reduce switching power.

 

 

 

 

Dynamic Power Management

 

 

 

Core Voltage Scaling

 

The DSP LDO and DSP logic support two voltage ranges to

 

 

allow voltage adjustments on-the-fly, increasing voltage during

 

 

peak processing power demand and decreasing during low

 

 

demand.

 

 

 

 

Static Power Management

 

 

 

DARAM/SARAM low power modes

 

The internal memory of the DSP can be placed in a low leakage

 

 

power mode while preserving memory contents.

 

 

 

Independent power domains

 

DSP Core (CVDD) and USB Core (USB_VDD1P3, USB_VDDA1P3)

 

 

can be shut off while other supplies remain powered.

 

 

 

 

I/O Management

 

 

 

I/O voltage selection

 

The operating voltage and/or slew rate of the I/O pins can be

 

 

reduced (at the expense of performance) to decrease I/O power

 

 

consumption.

 

 

 

USB power-down

 

The USB peripheral can be powered-down when not being

 

 

used.

 

 

 

1.5.2 Power Domains

The DSP has separate power domains which provide power to different portions of the device. The separate power domains allow the user to select the optimal voltage to achieve the lowest power consumption at the best possible performance. Note that several power domains have similar voltage requirements and, therefore, could be grouped under a single voltage domain.

SPRUFX5A –October 2010 –Revised November 2010

System Control

33

Submit Documentation Feedback

 

 

Copyright © 2010, Texas Instruments Incorporated

Page 33
Image 33
Texas Instruments TMS3320C5515 manual Power Domains, Power Management Features

TMS3320C5515 specifications

The Texas Instruments TMS3320C5515 is a highly specialized digital signal processor (DSP) designed for a wide range of applications, including telecommunications, audio processing, and other signal-intensive tasks. As part of the TMS320 family of DSPs, the TMS3320C5515 leverages TI's extensive experience in signal processing technology, delivering robust performance and reliability.

One of the main features of the TMS3320C5515 is its 32-bit architecture, which allows for a high level of precision in digital signal computation. The processor is capable of executing complex mathematical algorithms, making it suitable for tasks that require high-speed data processing, such as speech recognition and audio filtering. With a native instruction set optimized for DSP applications, the TMS3320C5515 can perform multiply-accumulate operations in a single cycle, significantly enhancing computational efficiency.

The TMS3320C5515 employs advanced technologies including a Harvard architecture that separates instruction and data memory, enabling simultaneous access and improving performance. Its dual data buses enhance throughput by allowing multi-channel processing, making it particularly effective for real-time applications where timely data manipulation is critical. The device supports a wide range of peripherals, facilitating connections to various sensors and communication systems, which is vital in embedded applications.

In terms of characteristics, the TMS3320C5515 operates at an impressive clock speed, providing the computational power necessary to handle demanding tasks. The device is optimized for low power consumption, making it ideal for battery-operated applications without sacrificing performance. Its flexibility in processing algorithms also allows it to be readily adapted for specific requirements, from audio codecs to modems.

Another noteworthy aspect is the extensive development ecosystem surrounding the TMS3320C5515, which includes software tools, libraries, and support resources designed to accelerate the development process. This allows engineers and developers to bring their projects to market more quickly while minimizing risk.

Overall, the Texas Instruments TMS3320C5515 stands out as a powerful DSP solution, equipped with features that cater to the needs of various industries. Its combination of performance, efficiency, and versatile application makes it an attractive choice for engineers working in signal processing.