Introduction
Global warming concerns and increasing cost of energy to end-consumers have increased energy consumption awareness among consumers. Governments and consumer agencies worldwide have been addressing this issue by formulating policies to decrease the waste and cost of energy along with adding capacity to build energy from cleaner sources and cheaper. This focus has led to many changes in market dynamics and thus has defined the technology needs as well.
One such policy that has far reaching impact is the introduction of star ratings on various appliances and consumer goods that run on-grid. The star ratings provided for appliances and consumer electronics goods are based on energy consumption and efficiency. The better a product is in-terms of energy ratings and performance – the higher ratings they get and thus making them preferable. Along with ratings, awareness on the direct impact on energy bills to end-consumers have made this appliances preferred by consumers given the overall economical and ecological benefits. For example, star ratings in appliances like air-conditioner or refrigerator in developing markets such as India has impacted the consumer behavior dramatically. Within few years of introduction and promotion of star ratings by Bureau of Energy Efficiency (BEE) for air conditioners, all of 80 Million air-conditioners that are sold in India are now star rated. Similar trends are observed for other appliances’ markets as well.
The implications of above trends for a design engineer can be simply stated - as the need for designing energy efficient system solutions which consume the least amount of energy, leading to power savings for the consumer as well as decrease the load at grid. Thus making loads smart and efficient while grid becomes smarter is a top global priority, which latest technology is set to address.
How design process technologies help
There is no single power reduction technique that is able to meet all the system requirements for energy minimization. The trick is to effectively combine architectural, platform and circuit techniques, system and application software, process technology and design methodology and tools to intelligently develop semiconductor designs for energy-efficient operation in all applications.
Freescale’s Energy Efficiency Target illustrates how these technologies and techniques are integrated into the full development process designed to achieve optimal energy efficiency. Each area of technology can be optimized toward more efficient operation while still contributing to overall performance goals.
Freescale Energy Efficiency
At the architectural level, energy efficiency technologies use circuit techniques to enable energy savings across the chip design. Using multiple power modes is a good example. On-chip power modes are designed to offer peak application performance (and attendant energy consumption) only when absolutely necessary. To deliver optimal energy efficiency over the life of the application, on-chip power modes such as run, wait, stop and standby, are used to manipulate power usage to get the most efficient use of the available energy source. This is a particularly effective strategy for portable hand-held devices and office automation systems on a LAN that periodically engage in short bursts of activity.
There are various other design techniques that can be implemented to ensure energy efficiency as well. This includes techniques such as Dynamic voltage and frequency scaling (DVFS) which allows on-the-fly frequency adjustment according to existing system performance requirements. Similarly Clock gating is another effective strategy that is widely used to help reduce power consumption while maintaining the same levels of performance and functionality. Clocks can consume as much as 40 percent of active power. By shutting off the clocks and stopping the data toggling in unused portions of the semiconductor we can realize sizable energy savings, particularly when the gating is engineered to control the toggling at the individual instruction level.
Along with various design techniques, design methodology and tools used plays an important role in ensuring overall energy efficiency. Support/analysis, design, implementation, architecture and power estimation tools help ensure system-to-silicon IC design optimization across power, throughput, latency and area constraints. They help designers create a reliable methodology for energy-efficient semiconductor design.
Just like various techniques at hardware level, software too plays a significant role in how efficiently a system performs. Software-based power management provides a flexible and scalable framework that communicates with hardware through device drivers, manages use-case policies, models performance requirements real-time and responds to external interfaces and event notifications.
Process technology is building block of any semiconductor product. Hence characteristics of process technology used play a critical role in energy consumption of a semiconductor product. Various process techniques can be employed to reduce the leakage at device level which in-turn can reduce the overall power consumption of a product. Some of the prevalent techniques include multi Vt process where the lower Vt device can give good performance but can also be leakier. Hence using multi Vt technique, lower Vt device can be placed at critical paths which need extra performance. The remainder can be implemented with higher Vt devices to ensure lower leakage, thus ensuring optimal performance along with energy efficient products.
Energy Efficient Solutions from Freescale
Let us look at the popular microcontroller and microprocessor products used in Industrial market applications today and are proven to consume low power.
ARM Cortex M0+ (low power) solution – Kinetis L 
:
The demand for ever lower-cost products with increasing connectivity (e.g. USB, Bluetooth, IEEE 802.15) and sophisticated analog sensors (e.g. accelerometers, touch screens) has resulted in the need to tightly integrate analog devices with digital functionality to pre-process and communicate data. Most 8-bit devices do not offer the performance to sustain these tasks without significant increases in MHz and therefore power. Hence embedded developers are required to look for alternative devices with more advanced processor technology. The 16-bit devices have previously been used to address energy efficiency concerns in microcontroller applications. However, the relative performance inefficiencies of 16-bit devices mean they will generally require a longer active duty cycle or higher clock frequency to accomplish the same task as a 32-bit device.
The Kinetis L series is an entry-level 32-bit MCU family built on the ARM Cortex
-M0+ core. It combines exceptional energy-efficiency and ease-of-use with the performance, peripheral sets, enablement and scalability of the Kinetis 32-bit MCU portfolio, while leveraging the inherent low-power and high-performance features of the ARM Cortex architecture.
Manufactured using Freescale’s low-leakage, 90 nm thin film storage (TFS) process technology, the Kinetis L series frees power-critical designs from 8- and 16-bit MCU limitations by combining excellent dynamic and stop currents with superior processing performance. A broad selection of on-chip flash memory densities and extensive analog, connectivity and HMI peripheral options enable increased energy efficiency and intelligence for a range of applications, including battery-operated devices, medical systems, smart meters and motor control systems.
Arm Cortex M4 based Kinetis K Family:
Freescale’s ARM Cortex-M4 range of Kinetis microcontrollers support up to ten low power modes, which when combined with high performance and parallel computing results in reduction of average current consumption.
Low power 8-bit Microcontroller – S08LL
The S08LL16 is Freescale’s most efficient 8-bit LCD controller, having less than 50 percent of the current draw of previous generation Freescale S08LC60 devices. Based on Freescale’s QE family LVLP
technology, the LL16 MCU demonstrates extreme low-power characteristics in power specs and power modes for battery-powered portable applications in the medical, consumer and industrial markets.
The LL16 MCU achieves extraordinary low-power results through:
• An improved time-of-day (TOD) module with reduced functionality from previous modules (a simple quarter-second counter): less switching = less power. In addition, the peripheral logic runs at
a reduced speed of 2 Hz rather than at the bus clock and the reset synchronizer is clock gated to reduce unnecessary power consumption.
• An improved internal clock source (ICS) with reduced-length clock traces between the ICS, oscillator and TOD modules: less length = less capacitance = less power.
• LVLP features shared with the QE family of MCUs include VLP oscillator, low-dropout standby regulator, 6 μs stop 3 wake up time, low-power run and wait modes, SATO, user-selectable
peripheral clock gating and clock tree synthesis.
These low-power features enable consumer, industrial and wireless LCD applications that provide ultra-long life (up to ten years) using a variety of batteries, including AA, AAA, lithium coin size and 3.6V lithium.
Optimal balance of power, performance and integration: i.MX Application Processors
Freescale ARM-based i.MX processors provide the most versatile platform which delivers an optimal balance of power, performance and integration to enable next-generation smart devices which includes power sensitive multimedia and display applications. i.MX solutions include processors based on ARM9, ARM11, ARM Cortex-A8 and ARM Cortex-A9 core technologies and are used in low-power multimedia processing and integration in general embedded, automotive, industrial and consumer applications. These processors are known for their industry leading power consumption figures and operate without a heat-sink making them Ideal for battery operated portable devices; some i.MX–based devices can last two or more weeks on a single charge. There are multiple low-power modes which help optimize power and performance. Power savings can also be obtained through integrated power management (i.MX233) or paired with external Freescale power management IC solutions.
Very high performance with low power – QorIQ and PowerQUICC
Freescale’s advanced QorIQ platforms and PowerQUICC processors are the at the heart of the world’s networks, the mobile wireless infrastructure, the smart grid, the automated factory, the intelligent hospital or aerospace and defense and provide highly integrated architecture that includes single or multiple cores, accelerators, security, connectivity and yet consume extremely low power. Low Power enables Fanless operation at 85°C. Typically, 1600 MIPS processors consume <1W and 2000 MIPS processors (P1010) consume only <3W.
Advanced Energy Management feature of these platforms allow implementation of complex networking solutions while consuming low power. These techniques include: Data path for network traffic during deep sleep, addressing packet loss/overflow on power down, using core clock frequency scaling jog mode to reduce power consumption for low work load and using separate power planes for platform vs. core to enable power off to core and cache for deep sleep mode.
Rewards of energy efficiency
As energy costs are on the rise, it is imperative for end systems to consume as minimal power as possible. Thanks to Freescale’s cutting edge technology, there is a broad portfolio of energy efficient products ranging from 8-bit to 32-bit microcontrollers and single to multi-core microprocessors which offer great flexibility in designing low power consuming end products. Energy Efficient products throughout the grid value chain benefits citizens of the world with both monetary and environmental benefits – making world smarter and greener!!
Authors:
- Meera Balakrishnan, Industrial Segment Lead, Freescale Semiconductor Inc.
- Viral Gosalia, Product Manager, Industrial & Multi-market, Freescale Semiconductor Inc.