EEWeb Pulse - Volume 8 by EEWeb Magazines

EEWeb

PULSE

EEWeb.com

Issue 8 August 23, 2011

Jeritt Kent

Analog Devices

Electrical Engineering Community

Itâ&#x20AC;&#x2122;s all about

connections. EEWeb Electrical Engineering Community students

hobbyists

discussions

technical documents

industry experts engineers

white papers links

resources

1.800.574.2791application notes

advertising@eeweb.com power

www.eeweb.com/advertising lighting

community sensor

microcontroller wireless

The user-to-user forum is for everyone, from design engineers to hobbyists, to discuss technology, products, designs and more. Join the discussions that match your interest or offer your expertise to others. Join the discussion now at:

www.digikey.com/techxchange Digi-Key is an authorized distributor for all supplier partners. New products added daily. ÂŠ 2011 Digi-Key Corporation, 701 Brooks Ave. South, Thief River Falls, MN 56701, USA

TA B L E O F C O N T E N T S TABLE OF CONTENTS

Jeritt Kent ANALOG DEVICES

Interview with Jeritt Kent - Senior Staff Applications Engineer, Analog Devices

8-Bit MCU Relevancy in 2011

BY STEVE DARROUGH WITH ZILOG As technology continually improves, the relevancy of 8-bit microcontrollers is drawn into question. Darrough discusses 8-bit microcontrollers and their future in various technology arenas.

Pervasive Power and Point of Load Delivery BY STEVE GRADY WITH CYMBET Based on fundamental power distribution and energy storage techniques, Grady introduces several new concepts for micro-power electronic system design.

RTZ - Return to Zero Comic

EEWeb | Electrical Engineering Community

Visit www.eeweb.com

INTERVIEW

How did you get into electronics/engineering and when did you start? Well, it is an interesting story. I started my collegiate studies in the late 1980s at Gonzaga University in Spokane, Washington on scholarship. As a junior in high school at Bishop Kelly in Boise, Idaho, I participated in the Junior Engineering Technical Society (JETS) program, and that certainly sparked my interest in becoming an engineer. While at Gonzaga, I originally considered mechanical or chemical engineering. I participated in core classes and an option class, “Introduction to Circuits.” There were actually two class options: one would satisfy the requirements for a mechanical engineering degree and the other was an advanced version that satisfied the requirements for an electrical engineering degree. I decided to take the advanced class thinking, “What if…?” As it turns out, I really enjoyed the class!

engineering program at Idaho was very strong. Many of my good friends planned to participate in Gonzaga’s Florence, Italy program for their junior year. This helped with the emotional side of my decision to transfer to Idaho. After transferring to Idaho, I had to learn a lot of math and related

During my sophomore year, I had a conversation with my advisor, Dr. Dennis Kelsh, about chemical engineering. As the chemical engineering program was not in place at Gonzaga, we discussed the idea of transferring to the University of Idaho. I visited the Moscow campus that summer and quickly realized that the

EEWeb | Electrical Engineering Community

Can you tell us more about working at AMI? It really was a tremendous opportunity, as jobs were very

Jeritt Kent - Senior Staff Applications Engineer, Analog Devices

Visit www.eeweb.com

FEATURED INTERVIEW

Jeritt Kent Analog Devices

material in order to succeed. The timing was very good, and it was a tremendous gift to have the opportunity to study under professors Earl Gray and Calvin Finn—two of Idaho’s premier teachers in analog design. I decided to stay in Moscow, and went on to receive my Bachelor’s Degree with a specialty in analog CMOS circuit design. I interviewed and accepted a position with American Microsystems (AMI) in Pocatello, Idaho as an Automotive Project Design Engineer.

INTERVIEW

A project engineer had primary responsibility for every aspect of a development: design, bond diagrams, specifications, test programs, and everything in between. While involved with many challenging and interesting projects at AMI, I was initially associated with the device that, at that time, was arguably the highestvolume Application Specific Integrated Circuit (ASIC) in the world; this was before cell phone Integrated Circuits (IC). This was called the PRNDL, which stands for Park Reverse Neutral Drive Low. The PRNDL is responsible for indicating the gear that a driver selects for the vehicle (usually via an LED). That IC went into almost every General Motors (GM) car made in the 1990s. Many might think that it is a simple device, but one thing about building a chip in the automotive industry is that it has to be at least 99 percent fault graded. This means that any internal fault needs to cause something at the pin level of the device to change. Since the only measurable interface on an IC is at its pins, the silicon and the test programs must work together to fully exercise the device to pass any potential flaws out to where they can be seen so that a bad part can be rejected. I am still very proud of this development; the test program and silicon modifications that I helped define with the team

allowed this part to be tested much more efficiently. Automotive devices need to meet tough environmental specifications like temperature and electrostatic discharge (ESD). While we were designing chips for GM, we spent a lot of time doing analysis on crystal oscillators and ceramic resonators. One of the big challenges is getting a circuit like that to work over a -55°C to 125°C environment. This is no small task, but we found ways to do just that. A lot of time was also spent developing some of the first high-voltage ESD tolerant circuits using punchthru devices and parasitic bipolar transistors. Jerry, Bob Klosterboer, and I spent a lot of time trying different structures on different test chips. One of our chips passed 4KV on every pin, and this was back at a time when this was very hard to do. My most memorable design was the first octal CMOS integrated smart carburetor driver for natural gas powered vehicles. Motorola had built individual bipolar drivers, but AMI’s Canadian customer wanted a single chip. This was a tremendous challenge. The simulation models that needed to be developed were cutting-edge. Saber Cadat was in its genesis. We used every Mentor tool we had: HSPICE, BSIM modeling, Cadence, and Synopsis. I was most proud when the seven state-variable, dual-feedback mixed signal circuit that I designed via Karnaugh maps successfully drove the TMOS IV power FETs in 60 nanoseconds while critically damped! The model of a low

EEWeb | Electrical Engineering Community

side solenoid driver with flyback control is still, in my opinion, one of the most encompassing electrical engineering challenges in the analog domain. Where did you go after AMI? In 1995, after four years in Pocatello, I was offered a position to replace one of my mentors, Art Tan, at the AMI representative office in Fort Wayne, Indiana. Having been a part of the teams that built a lot of the chips that were being integrated into vehicle systems at Delco and GM, I moved to Fort Wayne—the halfway point between Kokomo, Indiana and Flint, Michigan—the two main design locations for Delco Electronics. My stay was a little over a year as the Field Applications Engineer for the North Central Territory. During the summer of 1996, while celebrating with my grandparents for their 60th wedding anniversary on the Oregon coast, I noticed an ad in the Oregonian for semiconductor sales for the Pacific Northwest. I interviewed and accepted a position with Allegro Microsystems and moved back to Boise in late 1996 having various responsibilities in applications, marketing, distribution, and sales—a little bit of everything. When did you start with Analog Devices? On May 3, 1999, I went to work for Analog Devices. I remained in Boise for about ten months before moving to Seattle, Washington where I worked out of our Bellevue office for seven years. I was ADI’s first Linear Field Applications Engineer specifically

Visit www.eeweb.com

FEATURED INTERVIEW

scarce when I graduated. In January 1991, Jerry Downey hired me, and I was surrounded by some of the best Application Specific Integrated Circuit (ASIC) designers in the world. This design team gave me my first opportunity to work in mixed signal design.

INTERVIEW

As that happened, I had an opportunity to get involved in a specific area of RF as a result of work being done at some key customers who were adding communications capability to the electric, gas, and water metering infrastructure; the meter sometimes being referred to as a smart meter. One customer requested that I get much more involved with the key radio standards for this application space, namely 802.15.4g. My current role is RF and Systems Specialist for the Energy Segment within Analog Devices’ Industrial and Instrumentation division. What percentage of the nation has converted over to Smart Meters? It is evolving. There are three phases that companies are working on right now. There is Automatic Meter Reading (AMR), Automatic Meter Infrastructure (AMI), and then the Smart Grid. The AMI is more of a tactical application of adding twoway communication and some features to the structure of AMR, which is often a system of oneway communication to replace a meter reader. A main objective of

AMR was to keep the technician from having to go out and read the meters on your house. A primary objective of AMI is bidirectional communication allowing the utility to query and control the meter. The Smart Grid is really looking forward to the future to create a network that provides secure and robust means of measurement and control. The application possibilities for the Smart Grid are extensive, ranging beyond the simple task of

In order to be highly successful in analog you really have to have a rich history in cell development and maintain high team morale because a lot of what you are doing is based off of knowledge. gathering metrology information to saving lives. Japanese meter manufacturers, for example, have considered adding seismic measurement to a gas meter so in the case of a seismic event, the valve can be closed and possibly prevent a natural gas explosion.

EEWeb | Electrical Engineering Community

Another concept where the Smart Grid is likely to contribute is what some have called the Internet of Things (IOT): devices that range from vending machines to coffee pots in your home, connected into this dedicated grid-based network. Analog Devices is considered a world leader in metrology—the meter side of things—and RF. Some of my recent work is as a contributing member of 802.15.4g that meets every other month all over the world, defining the physical layer communications specifications for these Smart Grid applications. We are working diligently on the 4g standard because many believe that will represent a major part of the overall market. Where do you want to go from here? There will be career opportunities to “move up the ladder.” Right now though, the Smart Grid is, I believe, one of the most exciting areas in electronics and my team is right in the middle of it with global industry connections. I have been able to travel to many places because of these opportunities, visiting customers, learning from them, and engaging in successful business. The Smart Grid is a global objective; it will change the world. It is really exciting to imagine what it will do for countries like India, Brazil, China, and Japan, and how vital it could be for Europe. There are a whole lot of creative and valuable things you can do with the Smart Grid once the key building blocks are in place.

Visit www.eeweb.com

FEATURED INTERVIEW

assigned to the Northwest, transitioning later into a role as a communications engineer for the western U.S. My focus was Analog Devices’ GSM and EDGE radios and baseband devices before that part of the business was sold. After that business transaction, I became an RF specialist for the Americas, being involved with communication system design at many levels.

INTERVIEW In order to be highly successful in analog you really have to have a rich history in cell development and maintain high team morale because a lot of what you are doing is based off of knowledge. Analog Devices has been able to keep a great majority of the talent that it has harvested. It amazes me every month when I read the company website to see the number of personnel celebrating a 25th, 30th, or 35th anniversary. A lot of the reason Analog Devices is so successful is because of the people in its history that have designed the key intellectual property (IP) blocks that are shared with the new generation of engineers. Many of these designs are improved and managed to a new process or structure to further enhance and enable analog technology. Microelectromechanical systems (MEMS) are another key technology area, possibly best categorized as analog. We are in the very early days of understanding all of the possibilities that this technology will bring to the world. As digital gets better, faster, stronger, and more powerful, the requirement for the analog becomes more challenging. It gives you a chance to dig down and look at topics that were previously put on the back burner. Engineers really get a chance to operate “outside the box” looking at solutions. One of the certainties of the semiconductor industry is cyclicity. There have been groundbreaking discoveries and ideas in the past that have set

the next cycle. Gordon Moore, for example, threw down the gauntlet with “Moore’s Law” awhile back that still appears to hold. Breakthroughs will continue, I believe, mainly due to how effectively new-generation engineers can be educated through the previous efforts (successes and failures) of their mentors. What are some of the papers you have written?

Stability and Transient Analysis of the Miller-Compensated Linear Regulators on the ADP3178 w w w. a n a l o g . c o m / s t a t i c / imported-files/application_ notes/142101805AN593.pdf Interfacing the ADSP-BF533/ADSPBF561 Blackfin® Processors to High Speed Parallel ADCs w w w. a n a l o g . c o m / s t a t i c / imported-files/application_ notes/527407802AN813_0.pdf. ■

Following is a list of articles I have written or contributed to. They can be accessed at the URLs that accompany each title. The Smart Grid Communications Evolution: “Closing the Loop” for the Intelligent Electric Grid w w w. m w j o u r n a l . c o m / a r t i c l e . asp?HH_ID=AR_10003 Small Circuit Forms Programmable 4- to 20-mA Transmitter www.edn.com/ar ticle/480893Small_circuit_forms_ programmable_4_to_20_mA_ transmitter.php Encoder’s Spare Channel Embeds Whole-House Stereo Audio in Satellite Set-Top-Box Designs Stably and Cost-Effectively w w w. a n a l o g . c o m / l i b r a r y / analogDialogue/archives/39-07/ btsc.html Interfacing the ADSP-BF535 Blackfin® Processor to HighSpeed Converters (like those on the AD9860/2) Over the External Memory Bus www.analog.com/static/importedfiles/application_ notes/3EE162.pdf

EEWeb | Electrical Engineering Community

Visit www.eeweb.com

FEATURED INTERVIEW

Do you see the analog IC industry changing?

Avago Technologies Optocoupler Solutions

2 Times Faster, 50% Smaller, True Rail-to-Rail Output Voltage IGBT Gate Drives Applications Avago Technologies, an industry leader

in IGBT Gate Drive technology solutions IGBT/MOSFET Gate Drive introduces our Next Generation series!

AC and Brushless Based on BCDMOS technology Avago can DC Motor Drives deliver higher peak output current, better railRenewable Energy Inverters Industrial Inverters Switching Power Supplies

to-rail output voltage performance and faster speed than previous generation products. The increased drive and speed along with the very high CMR (common mode rejection) and isolation voltage will enable you to build more efficient and reliable motor drive and power conversion systems. In addition the SO6 package which is up to 50% smaller than conventional DIP packages facilitates smaller more compact design.

Benefits • Suitable for wide range of IGBT class for different market applications • High output peak current for fast and efficient IGBT operation • Rail-to-rail output voltage for reliable IGBT operation • Lower system power budget • Suitable for bootstrap power supply operation • Reduce dead time and improve system efficiency • Prevent erroneous driving of IGBT in noisy environment • 40%-50% smaller than DIP package for space and cost savings To request a free evaluation board go to:

www.avagotech.com/optocouplers

8-BIT MCU Relevancy in

2011

Steve Darrough

Director of Marketing, Worldwide

ndustry speculation continually exists across many past and present technologies. When looking at the evolution of microcontrollers, it is easy to wonder if and how long they will be relevant in such a fast-paced world that’s speeding down the technology highway. Certainly in some technology arenas, applications are ramping quickly into more and more sophisticated levels in which ARM and Atom are leading as powerful solutions and enjoying big buzz. Many in the industry would agree that as some products evolve into more richly-featured solutions, they are indeed the right solutions. But where does that leave the 8-bit micros that we all grew up on? Are they a thing of the past? Are they now just too archaic to have any real technical worth? Seemingly, there are plenty who believe 8-bits are now good for only the most basic functions, and that the future will have little use for them. Others may see this a bit differently – that although today’s fast pace has created a huge demand for better and more powerful microcontrollers, there’s also the larger arena that comprises many fast-growing technologies

EEWeb | Electrical Engineering Community

—technologies which compel the growing demand for many more 8-bits. Let’s just take one example in the energy management segment—although it’s certainly not a new arena or topic—and that’s the new emerging products which have recently gained far more traction in the marketplace. The change has resulted in broader awareness, justification for adoption, and far faster demand for more energy solutions than ever before. For the average person opening his or her electricity bill, the very thought of saving $30—$100 a month on home power consumption is a real wake-up call. When looking at it in terms of the smart home, intelligent appliances, or expansion of the smart grid, one can quickly see where there are tons of new product opportunities in which a certain level of smarts is needed to accomplish the energy and money-saving functions we demand. One might propose that as all of these solutions accelerate and expand across the globe, there will be several complex levels within each model. For instance, lighting control is quickly becoming one of the fastest areas in which people and businesses see instant return on their investments. Parking structures, factories,

Visit www.eeweb.com

TECHNICAL ARTICLE

Just one example of energy management, but one that’s easy to discern as the energy-hungry world looks for solutions, is that 8-bit MCUs are evolving alongside the unique embedded designs that are well-suited for the energy-saving tasks at hand—not overly so, but rather as just the right mix of peripherals and functionality for many of these new applications. The microcontroller market is growing, and although there are more types of solutions, my prediction is that 8-bit microcontrollers will be with us for a very long time!

TECHNICAL ARTICLE

schools, commercial and federal buildings consume huge amounts of power, and it doesn’t matter if anyone is present or not. Tomorrow’s solutions include relatively simple motion sensors along with lighting systems that are equipped with some form of unique embedded intelligence. Now, all of the sudden, a good deal of energy savings is possible and quite measurable. It seems that these systems solutions have several elements that will require some level of microcontroller management to work in harmony together. The system master or controlling device may well be a 32-bit controller, while the individual 8-bit end points that still beg for some intelligence will have a simpler function. When working as combined elements, the result is exactly what needs to be accomplished (i.e., when no one is around, the lights simply dim or shut off until someone reenters the area). Certainly, 32-bit or even 16-bit MCUs are not needed for end points such as wall plugs, switches, smart ballasts, even certain lighting fixtures. When examining deployment ratios, it is apparent that there are many more end points in many of these designs, and thus a large demand for our beloved 8-bit microcontrollers. They are not gone, and they most certainly are not forgotten!

Figure 1: Z16F Motor Controller Kit

About the Author Steve Darrough is Vice President of Marketing at IXYSZilog. Steve joined Zilog in 2008. Steve possesses more than twenty years of technical engineering and marketing management experience, leading branding and marketing programs. Prior to coming onboard with Zilog, Steve held marketing management and technical engineering roles at Intel Corporations for over 14 years where he had several teams driving new technologies directly relating to the current products initiatives. His teams drove worldwide programs in evangelizing new technologies and accelerate adoption. Steve has a Marketing Degree from the University of Oklahoma. ■

EEWeb Electrical Engineering Community

Join Today www.eeweb.com/register

EEWeb | Electrical Engineering Community

Visit www.eeweb.com

Steve Grady VP Marketing

PERVASIVE POWER and Point of Load Delivery

Pervasive Power Overview This article introduces several new concepts for micropower electronic system design. These concepts are based on the fundamental power distribution and energy storage techniques deployed in advanced power grid architectures. With the introduction of small solid state energy storage ICs, new Pervasive Power solutions can now be created by placing micro energy storage devices directly at the point of load (POL) where the energy is used. Point of load architectures have been deployed in various power architectures down to the circuit board level. Recently introduced breakthroughs in solid state energy device technology enable circuit designers to place energy storage directly inside a chip for true point of load powering.

• Micropower and nanopower trends with enabling technologies • Solid state energy storage devices and Embedded Energy • New point of load and Pervasive Power applications

In order to provide a foundation for understanding Pervasive Power and the advantages of Point of Load energy delivery, the following areas are discussed:

What is Pervasive Power? Pervasive Power is a recently introduced power distribution architecture that utilizes energy storage devices at the actual point of energy usage (point of load). This is accomplished by placing micro energy storage devices inside a complex device requiring power. Examples include microcontrollers, real-time clocks, SRAM memory, sensors, and multi-chip modules. The introduction of new solid state energy storage devices utilizing a silicon substrate is the enabling technology for this “Power on Chip” configuration. Co-packaged modules using a solid state energy storage device with other ICs are now commercially available.

• Defining Pervasive Power • The interconnected power grid hierarchy • Power distribution and energy storage techniques in the various grids

A Pervasive Power architecture is created when Power on Chip enabled devices are placed together on an electronic assembly. This new distributed energy storage configuration provides many advantages.

EEWeb | Electrical Engineering Community

Visit www.eeweb.com

TECHNICAL ARTICLE

• With Power on Chip for all major chips, overall system power is reduced. • Power saving techniques for “trickle charging” circuit boards can be used. • Damaging in-rush currents are reduced with Power on Chip. • Minimizes induction in the Power train. • Reduces current draw variability. • Reduces heat. • Peak energy shaving and energy shifting techniques can be utilized. • Lower power budget is realized through efficiencies. • Fewer voltage conversion interfaces. • “Dirty” power into chips is corrected, with “pure” power delivered. • Reduces bypass capacitors surrounding ICs.

• Lowers I/O switching noise • Scalable—energy storage on the System increases with each added device Pervasive Power and Power Grid Techniques In order to have a better feel for the benefits of Pervasive Power, it is instructive to look at Power Grid techniques and how Power Grids are interconnected. It turns out that almost every technique for power distribution, voltage conversion, energy storage, noise reduction, and energy loss avoidance can be utilized in each grid topology. Power Grid Hierarchy and Power Distribution Techniques When people think of power grids, they often think of the main Backbone Grid providing power from a Power Utility to a business or home. Recently, there has also been a great deal of news coverage on the “Micro-Grid” in either green buildings or home-based solar energy deployments. But if you look at power distribution from the point of energy creation to the actual point of load, there are actually five interconnected grids as shown in the following diagram.

Backbone Grid

Transmission Distribution Generation

Subtransmission Customer 26kV and 69kV

Transmission Lines 765, 500, 345, 230, and 138kV Substation Step-Down Transformer Generating Station Generator Step Up Transformer

Facility Grid

Transmission Customers 138kV or 230kV

Equipment Grid

Solar Modules

Board Grid

(optional)

Mounting Tracks

Supplies Power to AC Loads

Charge Controller

Supplies Power to DC Loads DC Service Panel

Inverter & DC/AC Disconnects

Primary Customer 13kV and 4kV Secondary Customer 120V and 240V

Backup Generator

(optional)

Battery Bank

DC Load Controller (optional)

AC Service Panel

Chip Grid

Figure 1

EEWeb | Electrical Engineering Community

Visit www.eeweb.com

TECHNICAL ARTICLE

Advantages of Pervasive Power A Pervasive Power architecture is realized when most of the major functional semiconductor chips on a circuit board have Embedded Energy capabilities. There are many advantages that are realized with a Pervasive Power architecture:

TECHNICAL ARTICLE

2. Facility Grid: Once AC power enters a facility (typically a business or residence), the power is distributed within the facility. The facility itself may have the ability to provide standalone power in the event the Backbone Grid power feed fails using local generators. The Facility Grid might be equipped with local power generators that could supply power back into the Backbone Grid; recent examples are home solar cells that create more energy than is used in the facility. There may be local energy storage in the Facility Grid, and new energy storage technologies and devices are being introduced. 3. Equipment Grid: This grid is most often seen in commercial facilities for data centers or manufacturing machinery. In most cases, this demarcation occurs at the point where input power from the Facility Grid is converted from AC to DC to run electronics. Often times, the power distribution in the Equipment Grid is a DC voltage (380V, 48V, 12V). Using DC power also enables more cost effective energy storage solutions (e.g., Uninterruptible Power Supplies) to be used to back the Rack Grid in the event of Facility Grid power failure. 4. Board Grid: This is the power grid used on a circuit board or electronic assembly. The electronic assembly might be autonomous like a portable handheld device that was powered from an AC/DC converter on the Facility Grid. Board Grids almost always use DC power distribution (12V, 5V) and step down the DC voltage from the Board Grid input source to the electronic devices on the board. 5. Chip Grid: This is the power grid that is inside a semiconductor-based device. Initially, power distribution in semiconductor devices was a fairly simple architecture. But with the advent of highly integrated large-scale devices and multi-chip module implementations, the Chip Grid power architectures have become very complex. Many of the power distribution and energy storage techniques used in the other grids now are being used in the Chip Grid.

EEWeb | Electrical Engineering Community

1. A physical hand-off from a one entity to another (e.g., power company to a consumer or a factory power center to a process machine). 2. A standardized physical power connection interface (e.g., wall outlet to electrical device or an electronics rack backplane to a circuit card). 3. A current conversion from AC to DC. 4. A DC-DC voltage step down. Reducing Grid Energy Losses There is currently a great deal of focus on the “Smart Grid” where many new solutions are being brought to bear to improve the performance of the Backbone Grid. The following chart of power uses by commercial enterprises is sourced from the DOE’s Energy Information Administration. Note how important it is to lower the Electrical Losses (wasted energy) on the Backbone Grid. Of course, lowering the amount of Electricity used in commercial enterprises will lower the corresponding amount of energy. In order to reduce Backbone Grid losses, new techniques such as active status monitoring, energy storage, more efficient energy conversion electronics, dynamic demand algorithms, and point of load energy delivery are being deployed. Commercial - By Major Source Coal

Quadrillion BTU

1. Backbone Grid: Power is generated at a power plant and distributed to facilities (businesses or homes) using high voltage AC power transmission lines that are stepped down to AC Power delivery at the facility. The Backbone Grid utilizes AC current and typically does not have large power storage elements due to the losses associated with AC to DC to AC power conversions. However, many new techniques for Backbone Grid power storage are now being explored.

Demarcations between each power grid are defined by the following:

Petroleum

Natural Gas

Electricity

Electrical Losses

2.5 BTU is lost for every 1 BTU delivered

8 6 4 2 0

1950

1960

1970

1980

1990

2000

Figure 2

Cross-grid Similarities and Point of Load Power Management Just like the Backbone Grid, the need to effectively manage power and reduce energy losses are just as

Visit www.eeweb.com

TECHNICAL ARTICLE

The Five Interconnected Grids

TECHNICAL ARTICLE Board Grid Power

TECHNICAL ARTICLE

Rack Grid Power

Chip Grid Power

Intermediate Bus Architecture (IBA)

48Vdc Bus (Typ)

BACKPLANE

Bulk AC/DC Power Supply

IBC

12V Intermediate Bus (Typ)

NiPCL

Load 1

NiPCL

Load 2

NiPCL

Load 4

NiPCL

Load 5

NiPCL

Load 6

Figure 3

important in the other four interconnected grids. Many of the same techniques and technologies implemented in the Backbone grid apply even down to the Chip Grid. The following figure shows the relationship between the Equipment, Board, and Chip grids. System level designs need to view these grids holistically in order to manage the energy used by the system effectively. One of the key methods of effectively managing power is to use Point of Load technologies. Point of Load management involves the following: • Measuring the power being used at the actual point of use • Characterizing all the points in the power delivery chain • Actively managing the power to the point of use through a closed loop control system • Implementing dynamic power demand algorithms to optimize the efficiency of the power used • Providing energy storage at the point of load to enable optimal energy saving profiles independent of the power input Placing rechargeable energy storage at the point of load in the Chip Grid has many advantages: • Minimize I2R losses. Because devices with Power on Chip can be trickled charged, less power needs to be presented to the electrics from the power supply. • Power Sources can be isolated. With Power on Chip, devices can be “lifted” off the grid and run on the

EEWeb | Electrical Engineering Community

• • •

•

• • •

pure power in the rechargeable energy storage device. When isolation is no longer required, the device can be “placed” back on the grid. Reduced I/O switching noise when using on chip power. Power where ever it is needed. Power on Chip can be placed in any type of device. Power Bridging. In the event of power brownout/ blackout, the on chip energy storage takes over and powers the device. Power Boosting. There may be times when a device needs additional power and can draw upon the on chip energy storage versus placing an additional demand on the main power supply. Effective in ultra low power applications which are typically also miniaturized. Warm start energy can be used for devices in deep sleep or standby. Bridging “error” or early termination conditions. There may be times when a power interruption would create device operation errors or even device failure. Having Power on Chip provides power to complete operations in an orderly fashion.

Chip Grid Trends There are key trends that are occurring at the Chip Grid level: • Lower power devices using lower voltages • Denser Devices

Visit www.eeweb.com

TECHNICAL ARTICLE Multi-chip modules System on Chip Lab on Chip Advanced power management techniques Same digital power control techniques as Board Grids and Rack Grids

All of these trends are pointing toward integration and miniaturization. Many technologies have progressed down this curve, but batteries have not kept pace. So what are the implications to the Chip Grid? One key implication is that we need to integrate intelligent rechargeable energy storage into the Chip Grid. In order to achieve this requirement, a new product technology has been introduced: solid state rechargeable energy storage devices. Solid State Rechargeable Energy Storage Devices Cymbet has introduced a solid state rechargeable energy storage device based on a silicon substrate called the EnerChip™. The following photo diagram shows how the EnerChip is created on a silicon wafer. The EnerChips are diced and then can be used as bare die or packaged in standard semiconductor packages. Mounted on tape and reel, the EnerChips are placed on circuit boards using Surface Mount Technology and then can be reflow soldered to the board. The EnerChips are treated like the other IC packages on the final board. Using the EnerChip bare die has unique advantages for Point of Load energy storage from a packaging perspective as they are small and can be co-packaged in many ways with other ICs or micro devices. The photo below is a millimeter-sized Solar Energy Harvesting sensor. The solar cell sits on an ultra low power microcontroller that sits on a solid state rechargeable energy storage device (EnerChip CBC012). The devices are wire bonded to each other. Integrated Energy Storage for Point of Load Power Delivery With the introduction of solid state rechargeable energy storage devices, it is now possible to co-package energy storage directly with other Integrated circuits. Examples are shown in the following diagrams:

EEWeb | Electrical Engineering Community

Rechargeable Solid State Energy bare die Co-packaged side-by side with an IC:

Rechargeable Solid State Energy bare die Co-packaged in “wedding cake” die stack:

EnerChip Bare Die µController, Sensor, RTC

Rechargeable Solid State Energy bare die in System on Chip module:

PASSIVES

FLIPCHIP

An important attribute of Solid State Energy Storage built on silicon wafer is that they can be solder attached to the circuit board surface using a “flip chip” technique. The flip chip attach mechanism opens up many new miniature packaging options.

Solder Bump

Mold Compound

Mold Compound Die Lead

Solder Bump

Die

Lead

Sn/Pb or Sn

Conclusions This paper introduced several key concepts which are summarized as follows: 1. Pervasive Power is a new power distribution architecture that provides enhanced use of power at the point of load that increases overall system energy efficiency. 2. There are five levels of interconnected Grids

Visit www.eeweb.com

TECHNICAL ARTICLE

• • • • •

TECHNICAL ARTICLE

3. The interconnected Grids can utilize digital power control techniques to optimize the end to end use of power and improve energy efficiencies. 4. In order to enable the Chip Grid, a new energy storage device that can be integrated into the Chip must be used. Rechargeable solid state energy devices ideally meet this need. 5. These rechargeable energy storage devices can be co-packaged with other ICs in the Chip Grid to create a miniature highly integrated package. 6. Once boards are populated with Chips with on-chip energy storage, Pervasive Power architecture is created.

For additional information on Pervasive Power for Integrated Energy Storage for Point of Load Delivery, the http://www.cymbet.com website has application notes, datasheets, videos and contact forms. About the Author Steve Grady is responsible for all strategic messaging, product roadmap, CRM, e-initiatives, collateral and lead generation at Cymbet. He has more than two decades of domestic and international experience in marketing, sales, business development, product management, engineering, and general management in the networking, hardware, and software industries. Steve has been in both startup and large company environments with global scope. Prior to joining Cymbet, Steve held senior management and technical positions at ADC, Marconi, TimeSys, Reltec, and AT&T Bell Labs. He holds BSEE and MSEE degrees from the University of Illinois Champaign-Urbana. â&#x2013;

EEWeb Electrical Engineering Community

1.800.574.2791 advertising@eeweb.com

www.eeweb.com/advertising EEWeb | Electrical Engineering Community

Visit www.eeweb.com

TECHNICAL ARTICLE

terminating in the Chip Grid. Each grid type shares the same principles of power generation, power distribution, energy storage and energy management using dynamic demand algorithms.

Get the Datasheet and Order Samples http://www.intersil.com

Digital DC/DC PMBus 12A Module ZL9101M

Features

The ZL9101M is a 12A variable output step-down PMBus-compliant digital power supply. Included in the module is a high performance digital PWM controller, power MOSFETs, an inductor, and all the passive components required for a complete DC/DC power solution. The ZL9101M operates over a wide input voltage range and supports an output voltage range of 0.6V to 4V, which can be set by external resistors or via PMBus. This high efficiency power module is capable of delivering 12A. Only bulk input and output capacitors are needed to finish the design. The output voltage can be precisely regulated to as low as 0.6V with ±1% output voltage regulation over line, load, and temperature variations.

• Complete Digital Switch Mode Power Supply

The ZL9101M features internal compensation, internal soft-start, auto-recovery overcurrent protection, an enable option, and pre-biased output start-up capabilities.

• Server, Telecom, and Datacom

• Fast Transient Response • External Synchronization • Output Voltage Tracking • Current Sharing • Programmable Soft-start Delay and Ramp • Overcurrent/Undercurrent Protection Get the Datasheet and Order Samples http://www.intersil.com

Digital DC/DC PMBus 12A Module ZL9101M

Features

• Complete Digital Switch Mode Power Supply • Fast Transient Response

• External Synchronization • Output Voltage Tracking • Current Sharing

• Programmable Soft-start Delay and Ramp • Overcurrent/Undercurrent Protection • PMBus Compliant

Applications

The ZL9101M features internal compensation, internal soft-start, auto-recovery overcurrent protection, an enable option, and pre-biased output start-up capabilities.

• Server, Telecom, and Datacom

• Industrial and Medical Equipment • General Purpose Point of Load

The ZL9101M is packaged in a thermally enhanced, compact (15mmx15mm) and low profile (3.5mm) over-molded QFN package module suitable for automated assembly by standard surface mount equipment. The ZL9101M is Pb-free and RoHS compliant.

Related Literature

• See AN2033, “Zilker Labs PMBus Command Set - DDC Products”

• See AN2034, “Configuring Current Sharing on the ZL2004 and ZL2006”

V DRV

4.7µF 16V

V25

VDRV

ENABLE

ZL9101M

DDC

SW (EPAD)

SCL

FB+

VTRK

VSET

PGND (EPAD)

V OUT

3 x 47µF 16V

RTN

FB-

SGND

SDA SA

2 x 22µF 16V

VOUT (EPAD)

SYNC

DDC Bus

I2C/SMBus

V IN 4.5V TO 13.2V

VIN (EPAD)

Ext Sync

10µF 16V

VDD

4.7µF 16V

10µF 16V

4.5V TO 6.5V

POWER GOOD OUTPUT

Notes: 1. The I2C/SMBus requires pull-up resistors. Please refer to the I 2C/SMBus specifications for more details. 2. The DDC bus requires a pull-up resistor. The resistance will vary based on the capacitive loading of the bus (and on the number of devices connected). The 10k default value, assuming a maximum of 100pF per device, provides the necessary 1µs pull-up rise time. Please refer to the Digital-DC Bus section for more details. 3. Additional capacitance may be required to meet specific transient response targets 4. The VR, V25, VDRV, and VDD capacitors should be placed no further than 0.5 cm from the pin.

FIGURE 1. 12A APPLICATION CIRCUIT NOTE: Figure 1 represents a typical implementation of the ZL9101M. For PMBus operation, it is recommended to tie the enable pin (EN) to SGND.

April 8, 2011 FN7669.2

Intersil (and design) is a registered trademark of Intersil Americas Inc. Copyright Intersil Americas Inc. 2010, 2011 All Rights Reserved. All other trademarks mentioned are the property of their respective owners.

• PMBus Compliant

Applications • Industrial and Medical Equipment • General Purpose Point of Load

Related Literature • See AN2033, “Zilker Labs PMBus Command Set - DDC Products” • See AN2034, “Configuring Current Sharing on the ZL2004 and ZL2006”

V DRV

VDD

ENABLE

4.7µF 16V

V25

VDRV

POWER GOOD OUTPUT

4.7µF 16V

10µF 16V

4.5V TO 6.5V

I C/SMBus

ZL9101M

DDC

SW (EPAD)

SCL

V OUT

3 x 47µF 16V

RTN

FB-

PGND (EPAD) FB+

SGND

VTRK

SDA VSET

2 x 22µF 16V

VOUT (EPAD)

SYNC

DDC Bus

V IN 4.5V TO 13.2V

VIN (EPAD)

EN Ext Sync

10µF 16V

Notes: 1. The I2C/SMBus requires pull-up resistors. Please refer to the I2C/SMBus specifications for more details. 2. The DDC bus requires a pull-up resistor. The resistance will vary based on the capacitive loading of the bus (and on the number of devices connected). The 10k default value, assuming a maximum of 100pF per device, provides the necessary 1µs pull-up rise time. Please refer to the Digital-DC Bus section for more details. 3. Additional capacitance may be required to meet specific transient response targets 4. The VR, V25, VDRV, and VDD capacitors should be placed no further than 0.5 cm from the pin.

FIGURE 1. 12A APPLICATION CIRCUIT NOTE: Figure 1 represents a typical implementation of the ZL9101M. For PMBus operation, it is recommended to tie the enable pin (EN) to SGND.

April 8, 2011 FN7669.2

RETURN TO ZERO RETURN TO ZERO

EEWeb | Electrical Engineering Community

Visit www.eeweb.com

RETURN TO ZERO RETURN TO ZERO

EEWeb EEWeb Electrical Engineering Community

Electrical Engineering Community

Join Today

www.eeweb.com/register

1.800.574.2791

advertising@eeweb.com www.eeweb.com/advertising EEWeb | Electrical Engineering Community

Visit www.eeweb.com