World’s lowest power capacitive sensors with auto-calibration NXP is a leader in low power capacitance touch sensors, which work based on the fact that the human body can serve as one of the capacitive plates in parallel to the second plate, connected to the input of the NXP capacitive sensor device. Thanks to a patented auto-calibration technology, the capacitive sensors can detect changes in capacitance and continually adjust to the environment. Things such as dirt, humidity, freezing temperatures, or damage to the electrode do not affect the device function. The rise of touch sensors in modern electronics has become a worldwide phenomenon, and with NXP’s low power capacitive sensors it’s never been easier to create the future.
Learn more at: touch.interfacechips.com
e Needs of High Speed ComputingCONTENTS FEATURED ENGINEER
4 6 8 16
d for high definition data transfers with n in computing and eo and data transfer et this demand, interspeeds, deliver excelidge to legacy resses these challenges plexer switches includerbolt I/O technology, f I2C peripherals and TM
DANIELA HALL - ALGORITHMS GROUP LEADER, MOVEA
One is a three-to-one mux and the other is a two-to-one mux. The Thunderbolt MUX is a three-to-one switch that selects between Thunderbolt data MEETING NEEDS OF HIGH SPEED COMPUTING path andTHE DisplayPort v1.2 control signals — either DDC or AUX. The port is backwards compatible and the DisplayPort data can be sent out when DisplayPort is connected or Thunderbolt WHAT DOES THE ARM CORTEX-A9 REALLY PROVIDE? data sent out wh Thunderbolt signals or peripherals are connected.
NXP INTERFACE TECHNOLOGY FEATURED ARTICLE
LEAD INTERVIEW
CBTL05024 also delivers advanced characteristics that en hance signal integrity and power efficiency. It is powered ROSS by BANNATYNE OF MCU LINE a 3.3 V- GENERAL supplyMANAGER and available inATa NXP small 3 × 3 mm nderbolt is a hard HVQFN24 package with 0.4 mm pitch. e interface developed ntel. The Thunderbolt The Thunderbolt controller acts as a junction in the daisy rface consists of two of Thunderbolt products. Current Intel “Ivy Bridge” 3 Gbps full-duplex TECH ARTICLEdata chain FEATURED ARTICLE Ridge” controllers, ween PCs,RASPBERRY laptops or PI PRIMER platforms use “Cactus NXP’S NEWThunderbolt AXP PART 1 FAMILY platforms use “Redwood while newly launched LOGIC “Haswell” vices. Thunderbolt Ridge” Thunderbolt controllers. Both devices, CBTL0502 video (DisplayPort) HOTused VIDEOS and CBTL05024, can be on Cactus Ridge and RednpluggingEMBEDDED of cablesWORKBENCH is wood Ridge platforms. For higher integration and a more anes can be running HIGHLIGHT simplified solution, theMCU WARS is the better solution CBTL05024 data ratesPRODUCT in a ThunderWANDBOARD OVERVIEWas it features integrated32-BIT MCU pull-up/pull-down resistors and a heral devices via simple COMPARISON LSRX (control signal for Thunderbolt channel) buffer to hunderbolt peripherals optimize the BOM and improve signal integrity. Devices on this chain derbolt cable can proAt Computex 2013 in Taipei, Intel officially branded its next-generation Thunderbolt as Thunderbolt 2, which will run at 20 Gbps and support 4K video (Ultra HD). The esign that needs some 3 ini-
12
26
15 22
30 34
In the Spotlight:
Daniela Hall
Algorithms Group Leader, Movea
How did you get into engineering? I was always interested in technology, proposing solutions for things that don’t exist yet. As a child I wanted to be an inventor, and during high school, the easiest subject for me was mathematics. I continued with an MSc degree in computer science and specialized in image processing for my PhD. For a while I worked in public research. I never regretted my choice to move to a private company. I enjoy the dynamics of a startup company, stimulated by the short time to bring products to the market. Our efforts as engineers go very quickly to the clients, and the interactions between departments are reactive. All this is very rewarding to engineers.
Why did you choose to focus on data fusion and motion sensing? Everybody needs to interact daily with consumer electronics devices, whether it is a computer, a smartphone, a tablet or a smartwatch. Up to now, most user interfaces are designed to have good interaction from the computer/machine side, but are not especially user friendly. Using motion allows a computer to understand the user in a more natural, less intrusive way. The goal is that the computer delivers a service to the user and the user forgets about the computer. Motion sensing and data fusion enable this kind of magic.
What exactly do you do for Movea? I lead the algorithms team, which one of five teams that form Movea’s R&D. The team is composed of 10 engineers, several holding a
4
PhD. My task is to be the interface between the other departments such as sales and marketing. Together we define the requirement specifications for new functionalities or improvements of existing functionalities. Once we have defined the requirements, I define with the team the work to be done. This can be an algorithm optimization, a proof of concept study of a new functionality, incremental improvements of existing modules, industrialization of code. My role is also to establish links inside and outside the team in order to grow the team competency and establish links between projects. My team also generates most of Movea’s IP.
Can you tell us about your favorite project? One of the most useful applications I worked on is navigation in unknown areas. Unfortunately, using GPS on a mobile device drains the battery, putting people at risk to really get lost. In addition, we spend most of our time inside buildings with no GPS reception so indoor pedestrian navigation offers a multitude of location-based services, enhancing users’ experience on mobile. We are currently working on an alternative navigation method without GPS that allows accurate and robust pedestrian location and trajectory, while maintaining ultra low power consumption. It is very exciting to see how much progress we have made and how Movea is enabling the industry with this solution. The first prototypes we demonstrated few months ago required that the phone be pointed in the walking direction; today, the phone can be placed anywhere, whether in hand, in the shirt pocket, the backpack – as long as it is carried – and can still accurately track the user’s location.
What was the trickiest bug you ever fixed? At Movea I don’t fix the bugs myself but work with our whole team to find the solution together to solve a problem that looked impossible before. I’ll take the example of our demonstration for CES 2013 last January. At that time, we had to hold
ENGINEER SPOTLIGHT the phone in the direction we were going to but this was not viable for customer deployment so we had to find a way to determine the walking direction with unknown phone position or even a phone position that is changing. Putting experts from different fields in one room paved the way toward a solution. The team’s collective ideas were criticized, defended, modified and finally implemented, tested and patented. In the end, we’re proud to have demonstrated a pedestrian navigation that works for unknown phone position at the demonstration we held in Paris and Seoul in May 2013.
What are your favorite tools? As a computer scientist, I like to program in C++. We have developed a very powerful signal processing design tool called MoveaStudio that is similar to Simulink. This tool enables us, through a simple graphical user interface, to drag and drop technology bricks to quickly create robust and extendable algorithms that can easily be deployed into customers’ products. We also use it as an implementation validation and debugging tool, simplifying our customers’ verification process, enabling them to bring their products to market faster. I think this is a nice platform for playing and extending existing algorithms. For rapid prototyping and new algorithm conceptions, matlab is my favorite.
“I do believe we will move toward more merging between the human body and the virtual world, as we see the move toward wearable devices.” How is data fusion going to change how we interact with our electronic devices? Data fusion techniques, such as the ability to enable position (whether indoors or outdoors), activity and environment detection by the merging of data from multiple sensors, devices, and the cloud, will make the dream of intelligent mobile apps and services a reality. We are getting close to the ideal user interface where the user forgets that he/she is interacting with a device and interacts in the most natural way— with simple, natural motion.
What are you currently working on? Key projects include indoor pedestrian navigation, activity monitoring and sports performance tracking, robust and accurate inertial orientation.
Do you have any note-worthy engineering experiences?
What do you do when you’re not working on electronics?
In my previous company, the engineering team built an optical inspection system that detects 3micron dust particles on MEMS sensors. This machine coupled to a robot arm automatically rejects bad MEMS parts before sealing. This system increases the quality of the MEMS production. This system had challenges in optics, mechanics, algorithm and system engineering. The team was rather small (8 engineers), and members transverse in engineering specialties. Collaboration was excellent and reactive, since every team member was aware that he needed the competency of his other teammates in order to make the system work.
Grenoble is a great place for mountaineering and I play saxophone in a jazz band. As a German national, I also appreciate the French lifestyle and in particular, French cuisine.
What do you think the next thing will be in electronics? I do believe we will move toward more merging between the human body and the virtual world, as we see the move toward wearable devices such as activity monitoring devices like Fitbit and Google Glass enhancing the human body and mind with technology. Placing electronics in clothing that connect to your smartphone in order to make life easier and more fun sounds promising to me. ■
5
NXP Interface Technology: ing the Needs of High Speed Computing
NXP Interface Technology: Meeting the Needs of High Speed Computing
One is a three-to-one mux and the owing demand for high definition other is a two-to-one mux. The Thunmedia and fast data transfers with derbolt MUX is a three-to-one switch g the evolution in computing and that selects between Thunderbolt data ology . As video and data transfer path and DisplayPort v1.2 control sigadvance to meet this demand, internals — either DDC or AUX. The port support higher speeds, deliver excelonsumers’ growing demand for high definition is backwards compatible and the Dismultimedia and fast data transfers with d providevideo, a bridge to legacy global access is driving the evolution in computing onductors addresses these challenges playPort data can be sent out when and communications technology. As video and data is connected or Thunderbolt data issent out when transfer protocols continue to includadvance to DisplayPort meet this lexer/demultiplexer switches One is a three-to-one mux and the other a two-to-one demand, interface technology must support higher mux. The Thunderbolt MUX is a three-to-one switch that porting Thunderbolt I/O technology, Thunderbolt signals or peripherals are connected. speeds, deliver excellent signal integrity and provide selects between Thunderbolt data path and DisplayPort ad selection I2C peripherals andSemiconductors v1.2 control signals — either DDC or AUX. The port is a bridge tooflegacy standards. NXP addresses these challenges with the fastest multiplexer/ compatible and characteristics the DisplayPort data CBTL05024backwards also delivers advanced thatcan endemultiplexer switches including mux switches supporting be sent out when DisplayPort is connected or Thunderbolt hance signaldata integrity power efficiency. is powered ThunderboltTM I/O technology, level shifters and a broad sent outand when Thunderbolt signals orItperipherals are selectionThunderbolt of I2C peripherals and bus enablers. connected. by a 3.3 V supply and available in a small 3 × 3 mm is a hard HVQFN24 package 0.4 mmadvanced pitch. characteristics that ware developed Thunderbolt is ainterface hardware interface developed by Intel. CBTL05024with also delivers The Thunderbolt interface consists of two 10.3 Gbps fullenhance signal integrity and power efficiency. It is powered by Intel. The Thunderbolt duplex data paths that enables fast transfers between by a 3.3 V supply and available in a small 3 × 3 mm HVQFN24 The Thunderbolt controller acts as a junction in the daisy interface consists two PCs, laptops or tablets and of peripheral and display package with 0.4 mm pitch. devices.10.3 Thunderbolt multiplexes data Express®) of Thunderbolt products. Current Intel “Ivy Bridge” Gbps full-duplex data(PCI chain and video (DisplayPort) onto one cable. Hot-plugging The Thunderbolt controller acts as a junction in the daisy “Cactus Ridge” Thunderbolt controllers, transfers betweenofPCs, laptops or by theplatforms and unplugging cables is supported protocol. use chain of Thunderbolt products. Current Intel “Ivy Bridge” two lanes can beThunderbolt running completely independent while newlyplatforms launched platforms use “Redwood use“Haswell” “Cactus Ridge” Thunderbolt controllers, and The display devices. at different data rates in a Thunderbolt cable. Users can while newly launched “Haswell” platforms use “Redwood ® Ridge” Thunderbolt controllers. Both devices, CBTL05023 ) and videodevices (DisplayPort) Express also add peripheral via simple daisy chaining; up Ridge” Thunderbolt controllers. Both devices, CBTL05023 to six and different Thunderboltof peripherals can be connected and CBTL05024, can be used Redugging unplugging cables is and CBTL05024, can on be Cactus used onRidge Cactusand Ridge and on a daisy chain. Devices on this chain also can be busRedwood Ridge platforms. For higher integration and wood Ridge platforms. For higher integration and a morea col.powered. The twoThe lanes can becable running Thunderbolt can provide up to ten more simplified solution, the CBTL05024 is the better solution the integrated CBTL05024 is the betterresistors solution of power. nt atwatts different data rates in a Thunder- simplified solution, as it features pull-up/pull-down and a LSRX (control signal for Thunderbolt channel) buffer integrated pull-up/pull-down resistors and ato lso This addkind peripheral devices via simple as it featuresoptimize of architecture leads to a design that needs some the BOM and improve signal integrity. LSRX (control signal for Thunderbolt channel) buffer to ix different Thunderbolt sort of muxing or splitting toperipherals enable both DisplayPort and PCIe® information to go through the same connector. Computex 2013 in Taipei, Intel officially branded its optimize theAtBOM and improve signal integrity. daisy chain. Devices on this chain NXP offers two high-speed solutions that enable the next-generation Thunderbolt as Thunderbolt 2, which will ed. The Thunderbolt cable canand pronecessary switching: CBTL05023 CBTL05024. run at 20 Gbps and support 4K video (Ultra HD). The initial production expected before the end ofbranded 2013 withits ramp At Computex 2013 inisTaipei, Intel officially power. The CBTL05023 is a multiplexer/demultiplexer switch device up in 2014. As we’ve seen, NXP has a sharp focus on next-generation Thunderbolt Thunderbolt 2, that which will for DisplayPort v1.2 signals and the control signals of a this market, offering as high-speed switches meet all 10 Gbit/s channel. The 10 Gbit/s channel does not pass Thunderbolt requirements, so without jumping the gun run at 20 Gbps and support 4K video (Ultra HD). The inire leads to a design that needs some through this switch. This device also provides a BIASOUT on future announcements it is safe to say that designers tialresistors production is expected before the end of 2013 with tingoutput to enable both DisplayPort and control signal and DC-biasing pull-down would do well to look to NXP for their high-speed interface to facilitate thesame external 10 Gbit/s channel. computing needs both today andhas going NXP has As we’ve seen, NXP a forward. sharp focus on go through the connector. NXP ramp up in 2014. a sharp focus on this market offering high speed switch this market, offering high-speed switches that meet all solutions that the10Gb/s necessary Featuring an enable integrated Thunderbolt signal with solutions that meet all Thunderbolt system requirements no CBTL05024. external PIN diodes, NXP’s next generation switch Thunderbolt requirements, without jumping thecomputing gun on and will continue toso offer high-speed interface 3 and CBTL05024 integrates two different muxes inside. solutions going forward. ■ future announcements it is safe to say that designers would do well to look to NXP for their high-speed interface commultiplexer/demultiplexer switch deputing needs both today and going forward. NXP has a 1.2 signals and the control signals of
C
TM
6
Portability & Power All In One... TECH ARTICLES Debug digital designs on an iPad, iPhone, iPod. Can a logic analyzer be sexy? Watch the video and weigh in...
Logiscope transforms an iPhone, iPad or iPod into a 100MHz, 16 channel logic analyzer. Not only is it the most intuitive logic analyzer available, the triggering is so powerful you’ll be able to count the hair on your bug.
See why our innovation keeps getting recognized.
7
Seductive Parallel Worlds:
What advantages do modules for ARM Cortex-A9 processors really provide? By Wolfgang Heinz-Fischer, Head of Marketing and PR, TQ-Group
Standardized processor modules promise their users that they will be able to change over from one manufacturer to the next. But the devil is in the details--no module can supply signals to the outside world that are not processed by the CPU. As TQ points out, what is important here is to differentiate clearly between mere marketing and technical facts.
With the Cortex-A9Ž processor, ARM and its licensees initially have a CPU that visibly encroaches upon the area of application of an x86 processor. Almost all module suppliers have reacted to this and have included corresponding modules amongst the products offered by them – even those suppliers that have hitherto concentrated exclusively on x86 modules. As is usually the case in the x86 world, the first standards for ARM modules have already arisen.
However, the term standard here is excessive and awakens false expectations amongst users. Therefore, this article sets out to clarify what is actually meant by a standard and what benefits the user derives. This is because ARM module standards do not meet all expectations and are fraught with pitfalls. In any case, it is worth taking a detailed look at the systems offered in order to be prepared for unpleasant surprises. As so often occurs at the time of new product presentations, marketing slogans stand out – during which many statements tend to over-embellish the technical facts.
Standards and Their Limitations Standards arise wherever it seems useful to set down all specifications on paper, for one reason, to guarantee compatibility between different suppliers. Whether standard or proprietary: each has its own specific advantages and disadvantages. Therefore customers must closely examine which is best suited to their particular use.
8
In addition, embedded modules are also characterized by scalability, i.e. compatibility between different performance classes. Successful standards are usually found in the electronics industry wherever special applications are involved.
TECH ARTICLES Here, the requirements made for the specifications and functions need to be clearly delineated. CompactPCI, AdvancedTCA and MicroTCA for the telephony market, and PC/104 and COM Express for the industrial PC sector, are without a doubt a few of the established standards. Standards promise the user reliable access to the technology regardless of the success or failure of an individual supplier. If a particular processor technology is no longer available, scalability promises to ensure the continued supply with corresponding modules characterized by a superior technology.
Look Before You Leap … In order to ensure that compliant products really are interchangeable with one another, there are comprehensive compatibility tests for genuine standards, which are conducted by leading non-profit organizations such as the PICMG and the PC/104 consortium. However, this also means higher costs and thus higher prices for genuine standards due to the increased work involved.
Successful standards are usually found in the electronics industry wherever special applications are involved. All standards define the mechanical dimensions and the corresponding connector system. Genuine, successful standards are usually limited to a small number of interface specifications and are thus capable of assuring real compatibility. So, for example, only ISA, PCI and/or PCIe are defined as busses for the PC/104. In contrast, the COM Express alone has ten different configurations for interface connectors. Here, caution should be exercised when selecting the right version and examining the right configuration. For example, 24 express lanes have been specified for Version 2 of the COM Express spec. 2.0. However, the module manufactured by one supplier, fitted with an Intel® Atom N2600/ N2800/ D2550 and NM10 chipset, provides two PCIe x1 slots whereas another module by the same supplier, equipped with an embedded Intel® Core i7/i5/i3 and QM67 chipset, provides five PCIe x1 slots. In addition, the module is equipped with a second LVDV interface and two SATA III interfaces.
Sometimes the chip fails to recognize a signal; sometimes the signal is not envisaged under the standard; standards, in view of the large variety of ARM processors, have their limits.
This example shows that in the case of a genuine standard such as COM Express, 100% compatibility and thus guaranteed interchangeability can only be guaranteed to a certain extent. If the design has been based on and optimized for the low-end Atom processor this means that the main board will have to be modified and a new design and layout will be needed if higher performance is required and the intention is to use a more powerful module. Almost certainly, it will be possible to carry over some parts of the circuit during this step and a software adaptation will also be relatively easy.
Two standards for ARM: Q7 and SMARC There are two competing standards on the market in the world of ARM modules: the Q7 Group on the one hand and the ULP-COM Group on the other hand. Note: The working name of the ULP-COM standard was recently redefined and presented as SMARC (“Smart Mobility ARChitecture”) It is clear that there is no universal, generally valid solution in the ARM area otherwise there would not be two strong groups on the market, each with its own particular solution, and its own standard.
9
A Closer Look at Specs & Definitions Here, it is also worth taking a closer look from the aforementioned points of view. What does the specification actually define and which functions, if any, are supported by the fully-equipped processor? And which functions are provided by the processor that are not connected, (i.e. not available in the application)? It is also worth taking a look behind the scenes: one module supplied by one manufacturer and equipped with a Freescale i.MX6 processor offers two CAN interfaces and a PCIe interface (among other things), while another supposedly compatible module supplied by the same manufacturer, equipped with a Nvidia Tegra 3, does not offer any CAN interface, although it does have two PCIe interfaces. The standard with three defined PCIe interfaces does not fulfill either of the two processors. Only a detailed comparison will reveal where the differences lie and that interchangeability isn’t possible. Similar considerations and comparisons may be performed for the solution offered by the Q7 Group – all with the same result.
The conclusion is-interchangeability is only possible in the case of ARM modules if an extremely limited number of signals are used. Standards Limitations in the ARM Market The obvious conclusion is that a real standard with guaranteed compatibility and thus interchangeability in the ARM market is only possible to a certain extent. The extremely differing characteristics that continue to exist between the processors supplied by individual chip manufacturers mean that any standard, such as the existing standard in the x86 module family with the COM Express, is only applicable with severe limitations. One of the main arguments for standards-interchangeability--is only possible in the case of ARM modules if an extremely limited number of signals are used. A user wishing to
10
In the TQMa28, TQ relies on proprietary technology in order to ensure that all the features of the Freescale i.MX28 can be used on the smallest possible space.
make use of the full range of capabilities offered by the currently most powerful ARM processor on the module, the Cortex-A9, must fall back on the ‘special cut’ provided by the module supplier. All of this means that it is a proprietary system that, nevertheless, continues to be bound to the standard as regards board size, memory system, connector system and other aspects. Even more limitations become evident if various manufacturer standards are scrutinized.
You’re better off planning for a proprietary system right from the start. In contrast, the freedom and advantage offered by a proprietary system is that the supplier can tailor it to a particular processor or processor family. In so doing, the board size can usually be significantly smaller than standard boards. All or the overwhelming majority of interfaces can be made available in the connector system. From the point of view of the ability to use a processor platform for as long as possible, such a module offers the greatest possible freedom to the user. The TQ modules were and continue to be developed with this in mind – maximum performance with the highest possible degree of integration, availability of all signals, a robust and reliable connector system, long-term availability and all this in the smallest possible module size. There is no doubt that in the ARM module market, just as it occurs in the x86 market, different solutions will continue to exist side-byside and find their users. ■
TECH ARTICLES
Join Today
eeweb.com/register 11
The Raspberry Pi Primer Part 1: Introduction and Required Hardware by Kyle Olive, EEWeb Contributing Author
R
ecently, interest in hobby electronics has grown dramatically. With platforms like the Arduino gaining popularity and achieving wide success in various retail markets, it’s no wonder that these kinds of hobby electronics have become more prevalent. While the Arduino is designed around a microcontroller like the Atmel AVR and allows hobbyists to acquire and build various shields and add-ons for more functionality however, the Raspberry Pi (though a similar piece of hobbyist electronics) takes a different approach. Rather than a microcontroller board, the Raspberry Pi is a complete computer about the size of a playing card. At its core is an ARM11 microprocessor, and it contains HDMI and audio output, 8 GPIO Pins, 2 USB Ports, Ethernet, a dedicated GPU, and more. It’s basically a standalone platform that you can fit in your pocket. The introductory model costs only $25 (the higher end model with more memory runs for $35).
Front view of the Raspberry Pi board.
12
The Pi is capable of running various distributions of Linux, and using it is very straightforward and easy. In fact, the idea behind the Raspberry Pi was to provide a cheap computer platform for schoolchildren to use as an educational tool. It was to be a tool that students with little to no experience using computers could use to learn more about programming, computers, and really any other topic that lends itself well to electronic education. The platform quickly took off however with hobbyists, and it even began to be used in some commercial applications. The device’s low price point, small size, relative power, and ease of use makes it an interesting and flexible development platform. Because the Raspberry Pi runs Linux, developers who are used to working in a Linux environment can easily get up to speed on the Pi, and they quickly come up with ideas and applications for the device. Many of the popular software packages available for traditional Linux desktops are available on ARM versions of Linux meant for use on the Raspberry Pi, and with the incorporation of a dedicated GPU, multimedia applications that were often limited to desktop environments or specialized embedded platforms are completely viable, and cheap. Finally, the large network of developers provides a good support network for those who are stuck on a problem, or looking for advice — a benefit that some hobbyist platforms and specialized platforms lack. Even if you’re not a hobbyist, and looking to develop a new embedded solution for a particular problem, the Raspberry Pi can be
TECH ARTICLES a good way to prototype a solution without having to go through the entire design and development process. Rather than spending a large amount of money and time into developing a custom built functioning prototype, you can use a Raspberry Pi to greatly reduce initial prototyping costs and get a feel for a potential idea. If you’re going to do development with a Raspberry Pi, you’re going to need some hardware. Below is a discussion of various pieces of hardware that you’ll need to get started using a Raspberry Pi, and a brief note on its purpose. Most of the parts are easy to acquire at your local electronics store, but the Raspberry Pi vendors listed below will also often sell you this equipment with your Raspberry Pi.
Raspberry Pi’s low price point, small size, relative power, and ease of use makes it an interesting and flexible development platform.
adapter) so you can view the device video and audio output. You can get one at your local electronics store, or bundled with a Raspberry Pi. An SD Card Reader/Writer – If you’re working on a laptop, odds are you’ll already have an SD Card slot to work with. If you don’t, you’ll have to get your hands on an SD Card Read/Write device that you can plug into your computer. You’ll need it to flash the SD Card with whatever operating system you’re going to be using. If you don’t plan on ever needing to reflash your Raspberry Pi, you can buy a preformatted SD Card from one of the vendors listed above. Otherwise, you can usually pick up an SD Card Reader at your local electronics store. A Micro-usb Power Supply – The Raspberry Pi is powered by 5V Microusb, so you’ll need to get a power supply. The manual recommends against powering your Raspberry Pi directly from the USB port of your computer, but unless you’re working with a lot of high power peripherals or USB hubs you’ll probably be OK doing so. In either case, the safest bet is to buy a power supply either at one of the retailers above or get a micro-usb power supply (the one used for your cellphone or tablet might be able to power the Raspberry Pi). Ethernet Cable – You’ll want an Ethernet cable for your Pi so you can give it an Internet connection; you’ll at least need to set it up on a local network for remote access. The cables are pretty cheap, and you probably have an extra one lying around somewhere. If you don’t you can pick one up online with your Raspberry Pi or at a local retail store. USB Keyboard/Mouse – This is fairly straightforward. You’ll need a keyboard and mouse to develop on the Pi.
The Raspberry Pi – You can’t have a Raspberry Pi project without a Raspberry Pi. As discussed above, the Raspberry Pi is a playing-card sized computer that starts at $25. You can get these from various locations, but some vendors that are listed on the Raspberry Pi Foundation’s website (www.raspberrypi.org) are: •E lement14 •A llied Electronics An SD Card – You’ll need a regular old SD card to store the operating system and any other files you’ll want to use on your Raspberry Pi. 2GB is the minimum size of SD card you’ll need to run the default operating system, Rhaspbian, but if you are going to be using the Pi heavily, you’ll probably want a bigger SD card to suit your needs. In that case, I would probably recommend at least an 8GB SD card. You can get one at your local electronics store, or bundled with a Raspberry Pi at the above listed vendors. An HDMI Cable – If you plan on doing more than SSH’ing into your Raspberry Pi, you’ll want an HDMI cable (or an HDMI-to-something
The default desktop of the Raspbian Wheezy Operating System running on a Raspberry Pi The parts listed are what you’ll need to get started doing anything with your Raspberry Pi. Once you’ve gotten a hold of these items, you can start looking into developing the next cool Pi Project. Intrepid developers have created everything from robots to low-powered FM transmitters and webservers; they’ve even creating mini-supercomputers by parallelizing many Raspberry Pis. ■
13
INDUSTRIAL
AEROSPACE
SYSTEM ON A CHIP
MEDICAL
AVIATION
CONSUMER
THREADX: WHEN IT
REALLY COUNTS When Your Company’s Success, And Your Job, Are On The Line You Can Count On Express Logic’s ThreadX® RTOS
Express Logic has completed 14 years of successful business operation, T H R E and our flagship product, ThreadX, has been used in over 1 billion electronic devices and systems, ranging from printers to smartphones, from single-chip SoCs to multiprocessors. Time and time again, when leading manufacturers put their company on the line, when their engineering team chooses an RTOS for their next critical product, they choose ThreadX. Our ThreadX RTOS is rock-solid, thoroughly field-proven, and represents not only the safe choice, but the most cost-effective choice when your company’s product
simply must succeed. Its royalty-free licensing model helps keep your BOM low, A D and its proven dependability helps keep your support costs down as well. ThreadX repeatedly tops the time-to-market results reported by embedded developers like you. All the while, Express Logic is there to assist you with enhancements, training, and responsive telephone support. Join leading organizations like HP, Apple, Marvell, Philips, NASA, and many more who have chosen ThreadX for use in over 1 billion of their products – because their products are too important to rely on anything but the best. Rely on ThreadX, when it really counts!
Contact Express Logic to fi nd out more about our ThreadX RTOS, FileX® fi le system, NetX™ Dual IPv4/IPv6 TCP/IP stack, USBX™ USB Host/Device/OTG stack, and PegX™ graphics toolkit for embedded GUI development. Also ask about our TraceX® real-time event trace and analysis tool, and StackX™, our stack size analysis tool that makestool stack overflows a thing of the patent-pending stack size analysis that makes stack overfl ows a past. And if you’re developing safety-critical products for aviation, industrial or medical applications, ask about our new Certification Pack™ for ThreadX.
Newnes
n
Second Editio
E REAL-TIM ED EMBEDD ADING RE MULTITH
adX for ARM, Coldfire, With Thre ices with append ctures Now archite PowerPC MIPS and
For a free evaluation copy, visit www.rtos.com • 1-888-THREADX L. Lam Edward Copyright © 2012, Express Logic, Inc. ThreadX, StackX,and andCertification CertificationPack Packare aretrademarks trademarksofofExpress ExpressLogic, Logic,Inc. Inc. ThreadX,FileX, FileX,and andTraceX TraceXare areregistered registeredtrademarks, trademarks,and andNetX, NetX,USBX, USBX,PrismX, PegX, StackX, All other trademarks are the property of their respective owners.
M CD-RO INCLU DED
ie
EMBEDDED WORKBENCH BeagleBoard-xM BeagleBoard-xM delivers extra MIPS with 1-GHz ARM® Cortex™-A8 performance and extra memory with 512MB of lowpower DDR RAM, enabling hobbyists, innovators and engineers to go beyond their current imagination and be inspired by the BeagleBoard.org community. Designed with the community inputs in mind, this open hardware design improves upon the laptop-like performance and expandability, while keeping at hand-held power levels. Direct connectivity is supported by the on-board four-port hub with 10/100 Ethernet, while maintaining a tiny 3.25” × 3.25” footprint. As with previous BeagleBoard.org offerings, the BeagleBoard-xM is not intended to be a complete development environment, but rather a community-supported platform that can be used as the basis for building more complete development systems and as a target for community software baselines. For a complete development system, please consider the Sitara™ AM37x Evaluation Module from Texas Instruments.
Tiva C Series Evaluation Kit The Tiva C Series TM4C123G LaunchPad Evaluation Kit is a lowcost evaluation platform for ARM® Cortex™-M4F-based microcontrollers from Texas Instruments. The design of the TM4C123G LaunchPad highlights the TM4C123GH6PM microcontroller with a USB 2.0 device interface and hibernation module. The EK-TM4C123GXL also features programmable user buttons and an RGB LED for custom applications. The stackable headers of the Tiva C Series TM4C123G LaunchPad BoosterPack XL Interface make it easy and simple to expand the functionality of the TM4C123G LaunchPad when interfacing to other peripherals with Texas Instruments´ MCU BoosterPacks.
LPC812-LPCXpresso Board The LPC812-LPCXpresso board with NXP´s LPC812 Cortex-M0+ microcontroller is designed to make it as easy as possible to get started with your project. Combined with the the full-featured, easy-to-use Eclipse-based LPCXpresso IDE, the entire product design cycle for the LPC800 is supported. The LPCXpresso development platform is jointly developed by NXP, Code Red, and Embedded Artists. Features: • NXP´s LPC812 Cortex-M0+ microcontroller in TSSOP20 package • 16kB of Flash, 4kB of Data Memory • 12.000 MHz crystal for CPU
15
FULL
ECOS
A
n
f
a
b
r
T
w
M
f
o
16
COVER INTERVIEW
L MCU
SYSTEM
Ross Bannatyne
As one of the top 20 leading semiconductor companies in the world, NXP continues to in-
novate and differentiate to maintain this status. Their expansive product offerings range
from analog and power management to logic devices and MCUs. In 2011, NXP hit
a significant milestone in that it had shipped over one billion ARM processor-
based chips, making them the only company to provide a microcontroller
roadmap based on the 32-bit ARM processor.
To find out more about this portion of NXP’s business, we spoke
with Ross Bannatyne, General Manager of the Mass Market
MCU Products at NXP, about the LPC Cortex-based MCU
family and how these products provide an all-in-
one solution for their customers.
17
Have you always worked in the MCU business? I started in the MCU business in 1991. Since then, I’ve worked in a variety of roles in engineering, marketing, and management. The thing I have always found very appealing about the MCU business is the diversity of the applications— it’s always great to see the many interesting applications that use MCUs, from anti-lock brakes to zoom lens controllers. An interesting thing about MCUs is that they start off as ‘general purpose’ devices and are customized to a unique application through the addition of software, so the challenge for our customers is writing the software to customize their own particular system. MCUs are at the heart of the system, so it’s always the first and most important design-in decision that engineers make, so it’s a special place to be on the bill of materials.
How long have you been at NXP? Although I have been at NXP for a relatively short time, I have been familiar with the MCU product lines for many years and have watched it grow (and competed with it) from the 8051 based products through the new Cortex-M based products. I have seen NXP products compete strongly in every region of the world. I was based in Europe until 1995 and have been based in the USA since then. NXP MCUs are ubiquitous and I look forward to contributing to that growth.
What is your role at NXP? I’m responsible for our mass market MCU business. That involves product management of a portfolio of MCUs, definition of new products and solutions, managing the introduction and their lifecycle, and working with customers to
18
get them into production quickly and smoothly. Ultimately it is about growing the revenue of the product line profitably – there are a lot of aspects to that job. It’s always a challenge for a customer to learn a new MCU architecture, so a big part of what we do is to provide the tools to make it as easy as possible. One of the advantages we have at NXP, because we were the first MCU adopters of the ARM platform, is that we’ve got a fantastic set of tools, reference designs, solutions, software and infrastructure designers need to make an easy transition. The ecosystem around the MCU has become a huge part of the total product. This easy transition really relates to the hardware and the software tools because if you spend a long time writing code, and you know it works and it’s fully tested, you want to reuse it as often as possible. The software development cycle is longer than that of the hardware. If software can be reused easily, then it helps reduce time to market, risk and development costs. We also extend our “ease of use” to the hardware as well because we know that inevitably, a lot of our customers run out of memory space and will want to add some new software features. Ideally, they’d like to drop another chip in to use all of their existing software without having to re-layout the board and make other system changes to accommodate a different chip.. In many product ranges, we provide a whole family of chips that are pin-to-pin compatible and have the registers in the same locations, so that it is possible to re-use software across many different microcontroller products. Many of our customers want to remain in production for a long time—often over ten years. This is because they spend a lot of time creating and testing the software. For this reason, we support a ‘longevity’ program that guarantees the supply of our products for many years.
COVER INTERVIEW Could you tell me about the products you offer? I mentioned that we were the original adopters of the ARM architecture, so over the last 10 years we’ve developed product families for ARM7, ARM9, ARM Cortex-M0, M0+, M3, and M4, resulting in over 3different ARM products covering the entire ARM microcontroller spectrum. We are very strong in applications that require driving displays as there are highly integrated peripherals for controlling displays on the MCUs. We take a systems approach to solving such problems with a set of integrated peripherals that work together to optimize the solution.USB connectivity is another area where we are particularly strong and can provide differentiation from others in the industry with a range of solutions – including software – that support high speed, full speed, host, device, on-the-go, dual USB controllers and so on. NXP is also particularly strong in industrial connectivity solutions like Ethernet and CAN (controller area network). We have the fastest Cortex-M4-based MCU available on the market, which runs at over 200 MHz. These high processing speeds are useful because they can simplify the problem where you’ve got to respond to lots of real-time interrupts or transfer lots of data in or out of an MCU at a high speed, or you’ve got significant real-time algorithm control like digital filtering. There are a lot of peripherals on these chips that are autonomous so that they can operate independently of the CPU. That gives us additional performance because it means your CPU can be dealing with one thing while your peripherals are dealing with data that’s coming in from the serial ports.
“It’s always a challenge for a customer to learn a new MCU architecture, so a big part of what we do is to provide the tools to make it as easy as possible.”
but have an interest in migrating into the ARM ecosystem - because the ARM ecosystem has a lot of obvious benefits in terms of the support and range of tools. We packed a lot of innovative features in the product that really makes it easy to use and versatile for those types of customers. This product family even includes an 8-pin DIP package to make it simple and friendly to traditional 8-bit MCU users. It is a simple migration because a lot of customers who haven’t used a 32-bit product before see as the transition from 8-bit as quite a daunting step. We’ve added things like the switch matrix peripheral that allows any peripheral function on the chip to be routed to any pin. It is a powerful feature and we’ve created a tool that basically generates the code automatically to do that, which makes it easy to deploy.
Bannatyne and team
Have there been any new products from NXP over the last year? Recently, one of the most interesting products that has been getting a lot of customer interest is our LPC800 family, which is our entry-level family, based on the Cortex-M0+. This is aimed at customers who have been using 8- and 16-bit products,
19
“ At the end of the day, our objective is to allow customers to solve problems as painlessly as possible. We’ve really got a big focus on creating full solutions for our customer rather than just offering chips.” These chips also have ROM on board, so that we can develop and test drivers, burn them into ROM and give customers the ability to use NXP developed software drivers. This simplifies code development for the customer, but it also means they need less code to implement their overall function, so they can keep their development time down.
Have you had any success in getting those legacy engineers and bring them up into your new chip families? Absolutely. We still have a lot of 8051customers, so when we were developing the new LPC800 family, we spent a lot of time talking to those customers and asking them how they see their own products migrating and how to make it easy for them to migrate to a 32-bit architecture. We got a great many ideas for the LPC800 by speaking to those existing 8-bit customers.
What resources are available for your customers? At the end of the day, our objective is to allow customers to solve problems as painlessly as possible. We’ve really got a big focus on creating full solutions for our customer rather than just offering chips. The objective of our mass-market business is to make our microcontroller solutions scalable – to make it as simple as possible for thousands of customers to use NXP microcontroller-based solutions. We’ve created a growing number of ‘solutions-in-a-box’ for many different applications including communications systems, lighting interfaces, digital audio streaming, and near-field communications, to name a few. There are many software and tools solutions available on our website as well as applications notes, forums and online support. This ease-of-use and support infrastructure enables the scalability and allows us to address thousands of customers.
Are these kits available on NXP’s website? Yes, they’re available on our site. In addition, there are lots of resources to solve problems available on the NXP website including tools ecosystems, design resources and training.
20
Let me give you an example of how customers need solutions rather than just chips and datasheets. Recently, I worked with a customer who develops weigh scales. The customer wanted to add a USB port to their weigh scale so that it could record data and send it to a PC to allow the user to keep track of their weight. The weigh scale could then become a piece of personal equipment for health and fitness. In that particular case, their core competence is how to measure analog data from load cells with a wide dynamic range. In this case, there is no need to reinvent the wheel by developing detailed USB knowledge in-house (writing USB driver code was not and would never be their focus), they simply want to add a USB port as painlessly as possible. By offering tools, driver code, and application notes, we let them focus on their core competency and we take care of the USB solution for them. Another example is some work that we have done with a metering customer who was developing a metering communications hub. They were taking on a big project and their biggest issue was that they didn’t know how much memory they were going to need. This was because by the time this project reaches fruition, the spec was probably going to change a few times and more software would be required. Their key design objective was the flexibility to be able to add more memory at a later date. On a number of our LPC microcontroller products, there’s a quad-spi interface that can interface to large external memories very efficiently. That means with a small microcontroller and a 4-pin interface, we can access many megabytes of memory. This allows the customer to come up with a design that is based on our processor that they knew would have enough horsepower, but it would allow them to simply drop in different sizes of memory chips without having to change anything on the board when more software is required. I think that every customer that we deal with has a particular set of priorities for the problems that they see as being the biggest ones to solve—whether it’s designing for flexibility or focusing on their core competency. Our objective is to make it as easy as possible to solve our customer’s problems. ■
COVER INTERVIEW
At Embedded Developer you can compare and . You can also compare ARM
速
s and development tools. Then you can buy them. www.embeddeddeveloper.com Everything you're looking for in one place.
21
Overview of the
Wandboard featuring the Freescale i-MX6 processor
The Wandboard is a non-for-profit modular development board that is created by Wandboard.org, whose goal was to create a low-cost, flexible, and accessible development board. The development board features the Freescale i.MX6 family of processors, and it comes in solo, dual, and quad core versions. As a modular board it includes both a processor board and an interface board, and comes with a custom enclosure for protecting the unit and for taking it outside the lab.
22
PRODUCT HIGHLIGHT
23
Included Hardware The Freescale i.MX6 series processors are pin-for-pin and software compatible, which means that if you need to go from the solo to the dual processor (or the dual to the quad processor) you won’t have to recode everything, or re-layout your board. You will probably have to do some coding to take advantage of the additional costs, but other than that, you’ll be ready to go. Besides the Freescale i.MX6 series processors, the Wandboard also comes with 4 DDR3DRM ICs, a Micro SD Connector, WiFi and Bluetooth integrated modules, and a 315 Pin Edge Connector (in compliance with the EDM standard). It also includes a reset button and expansion headers, which have your GPIOs, LBDS signals, I2C Busses, SPI Busses, and a connector for your stereoscopic camera interface. On the front side of the board you can find your 5V in, RJ45, HDMI, Digital Audio, Microphone In, Line In, Line Out, Serial, SATA, High Speed USB 2.0, a MicroSD slot for storage and applications, and USB On-the-Go connector ICs. Included in the development board you will also find your ADM 3202 2-Channel, low power high speed RS232 interface, the Freescale SGTL 5000 Ultra Low Audio CODEC, the AME 8816 LDO (which converts your 5V In to the 3.3V supply you need for the board) and the TPS 2061, a power switch for the On-the-Go USB. the AC+ and ACtest points, depending on what type of voltage you have.
24
PRODUCT HIGHLIGHT
Setup To set up your board, begin by downloading the Ubuntu 11.10 Version from the website and burn it onto a Micro SD card on the back. You should then be able to see Ubuntu Version 11.10 running on the Wandboard. At that point, you will want to make sure the HDMI output is plugged into the HDMI input on the laptop. Note that besides the Ubuntu Version 11.10, you can also download Android Jelly Bean. Make sure your WiFi antenna is hooked up. On mouser.com you can get the WiFi antenna kit that includes the appropriate cable (as well as a custom enclosure) if you do a keyword search for “Wandboard” or pull up the Wandboard product page, which should be listed in accessories.
Typical Uses and Setup A typical use for the Wandboard is something like a streaming music or media player. For these, or a similar uses, you’d want to put the board in the custom enclosure, hook it up to your TV via HDMI, and start streaming media from your USB drive or across your network. The advantage of this is, wherever you store your media, you should be able to access it.
Watch Video
Conclusion The Wandboard is a new take on development kits. By making the processor board and the essential functions of the unit more modular and the interface board a second board, the Wandboard allows you to customize your interface to your needs. It has basically everything you would need to design a system integrated into a single chip.
25
Introduces AXP Logic Family to Address the Need for Lower Power Consumption 26
PRODUCT HIGHLIGHT
We talked with NXP’s International Product Manager, Cliff Lloyd, about the new AXP Logic family released in July, and how it takes on the power-consumption problem.
27
[
]
“ We felt we had to address the need for faster performance... and also maintain or improve power savings performance.” - Cliff Lloyd, NXP
“The mobility markets have been exploding in recent years,” NXP’s International Product Manager, Cliff Lloyd told us. “That market is driven by two primary factors: one is the size of a device, and the other is battery life.” In the past, he explained, NXP focused on making smaller and smaller packages. NXP made a breakthrough last year in small package size when it released its Diamond package -- 0.8 by 0.8 in size with a 0.5mm pad pitch between pads -- which made it easier to mount in PCBs. After the Diamond package was released and package size addressed, NXP decided to focus on addressing low power consumption in the mobility market. “We felt we had to address the need for faster performance... and also maintain or improve power savings performance,” said Lloyd.
One of the advantages of the AXP family of devices is its configurable logic, which gives customers a lot of flexibility when selecting parts.
28
There are two very important parts to power savings performance, explained Lloyd. The first is static power consumption, which is the amount of power a device uses when it is simply connected to a battery. The other important part of power consumption is the power that’s consumed when the device is actually being used, which is called the Dynamic Power Consumption. Keeping these two parts of power consumption low was what NXP has aimed to address with the AXP family. What they have achieved with the AXP family is 15% lower power consumption for unique functions, as well as a top delay of 4 nanoseconds at 1.2V, which is about twice as fast as its previous AUP family at the same voltage node. One of the advantages of the AXP family of devices is its configurable logic, which gives customers a lot of flexibility when selecting parts. For instance, when a customer buys a device -- depending on how he hardwires it on the PCB -- the device could operate as a NAND Gate, an OR Gate, a NOR Gate, a buffer, or an inverter, and the customer doesn’t have to buy a different device to program these individual functions. The customer can get different functions from one device, and use a single qualification to cover multiple functions, which offers tremendous flexibility.
PRODUCT HIGHLIGHT
Although many people believe that the devices are programmable, Lloyd explained that that is a misconception. The devices are strictly hardwire configurable on the PCB. For example, if you hooked up pin 5 and pin 3, as in the figure below, you would get a twoinput AND gate. If you connected pin 2 and pin 1 to ground, you would get a two-input NOR gate. It all depends on how the board is laid out -- an inverter, or an exclusive NOR (XNOR) can also be achieved. Because of this flexibility, NXP decided to first release the configurable devices.
Features of AXP Devices • Very low dynamic power dissipation (CPD) • tpd of 2.9 ns at Vcc of 1.8 V • Wide supply voltage range (0.7 V to 2.75 V) • Fully specified at 0.8 V • Schmitt-trigger action on all inputs • 4.5 mA balanced output drive • Over-voltage tolerant I/Os • Fully specified (-40 to +85 °C) • Pb-free, RoHS compliant and Dark Green For more information, visit the NXP Tech Community.
With low static and dynamic power dissipation, a wide voltage range, true Schmitt-Trigger inputs, and configurable logic, NXP has aimed to address the need for low power consumption and reliable logic level switching with the AXP Logic family, and position itself for a mobile market where battery critical applications and battery life conservation is key. ■
29
USB - WHAT WOULD THE GURU DO?
LPC800: PACKING A 32-BIT PUNCH
SOLVING EMBEDDED GRAPHICS CONUNDRUMS
LPC: RAPID PROTOTYPING WITH MBED
KICKSTARTING NXP EMBEDDED DEVELOPMENT
NEAR FIELD COMMUNICATION MADE EASY
30
HOT VIDEOS
GET TO KNOW FREERTOS FROM THE CREATOR
LPC-LINK 2 DEBUG ADAPTER
LPC800 CHALLENGE - GETTING STARTED
CAPACITIVE TOUCH & PROXIMITY SENSORS
WHAT IS THE I2C BUS?
NXP USB 3.0 SIGNAL INTEGRITY SOLUTION
31
RX PERIPHERAL OVERVIEW
RX62T: IDEAL SOLUTION FOR MOTOR CONTROL
RENESAS CAPACITIVE TOUCH-KEY SOLUTIONS
RL78 SENSOR DEMO
RX SYSTEM INTERFACES
RX RDK NEW DEMOS
32
HOT VIDEOS
RX210 PROMOTIONAL BOARD
RENESAS’ SMART SOCIETY
ALI SEBT - DEVCON 2012 KEYNOTE
7C031: POWER FACTOR CORRECTION
RL78 FAMILY INTRODUCTION
RX200 FAMILY INTRODUCTION
33
Episode 3:
32-Bit Microcontr Comparison 34
roller
TECH ARTICLES
In this episode of MCU Wars, Ritesh Tyagi and Chris Anderson compare two 32-bit microcontrollers—the RX100 from Renesas, and the STM32 from STMicroelectronics. We chose these two devices because of the increasing demand for 32-bit microcontrollers in a variety of industries. Which one reigns supreme?
35
VS. RX100 by Renesas The RX100 series is part of Renesas’ RX family that was launched in 2007. In the same family, there is the RX600, which is a high-performance series, and the RX200, which is the low-cost version of the RX600. The recently launched RX100 has the same CPU core as the 200 and 600, but was developed with low-power battery operated applications in mind. The RX100 has 1.56 megahertz of CPU performance and runs at very low power, both in active mode as well as in standby mode. The device comes in a variety of packages—from 40-pin to 64-pin—and all the peripherals are included within. In terms of memory, the RX100 ranges from as low as 8 kilobytes to 128 kilobytes in on-board flash, with additional data flash available.
36
STM32 by STMicroelectronics The STM32 L1 series from STMicroelectronics is based on a 32-bit ARM Cortex-M3. The device has multiple low-power modes and is available in 4 different lines: the STM32L100 Value line, STM32L151, STM32L152 (LCD), and the STM32L162 (LCD and AES-128). The device can get up to 384K of flash and up to 48K of SRAM as well as up to 12K of EEPROM. Like the RX100, the STM32 comes in multiple different packages as small as a wafer-level chip scale package. It can get up to 114 I/Os with all of the standard communication interfaces, up to 40 12-bit ADCs, and up to two 12-bit DACs. There is also an integrated DMA controller as well as an integrated LCD driver.
TECH ARTICLES
Featured in this episode:
Ritesh Tyagi
Chris Anderson
Sr. Director of Product Marketing at Renesas Electronics America
Electrical Engineering Consultant and Developer
Optimized for Low-Power
The STM32 L1 Toolchain
According to Tyagi, Renesas has been “using a lot of proprietary technology to optimize the RX100 power consumption.” In order to do this, Renesas implemented a 130mm, low-leakage, low-power transistor flash process. “Compared to the RX600 or RX200 where we used their proprietary MONOS technology,” Tyagi explained, “we decided to use a completely different technology for the RX100.”
When asked about the STM32 toolchain, Chris Anderson offered his preferences for ST products. Anderson explained, “ST offered a number of choices with quickstart guides for Embedded Workbench, but I settled on the Kiel Microvision 4.” Anderson stated that the Microvision had a number of example files that had all the source code needed to start tweaking in order to find something close to what he was trying to do. It also has a project manager compiler, an integrated debugger, and an abundance of example projects for dev kits of all the other major ARM vendors.
Another differentiating factor is the redesign of the clock circuitry to offer a much faster wake-up. This allows the device to remain low-power in active and standby modes. The voltage regulator was built specifically for this device, offering very low power consumption. “Overall, there are a lot of different techniques and tricks that we employed to achieve a very low power standby current and active mode current,” Tyagi stated.
“Overall, the RX100 comes with a pretty packed kit with an entire toolchain inside.” – Ritesh Tyagi
37
Renesas’ RX RDK Development Kit Analog Devices Temperature Sensors and Accelerometer
Segger J-Link OKAYA LCD
Micron PCM Memory
38
Total Phase Debug Connector
Analog Devices Audio Amp
NDK Crystals
STAR Speaker
National Semiconductor Ethernet PHY
TECH ARTICLES
ST’s Discovery Development Kit ST-Link/V2 Debugger and Programmer
LED Display
Based on the STM32L152 (ARM Cortex-M3)
User Push Button
Reset Push Button
Capacitive Sensor
39
RX100
STM32
Renesas' Eclipse-based IDE, e2studio
• CPU: 32-bit (RX)
• CPU: 32-bit (Cortex-M3)
• Max. Frequency: 32MHz
• Max. Frequency: 32MHz
• Performance - DMIPS/MHz: 1.56 - MAC/DIV: Yes - DSP Instructions: Yes - Coremarks/MHz: 3.08 - Max. DMIPS: 50 - Interrupt Latency: 5 cycles - Run μA/MHz: 100 - Lowest Power Mode with RTC on (μA): 0.6 - Wake-up Time from Previous Mode: 5μs
• Performance - DMIPS/MHz: 1.03 - MAC/DIV: Yes - DSP Instructions: No - Coremarks/MHz: 2.17 - Max. DMIPS: 33 - Interrupt Latency: 12 cycles - Run μA/MHz: 290 - Lowest Power Mode with RTC on (μA): 0.9 - Wake-up Time from Previous Mode: 2.6ms
• Voltage Range: 1.8-3.6
• Voltage Range: 1.65-3.6
• Lineup - Pin: 36-64 - Flash Size: 16KB-128KB
• Lineup - Pin: 48-100 - Flash Size: 32KB-384KB
• Peripherals - USB Device: Yes - USB Host: Yes - UART/USART: 3 - I2C: 4 (1+3UART) - SPI: 4 (1+3UART) - ADC: 8 x 12-bit - DAC: 2 x 8-bit - RTC: Yes
• Peripherals - USB Device: Yes - USB Host: No - UART/USART: 3 - I2C: 2 - SPI: 3 - ADC: 24 x 12-bit - DAC: 2 x 12-bit - RTC: Yes
40
Getting Started the RX100 For the RX family, Renesas made a big company-wide push to launch the Eclipse-based IDE, called the e2studio. The RX100 will have the same toolchain based on the e2studio. Underneath that, Renesas offers a compiler as well as a compiler from IAR and other third parties. “In fact,” Tyagi explained, “we are doing a pretty big joint promotion with IAR where they start offering a 64K version of an IAR workbench absolutely free.” The kit also comes with multiple sample projects and application examples. The board comes with a CD with a compiler and a debugger all within the kit. “I can guarantee,” Tyagi said, “that with the RX kit, you can be up and running within an hour.” Editor’s Pick Overall, the RX100 by Renesas offered significant low-power advantages over the STM32. The RX RDK Development Kit also offered significant features over the ST Discovery Kit, made possible by Renesas’ many partnerships in on-board components. ■
TECH ARTICLES
Click below to watch Episode 3:
Click below to watch Episode 2:
Click below to watch Episode 1:
41