Nexperia Takes the
POLE POSITION in Automotive Logic Applications Exclusive interview with Michael Lyons, Nexperia’s Technical Marketing Manager for BL Logic
March 2017
Automated Petrol Stations TQ COM Express Modules
Smart & efficient
Logic for automotive applications
Nexperia brings over 50 years of proven quality, commitment, and efficiency to automotive Logic applications. Our product portfolio meets and even exceeds the AEC-Q100 standard. Nexperia -Q100 Logic is guaranteed for automotive applications by meeting OEM zero defect requirements, supporting higher visibility and traceability in automotive production. With over 1,000 automotive certified -Q100 types, Nexperia delivers the largest product portfolio in the industry. We offer a wide range of functional categories (gates, switches, shift registers, etc.), traditional HC(T), AHC(T), full feature LVC, AUP logic families, and PnT translating gates, in either Standard Logic (> 10 pin) or Mini Logic (≤ 10 pin) types. With globally located sales and support teams, you’re assured of priority technical support, priority design-in assistance, PPAP Q100 qualification data, and fast turn-around on support issues. To learn more, please visit www.nexperia.com/logic
nexperia.com
March 2017
Embedded Developer
EDITORIAL STAFF Content Editor Kathleen West kathleen@convergencepromotions.com Digital Content Manager Heather Hamilton hhamilton@aspencore.com Director, Creative Development Jeff Chavez jchavez@aspencore.com Graphic Designer Carol Smiley csmiley@aspencore.com Audience Development Claire Hellar chellar@aspencore.com Register at EEWeb http://www.eeweb.com/register/
Published by AspenCore 950 West Bannock Suite 450 Boise, Idaho 83702 Tel | 208-639-6464
CONTENTS
4 8 12 22 28 36
Open Source Reaches Processing Core Automated Petrol Stations Deep Thoughts Regarding the Bodacious Brain Speed Optimizations in 8-bit MCUs TQ COM Express Modules Optimize Industrial Automation, IoT and Defense Applications Nexperia Takes the Pole Position in Automotive Logic Applications
4
8
Victor Alejandro Gao General Manager Executive Publisher Cody Miller Global Media Director Group Publisher Glenn ImObersteg Publisher Contributing Editor Embedded Developer
EEWeb
28
36
Embedded Developer
OPEN SOURCE
Reaches Processor Core By Richard Quinnell, EDN
4
March 2017
Whether for budgetary, philosophical, or other reasons, an increasing number of embedded systems are being designed using open source elements. For the most part, these elements are software based, although there are some open source board designs in use as well. Now, the microcontroller that empowers a PCB design is available as an open source design.
www.EmbeddedDeveloper.com
5
Embedded Developer
HiFive1 Development Board
(Source SiFive)
The board is open source, has an associated software development kit, and is Arduino compatible.
6
A little over a month ago, startup SiFive announced a milestone product in the development of the RISC-V (pronounced risk-five) open source microprocessor instruction set architecture (ISA). Originally developed for research and education, the architecture began moving toward industry implementation with the creation of the RISC-V Foundation in 2015. SiFive advanced that movement by developing a microcontroller design implementing the RISC-V ISA. The company has now proven that design in silicon and donated the RTL code for the design to the open source community. SiFive’s Freedom E310 MCU is implemented in the TSMC 180G process, runs at better than 300 MHz, and includes 16kB each of L1 cache and of data scratchpad memory. It also has 8kB of mask ROM and 8kB of in-circuit one-time programmable memory as well as a quadSPI Flash controller for off-chip, executein-place code. The E310’s IO offerings include a GPIO complex, watchdog, and a JTAG-connected debug module. Developers interested in exploring the RISC-V architecture, and the Freedom E310 in particular, can use the company’s $59 HiFive1 development board. The board is also open source, has an associated software development kit, and is Arduino compatible. Developers can thus use the Arduino IDE to create code for the E310.
March 2017
Traditional processor designs are either proprietary, requiring developers to select among and purchase stock chips, or are based on licensed IP, as with the ARM architecture. The RISC-V, being open source, does not readily support either of those traditional business models. The whole idea of making the microcontroller open source, after all, is to allow developers to customize the design for their specific needs rather than compromising on a bestmatch stock model. And there is no cost to license the base design. The business model that SiFive chose in order to turn a profit from their opensourced design is based on the idea of building customized systems on chip (SoCs) for customers. Its goal is to establish a “chip design factory” that can handle 1000 new chip designs a year. For under $100k, the company claims, it can in three to six months create and start manufacturing a custom MCU for a customer while ensuring the SoC remains compatible with the Freedom E310 core. This compatibility means that a team can develop software on the HiFive1 board and then execute it on their custom MCU with little to no porting effort. It also means that any other developments that the open source community creates for the E310 will work with the custom MCU as well.
Behind this whole RISC-V movement is the belief that Moore’s Law has reached its economic limits. No longer is the cost per transistor for an MCU design going down, the thinking goes, and the development cost for a new chip has exploded. Processor vendors are thus concentrating on winning a few big customers so that they can have production volumes sufficient to amortize the high design cost while minimizing the cost of providing development support to their customer base. The small, entrepreneurial company ends up being hostage to this trend, having to deal with “black box” hardware and software that it cannot optimize for its specific needs. It has to settle for whatever it can get. By moving processor design to open source, supporters believe, the SME (small to medium enterprise) design community will get the ability to compete more directly with the big guys. Development teams no longer have to start out as major customers to gain the benefits of customization. Chips optimized for the application become affordable and will enjoy widespread development support. Open source MCUs could thus be a game-changer in the emerging Internet of Things and other embedded markets.
The whole idea of making the microcontroller open source, after all, is to allow developers to customize the design for their specific needs rather than compromising on a best-match stock model. www.EmbeddedDeveloper.com
7
Embedded Developer
Automated Petrol Stations By Pavel Oklestek, Oaxis
The AVK Automatic Fuel Dispensing Machine by PETROCard Czech Ltd is a modern device for self-service dispensing of various types of fuel and other liquids. The custom designed AVK building forms a compact unit that consists of the following components: • Built-in Double Shell Tanks • Hydraulic Refueling Units • Fuel Dispensing Unit • Automated Control and Transfer Unit • Payment Terminal
8
March 2017
www.EmbeddedDeveloper.com
9
Embedded Developer
Built-in Double Shell Tanks Built-in double shell operational tanks are inserted into special ruggedized containers. There is the space between the outer wall of the container (third shell) and operational tank for visual inspection and safe operation. This space also serves as the emergency sump.
Hydraulic Refueling Units Each AVK is equipped with refueling suction pumps for both petrol and diesel delivery A petrol vapor recovery system is added conform to environmental standards. Connection ports are fitted with a non-return valve and suction pumps to minimize fuel spillage.
Fuel Dispensing Unit The AVK is certified to dispense both petrol and diesel fuel, and tanks can be configured for either singular or dual capacity. The dispensing device generates a flow of 40 l/min, 80 l/min and 120 l/min. The dispenser is equipped with an ATC temperature compensation (conversion to a reference value of 15° C) which gives the operator maximal fuel inventory control.
Automated Control and Transfer Unit A special electronic system was developed for control and safety management of the AVK. The system provides:
10
•
Fuel dispensing management
•
Refueling including signalization and overfill protection
•
Fuel level measurement
•
APU (Auxiliary Power Unit) start up
•
Safety management and evaluation
•
Lightening control
•
Data transfer between AVK and remote central office or surveillance center
March 2017
Payment Terminal For security, and to protect customers during transactions and fueling, the stations are monitored by motion-activated recording cameras. The fuel pump is activated with a credit card or authorized petrol card, and consists of three main user interfaces: •
Touchscreen monitor for main automatic dispenser control, displaying of registration status, video or entertainment presentation and selection of dispenser functions.
•
Control panel with bank terminal and the space for contactless card sensing. Bank cards sensor serves as the sensor of magnetic local cards. The keyboard with display accepts PIN codes, and a contactless cards sensor area is intended for capture of RF cards.
•
Receipt slot for generating payment information
The AVK Automatic Fuel Dispensing Machine
The AVK Automatic Fuel Dispensing Machine is powered by a STKa53, TQMa53 and I/O extension board (above). The STKa53 is used for security, including checking CB status, video cameras, video recorder, etc.
About the Author Pavel Oklestek is the CEO of Oaxis Ltd, a leading supplier of TQ Components and design in the Czech Republic. The electronics design and implementation of the AVK Automatic Fuel Dispensing Machine by PETROCard Czech Ltd was provided by Oaxis. More information may be obtained by contacting pavel.oklestek@oaxispro.cz.
The TQMa53 embedded module from TQ-Group supports the NXP i.MX537 processor based on the ARM® Cortex®-A8 core. The i.MX537 is an NXP Energy-efficient Solutions product and is optimized for both performance and power to meet the demands of high-end, advanced applications. Ideal for a broad range of applications in the consumer, automotive, medical and industrial markets, the i.MX537 includes an integrated display controller, full HD capability, enhanced graphics and connectivity features. For more information: www.embeddedmodules.net
www.EmbeddedDeveloper.com
11
Embedded Developer
12
March 2017
Deep Thoughts Regarding the
Bodacious
BRAIN There’s more to building a brain than one might think. Should each cell reside in blissful ignorance or have some level of spatial awareness, for example?
By Max Maxfield, Designline Editor, EE Times
www.EmbeddedDeveloper.com
13
Embedded Developer
One of my problems is that... “Squirrel!” I’m sorry, what were we talking about? Oh, yes; I’m easily distracted. I’m also given to wild enthusiasms and I do tend to over-engineer things, but “to know me is to love me” (as I always say).
Have you got the domes? (No, I always walk this way!) My current “project du jour” is the Bodacious Brain. Considering the fact that I only really started thinking about this little rascal in earnest last Friday, it’s actually progressing rather rapidly. For example, we already have two possible enclosures for the little beauty as illustrated below.
container on the right at a Pottery Barn store on sale for only $44. I’m currently planning on using the smooth-sided dome for my first implementation because it looks more “scientific,” but I also have plans for a future incarnation in the curvaceous container.
When I say jump... Before we hurl ourselves deeper into the fray, let’s briefly remind ourselves that we’re going to have a master controller in the base of the tower and lots of semiautonomous cells (neurons) filling the tower. In addition to some other “stuff,” each cell will feature a tri-color LED that will indicate the cell’s current state.
I ordered the smooth-sided 25" tall, 10" diameter dome on the left from eBay last Friday before heading home. The following day, while doing some Christmas shopping with my wife (Gina, the Gorgeous), I ran across the curvaceous
Figure 1.
14
(Source: Max Maxfield)
Figure 2.
(Source: Max Maxfield)
March 2017
The master controller is responsible for providing the initial stimulus into the bottom layer of cells. Using an Arduino Nano, we currently have 12 signals available for this purpose, but we could use a different microcontroller (MCU) or create a port expander to provide more signals if we so desire.
cheerful (five for ~$22) Arduino Nanos from Amazon.com along with some meaty breadboards from MPJA.com. Just for giggles and grins, I’ve also ordered a couple of hundred common-cathode tri-color LEDs from Amazon.com plus associated currentlimiting resistors from Digikey.com. Phew!
Analog versus digital
Structured versus higgledy-piggledy
The first step is to start prototyping the cells that will eventually form the Bodacious Brain. Each cell will act as an individual, self-contained, and selfmotivated entity, accepting signals from, and sending signals to, surrounding cells.
I keep on bouncing back and forth between (a) placing and connecting the cells in a structured manner in the threedimensional space inside the brain versus (b) placing and connecting them in a more higgledy-piggledy (random) fashion.
I had originally toyed with the idea of creating analog cells, but I quickly evolved to each cell being digital based on a small MCU. Each of these MCU-based cells will be running the same program. The current plan is for this program to be preloaded into the cells, and—at run time—for the master controller to upload various mode control values and parameters into all the cells in parallel using a shared I2C bus. Given a choice, I’d prefer to be able to upload complete programs into the cells via the I2C bus, but I’m still trying to wrap my brain around the way in which this could be achieved.
Let’s start by considering a structured approach. I just measured the inner diameter of the smooth-sided glass tube as being 9.25". Since we’re planning on ‘airwiring’ the components, creating each cell as an approximately 1.75” x 1.75" x 1.75" cube is probably about as good (small) as we can get. If we stick to pure cubes, however, this means that we can only manage 13 cells per layer as illustrated in Figure 3.
I’m going to start prototyping the cells using Atmel MCU-based Arduino MCUs because these are easy for me to play with (the final implementation may end up using Microchip PIC MCUs). Based on this, I’ve ordered a bunch—I think that’s the technical term—of the cheap-and-
Figure 3.
(Source: Max Maxfield)
www.EmbeddedDeveloper.com
15
Embedded Developer
(On the other hand, if we allow ourselves a little more freedom regarding our definition of a cube, we could increase this to 25 cells per layer, two possible implementations of which are in Figure 4. What about the vertical plane? We could replicate the layers and simply stack them one on top of the other, or we could offset them layer by layer by one third of each cell’s arc as shown in Figure 5. Remember that I’m just throwing ideas at the wall to see which ones will stick. Maybe this structured approach is flawed from its conception; perhaps a higgledy-piggledy layout would be more aesthetically pleasing. As always, I welcome any and all feedback; what do you think about this?
Figure 4.
(Source: Max Maxfield)
Pseudo cubic cells; Inputs and outputs At some point, of course, we have to draw a metaphorical line in the sand. Thus, irrespective of the brain’s coarse cellular configuration (structured or random), let’s visualize this classic cognitive creation as being cylindrical, and let’s also envision each cell forming the brain as occupying a vaguely cubic region of space. Based on these assumptions, let’s say that each cell can—at most—connect to one cell above, one cell below, one cell inwards (toward the central axis of the cylinder), one cell outwards
16
Figure 5.
(Source: Max Maxfield)
(away from the central axis toward the outside world), one cell clockwise (looking down from the top of the cylinder), and one cell widdershins (counterclockwise).
March 2017
As an aside, if we decide to use terms like “Up” (U), “Down” (D), “Clockwise” (C), and “Widdershins” (W) to describe the locations of adjacent cells, then what would be good terms to use for pointing into and out of the cylinder, where these directions are indicated by ‘?’ in the illustration below?
Moving on, one possible highlevel cell implementation boasting six inputs and one output could be as illustrated in Figure 7.
Figure 7.
Figure 6.
(Source: Max Maxfield)
Furthermore, what terms would we use in the case of a cell located on the cylinder’s central axis, where these directions are indicated by ‘??’ in the above illustration?
(Source: Max Maxfield)
The idea is that each of the six inputs comes from one of the surrounding cells (above, below, clockwise, etc. as we just discussed). In addition to using three pins to control its tri-color LED, the cell also has one output that’s used to indicate if it’s on/active or off/ inactive. This output would be fed back to all of the surrounding cells. In this case, our MCU would have to have 14 pins. In addition to the six inputs and one output, we also need +5V and GND, SCL and SDA (for the I2C bus), and R, G, and B (for the LED). An alternative implementation could be to have six outputs feeding back to the cells driving the six inputs, in which
www.EmbeddedDeveloper.com
17
Embedded Developer
case our MCU will require 19 pins (which really means 20 pins with one spare).
use MCUs presented in 20-pin DIL (dualin-line) packages and solder directly onto their pins. Alternatively, we could use MCUs presented in surface mount packages, in which case we’d mount them on little break-out boards, but we’ll still air-wire the through-hole resistors.
Where am I? Who’s there? Who are you? As was previously noted, we’re planning on running the same program on all the cells. We can envisage this program as making decisions and performing actions like: “If one of my inputs is active, I’ll make my LED red; if two of my inputs are active, I’ll make my LED green; ... if all six of my inputs are active, I’ll make my LED white.” Figure 8.
(Source: Max Maxfield)
Having these matched inputs and outputs would give us much more control. If a cell receives a stimulus from the cell below, for example, it may decide to propagate this signal onwards and upwards, but it may be preferable to not propagate it back down again lest the two cells become locked in a feedback loop (addressing this issue is trickier if we have only a single output). As another aside, remembering—once again—that we’re planning on ‘air-wiring’ everything together, we could simply
18
But what about cells located on the outer edge of the brain? They can only ever have a maximum of five surrounding cells. How about the cells residing on the uppermost layer. They won’t have any cells connected to them from above. Things will be even worse at the beginning when I’m prototyping things and can play with only a handful of cells. One way around this is for all the cell’s inputs to have their internal pull-up resistors made active. On power-up, each cell could drive its output(s) to a logic 0, wait for a fraction of a second, then
March 2017
check its six inputs to see which ones are being pulled low by other cells. This technique would at least let each cell know how many other cells are connected to it, and which ones, and it could use this information to modify its “if this, do that” decision tree. Furthermore, using this approach, each cell could also infer some amount of spatial information, such as whether it was located on the top layer, the bottom layer, or an inner layer; also, whether it was on the outer surface of the brain or buried deep inside. Generally speaking, we probably won’t be too bothered about the cells having spatial awareness. We’ll be quite happy if some primary stimulus triggers a ripple of responses throughout the brain. On the other hand, suppose we decide to implement a “music mode” in which the amplitudes of the various frequencies are reflected by the vertical number of layers that are illuminated (similar in concept to the BADASS Display)? In this case, it would be useful for every cell to know its location in relation to the whole. One way to do this would be to augment the initial “discovery” process discussed previously. Once the cells have determined how many of their inputs are connected to other cells, the master controller in the base of the tower could transmit a signal to a
“prime cell” on the bottom layer. It might be that this signal is a coded series of pulses saying “You are cell 0 on level 0.” This prime cell then could then send a message to the cell clockwise saying “You are cell 1 on level 0.” This would continue around the brain until the final cell tried to communicate with cell 0, at which point cell 0 might pass a signal to the cell above saying “You are cell 0 on level 1,” and so on and so forth.
You are the light of my life As we previously discussed, one of the possible display modes for a cell might be based on decisions along the lines of: “If one of my inputs is active, I’ll make my LED red; if two of my inputs are active, I’ll make my LED green; ... if all six of my inputs are active, I’ll make my LED white.” One question to consider is whether the cell’s LED/output(s) should remain active so long as its inputs stay in this configuration, which could potentially lead to the entire brain becoming locked up in a certain state. Alternatively, the cell might direct its LED/output(s) to only remain active for a specified time, where this time could be fixed, or vary based on the number of active inputs, or even have some random element to it. Similarly, once a cell moves to its inactive state, it may be a good idea to force it to “take a break” for some amount of time before it once again returns to check its inputs.
www.EmbeddedDeveloper.com
19
Embedded Developer
This also leads us to consider whether— when in general use—the output(s) from a cell should be simple levels (0 and 1), in which case the only information available to surrounding cells is if this cell is currently on/active or off/ inactive. But suppose we want to use a different display mode whereby a cell displaying say a yellow color wishes to convey this information to its surrounding cells? In this case, it might be preferable for the cell to transmit a series of pulses on its output(s) saying something like: “I just started displaying yellow.”
On the other hand, if we adopt a more structured architecture, then this opens the door to a lot of possibilities. At a minimum, I could create a series of small jigs to allow me to create series of closely matched common parts. However, I must admit that I’m currently lusting over the idea of a DIWIRE—or even (be still, my beating heart) a DIWIRE PRO—from the guys and gals at PensaLabs.com.
Getting bent out of shape One thing I know for sure is that there’s going to be an awful lot of wire bending involved in this project. If we go for a higgledy-piggledy topology, then each cell will be a custom creation, in which case I’ll be pretty much on my own (oh, the loneliness of the long-distance bender).
20
Sad to relate, I fear that both of these little beauties are well outside my price range, but I live in hopes that one day someone I know will acquire one, at which time they will discover that they have a new best friend, LOL.
There is a new name in Discretes, Logic and MOSFETs.
Nexperia is a dedicated global leader in Discretes, Logic and MOSFETs devices. Originally part of Philips and more recently NXP, we became independent at the beginning of 2017. Focused on efficiency, Nexperia produces consistently reliable semiconductor components at high volume: 85 billion annually. Our extensive portfolio meets the stringent standards set by the Automotive industry. And industry-leading small packages, produced in our own manufacturing facilities, combine power and thermal efficiency with best-in-class quality levels. Built on over half a century of expertise, Nexperia has 11,000 employees across Asia, Europe and the U.S. supporting customers globally. Introducing: Nexperia, the Efficiency Company.
nexperia.com
Embedded Developer
SPEED Optimizations
in 8-bit MCUs By Jan NyrĂŠn, IAR Systems
22
March 2017
Many engineers think that everything that could be gotten out of 8-bit devices has already been done and the only way to further improve performance is to move to a more powerful device such as a 32-bit device. However, moving to a 32-bit device can be associated with penalties, for instance on power consumption and ease of use. Regardless of how efficient your 32-bit device is, an 8-bit device normally consumes much less static power than a 32-bit. Although the price of 32-bit devices is constantly decreasing, there is still a significant price premium of a 32-bit compared to an 8-bit. Performance improvements on an 8-bit device, both in size and speed, can be achieved by some simple means that are not obvious to all. This article gives you some basic tips and examples of how to get even more out of these devices.
www.EmbeddedDeveloper.com
23
Embedded Developer
Make Full Use of Your Compiler
Structuring the Code
The first and most obvious path to improving performance is to use the full set of functionality provided by your compiler. Compilers today are immensely advanced and many of them offer optimizations that previously only could be achieved by writing smart code. With the functionality provided by the smart compiler you can focus on writing understandable, readable and maintainable code.
It is important that you structure your program so that you can optimize different parts of the code in different ways. At this stage, knowledge of the final application and how it will be used is crucial. Consider for example a program where one specific part of the code is used frequently and which is very time critical, for instance, a wireless communication stack of an Internet of Things application. To avoid unnecessary lag in the communication channel, it is imperative that this section of the code is fast.
All decent compilers today offer function inlining, unroll loops when applicable, eliminate common subexpressions and perform hoisting. Depending on the application and the available memory, compilers can optimize on speed at the expense of code size and vice versa. Make sure that you know your device and use the compiler accordingly.
Code Handling However advanced your compiler is, there are optimizations that it just cannot do for you. This is where you as a programmer need to carefully consider how you write your code. I will list some important considerations and pitfalls, assuming that you are running on an 8-bit device.
24
You must thus structure the code and separate the communication stack part of the code from the rest so that it can be optimized based on speed, while the rest of the code is optimized on other parameters such as code size. Any seasoned developer also knows the positive side effects of structuring the code: added maintainability and portability.
March 2017
Use the Most Efficient Data Type Different architectures have different natural data sizes. 32-bit operations are for instance more resource consuming for an 8-bit device than for a 32-bit device. On the other hand, 8-bit operations are less efficient on a 32-bit architecture. As can be seen in the following example,
the simple addition of two chars requires 2 cycles on a 32-bit machine, whereas the same operation only requires 1 cycle on an 8-bit machine. If the same operation is performed on two ints, the 32-bit machine can make it in 1 cycle and the 8-bit machine requires 2 cycles.
Any seasoned developer also knows the positive side effects of structuring the code: added maintainability and portability. www.EmbeddedDeveloper.com
25
Embedded Developer
Data size is also important in memorymapped I/O and communication protocols. The recommendation is therefore to use typedefs to define fixed size, as shown to the right.
Other Recommendations on Data Types The ANSI C standard prefers signed types, with the implication that any type without a signed/unsigned keyword will automatically be treated as signed. Signed types are more expensive than unsigned types, in particular because they inhibit optimization when arithmetic operations are substituted with cheaper bit operations. So unless you specifically need to handle negative numbers, use unsigned types. Floating-point numbers are expensive. As a rule of thumb, calculations with single-precision floating-point numbers generate three times larger and slower code. Double-precision numbers are about three times larger and slower than singleprecision numbers. However, there are obviously times where you really need to handle numbers with fractions. In most cases you do not need the high precision that floating-point arithmetic provides. For instance, let’s assume that you have a temperature sensor with a precision
26
to 1/100th of a degree. If you use a uint16_t, measuring 1/100th degrees, you will still have a range from 0 to 650 degrees Kelvin. If that’s not enough, use an uint32_t.
Conclusions Even though you may think that you’ve already gotten the most out of your 8-bit device, there are some additional things you can do to get even more out of it. In this text, I have highlighted some methods for this: •
Make use of your compiler. Make sure that you understand the full set of features provided by your compiler and trust it so that you can focus on writing structured, understandable and maintainable code.
•
Optimize subsections of the code individually. Structure your code so that you can optimize different sections of the code separately, depending on how the code is being executed.
•
Use the most efficient data type for your architecture. The natural data type is normally more resource efficient.
•
Unless you specifically need to handle negative numbers, use unsigned types.
Bearing these tips in mind during development aids you in achieving the full potential of your 8-bit device.
March 2017
Connected Vehicles. Connected Infrastructure. Connected Auto. Connected Auto is your channel to reach the engineers, companies and auto manufacturers who are designing connected cars. Our newsletters,magazines,website and conferences are valuable marketing channels for companies selling products to this market. Make sure Connected Auto is on your radar for 2017.
ConnectedAuto.org
www.EmbeddedDeveloper.com
27
Embedded Developer
TQ COM Express Modules
Optimize Industrial Automation, IoT and Defense Applications
28
March 2017
TQ releases a new line of Intel® CPUs including Intel® Atom E3900 (Apollo Lake) and 7th Generation Intel® CoreTM processors (Kaby Lake).
The explosion in Industrial Automation, IoT and Defense applications requires smart embedded solutions that are robust, high-performance, flexible and available for decades.
TQ-Group in Germany has released a complete line of COM Express® modules (SOMs) for the new Intel® Atom E3900 (Apollo Lake) and 7th Generation Intel® CoreTM processor (Kaby Lake). With this release, TQ now has modules that cover the Intel® AtomTM Bay Trail, Apollo Lake and the 5th, 6th and 7th generation Intel® CoreTM processors.
www.EmbeddedDeveloper.com
29
Embedded Developer
These new low-power processors from Intel® offer significantly higher performance and lowerpower consumption with greatly improved graphics capabilities. Using TQ mainboards in conjunction with these modules provide customers with significant growth and flexibility options while continuing to use the same baseboard in their product.
In addition to the Board Controller, TQ also supplies a feature-rich BIOS interface (UEFI) for: • Touch support in the BIOS setup screen • Configuration for default setup • Multiple setup configurations within one BIOS image • Security administration Two optional Hardware/ Software features exclusively offered by TQ include:
The following product brief details the advantages of using the TQ COM Express® Modules in Industrial Automation, IoT, Defense and other robust embedded applications.
TQ Intel® Modules are pre-packaged with their own software We provide ‘Board Controller’ software with every TQ module to give designers control over: • Power management • Fan and thermal control • I2C (master/slave) • UARTS and GPIO • ‘Green ECO-off’ – reducing current consumption in S5 standby > 0.1 watts
30
• Trusted Platform Module (TPM): A dedicated microcontroller designed to secure hardware by integrating cryptographic keys into devices. • Industrial RTC (iRTC): Highly accurate and stable RTC over the complete temperature range with only a 250 nA current consumption
Extended Board functionality with patented COM Express® risers for PCIe I/O Cards TQ‘s patented risers for the COM Express® 10 carrier board extend functionality with PCIe I/O cards.
Simplified installation with the TQ COMKit-HHD/SSD COMKit-HDD enables easy installation of HDD and SSD to the underside of the baseboard without interfering with processor cooling.
March 2017
TQ Intel® AtomTM Embedded Modules
TQMxE38M-xx – Intel® AtomTM Bay Trail Modules
TQMxE39M-xx – E3900 Intel® AtomTM Apollo Lake Modules
TQMxE38C-xx – Intel® AtomTM Bay Trail Modules
The TQMxE38M-xx embedded modules enable the implementation of powerful and economical Atom-based systems based on the standard COM Express® Mini Type 10 form factor (84mm x 55mm) with a temperature range of -40° C to +85° C. With the Type 10 compliant pin out of 1x 220 pins, the user has access to all the basic interfaces of the CPU such as the digital display interface (DDI) / embedded DisplayPort (eDP), LVDS, up to 4x PCIe, 2x SATA and 4x USB (2.0 and 3.0).
This TQ module (TQMxE39M-xx) is based on the latest generation of the Intel® Atom™. Code name Apollo Lake, these TQ modules achieve a new level of compute performance together with security and media processing. It is a COM Express® Mini Type 10 form factor (84mm x 55mm) with a temperature range of -40° C to +85° C and a typical power envelope of 3 to 8 watts. With the latest Intel® integrated graphics processor, the module delivers 4K high resolution graphics, immersive 3D processing, and greatly increased video encode and play-back performance. The module interfaces include 8 USB ports—including 2x USB 3.0— and up to 4 PCIe lanes, enabling high bandwidth communication with peripherals and additional interfaces on the carrier board.
The TQMxE38C-xx embedded modules differ from the TQMxE38M-xx only in the pin out, 2x 220 pins. The TQMxE38Cxx modules are COM Express® Type 6. The TQMxE38C-xx embedded module enables the implementation of the powerful and economical AtomTM platform based on the PICMGTM standard COM Express® Compact, (95mm x 95mm) Mini ITX (COM.0 R2.1). From the Type 6 compliant pin out, the user has access to all the interfaces of the CPU, so all the features of the Intel® AtomTM processor can be used. Direct access to interfaces such as the digital display interface (DDI)/ embedded DisplayPort (eDP) and USB devices gives you the freedom to use the features of the CPU in the way that suits the application.
With options such as conformal coating, optimized cooling solutions, TQMx Board Controller and UEFI interface, the TQMxE39M is ideal to power mobile (15% longer battery life) rugged real-time computing in industrial automation, digital surveillance, aviation, medical, retail and more.
Low power consumption enables very compact, passively cooled system designs that are ideal for vending and POS terminals, medical imaging, compact diagnostic workstations as well as IoT-connected industrial controls and extremely compact, robust box PCs and thin clients.
Low power consumption enables very compact, passively cooled system designs that are ideal for vending and POS terminals, medical imaging, compact diagnostic workstations as well as IoT-connected industrial controls, extremely compact, robust box PCs and thin clients.
www.EmbeddedDeveloper.com
31
Embedded Developer
TQ Intel® CoreTM i3/i5/i7 Embedded Modules
TQMx50UC-xx – Intel 5th Generation Broadwell Modules
TQMx60EB-xx – Intel® 6th Generation Skylake Modules
The TQMx50UC-xx modules are COM Express® Compact Type 6 (95mm x 95mm) and have CPUs with outstanding graphics performance and features such as 3D scanning, gesture control, and voice commands. The integrated Intel® graphics core HD5500/HD6000 supports up to three independent display interfaces with up to 4K resolution. It’s possible to drive multiple external HD displays from one single output with Multi-Stream Transport (MST) in accordance with the DisplayPort 1.2 standard. The overall performance also benefits from the ultra-fast DDR3L-1600 onboard memory and up to 4 MB cache. For industrial applications, the module delivers high bandwidth interconnection with PCI Express, SATA Gen3, USB 3.0 and Gigabit Ethernet with IEEE 1588 support.
Dual-channel DDR4-2133 support, high bandwidth PCIe, 4x SATA (6 Gb/s) and 4x USB 3.0 make the TQMx60EB series of modules ideal for applications that require high-end performance and graphics. The COM Express® Basic Type 6 (125mm x 95mm) module can be equipped with 6th Generation Intel® Core™ i3/i5/ i7 or Intel® Xeon® E3-15xx v5 processors. The integrated Intel® GT1/GT2/GT4e graphics controller supports up to three independent display outputs with up to 4K resolution at 60 Hz and excellent 3D / rendering performance. Power and performance optimizing functions like Intel® Turbo Boost Technology 2.0 and Enhanced Intel® SpeedStep® Technology provide an outstanding user experience, but with moderate power consumption.
Typical applications include medical and industrial imaging, centralized control room technology, shop floor terminals, HMIs, robotics, professional gaming, infotainment, professional AV, smart video surveillance, autonomous vehicle control, and high-end digital signage applications. The ability to drive multiple displays makes the module ideal for applications in retail and kiosks where control of multiple screens of up to three independent cash or vending machines is necessary.
32
Key applications for the TQ modules include 4k multi-screen solutions where one embedded module controls up to three independent screens without the need for a dedicated graphics card. Applications can be found in operating rooms and medical diagnostics terminals, HMIs in industrial machinery and facilities, digital signage, vending and retail applications as well as gaming and infotainment terminals.
March 2017
TQMx70EB-xx – Intel® 7th Generation Kaby Lake Modules The Kaby Lake 7th Generation modules from TQ deliver high-end computing with increased CPU and graphics performance compared with Skylake. Dual-Channel DDR4-2133 support, high bandwidth PCIe, 4x SATA (6 Gb/s) and 4x USB 3.0, the TQMx70EB guarantees best-in-class system performance. The COM Express® Basic Type 6 (125mm x 95mm) module TQMx70EB supports the Intel® Core™ i3, i5 and i7 7000E series processors with higher clock speeds than Skylake (up to quad core 3.7 GHz / 8 MB cache). For server-like applications, the module also supports quad core Intel® Xeon® E3-1500 v6 processors. Equipped with 10-bit codecs (HEVC/VP9), the module can decode and encode HEVC video at 4K resolution with 10-bit color in hardware. The latest SSD mass storage devices like the Intel® Optane™ with 3D XPoint™ technology are supported with better results for read/write latency and high data throughput. With options such as conformal coating, optimized cooling solutions, TQMx Board Controller and UEFI interface, the TQMx70EB sets a new level of performance, lowers power consumption by up to half and is ideal for applications where data intensive streams need to be processed and displayed in realtime. Target markets include large data processing systems such as servers, medical imaging systems, video surveillance and vision-based quality control, simulation equipment, vision systems in industrial control rooms and other plant-wide supervision systems or high-end professional gaming and digital signage.
www.EmbeddedDeveloper.com
33
Embedded Developer
Production-ready Carrier Boards with COM Express® Form Factors TQ delivers a number of mainboards for the different COM Express® form factors in accordance with the PICMG™ standards and built to the same high standard as their modules. They make ideal production-ready carrier boards for the different modules that TQ manufacture. In addition, TQ‘s patented risers for the COM Express® 10 and COM Express® 6 carrier boards extend functionality with PCIe I/O cards, and the COMKit-HDD which enables easy installation of HDD and SSD to the underside of the baseboard without interfering with processor cooling.
MB-M10-1 This carrier board is a 100mm x 100mm eNUC board with 220 pins for the TQ AtomTM COM Express® Type 10 Modules. It is scalable throughout the TQMxE38M (Bay Trail) and TQMxE39M (Apollo Lake) Intel® AtomTM module range and forms a very compact hardware system. This means you can go from a single core 1.46 GHz AtomTM to a quad core 4x 2.0 GHz AtomTM on the same carrier board.
34
MB-COME10-1 The MBCOME10-1 carrier board is a 170mmx 170mm COM Express® Type 10 Mini compliant with 220 pins. This carrier board supports both the TQMxE38M (Bay Trail) and the TQMxE39M (Apollo Lake) Intel® AtomTM COM Express® Type 10 modules. This means you can go from a single core 1.46 GHz AtomTM to a quad core 4x 2.0 GHz AtomTM on the same carrier board.
MB-COME6-1 The MBCOM6-1 carrier board is also 170mm x 170mm, but is COM Express® Type 6 Compact compliant with 2x 220 pins. This carrier board supports the TQMxE38C (Bay Trail) and the TQMx50UC (Broadwell) modules. This means you can go from a single core 1.46 GHz AtomTM to a dual core 2x 2.2 GHz i7 on the same carrier board.
MB-COME6-2 The MBCOM6-2 carrier board is also 170mm x 170mm in size, but is COM Express® Type 6 Basic compliant with 2x 200 pins. This carrier board supports the TQMx60EB (Skylake) and the TQMx70EB (Kaby Lake) modules. This means you can go from a quad core 4x 1.9 GHZ i5 to a quad core 4x 3.0 GHz i7 on the same carrier board.
March 2017
Extend Board Functionality with Patented COM Express Risers for PCIe I/O Cards
TQ risers extend I/O functionality without compromising cooling
TQ’s Coast-to-Coast North American Sales and Support TQ products are distributed in North America by Convergence Promotions LLC. TQ-Group is a $250M electronics company based in Munich, Germany. TQ has a significant range of modules supporting Intel®, ARM® and PowerPC®. All of TQ’s modules are designed for use in embedded applications where low power, fan-less, rugged designs are a key requirement. TQ’s obsolescence management program guarantees support and availability for ten years after purchase.
HDD/SSD installation simplified with the TQ COMKit
FOR MORE INFORMATION, CONTACT: Vaughn Orchard vaughn@convergencepromotions.com +1 (508) 209-0294 www.embeddedmodules.net
www.EmbeddedDeveloper.com
35
Embedded Developer
Nexperia Takes the
POLE POSITION
in Automotive Logic Applications An exclusive interview with Michael Lyons, Nexperia’s Technical Marketing Manager for BL Logic.
36
March 2017
ON FEBRUARY 7, 2017 THE FORMER NXP STANDARD PRODUCT BUSINESS BECAME A NEW COMPANY WITH THE TRADENAME NEXPERIA. NEXPERIA IS AN INDUSTRY LEADING SUPPLIER OF DISCRETE, LOGIC AND POWERMOS SEMICONDUCTORS WITH ITS FOCUS ON THE AUTOMOTIVE, INDUSTRIAL, COMPUTING, CONSUMER AND WEARABLE APPLICATION MARKETS.
Logic allows different types of chips or circuits to work together by acting as an interface between them, and as such even though it has been around since the days when engineers used slide rules, it is still an essential part of embedded design. As a leading supplier of logic products to the automotive industry, Nexperia’s logic products are employed in a wide variety of automotive applications including instrument clusters, body control modules and engine control units. The company offers a complete range of standard and Mini-Logic functions, including analog switches, buffers/inverters, bus switches, counters, decoders/ de-multiplexers, multiplexers, flip-flops, gates, latches, level shifters, multivibrators, Schmitt-triggers, shift registers and transceivers. Along with this broad choice, Nexperia helps speed the design process with the versatility of its configurable logic and offers advanced logic solutions with low dynamic and static power consumption. And its industry-leading small packages, produced in-house, combine power and thermal efficiency with best-in-class quality levels. Recently, we had a chance to talk with Nexperia’s Michael Lyons about this new company with a long history, broad experience and a global customer base. Lyons himself has over 25 years of experience in new product development functions within the semiconductor industry, having held various marketing, business development and product engineering management positions. Michael Lyons
www.EmbeddedDeveloper.com
37
Embedded Developer
How has the launch of Nexperia gone so far? What has the response been from your automotive customers to the new company? The launch of Nexperia has gone well. The planning and long hours put in by the Management Team and all teams over the globe paid off. We did have start-up issues with a few interface systems at some customers, but the customer response has generally been very cooperative and positive. The slogan of the new company seems to be “the company where efficiency wins”. Could you explain this a bit more? “Nexperia—the company where efficiency wins” captures how we strive to be efficient or improve efficiency in our way of working and our products. We deliver over 70 billion semiconductor products annually to thousands of customers, either directly and via our channel partners. As the efficiency semiconductor company, we offer power efficient devices in industry leading, small, space efficient packages. Logic gates are the fundamental building blocks of any digital circuit. Your comprehensive range of logic gates allow for a variety of control logic solutions in automotive functions such as engine monitoring, navigation, interior lighting, battery monitoring, etc. Are there any interesting applications you’ve come across that you can tell us about?
Continuing with the efficiency theme, we have seen a growing increase of voltage level translators within automotive applications. As the density of electronics increases within vehicles, the need to lower the power consumption of applications has increased. As power is directly proportional to voltage, moving to lower supply-voltage achieves the goal of power reduction. Reducing the supply from the 5 V to the 3.3 V supply node or the 3.3 V to 1.8 V supply node results in a 34% or 30% system level power reduction. In practice, not all elements of an application have lower voltage equivalents or there is a requirement to interface with legacy higher voltage systems. Low-power voltage level translators such as LVCnT, AVCnT and AXPnT enable system level power reduction while maintaining compatibility with existing higher voltage systems. What about the trend to go from gold to copper interconnects? Will we see this in the automotive Q100 portfolio? Yes, we now have the reliability data required to allow us to introduce copper wire products into the automotive portfolio. Copper wire has been used in our non-automotive portfolio since 2012 with a failure rate < 7 PPB. This is better than our gold wire products and meets our safe launch criteria. We have issued process change notifications to begin the conversion of the automotive portfolio to copper wire.
As the efficiency semiconductor company, we offer power efficient devices in industry leading, small, space efficient packages.
38
March 2017
Last year NXP introduced Industry’s smallest 8-pin GX logic package for mobile, portable and IoT applications. Will there be automotive applications as well? The 8-pin GX package is fully AEC-Q100 qualified and we will be able to release it into our automotive range of products once we have met the safe launch criterion required by our zerodefect automotive strategy. A move to smaller packages is expected in automotive applications as electronic content increases in vehicles. The preference for optical inspection of solder joints on all but ball grid array packages may limit the use of the land grid array GX8 to only space critical applications in automotive. How have design engineers used Dual PCB Configurable Logic to improve the way they implement control logic solutions into their designs? We see several innovative designs taking advantage of the nontraditional functions permitted by PCB configurable gates. With traditional functions a design engineer would require two devices to create a 2-input NAND gate with 1-input inverted. If there was a requirement to interface to slowly transitioning signals a Schmitt-trigger device may also be required. With integrated Schmitttrigger inputs and non-traditional functions the configurable gates allow space and power savings by replacing up to three discrete logic devices.
DQFN represents the industry’s smallest logic packages for standard logic gates. These leadless packages provide space savings up to 76% over traditional leaded TSSOP packages. Talk about the future of DQFN in the automotive space. The use of our automotive DQFN packages allows our customers to reduce the footprint of their solutions while maintaining the ability to automatically optically inspect the quality of solder joints. We continue to see a migration within automotive applications away from the larger SO packages towards TSSOP and now the DQFN packages.
As a family of the future, AXP is set to grow in sub 3.3 V supply node applications.
What’s going on with the advanced extremely low voltage and power (AXP) logic family? We continue to see the growth of AUP in low power applications, primarily due to compatibility with legacy 3.3 V applications. AXP is replacing AUP in applications that require higher speed but do not need 3.3 V compatibility. As a family of the future, AXP is set to grow in sub 3.3 V supply node applications. It has lower power, higher output drive and is faster than AUP at the 1.2, 1.5, 1.8 and 2.5 V supply nodes. AXP is the only logic family fully specified at 0.8 V. The introduction of the AXP translators (AXPnT) has facilitated the reduction of the system voltage to the range 0.7 to 2.75 V, while ensuring outputs can be used with legacy interfaces at 1.65 to 5.5 V.
www.EmbeddedDeveloper.com
39
Embedded Developer
Finally, talk about your plans for product design support for your customers.
A Nexperia will use its depth of engineering expertise to assist customers in finding robust, lasting solutions. Within the sales and marketing community, our product application engineers help with the design-in of existing products and the definition of future solutions. They have access to and the support of a team of application marketing managers who in turn have direct contact to the engineering community within the business groups. Regular internal training ensures that the technical competence of the team is not only maintained at a very high standard, but continually improves. As the efficiency company our goal is to ensure we equip our team with the tools and information they need to provide quality customer support quickly. For more information about Nexperia go to http://www.nexperia.com.
Nexperia continually invests in new process and package technologies, as well as new packaging facilities, and is focused on increasing performance, lowering power consumption, and reducing size. The company has the largest portfolio of dedicated Q100 devices. As an example, consider its 74LVC08A-Q100 providing four 2-input AND gates. Inputs can be driven from either 3.3 V or 5 V devices allowing use of these devices as translators in mixed 3.3 V and 5 V applications. Features include: »» Automotive product qualification in accordance with AEC-Q100 (Grade 1) »» Specified from -40 °C to +85 °C and from -40 °C to +125 °C »» 5 V tolerant inputs for interfacing with 5 V logic »» Wide supply voltage range from 1.2 V to 3.6 V »» CMOS low power consumption »» Direct interface with TTL levels »» Complies with JEDEC standard: »» JESD8-7A (1.65 V to 1.95 V) »» JESD8-5A (2.3 V to 2.7 V) »» JESD8-C/JESD36 (2.7 V to 3.6 V) »» ESD protection: »» MIL-STD-883, method 3015 exceeds 2000 V »» HBM JESD22-A114F exceeds 2000 V »» MM JESD22-A115-A exceeds 200 V (C = 200 pF, R = 0 Ω) »» Multiple package options
40