+ All Categories
Home > Documents > EEWeb Pulse 122 - Crossbar

EEWeb Pulse 122 - Crossbar

Date post: 23-Mar-2016
Category:
Upload: eeweb-magazines
View: 230 times
Download: 0 times
Share this document with a friend
Description:
Raising the Bar: A look at Crossbar's entirely new type of memory; Introducing Fairchild 2.0; TI's Tiva C Series Connected Launchpad; Intel Galileo Overview
Popular Tags:
56
George Minassian, CEO of Crossbar CROSSBAR: Pioneering a New type of Memory TI’s New Launch Pad Ecosystem A Look at Fairchild 2.0
Transcript
Page 1: EEWeb Pulse 122 - Crossbar

George Minassian, CEO of Crossbar

CROSSBAR: Pioneering a New type of Memory

TI’s New Launch Pad Ecosystem

A Look at Fairchild 2.0

Page 2: EEWeb Pulse 122 - Crossbar

concepts to reality Bringing your

is as easy as...

Copyright ©2013 Aspen Labs LLC.

Visit: digikey.com/schemeit • partsim.com • pcbweb.com

1.

Create schematics, technical diagrams, and flowcharts using your browser.

• 600+ Symbol Library• Share Schematics Online• Export High Quality Images

digikey.com/schemeit

2.

Free and easy-to-use circuit simulator that runs in your browser.

• SPICE Simulator• AC/DC/Transient Sims• Waveform Viewer

partsim.com

3.

Full featured online CAD application for designing and manufacturing electronics hardware.

• Schematic Capture• PCB Layout• BOM Integration

pcbweb.com

Page 3: EEWeb Pulse 122 - Crossbar

CONTENTSPULSE

How Product Innovation Happens:

Engineers who use oscilloscopes are typically focused on solving problems and looking for ways to make better products. It turns out that engineers who design oscilloscopes are essentially doing the same thing.

More than a decade ago, an Agilent Technologies oscilloscope designer named Scott Genther was getting ready to launch a massive, multiyear project to design a new chip for an advanced algorithmic oscilloscope trigger—affectionately known in the Agilent R&D lab as the “nerd trigger.” Before the project even got off the ground, he came up with a remarkably simple idea for a trigger that ultimately redefined how engineers use Agilent oscilloscopes.

Where do innovative ideas like Genther’s come from? And how do you get from the original spark of inspiration to its implementation in an innovative product? Genther’s story sheds light on how product innovations happen.

The Origin of Agilent’s Oscilloscope Trigger

CONNECTING WITH CUSTOMER FRUSTRATIONS

In a quest to define the features he needed to include in his trigger ASIC, Genther visited oscilloscope customers to understand how they used triggering and what stumbling blocks they faced. “In the lab here at Agilent, R&D engineers know how to use all the triggers on their scopes,” said Genther. “But I quickly discovered our customers don’t. They don’t like thinking about triggers. A lot of customers just pick a mode and turn the knob to clean up the display. They don’t have time to sit down and think about it.”

The first customer Genther visited was having trouble defining a trigger to capture the clock edge reads he wanted to see. In frustration, he jabbed a finger at the screen and wondered why he couldn’t just click on a piece of a waveform and get what he wanted. At the time, Genther just assumed it was impossible.

The second customer Genther visited had come back from lunch and discovered a bad trace on his oscilloscope screen. The scope’s infinite persistence showed him something had happened when he was away. To figure out what caused it, he wanted to get a view of the trace by itself so he could correlate it to activity on other channels.

The customer had been doing single-shot acquisitions for days trying to locate the infrequently occurring glitch. He pushed the “single” button over and over and still couldn’t locate the problem. Watching the engineer laboriously pecking away at the button, Genther experienced his big “aha” moment. He could see that pressing the “single” button a million times is a painful way to find a one-in-a-million glitch, and he realized he could make both customers’ jobs easier using Agilent’s existing oscilloscope technology.

“The nerd trigger was going to take forever to complete, but I figured we could create very similar functionality using the scope’s mask test infrastructure,” Genther said. “We could qualify acquisitions with mask testing and use that feature to push the button for him.”

Mask testing shows all the waveforms that fail or exceed the limits of the mask. Genther’s brainstorm was to ignore all the waveforms that fail and just display the ones that do meet the mask limits. This process could happen behind the scenes, using digital technology to accomplish what the customer was doing manually.

CHAMPIONING THE IDEA

Coming up with the idea to use a reverse mask test to achieve an easier version of the trigger was just the first step in the process of bringing it to market. Next, Genther had to convince his boss and teammates the project was worth pursuing.

Figure 1: Example of a hard-to-capture signal anomaly with infinite persistence turned on

Scott GentherKey inventor of Agilent Zone Trigger

Propels Development

CONNECTED LAUNCHPAD

TI’s Tiva™ C Series

in the Cloud

The microcontroller (MCU) lies at the heart of the interconnected network of devices known as the Internet of Things (IoT). These MCUs provide the intelligence needed for the myriad of devices in the network to communicate with each other. With more and more devices being added to the IoT, the need for the development of increasingly sophisticated MCUs is created. Texas Instruments’ (TI) MCU LaunchPad evaluation kits have a history of being both low-cost and fully featured. Their latest LaunchPad—the Tiva™ C Series Connected LaunchPad—provides developers with just about everything they need to give their clever ideas a cloud-based life. TI Tiva™ Connected Launchpad

3Visit: eeweb.com

410

Featured ProductsThis week’s latest products from EEWeb.

1220

Back to Basics: JitterA look at some of the tools built into modern

digital oscilloscopes for jitter measurement and analysis.

2834

28

TI's Tiva C Series Connected Launchpad

This new board provides developers with everything they need to give their clever ideas

a cloud-based life.

20

Intel GalileoDevelopment Board

The Intel Galileo board is the first ever development board to be available from the Arduino Certified program. The Galileo board features the Intel Quark SoC X1000 processor, making it a great development board for hobbyist, student, and even professional use. The combination of the Quark SoC and the extensibility of the Arduino environment make the Galileo a powerful, easy-to-use, board for taking designs to the next level.

Overview of the

40

Introducing Fairchild 2.0By combining Fairchild's rich history of successes

with the spirit that ignited the company over 50 years ago, Fairchild 2.0 promises to drive

innovation well into the future.

34

How Product Innovation Happens

A look at the origen of Agilent's advanced algorithmic oscilloscope trigger.

George MinassianCEO of Crossbar

A conversation about the company's groundbreaking new type of memory and how it

will change the industry as a whole.

RTZReturn to Zero Comic

EEWeb Tech Lab Product Overviews

Altera Cyclone V FPGA; Fairchild Buck Converter; Intel Galileo Board; IR Audio Amp

4056

Page 4: EEWeb Pulse 122 - Crossbar

4 Visit: eeweb.com

PULSE

MBC450 Series AC-DC Power Supply 155 W to 450 W steady state depending upon output voltage. Peak power of 475 W for 24 V, 30 V & 500 W for 48 V model for 5 seconds, maximum duty cycle 10%. Goes high approximately 100 mSec after V1 reached regulation limit. Upon failure this TTL signal goes low after V1 drops more than 5% of nominal (10% for 5 V output). Signal has a 10 k pull-up resistor between the signal and the +5 Vs/b. Compensates for 200 mV of cable drop. Allows sharing of 2 or more power supplies within +/-10%. Current sharing with internal OR-ing diodes /MOSFETS for redundancy...Read More

Arbitrary Waveform Function GeneratorsDG1000Z series Waveform Generators, adding to its extensive portfolio of feature rich test and measurement equipment. Powered by Rigol’s innovative SiFi technology, the DG1000Z is ideal for low frequency testing applications requiring high signal fidel-ity, low noise floor and long arbitrary waveform length. The 2 channel DG1000Z series features Best-In-Class memory depth of 8Mpts (16 Mpts. optional) to facilitate deep-er, more intricate arbitrary waveforms and includes up to 160 built-in waveforms for quick and easy access to more predefined signals...Read More

Miniature SMT Switching POL RegulatorThe LSR7805 Miniature SMT 0.5A Switching POL Regulator Series has an input voltage range of 4.5 ~ 28.0 Vdc and an output voltage range of 3.3 ~ 15.0 Vdc, with an out-put current accuracy of ±2.0 %. The device’s line regulation is at ±0.2 % and its load regulation is at ±0.3 %. The LSR7805 Series’ operating temperature range is at -40 ~ +85 degrees Celsius...Read More

Intelligent Battery SensorThe ZSSC1856 is a dual-channel ADC with an embedded microcontroller for bat-tery sensing/management in automotive, industrial, and medical systems. One of the two input channels measures the battery current IBAT via the voltage drop at the external shunt resistor. The second channel measures the battery voltage VBAT and the temperature. An integrated flash memory is provided for customer-specific software; e.g., dedicated algorithms for calculating the battery state. During Sleep Mode (e.g., engine off), the system makes periodic measurements to monitor the discharge of the battery...Read More

Dual Cool™ Package for DC-DC ApplicationsDesigners of DC-DC conversion applications are challenged with improving power density while saving board space and reducing thermal resistance. Fairchild Semi-conductor’s new mid-voltage PowerTrench® MOSFETs with Dual Cool™ packaging technology are well suited to help with these design challenges. Fairchild has ex-panded and improved its Dual Cool packaging portfolio, an industry-standard pin-out package with top-side cooling, to include a 40-100V mid-voltage series. Silicon technology advancements, paired with the Dual Cool technology, provide excel-lent switching performance...Read More

Page 5: EEWeb Pulse 122 - Crossbar

5Visit: eeweb.com

FEATURED PRODUCTS

High Temp Shielded Power InductorsCoilcraft’s XAL 1010 Series of high-temperature, shielded power indcutors combines excellent current handling and exceptionally low DC resistance in a rugged, molded package. The XAL1010 is available with thirteen inductance values from 0.22 to 15 µH, current ratings to 98.8 Amps and DCR as low as 0.45 mOhms. The popular, cube-shaped package measures just 10.0 × 11.3 × 10.0mm. For applications with limited height restrictions, Coilcraft also offers the lower-profile XAL1060 Series. Both versions are AEC-Q200 Grade 1 qualified (-40°C to +125°C), making them suitable for harsh automotive environments...Read More

High Temperature Resistant EpoxyMaster Bond EP30HT is designed for high performance general purpose bonding, sealing, coating and casting. This optically clear, two component epoxy features high temperature resistance and a room temperature cure. It conforms to Title 21, U.S. Code of Federal Regulations, FDA Chapter 1, Section 175.105 for Food Applica-tions. Featuring a superior combination of performance properties, Master Bond Polymer Adhesive EP30HT is widely used in the electronic, electrical, computer, opti-cal, fiber-optic, medical, automotive and chemical industries...Read More

Power Supply Solution ProductsDigi-Key Corporation announced a global distribution agreement with XP Power, a power-solution leader offering the widest range of power products available from one source. XP Power is one of the world’s leading developers and manufacturers of power supply solutions for the electronics industry, continuously updating their product range to provide leading-edge power supply technology. This constant refreshing of their product line assures customers that they can source the newest, most cost-efficient products on the market...Read More

150 W Single Output LED Driver with PFCOne of the most attractive and important benefits of LED lighting technology is its energy efficiency. However, to maximize LED lighting system efficiency and reliability, the power source – the lifeblood of the LED – must be equally energy efficient. Our LED power supply series are designed with the latest in switched-mode technology, maximizing both efficiency and reliability. Finally, our PLN series devices are limited power sources which use the latest PFC topology, delivering efficiencies up to 86%. Astrodyne LED drivers include universal AC input of 120/240VAC...Read More

Extremely Efficient Cortex-A53 ProcessorThe Cortex®-A53 processor is an extremely power efficient ARMv8 processor capa-ble of supporting 32-bit and 64-bit code seamlessly. It makes use of a highly efficient 8-stage in-order pipeline balanced with advanced fetch and data access tech-niques for performance. It fits in a power and area footprint suitable for entry level smartphones, at the same time, capable of delivering high aggregate performance in scalable enterprise systems via high core density. It delivers significantly higher performance than the highly successful Cortex-A7...Read More

Page 6: EEWeb Pulse 122 - Crossbar

6 Visit: eeweb.com

PULSE

Ultra Miniature 2-Watt DC-DC ConverterMurata announced the MTU2 series of ultra miniature surface mounted 2 W DC-DC converters from Murata Power Solutions. Believed to be the industry’s smallest 2W converter and measuring just 8.2 × 8.4 × 8.5 mm with a 0.69 cm2 footprint, it is 50% smaller than the current 1.67 cm2 industry standard. With a typical conversion ef-ficiency of 85% across the full load range and a power density of 3.403 Watts/cm3, the MTU2 series is available with either a single or dual output voltage. Input volt-ages cover the popular nominal inputs from 3.3 to 24 VDC. Output voltage options include 5, 12, or 24 VDC...Read More

Audio DSPs with Mixed-Signal IntegratedThe CS470xx family is a new generation of audio system-on-a-chip (ASOC) proces-sors targeted at high fidelity, cost sensitive designs. Derived from the highly successful CS48500 32-bit fixed-point audio enhancement processor family, the CS470xx further simplifies system design and reduces total system cost by integrating the S/PDIF Rx, S/PDIF Tx, analog inputs, analog outputs, and SRCs. For example, a hardware SRC can down-sample a 192 kHz S/PDIF stream to a lower Fs to reduce memory and MIPS requirements for processing...Read More

Omnipolar Detection Hall Effect Sensor ICsThe demands for extended battery operation, greater reliability and increased fea-tures are driving the designs of mobile phones, notebook computers, video cam-eras, navigation systems and game controllers to use smaller, higher performance components. Ultra-small, hall effect, non-contact switches from ROHM Electronics can simplify and enhance your designs while offering the benefits of high-reliability and low power consumption...Read More

Continuous-Time Hall Effect LatchesThe Allegro A1210-A1214 Hall-effect latches are next generation replacements for the popular Allegro 317x and 318x lines of latching switches. The A121x family pro-duced with BiCMOS technology consists of devices that feature fast power-on time and low-noise operation. Device programming is performed after packaging to ensure increased switchpoint accuracy by eliminating offsets that can be induced by package stress. Unique Hall element geometries and low-offset amplifiers help to minimize noise and to reduce the residual offset voltage normally caused by device overmolding, temperature excursions, and thermal stress...Read More

High-performance 24-bit ADCAK5365 is a high-performance 24-bit, 96 kHz sampling ADC for consumer audio and digital recording applications. Thanks to AKM’s Enhanced Dual-Bit modulator architecture, this analog-to-digital converter has an impressive dynamic range of 103 dB with a high level of integration. The AK5365 has a 5-channel stereo input selector, an input Programmable Gain Amplifier with an ALC function. All this in-tegration with high-performance makes the AK5365 well suited for CD and DVD recording systems...Read More

Page 7: EEWeb Pulse 122 - Crossbar

7Visit: eeweb.com

FEATURED PRODUCTS

Triple Air Core Gauge DriverThe MLX10420 is a µP peripheral for air-core meters control using SIN/COS PWM com-mands. The circuit controls two independent sets of CMOS power bridges. A ten bit angle is displayed with 9 bits per quadrant. The PWM frequency is set by an on chip oscillator. A power-on self test detects open or short-circuits outputs for each air-core meter and a real time angle tracking avoids display errors. The chip can also drive one small angle air-core meters (90°). The communication with the µP is done via a three wires serial link...Read More

Intel Galileo Development BoardMouser Electronics, Inc. is now taking pre-orders for Intel’s hotly anticipated Galileo Development Board, based on the Intel Quark Application Processor. The Galileo Development Board available from Mouser Electronics is based on the Intel Quark SoC X1000 Processor, a 32-bit Intel Pentium-class system on a chip (SoC). The Intel Galileo board is the first product in a new family of Arduino certified boards featuring Intel architecture and is designed to be hardware and software pin-compatible with Arduino shields...Read More

Ultra Low Power Embedded TCP/IP ModuleThe RN171 is a small form factor, ultra-low power embedded TCP/IP module meas-uring only 27 × 18 × 3.1 mm. The RN171 is a full-featured 802.11 b/g surface mount module. Due to its small form factor and extremely low power consumption, it is perfect for mobile wireless applications such as asset monitoring, sensors, and port-able battery operated devices. For quickly and easily evaluating the RN171 module, Microchip offers the RN-171-EK, a compact, battery and USB-powered kit with push-button for AP and WPS mode...Read More

Dual-mode Wireless Power ReceiverIDT announced the industry’s first dual-mode wireless power receiver compatible with both the Wireless Power Consortium’s (WPC) 1.1 Qi standard, as well as the Power Matter’s Alliance (PMA) 1.1 standard. The innovative solution enables OEMs to use a single wireless power receiver IC to develop mobile devices fully compat-ible with the latest versions of both Qi and PMA charging bases. The IDTP9023 is a WPC 1.1- and PMA 1.1-compliant single-chip wireless power receiver consisting of a synchronous full-bridge rectifier...Read More

Ultra-Compact IGBT Drivers1SD536F2 single-channel, plug-and-play IGBT drivers are ultra-compact, high-perfor-mance and intelligent. They support 2-level, 3-level and multilevel converters and have been designed to precisely and reliably drive high-power, high-voltage IGBT modules packaged in standard 130 × 140 mm or 190 × 140 mm housings while deliv-ering optimum protection. 1SD536F2 drivers are based on CONCEPT’s sophisticated SCALE-1 driver chipset. 1SD536F2 series drivers feature an active clamping function, offering rugged and high-performance protection for costly IGBTs...Read More

Page 8: EEWeb Pulse 122 - Crossbar

8 Visit: eeweb.com

Rad-Hard 12V POL Switching RegulatorIntersil Corporation introduced the industry’s first radiation hardened 12V input point-of-load (POL) synchronous buck switching regulator capable of operating over an input voltage range of 3.0 to 13.2V for aerospace and satellite power applications. The ISL70003SEH synchro-nous buck switching regulator is a highly integrated single-chip power management solution that increases power system efficiency. The ISL70003SEH improves thermal performance while reducing the number of required external components to streamline design and qualification time. Leveraging proprietary design, process and testing techniques that Intersil has refined over many decades...Read More

FRAM Chip for Memory CardMB89R111 was developed for applications in which low power and fast, durable read/write capability are primary requirements. MB89R111 includes 2 Kbytes, 128-block FRAM, which consists of eight pages. The FRAM memory map is divided into a 32-byte system area and a user area. Manufacturer-assigned device serial number and customer-specified user ID and application ID can be programmed before shipping. For maximum security, these IDs cannot be rewritten irrespective of assignment of the Write Access Control Area. The de-vice incorporates a sophisticated data-protection scheme to prevent a writing error to FRAM...Read More

Cascadable InGaP MMIC AmplifierThe MHA-054020-89 is a broadband MMIC amplifier utilizing high reliability InGaP/GaAs HBT technology. Packages in low cost SOT-89 lead-free Green Package, the MMIC is ideally suited for driver amplifier or gain block in wireless applications such as Cellular, PCS, GSM and UMTS base stations as well as CATV, wireless IPTV, microwave radio, instrumentation, homeland security systems etc. It has excellent, typcially 19 dB above P-1 dB below 500 MHz and 18 dB above P-1dB at 900 MHz. It has on-chip bias circuit to provide bias stability and ease of use...Read More

Logic Controlled High-Side Power SwitchThe NX5P2924 is a high-side load switch which features a low ON resistance N-channel MOSFET with controlled slew rate that supports 2.5 A of continu-ous current. Designed for operation from 0.8 V to 5.5 V, it is used in power domain isolation applications to reduce power dissipation and extend bat-tery life. The enable logic includes integrated logic level translation making the device compatible with lower voltage processors and controllers. The NX5P2924 is ideal for portable, battery operated applications due to low ground current...Read More

PULSE

Page 9: EEWeb Pulse 122 - Crossbar

9Visit: eeweb.com

FEATURED PRODUCTS

FM+ I2C-Bus Stepper Motor ControllerThe PCA9629A is an I²C-bus controlled low-power CMOS device that provides all the logic and control required to drive a four phase step-per motor. PCA9629A is intended to be used with external high current drivers to drive the motor coils. The PCA9629A supports three stepper motor drive formats: one-phase (wave drive), two-phase, and half-step. In addition, when used as inputs, four General Purpose Input/Out-puts (GPIOs) allow sensing of logic level output from optical interrupter modules and generate active LOW interrupt signal on the INT pin of PCA9629A...Read More

World-Class 1-Wire SHA-1 AuthenticatorThe DS28E10 combines secure challenge-and-response authentication functionality based on the FIPS 180-3 specified Secure Hash Algorithm (SHA-1) with 224 bits of one-time programmable user EPROM in a single chip. Once written, the memory is automatically write protected. Addi-tionally, each device has its own guaranteed unique 64-bit ROM iden-tification number (ROM ID) that is factory programmed into the chip. Memory writes are performed 4 bytes at a time. A secure and low-cost factory programming service is available to preprogram device data, including the SHA-1 security data components. The device communi-cates over the single-contact 1-Wire® bus...Read More

Ultra-precision Absolute EncoderThe AEAT-9000 series are high resolution single turn optical absolute encoders. The 17-bit AEAT-9000 encoder code disc consists of 13 pairs of differential absolute tracks and 2 pairs of sinusoidal tracks to per-form 4 bits interpolation. In addition, the encoder incorporates photo detectors for electrical alignment on the radial and tilt. AEAT-9000 also comes with 2 channel incremental output with the basic of 2048 counts per rotation. The AEAT-9000 is a modular absolute encoder that consists of a read head module and a high-precision code disc. The modular design allows for better flexibility to system designers to easily design-in the encoder feedback system...Read More

100 MHz, 32 bit RX63T MCUsThe RX63T Group has an expanded pin package lineup and memory lineup, features enhanced safety functions over the RX62T Group, and is also equipped with a max. 312.5 psec high-resolution PWM output function and a digital power management controller (DPC). The com-bination of this high-resolution PWM output and the high CPU perfor-mance of the RX core allows more precise inverter/converter control, enabling highly efficient power management for digital power control applications such as server power supplies and general-purpose power supply systems...Read More

Page 10: EEWeb Pulse 122 - Crossbar

10 Visit: eeweb.com

PULSE

BACK TO BASICS:JITTER

Anyone working in applications that involve digital data, clocks, and serial data in general will eventually bump up against issues concerning jitter. Jitter is a subject of keen interest to every strata of the electronics industry. Chip makers, board integrators, system integrators, you name it: Everybody wants, and needs, to come to terms with jitter. It impacts reliability, manufacturability, and cost at all levels. And, of course, it’s of keen interest to purveyors of test instruments, including us here at Teledyne LeCroy. In this first post of a projected series on jitter, we’ll look at some of the tools built into modern digital oscilloscopes for jitter measurement and analysis.

Figure 1: Jitter is short-term variationof a signal with respect to itsideal position in time

Figure 2: Persistence display can serve for a quick eyeball estimate of total jitter in a signal

Figure 3: Time Interval Error (TIE) is the difference between the measured clock edge and the ideal clock edge locations

David MaliniakTechnical Marketing Communication SpecialistTeledyne LeCroy

What’s all this “jitter” stuff, anyhow? A broad definition is “the short-term variations of a signal with respect to its ideal position in time.” Jitter can manifest as instability in signal period, frequency, phase, duty cycle, or other timing characteristics. It’s of interest from pulse to pulse, over many consecutive pulses, or as a longer-term variation.

There are a number of measurement and analysis techniques one can turn to for quantifying and estimating jitter. Among them are:

Histograms

Measurement track

Time Interval Error (TIE)

Jitter breakdown and extrapolation

Persistence display of jitter

A traditional means of investigating a signal for jitter is to use the oscilloscope’s persistence display mode (Figure 2). It’s relatively easy to set up and get a quick eyeball estimate of total jitter. However, it’s not a method you’d use for serious diagnostics. The results are skewed by any trigger jitter that is present, and it’s also subject to human interpretation.

Some of the persistence analysis tools one finds in some Teledyne LeCroy oscilloscopes include:

Ptrace mean: Generates a mean waveform from a persistence map

Phistogram: Provides a histogram of a vertical or horizontal slice through a persistence map

Ptrace range: Generates a waveform with a vertical width derived from a population range (estimate of confidence interval) of a persistence map

Ptrace sigma: Generates a waveform with a width derived from the sigma of a persistence map.

A broad range of clock and timing measurement parameters are available on most digital oscilloscopes. These include items such as:

Hold time: Time from the clock edge to the data edge

Setup time: Time from the data edge to the clock edge

Skew: Time of Clock1 edge minus time of nearest Clock2 edge

Dperiod@level: Adjacent cycle deviation (cycle-to-cycle jitter) of each cycle in a waveform

Dwidth@level: Difference of adjacent width above or below a specified level

Duty@level: Percent of period for which data are above or below a specified level

Edge@level: Number of edges in a waveform

Freq@level: Frequency at a specific level and slope for every cycle in a waveform

Period@level: Period at a specific level and slope for every cycle in a waveform

TIE@level: Difference between the measured times of crossing a given slope and level and the ideal expected time

Width@level: Width measured at a specific level

Half period: Half period of a waveform

dVdt: Local slope or “slew” rate

Bit rate: Bit rate of a serial data streamTime Interval Error

To look at one of these parameters a little closer, TIE@level, or Time Interval Error, is the difference between the measured clock edge and the ideal clock edge locations (Figure 3). Basically, TIE is the instantaneous phase of the signal. References can be chosen from a standard telecom frequency or set to a user-defined value. TIE can be plotted versus time or unit intervals (UIs), the latter being the period of time corresponding to one cycle of a given frequency.

A valuable tool is tracking of measurements, such as period, frequency, width, or duty cycle, to get a handle on how these measurements might be changing over time. Further, there is the ability to take a statistical view of jitter, which affords more reliable measurements thanks to the acquisition of more data.

Page 11: EEWeb Pulse 122 - Crossbar

11Visit: eeweb.com

TECH ARTICLE

BACK TO BASICS:JITTER

Anyone working in applications that involve digital data, clocks, and serial data in general will eventually bump up against issues concerning jitter. Jitter is a subject of keen interest to every strata of the electronics industry. Chip makers, board integrators, system integrators, you name it: Everybody wants, and needs, to come to terms with jitter. It impacts reliability, manufacturability, and cost at all levels. And, of course, it’s of keen interest to purveyors of test instruments, including us here at Teledyne LeCroy. In this first post of a projected series on jitter, we’ll look at some of the tools built into modern digital oscilloscopes for jitter measurement and analysis.

Figure 1: Jitter is short-term variationof a signal with respect to itsideal position in time

Figure 2: Persistence display can serve for a quick eyeball estimate of total jitter in a signal

Figure 3: Time Interval Error (TIE) is the difference between the measured clock edge and the ideal clock edge locations

David MaliniakTechnical Marketing Communication SpecialistTeledyne LeCroy

What’s all this “jitter” stuff, anyhow? A broad definition is “the short-term variations of a signal with respect to its ideal position in time.” Jitter can manifest as instability in signal period, frequency, phase, duty cycle, or other timing characteristics. It’s of interest from pulse to pulse, over many consecutive pulses, or as a longer-term variation.

There are a number of measurement and analysis techniques one can turn to for quantifying and estimating jitter. Among them are:

Histograms

Measurement track

Time Interval Error (TIE)

Jitter breakdown and extrapolation

Persistence display of jitter

A traditional means of investigating a signal for jitter is to use the oscilloscope’s persistence display mode (Figure 2). It’s relatively easy to set up and get a quick eyeball estimate of total jitter. However, it’s not a method you’d use for serious diagnostics. The results are skewed by any trigger jitter that is present, and it’s also subject to human interpretation.

Some of the persistence analysis tools one finds in some Teledyne LeCroy oscilloscopes include:

Ptrace mean: Generates a mean waveform from a persistence map

Phistogram: Provides a histogram of a vertical or horizontal slice through a persistence map

Ptrace range: Generates a waveform with a vertical width derived from a population range (estimate of confidence interval) of a persistence map

Ptrace sigma: Generates a waveform with a width derived from the sigma of a persistence map.

A broad range of clock and timing measurement parameters are available on most digital oscilloscopes. These include items such as:

Hold time: Time from the clock edge to the data edge

Setup time: Time from the data edge to the clock edge

Skew: Time of Clock1 edge minus time of nearest Clock2 edge

Dperiod@level: Adjacent cycle deviation (cycle-to-cycle jitter) of each cycle in a waveform

Dwidth@level: Difference of adjacent width above or below a specified level

Duty@level: Percent of period for which data are above or below a specified level

Edge@level: Number of edges in a waveform

Freq@level: Frequency at a specific level and slope for every cycle in a waveform

Period@level: Period at a specific level and slope for every cycle in a waveform

TIE@level: Difference between the measured times of crossing a given slope and level and the ideal expected time

Width@level: Width measured at a specific level

Half period: Half period of a waveform

dVdt: Local slope or “slew” rate

Bit rate: Bit rate of a serial data streamTime Interval Error

To look at one of these parameters a little closer, TIE@level, or Time Interval Error, is the difference between the measured clock edge and the ideal clock edge locations (Figure 3). Basically, TIE is the instantaneous phase of the signal. References can be chosen from a standard telecom frequency or set to a user-defined value. TIE can be plotted versus time or unit intervals (UIs), the latter being the period of time corresponding to one cycle of a given frequency.

A valuable tool is tracking of measurements, such as period, frequency, width, or duty cycle, to get a handle on how these measurements might be changing over time. Further, there is the ability to take a statistical view of jitter, which affords more reliable measurements thanks to the acquisition of more data.

Page 12: EEWeb Pulse 122 - Crossbar

12 Visit: eeweb.com

PULSE

DELIVERS POWER TO AMAZE

FRESH. FOCUSED. FAIRCHILD.

Founded in 1957 by eight Shockley Semiconductor defectors, Fairchild Semiconductor pioneered the chip manufacturing industry and established the successful Silicon Valley start-up model still thriving today. Fueled by a shared goal, intellectual deftness, and an unrelenting spirit, the Shockley “treacherous eight” quickly grew Fairchild into a semiconductor industry leader, and by 1960, it was an incubator for Silicon Valley startups, such as AMD and Intel. Today, after a storied path, Fairchild 2.0 has emerged. By combining Fairchild’s rich history of successes with the spirit that ignited the company over 50 years ago, Fairchild 2.0 promises to drive innovation well into the future.

FAIRCHILD 2.0

Page 13: EEWeb Pulse 122 - Crossbar

13Visit: eeweb.com

TECH ARTICLE

DELIVERS POWER TO AMAZE

FRESH. FOCUSED. FAIRCHILD.

Founded in 1957 by eight Shockley Semiconductor defectors, Fairchild Semiconductor pioneered the chip manufacturing industry and established the successful Silicon Valley start-up model still thriving today. Fueled by a shared goal, intellectual deftness, and an unrelenting spirit, the Shockley “treacherous eight” quickly grew Fairchild into a semiconductor industry leader, and by 1960, it was an incubator for Silicon Valley startups, such as AMD and Intel. Today, after a storied path, Fairchild 2.0 has emerged. By combining Fairchild’s rich history of successes with the spirit that ignited the company over 50 years ago, Fairchild 2.0 promises to drive innovation well into the future.

FAIRCHILD 2.0

Page 14: EEWeb Pulse 122 - Crossbar

14 Visit: eeweb.com

PULSE

For Fairchild, it has always been about the customer. The company’s commitment to its customers’ continued success has long served as the impetus behind the design of its technologies. The emphasis on quality improves the efficiency of home appliances and industrial products, and enables manufacturers of mobile devices to deliver innovative new features. But in the 50 years Fairchild has been serving its customers, the challenges facing engineers have evolved significantly. As Sajal Sahay, Vice President of Worldwide Corporate Marketing at Fairchild Semiconductor told EEWeb, “The two key things I keep hearing from our customers is that there is too much new information coming to them all the time, and that time to market is getting smaller and smaller. Engineers are always stressed to know more than what they do, mixed with the stress to get things done in record time.”

Fairchild is well acquainted with the challenges facing its customers and recognizes that upholding an advantage in a highly competitive industry comes down to finding ways to help their

customers work faster, smarter, and cheaper. The launch of Fairchild 2.0 is testimony of their efforts and of Fairchild’s long term commitment to its customers. “Our aspiration is to be transformative going forward,” Sahay said. “We’re focusing on individual products, expertise in design, expertise in pain-point resolution, and supply chain efficiency. As a company, we are centered on one simple goal: to make us the best company to work with as a design engineer in the power management field.”

The tenets underlying Fairchild 2.0 hearken back to the company’s roots. The founders derided the prevalence of hierarchically structured companies common in the ‘50s and instead sought to create a company that took risks, fostered creativity, and rewarded contributions. The egalitarian culture pioneered by Fairchild—and now ubiquitous throughout Silicon Valley—is fundamental to Fairchild’s renewal. Fairchild 2.0, as Sahay expressed, “Is not only about financial gain. Financial gain is a byproduct of doing what’s right and making the world better. We’re bringing back what made the company great.”

With Fairchild 2.0, Fairchild has refocused the company’s attention on the customer and the accompanying launch of their new tagline that aptly characterizes their promise to truly deliver the Power to Amaze. As Sahay told EEWeb, “’Power’ refers to our focus on power management, empowering our people, and having the power to take care of our customers. It’s about giving laser focus on the area the company excels in and it also galvanizes our entire organization to put the customer first. ‘Amaze’ stands for going above and beyond and making customers very delighted in the service they get. We’re not going to be happy with simply supplying products.”

Accompanying the reincarnation of Fairchild is a new logo. The debut of their new logo differentiates Fairchild 2.0 from its predecessor as a semiconductor-centric company and reinforces their dedication to supplying their customers with amazing products, design expertise, logistics, and service. The old Fairchild logo had the two bars on top with the word “Semiconductor” below—the two bars were meant to be the first wafer. “What we’ve done with the new logo is made it cleaner,” Sahay explained. “We’ve taken the word ‘Semiconductor’ out because we are much more than a semiconductor component seller, we’re about providing comprehensive design solutions to our customers. We’ve also made the two red bars much wider and leaning forward. The intent here is to build from the past and actually drive towards the future.”

FAIRCHILD 2.0

THE POWER TO AMAZE

“As a company, we are centered on one

simple goal: to make us the best company to

work with as a design engineer in the power management field.”

“Financial gain is a byproduct of doing

what’s right and making the world

better. We’re bringing back what made the

company great.”

Page 15: EEWeb Pulse 122 - Crossbar

15Visit: eeweb.com

TECH ARTICLE

For Fairchild, it has always been about the customer. The company’s commitment to its customers’ continued success has long served as the impetus behind the design of its technologies. The emphasis on quality improves the efficiency of home appliances and industrial products, and enables manufacturers of mobile devices to deliver innovative new features. But in the 50 years Fairchild has been serving its customers, the challenges facing engineers have evolved significantly. As Sajal Sahay, Vice President of Worldwide Corporate Marketing at Fairchild Semiconductor told EEWeb, “The two key things I keep hearing from our customers is that there is too much new information coming to them all the time, and that time to market is getting smaller and smaller. Engineers are always stressed to know more than what they do, mixed with the stress to get things done in record time.”

Fairchild is well acquainted with the challenges facing its customers and recognizes that upholding an advantage in a highly competitive industry comes down to finding ways to help their

customers work faster, smarter, and cheaper. The launch of Fairchild 2.0 is testimony of their efforts and of Fairchild’s long term commitment to its customers. “Our aspiration is to be transformative going forward,” Sahay said. “We’re focusing on individual products, expertise in design, expertise in pain-point resolution, and supply chain efficiency. As a company, we are centered on one simple goal: to make us the best company to work with as a design engineer in the power management field.”

The tenets underlying Fairchild 2.0 hearken back to the company’s roots. The founders derided the prevalence of hierarchically structured companies common in the ‘50s and instead sought to create a company that took risks, fostered creativity, and rewarded contributions. The egalitarian culture pioneered by Fairchild—and now ubiquitous throughout Silicon Valley—is fundamental to Fairchild’s renewal. Fairchild 2.0, as Sahay expressed, “Is not only about financial gain. Financial gain is a byproduct of doing what’s right and making the world better. We’re bringing back what made the company great.”

With Fairchild 2.0, Fairchild has refocused the company’s attention on the customer and the accompanying launch of their new tagline that aptly characterizes their promise to truly deliver the Power to Amaze. As Sahay told EEWeb, “’Power’ refers to our focus on power management, empowering our people, and having the power to take care of our customers. It’s about giving laser focus on the area the company excels in and it also galvanizes our entire organization to put the customer first. ‘Amaze’ stands for going above and beyond and making customers very delighted in the service they get. We’re not going to be happy with simply supplying products.”

Accompanying the reincarnation of Fairchild is a new logo. The debut of their new logo differentiates Fairchild 2.0 from its predecessor as a semiconductor-centric company and reinforces their dedication to supplying their customers with amazing products, design expertise, logistics, and service. The old Fairchild logo had the two bars on top with the word “Semiconductor” below—the two bars were meant to be the first wafer. “What we’ve done with the new logo is made it cleaner,” Sahay explained. “We’ve taken the word ‘Semiconductor’ out because we are much more than a semiconductor component seller, we’re about providing comprehensive design solutions to our customers. We’ve also made the two red bars much wider and leaning forward. The intent here is to build from the past and actually drive towards the future.”

FAIRCHILD 2.0

THE POWER TO AMAZE

“As a company, we are centered on one

simple goal: to make us the best company to

work with as a design engineer in the power management field.”

“Financial gain is a byproduct of doing

what’s right and making the world

better. We’re bringing back what made the

company great.”

Page 16: EEWeb Pulse 122 - Crossbar

eeweb.com/register16 Visit: eeweb.com

PULSE

The power semiconductor industry has been growing two to three percent annually and is projected to explode into a 40 billion dollar business by 2016. Driving this exponential growth is the global demand for products that increasingly require more energy. Drawing upon its deep portfolio of products and legacy of expertise, Fairchild 2.0 is poised to meet the needs of its customers in terms of products and support. “We have expertise in multiple key markets,” Sahay explained. “We have expertise in mobile, cloud storage, industrial, and automotive. Our key promise to the customer is to offer them a breadth of power products from super low voltages to super high voltages.”

However, supporting customers goes beyond just selling innovative products. To complement their vast product line, Faichild 2.0 will also offer designers ample support such as power design centers, product seminars, and online design and educational tools that will enable customers to solve challenges and speed up their time to market. Additionally, customers can be assured supply chain excellence. “Our supply lead time has decreased significantly from where it was twelve months ago,” Sahay said. “And our on-time delivery is best in class.”

As a global company, Fairchild 2.0 is well suited to meet the power demands of the future. With 27 sales offices, five worldwide distributers, five global distribution centers, and 100 shipping lanes, combined with global power resource centers, Fairchild 2.0 customers can expect to be amazed.

MEETING THE POWER DEMANDS OF THE FUTURE

“As a global company, Fairchild 2.0 is well

suited to meet the power demands of the future.”

Page 17: EEWeb Pulse 122 - Crossbar

eeweb.com/register

Join Today

Transform iPad, iPhone, & iPod Into An Oscilloscope

o s c i u m . co m

• Ultra Portable

• Two Analog Channels

• 4x Faster Than iMSO-104

Make TheWorld Your Lab

The power semiconductor industry has been growing two to three percent annually and is projected to explode into a 40 billion dollar business by 2016. Driving this exponential growth is the global demand for products that increasingly require more energy. Drawing upon its deep portfolio of products and legacy of expertise, Fairchild 2.0 is poised to meet the needs of its customers in terms of products and support. “We have expertise in multiple key markets,” Sahay explained. “We have expertise in mobile, cloud storage, industrial, and automotive. Our key promise to the customer is to offer them a breadth of power products from super low voltages to super high voltages.”

However, supporting customers goes beyond just selling innovative products. To complement their vast product line, Faichild 2.0 will also offer designers ample support such as power design centers, product seminars, and online design and educational tools that will enable customers to solve challenges and speed up their time to market. Additionally, customers can be assured supply chain excellence. “Our supply lead time has decreased significantly from where it was twelve months ago,” Sahay said. “And our on-time delivery is best in class.”

As a global company, Fairchild 2.0 is well suited to meet the power demands of the future. With 27 sales offices, five worldwide distributers, five global distribution centers, and 100 shipping lanes, combined with global power resource centers, Fairchild 2.0 customers can expect to be amazed.

MEETING THE POWER DEMANDS OF THE FUTURE

“As a global company, Fairchild 2.0 is well

suited to meet the power demands of the future.”

Page 18: EEWeb Pulse 122 - Crossbar

www.partsim.com

PartSim includes a full SPICE simulation engine, web-based schematic capture tool, and a graphical waveform viewer.

Some Features include:• Simulate in a standard Web Browser• AC/DC and Transient Simulations• Schematic Editor• WaveForm Viewer• WaveForm Viewer• Easily Share Simulations

Online Circuit Simulator

Try-it Now!

Page 19: EEWeb Pulse 122 - Crossbar

www.partsim.com

PartSim includes a full SPICE simulation engine, web-based schematic capture tool, and a graphical waveform viewer.

Some Features include:• Simulate in a standard Web Browser• AC/DC and Transient Simulations• Schematic Editor• WaveForm Viewer• WaveForm Viewer• Easily Share Simulations

Online Circuit Simulator

Try-it Now!

Page 20: EEWeb Pulse 122 - Crossbar

20 Visit: eeweb.com

PULSE

Page 21: EEWeb Pulse 122 - Crossbar

21Visit: eeweb.com

TECH ARTICLE

Page 22: EEWeb Pulse 122 - Crossbar

22 Visit: eeweb.com

PULSE

INTRODUCING THE TIVA C SERIES CONNECTED LAUNCHPAD

The IoT, combined with the cloud, has the potential to dramatically magnify the efficiency of appliances and embedded systems worldwide. But up until now, creating systems and developing applications that can communicate with one another has proven to be a non-trivial task. The recent announcement of TI’s Tiva C Series Connected LaunchPad has eliminated many of the obstacles that have stifled the development of IoT applications and offers everyone from the hobbyist to the professional engineer an easy (and low-cost) means to start developing applications that would have been inconceivable just a few years ago. EEWeb spoke with Joe Folkens, Product Marketing Engineer, Tiva C Series MCUs at TI, about the benefits of their newest LaunchPad device. “Based on the Tiva C Series TM4C1294NCPDT MCU, the Tiva C Series Connected LaunchPad will enable the IoT designs to get the desired results.” Engineers are still trying to determine what the IoT will actually be, meaning that a certain amount of flexibility is needed when developing these devices.

TI’s latest LaunchPad is the first in TI’s MCU LaunchPad and BoosterPack ecosystem to connect the user to the cloud right out of the box. Affordable and easy to use, the Tiva C Series Connected LaunchPad offers multi-type connectivity to all the interfaces on the board and allows embedded control to be extended through the cloud. “The kit comes with access to a cloud-based scale of technology from a third party partner named Exosite,” Folkens explained. “This allows the user to connect to and remotely interact with the Connected LaunchPad applications through embedded software operating on the board. It actually allows the user to essentially interact with the website or browser and other applications that operate on the Connected LaunchPad.”

THE DETAILS

The Tiva C Series LaunchPad includes a TM4C1294NCPDTI MCU with a 120 MHz ARM Cortex-M4F CPU, 1 MB Flash 256KB SRAM, and 6KB EEPROM. When it comes to connectivity, this Tiva C MCU integrates 10/100 Ethernet (both MAC and PHY), 8 x 32-bit (16 x 16-bit) timers, 10 x I2C, 8 x UART, 4 x QSPI, 2 x CAN, EPI, USB, and 2 x 12-bit ADCs. An IO connection grid provides access to the MCUs IO pins and the ability to interface to external circuitry, breadboards, and custom baseboards. Meanwhile, the RJ45 Ethernet jack, USB Host/Device Port, combined with user buttons and LEDs, make the Tiva C Series Connected LaunchPad an easy to use, versatile development platform.

The feature-rich Tiva C Series LaunchPad makes it an ideal choice for the development of IoT applications. As Folkens told EEWeb, “The IoT application comes already pre-programmed on the board, which gives access to scalable, cloud-based solutions from Exosite. For a free software package, we have TivaWare, which is a very extensive, mature software base that has been developed over many years giving a strong foundation so users can focus on building their

value-add rather than spending the time getting the basic functions to work.”

The Tiva C Series Connected LaunchPad comes with an integrated USB-powered, in-circuit, debug interface (ICDI) supported by the prior generation TM4C123 microcontroller also included on the board. BoosterPack plug-in modules that fit on top or bottom of the LauchPad using a standardized connector allow developers to expand the capabilities of the LaunchPad to explore different applications. The BoosterPacks are available from TI, third parties, as well as the community, and include functions such as capacitive touch, wireless communication, LCD panels, sensor readings, LED lighting control and more. The connectors that TI is using are stackable and dual-gender, which means that the board has a back side for the female version of the header so you can plug-on to something else as well. “Theoretically, you can plug-in four different booster packs into the two headers,” Folkens explained, “but you must also consider that there is shared GPIO or functionality so you may have to make some minor modifications in your firmware. Nevertheless, it is there for you to use,” Folkens said.

“Affordable and easy to use, the Tiva C Series Connected LaunchPad offers multi-type connectivity to all the interfaces on the board and allows embedded control to be extended through the cloud.”

Page 23: EEWeb Pulse 122 - Crossbar

23Visit: eeweb.com

TECH ARTICLE

INTRODUCING THE TIVA C SERIES CONNECTED LAUNCHPAD

The IoT, combined with the cloud, has the potential to dramatically magnify the efficiency of appliances and embedded systems worldwide. But up until now, creating systems and developing applications that can communicate with one another has proven to be a non-trivial task. The recent announcement of TI’s Tiva C Series Connected LaunchPad has eliminated many of the obstacles that have stifled the development of IoT applications and offers everyone from the hobbyist to the professional engineer an easy (and low-cost) means to start developing applications that would have been inconceivable just a few years ago. EEWeb spoke with Joe Folkens, Product Marketing Engineer, Tiva C Series MCUs at TI, about the benefits of their newest LaunchPad device. “Based on the Tiva C Series TM4C1294NCPDT MCU, the Tiva C Series Connected LaunchPad will enable the IoT designs to get the desired results.” Engineers are still trying to determine what the IoT will actually be, meaning that a certain amount of flexibility is needed when developing these devices.

TI’s latest LaunchPad is the first in TI’s MCU LaunchPad and BoosterPack ecosystem to connect the user to the cloud right out of the box. Affordable and easy to use, the Tiva C Series Connected LaunchPad offers multi-type connectivity to all the interfaces on the board and allows embedded control to be extended through the cloud. “The kit comes with access to a cloud-based scale of technology from a third party partner named Exosite,” Folkens explained. “This allows the user to connect to and remotely interact with the Connected LaunchPad applications through embedded software operating on the board. It actually allows the user to essentially interact with the website or browser and other applications that operate on the Connected LaunchPad.”

THE DETAILS

The Tiva C Series LaunchPad includes a TM4C1294NCPDTI MCU with a 120 MHz ARM Cortex-M4F CPU, 1 MB Flash 256KB SRAM, and 6KB EEPROM. When it comes to connectivity, this Tiva C MCU integrates 10/100 Ethernet (both MAC and PHY), 8 x 32-bit (16 x 16-bit) timers, 10 x I2C, 8 x UART, 4 x QSPI, 2 x CAN, EPI, USB, and 2 x 12-bit ADCs. An IO connection grid provides access to the MCUs IO pins and the ability to interface to external circuitry, breadboards, and custom baseboards. Meanwhile, the RJ45 Ethernet jack, USB Host/Device Port, combined with user buttons and LEDs, make the Tiva C Series Connected LaunchPad an easy to use, versatile development platform.

The feature-rich Tiva C Series LaunchPad makes it an ideal choice for the development of IoT applications. As Folkens told EEWeb, “The IoT application comes already pre-programmed on the board, which gives access to scalable, cloud-based solutions from Exosite. For a free software package, we have TivaWare, which is a very extensive, mature software base that has been developed over many years giving a strong foundation so users can focus on building their

value-add rather than spending the time getting the basic functions to work.”

The Tiva C Series Connected LaunchPad comes with an integrated USB-powered, in-circuit, debug interface (ICDI) supported by the prior generation TM4C123 microcontroller also included on the board. BoosterPack plug-in modules that fit on top or bottom of the LauchPad using a standardized connector allow developers to expand the capabilities of the LaunchPad to explore different applications. The BoosterPacks are available from TI, third parties, as well as the community, and include functions such as capacitive touch, wireless communication, LCD panels, sensor readings, LED lighting control and more. The connectors that TI is using are stackable and dual-gender, which means that the board has a back side for the female version of the header so you can plug-on to something else as well. “Theoretically, you can plug-in four different booster packs into the two headers,” Folkens explained, “but you must also consider that there is shared GPIO or functionality so you may have to make some minor modifications in your firmware. Nevertheless, it is there for you to use,” Folkens said.

“Affordable and easy to use, the Tiva C Series Connected LaunchPad offers multi-type connectivity to all the interfaces on the board and allows embedded control to be extended through the cloud.”

Page 24: EEWeb Pulse 122 - Crossbar

eeweb.com/register24 Visit: eeweb.com

PULSE

READY TO LAUNCH

For the launch of the Tiva C Series Connected LaunchPad, TI has partnered with Exosite, mentioned briefly above, to provide easy access to the LaunchPad from the Internet. The LaunchPad takes about 10 minutes to set up and you can immediately interact with it across the Internet and do things like turn an LED on and off remotely from the website and see the reported temperature as well. It can also display approximate geographic location based on the assigned IP address and display a map of all other connected LaunchPad owners if they are active and plugged-in to Exosite. “In addition, it supports a basic game by enabling someone to interface to the Connected LaunchPad through a serial port from a terminal while someone else is playing with them through their browser. It is basically showing how you can interact remotely with this product and a user even if you are across the globe,” Folkens explained.

START DEVELOPING

The Tiva C Series Connected LaunchPad is shipping now and the price is right; at $19.99 USD, it is less than half the price of other Ethernet-ready kits. The LaunchPad comes complete with quick start and user guides, and ample online support to ensure developers of all backgrounds are well equipped to begin creating cloud-based applications. “We have assembled an online support team to monitor the Engineering-to-Engineering (or E2E) Community,” Folkens said. “Along with this, you also got a free Code Composer Studio Integrated Development Environment, which allows developers to use the full capability. We also support other tool chains like Keil, IAR and Mentor Embedded.

Affordable, versatile, and easy to use, the Tiva Series Connected LaunchPad is well suited for a broad audience and promises to facilitate the expansion of ingenious IoT applications in the cloud. As Folkens concluded, “The target audiences actually are the hobbyists, students and professional engineers. A better way of looking at it is that we are targeting people with innovative ideas and trying to help them get those ideas launched into the cloud.”

Page 25: EEWeb Pulse 122 - Crossbar

eeweb.com/register

Join Today

READY TO LAUNCH

For the launch of the Tiva C Series Connected LaunchPad, TI has partnered with Exosite, mentioned briefly above, to provide easy access to the LaunchPad from the Internet. The LaunchPad takes about 10 minutes to set up and you can immediately interact with it across the Internet and do things like turn an LED on and off remotely from the website and see the reported temperature as well. It can also display approximate geographic location based on the assigned IP address and display a map of all other connected LaunchPad owners if they are active and plugged-in to Exosite. “In addition, it supports a basic game by enabling someone to interface to the Connected LaunchPad through a serial port from a terminal while someone else is playing with them through their browser. It is basically showing how you can interact remotely with this product and a user even if you are across the globe,” Folkens explained.

START DEVELOPING

The Tiva C Series Connected LaunchPad is shipping now and the price is right; at $19.99 USD, it is less than half the price of other Ethernet-ready kits. The LaunchPad comes complete with quick start and user guides, and ample online support to ensure developers of all backgrounds are well equipped to begin creating cloud-based applications. “We have assembled an online support team to monitor the Engineering-to-Engineering (or E2E) Community,” Folkens said. “Along with this, you also got a free Code Composer Studio Integrated Development Environment, which allows developers to use the full capability. We also support other tool chains like Keil, IAR and Mentor Embedded.

Affordable, versatile, and easy to use, the Tiva Series Connected LaunchPad is well suited for a broad audience and promises to facilitate the expansion of ingenious IoT applications in the cloud. As Folkens concluded, “The target audiences actually are the hobbyists, students and professional engineers. A better way of looking at it is that we are targeting people with innovative ideas and trying to help them get those ideas launched into the cloud.”

Page 28: EEWeb Pulse 122 - Crossbar

28 Visit: eeweb.com

PULSE

How Product Innovation Happens:

Engineers who use oscilloscopes are typically focused on solving problems and looking for ways to make better products. It turns out that engineers who design oscilloscopes are essentially doing the same thing.

More than a decade ago, an Agilent Technologies oscilloscope designer named Scott Genther was getting ready to launch a massive, multiyear project to design a new chip for an advanced algorithmic oscilloscope trigger—affectionately known in the Agilent R&D lab as the “nerd trigger.” Before the project even got off the ground, he came up with a remarkably simple idea for a trigger that ultimately redefined how engineers use Agilent oscilloscopes.

Where do innovative ideas like Genther’s come from? And how do you get from the original spark of inspiration to its implementation in an innovative product? Genther’s story sheds light on how product innovations happen.

The Origin of Agilent’s Oscilloscope Trigger

CONNECTING WITH CUSTOMER FRUSTRATIONS

In a quest to define the features he needed to include in his trigger ASIC, Genther visited oscilloscope customers to understand how they used triggering and what stumbling blocks they faced. “In the lab here at Agilent, R&D engineers know how to use all the triggers on their scopes,” said Genther. “But I quickly discovered our customers don’t. They don’t like thinking about triggers. A lot of customers just pick a mode and turn the knob to clean up the display. They don’t have time to sit down and think about it.”

The first customer Genther visited was having trouble defining a trigger to capture the clock edge reads he wanted to see. In frustration, he jabbed a finger at the screen and wondered why he couldn’t just click on a piece of a waveform and get what he wanted. At the time, Genther just assumed it was impossible.

The second customer Genther visited had come back from lunch and discovered a bad trace on his oscilloscope screen. The scope’s infinite persistence showed him something had happened when he was away. To figure out what caused it, he wanted to get a view of the trace by itself so he could correlate it to activity on other channels.

The customer had been doing single-shot acquisitions for days trying to locate the infrequently occurring glitch. He pushed the “single” button over and over and still couldn’t locate the problem. Watching the engineer laboriously pecking away at the button, Genther experienced his big “aha” moment. He could see that pressing the “single” button a million times is a painful way to find a one-in-a-million glitch, and he realized he could make both customers’ jobs easier using Agilent’s existing oscilloscope technology.

“The nerd trigger was going to take forever to complete, but I figured we could create very similar functionality using the scope’s mask test infrastructure,” Genther said. “We could qualify acquisitions with mask testing and use that feature to push the button for him.”

Mask testing shows all the waveforms that fail or exceed the limits of the mask. Genther’s brainstorm was to ignore all the waveforms that fail and just display the ones that do meet the mask limits. This process could happen behind the scenes, using digital technology to accomplish what the customer was doing manually.

CHAMPIONING THE IDEA

Coming up with the idea to use a reverse mask test to achieve an easier version of the trigger was just the first step in the process of bringing it to market. Next, Genther had to convince his boss and teammates the project was worth pursuing.

Figure 1: Example of a hard-to-capture signal anomaly with infinite persistence turned on

Scott GentherKey inventor of Agilent Zone Trigger

Page 29: EEWeb Pulse 122 - Crossbar

29Visit: eeweb.com

TECH ARTICLE

How Product Innovation Happens:

Engineers who use oscilloscopes are typically focused on solving problems and looking for ways to make better products. It turns out that engineers who design oscilloscopes are essentially doing the same thing.

More than a decade ago, an Agilent Technologies oscilloscope designer named Scott Genther was getting ready to launch a massive, multiyear project to design a new chip for an advanced algorithmic oscilloscope trigger—affectionately known in the Agilent R&D lab as the “nerd trigger.” Before the project even got off the ground, he came up with a remarkably simple idea for a trigger that ultimately redefined how engineers use Agilent oscilloscopes.

Where do innovative ideas like Genther’s come from? And how do you get from the original spark of inspiration to its implementation in an innovative product? Genther’s story sheds light on how product innovations happen.

The Origin of Agilent’s Oscilloscope Trigger

CONNECTING WITH CUSTOMER FRUSTRATIONS

In a quest to define the features he needed to include in his trigger ASIC, Genther visited oscilloscope customers to understand how they used triggering and what stumbling blocks they faced. “In the lab here at Agilent, R&D engineers know how to use all the triggers on their scopes,” said Genther. “But I quickly discovered our customers don’t. They don’t like thinking about triggers. A lot of customers just pick a mode and turn the knob to clean up the display. They don’t have time to sit down and think about it.”

The first customer Genther visited was having trouble defining a trigger to capture the clock edge reads he wanted to see. In frustration, he jabbed a finger at the screen and wondered why he couldn’t just click on a piece of a waveform and get what he wanted. At the time, Genther just assumed it was impossible.

The second customer Genther visited had come back from lunch and discovered a bad trace on his oscilloscope screen. The scope’s infinite persistence showed him something had happened when he was away. To figure out what caused it, he wanted to get a view of the trace by itself so he could correlate it to activity on other channels.

The customer had been doing single-shot acquisitions for days trying to locate the infrequently occurring glitch. He pushed the “single” button over and over and still couldn’t locate the problem. Watching the engineer laboriously pecking away at the button, Genther experienced his big “aha” moment. He could see that pressing the “single” button a million times is a painful way to find a one-in-a-million glitch, and he realized he could make both customers’ jobs easier using Agilent’s existing oscilloscope technology.

“The nerd trigger was going to take forever to complete, but I figured we could create very similar functionality using the scope’s mask test infrastructure,” Genther said. “We could qualify acquisitions with mask testing and use that feature to push the button for him.”

Mask testing shows all the waveforms that fail or exceed the limits of the mask. Genther’s brainstorm was to ignore all the waveforms that fail and just display the ones that do meet the mask limits. This process could happen behind the scenes, using digital technology to accomplish what the customer was doing manually.

CHAMPIONING THE IDEA

Coming up with the idea to use a reverse mask test to achieve an easier version of the trigger was just the first step in the process of bringing it to market. Next, Genther had to convince his boss and teammates the project was worth pursuing.

Figure 1: Example of a hard-to-capture signal anomaly with infinite persistence turned on

Scott GentherKey inventor of Agilent Zone Trigger

Page 30: EEWeb Pulse 122 - Crossbar

30 Visit: eeweb.com

PULSE

mockup to his desk. “I used it on the spot,” he said. “It was clear to me this was going to be a useful feature.”

“Once you have a mockup, it is an easier sell than if you are just trying to sell an idea,” said Genther. “You can say, ‘Come on over here and take a look,’ and immediately show it to your boss or teammates.”

At the last minute, the team agreed to use the new feature in the original release of the InfiniiScan event identification software package for Agilent DSO80000B Series Infiniium oscilloscopes, introduced in February 2006. The software is designed to help engineers identify signal integrity issues in their electronic designs more easily.

Before the feature could be released, the team had to figure out what to call it. “We had a huge debate about whether it was OK to call it a trigger,” said Genther. “Technically, it is a mask, not a trigger, and we didn’t want to confuse customers.”

The team did some research and discovered that most engineers think of triggers as “anything that gives them the picture they want.” The team ultimately decided to call it a software trigger. “It ended up being the star feature of the InfiniiScan software package,” Genther said.

MOVING BEYOND SOFTWARE

The software trigger’s success as part of the InfiniiScan package made the next step easy: incorporating the feature directly in an oscilloscope. When the scope designers were defining the high-performance Infiniium 90000 Series oscilloscopes, the team didn’t hesitate to include it. In fact, this technology enabled the world’s first three-stage triggering system, where the first two stages are hardware-based triggers and the third stage is a software trigger.

At that point, advanced trigger features in general, and the software trigger in particular, were incorporated only in high-end oscilloscopes.

Several years later, Agilent’s oscilloscope team designed a new chip that dramatically changed

the game by allowing engineers to control and use a scope’s features via a capacitive touch screen. Shown in Figure 4 the new “MegaZoom IV” chip enabled scopes with extremely fast update rates -- a million waveforms per second – and it implemented mask test and Genther’s inverse mask test, rebranded as InfiniiScan Zone triggering, as hardware features. Implementing the zone triggering functionality using high-speed digital ASIC hardware made it much faster. It could capture 1,000,000 waveforms per second, which made it much more likely that engineers would be able to see the glitch or anomaly they were looking for. The InfiniiScan Zone trigger, even when it is implemented in hardware, can produce the signal of interest only as fast as the oscilloscope can capture it.

The chip made its debut in the InfiniiVision 4000 X-Series oscilloscopes, introduced November 13, 2012, and it revolutionized how engineers use oscilloscopes. The combination of the capacitive touch screen capability and the hardware-based InfiniiScan Zone trigger makes it easy to trigger on signals by drawing a box around the

ones of interest. The 4000 X-Series oscilloscope then creates the trigger based on the mask. “It would be very clumsy to try to draw a box by turning knobs and pushing buttons,” Genther said. “With the touch screen, if you can see it, you can trigger on it simply by pointing at it- exactly what the engineer from my original customer visit was asking for. Using this method, you can easily isolate things you see only infrequently.

It is the combination of these features that sets the InfiniiVision 4000 X-Series apart. The hardware-based InfiniiScan Zone trigger and capacitive touch screen are the most obvious innovations, but without the fast update rate of the MegaZoom IV ASIC, you wouldn’t even know the infrequent events are there.

“Any engineer who does digital design will appreciate the InfiniiScan Zone touch trigger,” said Genther. “The classic case is working on memory devices. You have a clock, and you have reads and writes on top of each other. It is a mess. The bus goes one way with the reads and the other way with the writes, and the timing is completely different. You are probing the bus and seeing both. So you draw a box around something that happens only with the writes, and boom—you have just the writes.”

MEANINGFUL INNOVATIONS

The most important product innovations happen when designers respond to customer frustrations and work toward ways to help them solve their problems more easily.

“With zone triggering, it was like figuring out we could slice the bread before we sell it to customers so they don’t have to slice it themselves,” said Genther. “The whole thing was about the user interface. The result is that customers no longer have to think about how to set up an oscilloscope trigger.”

Combing the trigger innovation with the MegaZoom IV chip takes it to the next level. Now Agilent’s customers can see their glitches and trigger on them without needing to understand nerdy triggering concepts.

Figure 4: Agilent MegaZoom IV” chip

Figure 2 & 3 Demonstrate how thw zone trigger was eventually implemented with a touch screen. Touch the signal of interest and then select must intersect - the scope then displays only the signal of interest.

Figure 5: InfiniiVision 4000 X-Series Oscilloscope

He got software designers and hardware designers together in the same room and lobbied for his idea. He stressed that the new feature required no new technology to be developed. “We already had the infrastructure in place for qualified triggers, so I had to convince the team we could use that infrastructure to address a user-interface problem,” said Genther.

The breakthrough came when one of the software developers on his team agreed to create a mockup of the zone trigger.

Genther was doing some troubleshooting, looking for bad signals on a bus, when the software developer brought the completed

Page 31: EEWeb Pulse 122 - Crossbar

31Visit: eeweb.com

TECH ARTICLE

mockup to his desk. “I used it on the spot,” he said. “It was clear to me this was going to be a useful feature.”

“Once you have a mockup, it is an easier sell than if you are just trying to sell an idea,” said Genther. “You can say, ‘Come on over here and take a look,’ and immediately show it to your boss or teammates.”

At the last minute, the team agreed to use the new feature in the original release of the InfiniiScan event identification software package for Agilent DSO80000B Series Infiniium oscilloscopes, introduced in February 2006. The software is designed to help engineers identify signal integrity issues in their electronic designs more easily.

Before the feature could be released, the team had to figure out what to call it. “We had a huge debate about whether it was OK to call it a trigger,” said Genther. “Technically, it is a mask, not a trigger, and we didn’t want to confuse customers.”

The team did some research and discovered that most engineers think of triggers as “anything that gives them the picture they want.” The team ultimately decided to call it a software trigger. “It ended up being the star feature of the InfiniiScan software package,” Genther said.

MOVING BEYOND SOFTWARE

The software trigger’s success as part of the InfiniiScan package made the next step easy: incorporating the feature directly in an oscilloscope. When the scope designers were defining the high-performance Infiniium 90000 Series oscilloscopes, the team didn’t hesitate to include it. In fact, this technology enabled the world’s first three-stage triggering system, where the first two stages are hardware-based triggers and the third stage is a software trigger.

At that point, advanced trigger features in general, and the software trigger in particular, were incorporated only in high-end oscilloscopes.

Several years later, Agilent’s oscilloscope team designed a new chip that dramatically changed

the game by allowing engineers to control and use a scope’s features via a capacitive touch screen. Shown in Figure 4 the new “MegaZoom IV” chip enabled scopes with extremely fast update rates -- a million waveforms per second – and it implemented mask test and Genther’s inverse mask test, rebranded as InfiniiScan Zone triggering, as hardware features. Implementing the zone triggering functionality using high-speed digital ASIC hardware made it much faster. It could capture 1,000,000 waveforms per second, which made it much more likely that engineers would be able to see the glitch or anomaly they were looking for. The InfiniiScan Zone trigger, even when it is implemented in hardware, can produce the signal of interest only as fast as the oscilloscope can capture it.

The chip made its debut in the InfiniiVision 4000 X-Series oscilloscopes, introduced November 13, 2012, and it revolutionized how engineers use oscilloscopes. The combination of the capacitive touch screen capability and the hardware-based InfiniiScan Zone trigger makes it easy to trigger on signals by drawing a box around the

ones of interest. The 4000 X-Series oscilloscope then creates the trigger based on the mask. “It would be very clumsy to try to draw a box by turning knobs and pushing buttons,” Genther said. “With the touch screen, if you can see it, you can trigger on it simply by pointing at it- exactly what the engineer from my original customer visit was asking for. Using this method, you can easily isolate things you see only infrequently.

It is the combination of these features that sets the InfiniiVision 4000 X-Series apart. The hardware-based InfiniiScan Zone trigger and capacitive touch screen are the most obvious innovations, but without the fast update rate of the MegaZoom IV ASIC, you wouldn’t even know the infrequent events are there.

“Any engineer who does digital design will appreciate the InfiniiScan Zone touch trigger,” said Genther. “The classic case is working on memory devices. You have a clock, and you have reads and writes on top of each other. It is a mess. The bus goes one way with the reads and the other way with the writes, and the timing is completely different. You are probing the bus and seeing both. So you draw a box around something that happens only with the writes, and boom—you have just the writes.”

MEANINGFUL INNOVATIONS

The most important product innovations happen when designers respond to customer frustrations and work toward ways to help them solve their problems more easily.

“With zone triggering, it was like figuring out we could slice the bread before we sell it to customers so they don’t have to slice it themselves,” said Genther. “The whole thing was about the user interface. The result is that customers no longer have to think about how to set up an oscilloscope trigger.”

Combing the trigger innovation with the MegaZoom IV chip takes it to the next level. Now Agilent’s customers can see their glitches and trigger on them without needing to understand nerdy triggering concepts.

Figure 4: Agilent MegaZoom IV” chip

Figure 2 & 3 Demonstrate how thw zone trigger was eventually implemented with a touch screen. Touch the signal of interest and then select must intersect - the scope then displays only the signal of interest.

Figure 5: InfiniiVision 4000 X-Series Oscilloscope

He got software designers and hardware designers together in the same room and lobbied for his idea. He stressed that the new feature required no new technology to be developed. “We already had the infrastructure in place for qualified triggers, so I had to convince the team we could use that infrastructure to address a user-interface problem,” said Genther.

The breakthrough came when one of the software developers on his team agreed to create a mockup of the zone trigger.

Genther was doing some troubleshooting, looking for bad signals on a bus, when the software developer brought the completed

Page 32: EEWeb Pulse 122 - Crossbar

Copyright 2013, Silicon Frameworks, LLC PCBWeb.com

ep page master fp.indd 1 3/11/13 10:34:41 AM

Page 33: EEWeb Pulse 122 - Crossbar

Copyright 2013, Silicon Frameworks, LLC PCBWeb.com

ep page master fp.indd 1 3/11/13 10:34:41 AM

©2013 Intersil Americas LLC. All rights reserved.

PURE DIGITAL POWER

►INTERSIL.COM/ZL8800

Download Datasheet

Read App Note

Order Demo Board

Watch Videos

Not only can you now track and control every aspect of your power supply without additional monitors or sequencers, the ZL8800’s ChargeMode™ control loop technology delivers best-in-class transient response and eliminates the need for complex compensation. This all adds up to a pure digital solution that lowers system cost, reduces design time and increases available board space.

ZL8800 Combines Digital Control Loop Technology and PMBus™ Interface for the Industry’s Most Advanced Digital Power Solution

Stop Coding and Start Designing with Intersil’s PowerNavigatorTM Download free at intersil.com/powernavigator

New Digital Power Knowledge CenterWatch and learn at intersil.com/DigiPowerTraining

Page 34: EEWeb Pulse 122 - Crossbar

34 Visit: eeweb.com

PULSE

Crossbar is a groundbreaking memory technology company based in Santa Clara, California. Crossbar turned heads last year when they unveiled a new category of high performance resistive RAM (RRAM) technology that can scale up to 1 terabyte on a chip the size of a postage stamp. This highly scalable technology performs around 20x faster than today’s NAND Flash memory and has 10x the endurance at half the die size. With these unparalleled specifications, Crossbar’s technology will enable the next generation of high-performing electronic devices.

EEWeb spoke with George Minassian, CEO of Crossbar, about the current limitations in the Flash memory industry, how the company plans to overcome them, and how their new RRAM technology will redefine the industry as we know it.

Raising the Bar:Crossbar’s Entirely New Type of Memory

George Minassian, CEO of Crossbar

Page 35: EEWeb Pulse 122 - Crossbar

35Visit: eeweb.com

INTERVIEW

Crossbar is a groundbreaking memory technology company based in Santa Clara, California. Crossbar turned heads last year when they unveiled a new category of high performance resistive RAM (RRAM) technology that can scale up to 1 terabyte on a chip the size of a postage stamp. This highly scalable technology performs around 20x faster than today’s NAND Flash memory and has 10x the endurance at half the die size. With these unparalleled specifications, Crossbar’s technology will enable the next generation of high-performing electronic devices.

EEWeb spoke with George Minassian, CEO of Crossbar, about the current limitations in the Flash memory industry, how the company plans to overcome them, and how their new RRAM technology will redefine the industry as we know it.

Raising the Bar:Crossbar’s Entirely New Type of Memory

George Minassian, CEO of Crossbar

Page 36: EEWeb Pulse 122 - Crossbar

36 Visit: eeweb.com

PULSE

Could you tell us about how you came to co-found Crossbar?

I started at AMD where I spent time working on communication devices, microprocessors and system architectures. I then joined the Flash JV division of AMD and Fujitsu, where I worked in strategic marketing. While I was there, I heard about this new technology from a client and got very excited about it. The more I learned about it, the more excited I became. The technology was produced by a team of researchers at the University of Michigan, and they had already passed a number of milestones needed to achieve credibility. After I had a chance to see the technology up close, I got involved with the team, which included financial and corporate support from Kleiner Perkins Caufield & Byers, a Bay Area venture firm. Together we knew we had to start a company around the underlying technology. I brought in a few other colleagues and that is how we founded Crossbar.

How important was that initial phase in terms of gaining investor interest?

The initial part of starting a company like this is getting from lab to fab. When you set up a lab, you take the best of the “recipes” for your technology because everything is available to you. The problem is, the toolset for each of the “recipes” can be slightly different, and some might not be manufacturable. So, using commercially available fabrication tools, we tried to duplicate the outcomes. In the first couple of runs, we discovered some things weren’t working the way they should. We spent about a year fine-tuning, taking measurements and making accurate adjustments before we got to a point where the array was fully functioning. After that, we used CMOS from TSMC, put the array on top of it and put the rest of the pieces together – all over the course of about three years.

What is Crossbar currently offering as a company?

We are providing licensable IP technology for embedded to start with. We are also transferring the existing technologies we have with our pilot line fab into our commercial fabs. As we are developing the sub-20nm arrays and characterizing those, we are also preparing our 55nm and 40nm technology for production. When the transfer is completed, over the next nine months, the fab will be qualified to accept customers for business. Our product is very unique—it’s CMOS compatible, and it has embedded capabilities in terms of random access and CMOS compatible process. It is also suited to handle higher densities because the cell itself is extremely small. It has MLC and stacking capabilities. Stacking is very simple—each layer just repeats itself without the complexity of 3D NAND implementations.

Given all of that, we can do very high-density devices in a 40nm or even 65nm technology nodes, meaning it will cover a wide range of technologies. Of course, the smaller the node, the denser the device can get. The interesting part is that all of these types of technologies are being transferred to our fab. When you transfer this to a fab, you have the opportunity to do embedded, standalone NOR, or standalone NAND. For us, the complexity of the controller is the only thing that makes it one or the other. In terms of Flash today, you need to change the entire factory and manufacturing process to go from one to the other. In our case, you just have to change the controller.

Current Flash is based on the presence of electrons to detect the logic. For example, in 30nm technology, you have a couple dozen electrons for a 1-volt differential. When you shrink your device, the number of electrons becomes fewer and fewer. At some point, for very small nodes, you have to detect ten to 20 electrons to know if the device is ON or OFF. What we have witnessed is, as theses nodes get smaller and the number of electrons diminishes, the controller logic needs to become a lot more sophisticated. You put a lot of these cells next to each other, then test them by leaving them in your car to see if the electrons behave differently at higher temperatures. This is especially true for parts destined for commercial devices. Those elements make the detectors even more sensitive, which is why NAND and NOR are having difficulty.

Crossbar’s scalable cell structure

CROSSBAR’S SIMPLE AND VERY SCALABLE MEMORY CELL STRUCTURE ENABLES AN ENTIRELY NEW CLASS OF 3D RRAM, WHICH CAN BE EASILY INCORPORATED INTO THE BACK END OF ANY STANDARD CMOS MANUFACTURING FAB.

What are the current limitations of the standard non-volatile Flash memory and how does your technology differ?

Page 37: EEWeb Pulse 122 - Crossbar

37Visit: eeweb.com

INTERVIEW

Could you tell us about how you came to co-found Crossbar?

I started at AMD where I spent time working on communication devices, microprocessors and system architectures. I then joined the Flash JV division of AMD and Fujitsu, where I worked in strategic marketing. While I was there, I heard about this new technology from a client and got very excited about it. The more I learned about it, the more excited I became. The technology was produced by a team of researchers at the University of Michigan, and they had already passed a number of milestones needed to achieve credibility. After I had a chance to see the technology up close, I got involved with the team, which included financial and corporate support from Kleiner Perkins Caufield & Byers, a Bay Area venture firm. Together we knew we had to start a company around the underlying technology. I brought in a few other colleagues and that is how we founded Crossbar.

How important was that initial phase in terms of gaining investor interest?

The initial part of starting a company like this is getting from lab to fab. When you set up a lab, you take the best of the “recipes” for your technology because everything is available to you. The problem is, the toolset for each of the “recipes” can be slightly different, and some might not be manufacturable. So, using commercially available fabrication tools, we tried to duplicate the outcomes. In the first couple of runs, we discovered some things weren’t working the way they should. We spent about a year fine-tuning, taking measurements and making accurate adjustments before we got to a point where the array was fully functioning. After that, we used CMOS from TSMC, put the array on top of it and put the rest of the pieces together – all over the course of about three years.

What is Crossbar currently offering as a company?

We are providing licensable IP technology for embedded to start with. We are also transferring the existing technologies we have with our pilot line fab into our commercial fabs. As we are developing the sub-20nm arrays and characterizing those, we are also preparing our 55nm and 40nm technology for production. When the transfer is completed, over the next nine months, the fab will be qualified to accept customers for business. Our product is very unique—it’s CMOS compatible, and it has embedded capabilities in terms of random access and CMOS compatible process. It is also suited to handle higher densities because the cell itself is extremely small. It has MLC and stacking capabilities. Stacking is very simple—each layer just repeats itself without the complexity of 3D NAND implementations.

Given all of that, we can do very high-density devices in a 40nm or even 65nm technology nodes, meaning it will cover a wide range of technologies. Of course, the smaller the node, the denser the device can get. The interesting part is that all of these types of technologies are being transferred to our fab. When you transfer this to a fab, you have the opportunity to do embedded, standalone NOR, or standalone NAND. For us, the complexity of the controller is the only thing that makes it one or the other. In terms of Flash today, you need to change the entire factory and manufacturing process to go from one to the other. In our case, you just have to change the controller.

Current Flash is based on the presence of electrons to detect the logic. For example, in 30nm technology, you have a couple dozen electrons for a 1-volt differential. When you shrink your device, the number of electrons becomes fewer and fewer. At some point, for very small nodes, you have to detect ten to 20 electrons to know if the device is ON or OFF. What we have witnessed is, as theses nodes get smaller and the number of electrons diminishes, the controller logic needs to become a lot more sophisticated. You put a lot of these cells next to each other, then test them by leaving them in your car to see if the electrons behave differently at higher temperatures. This is especially true for parts destined for commercial devices. Those elements make the detectors even more sensitive, which is why NAND and NOR are having difficulty.

Crossbar’s scalable cell structure

CROSSBAR’S SIMPLE AND VERY SCALABLE MEMORY CELL STRUCTURE ENABLES AN ENTIRELY NEW CLASS OF 3D RRAM, WHICH CAN BE EASILY INCORPORATED INTO THE BACK END OF ANY STANDARD CMOS MANUFACTURING FAB.

What are the current limitations of the standard non-volatile Flash memory and how does your technology differ?

Page 38: EEWeb Pulse 122 - Crossbar

38 Visit: eeweb.com

PULSE

In our case, our design is based on a metal filament. We have a layer of isolated silicon and we drive metal through it by applying some polarities across that sandwich of silicon layers. Once the metal goes inside, you make a connection. This is based on the ions of metal being forced into a non-conducting layer. There are three layers in total: one conducting layer at the bottom, one non-conducting in the center and the metal on the top.

We saw how the physics works against NAND, because of the diminishing number of electrons at the smaller geometries. Compared to NAND, our technology differs because of a very large ON current to OFF current ration which makes the detection of ON and OFF states much simpler. For us when the device is ON, that current is a function of the filament and, regardless of the size of the device, it stays the same. The current on the bottom of the equation—which is the OFF state—is when there are no filaments in the device. This current is caused by the leakage in the device. As you shrink the device, that leakage current gets smaller and smaller. As a result the ON/OFF ratio for us becomes larger and the ability to detect when the device is ON, versus when it is OFF, becomes easier. The physics actually work in our favor; as you shrink the device, the performance gets better.

As a consequence of that, the 20nm small arrays we are doing in our R&D fab start with four levels per cell. We have tested tens of thousands of cycles without any difficulty.

For investors, a technology like this would seem very attractive. What’s it going to take for this technology to become ubiquitous?

The steps we are taking right now lead toward mainstream adoption. The commercialization process involves bringing our existing technology to a fab and repeating the same process at 40nm and 50nm technology nodes, which we are doing right now. Once we have that completed we will start the embedded licensing business, as well as the standalone code and data business. We have done transfers from fabs multiple times—starting from the University to our fab, and then from our fab to several commercial grade fabs. Each time we set up, we learn how quickly we can do this. Each time, within three months of transferring, we had our arrays working. We now have to repeat the transfer again as we move to our new commercial fab, and we’re confident it can be done. This means within nine months we’ll be ready for a qualification cycle, and at that point, we’ll have customers coming in with their requirements.

Do you have any technology partners that you are working with to bring this to market?

Crossbar has a complete technology, so in terms of partnerships, we are set. In fact, we already have about 45 issued patents for our technology, with another 55 pending. From a manufacturing standpoint and implementing at specific nodes, we’ll be partnering with large, commercial fab partners.

What is Crossbar’s roadmap for competing with Flash technologies?

There are several requirements to put our technology into a competitive position versus Flash. For our one terabyte devices, we will need a 20nm fab, and we will have had to perfect the stacking of MLC. Once we transfer the cells and arrays to a commercial fab, we can move from an embedded design to a standalone design. To do this, we will need to change the controller because standalone designs need to have different interfaces to the outside devices. Switching the controller from an embedded architecture, which uses someone else’s controller logic, to a standalone controller, is the biggest hurdle.

The other requirement is moving from a 40nm node to a 28nm node. Currently, we plan to use a 55nm node, based on the toolsets available right now. After that, we will convert to a smaller node. Most of the 28nm fabs are in various stages of availability, but are ready to be introduced to a new technology. We expect it to take about two years to completely transfer our technology to a 28nm fab.

Could you talk about some of the non-volatile aspects of this memory?

When it comes to Flash, there are several parameters that people use. The most critical parameters are voltage, current, endurance and temperature. For us, the voltages are all well below 5V because we are 100 percent CMOS and very close to 2V for read and write. The power has to be extremely low, and in our case, we are 20 times lower than anyone else. In terms of endurance, we are pushing 100,000 cycles.

What is Crossbar most excited about with this technology?

We are a highly technical group, so we tend to get excited about the technology. But, we get even more excited about it when we meet with potential customers and we see they’re excited. We feel we are at the leading edge of a new memory landscape, and we strongly believe our technology will bring great improvements to current systems and will enable a new era of applications and products.

“THE PHYSICS ACTUALLY WORKS IN OUR FAVOR; AS YOU SHRINK THE DEVICE, THE PERFORMANCE GETS BETTER.”

CROSSBAR RRAM TECHNOLOGY: SIMPLE CMOS INTEGRATIONIts simplicity, stackability, and CMOS compatibility enables logic and memory to be easily integrated onto a single chip at the latest technology node.

Page 39: EEWeb Pulse 122 - Crossbar

39Visit: eeweb.com

INTERVIEW

In our case, our design is based on a metal filament. We have a layer of isolated silicon and we drive metal through it by applying some polarities across that sandwich of silicon layers. Once the metal goes inside, you make a connection. This is based on the ions of metal being forced into a non-conducting layer. There are three layers in total: one conducting layer at the bottom, one non-conducting in the center and the metal on the top.

We saw how the physics works against NAND, because of the diminishing number of electrons at the smaller geometries. Compared to NAND, our technology differs because of a very large ON current to OFF current ration which makes the detection of ON and OFF states much simpler. For us when the device is ON, that current is a function of the filament and, regardless of the size of the device, it stays the same. The current on the bottom of the equation—which is the OFF state—is when there are no filaments in the device. This current is caused by the leakage in the device. As you shrink the device, that leakage current gets smaller and smaller. As a result the ON/OFF ratio for us becomes larger and the ability to detect when the device is ON, versus when it is OFF, becomes easier. The physics actually work in our favor; as you shrink the device, the performance gets better.

As a consequence of that, the 20nm small arrays we are doing in our R&D fab start with four levels per cell. We have tested tens of thousands of cycles without any difficulty.

For investors, a technology like this would seem very attractive. What’s it going to take for this technology to become ubiquitous?

The steps we are taking right now lead toward mainstream adoption. The commercialization process involves bringing our existing technology to a fab and repeating the same process at 40nm and 50nm technology nodes, which we are doing right now. Once we have that completed we will start the embedded licensing business, as well as the standalone code and data business. We have done transfers from fabs multiple times—starting from the University to our fab, and then from our fab to several commercial grade fabs. Each time we set up, we learn how quickly we can do this. Each time, within three months of transferring, we had our arrays working. We now have to repeat the transfer again as we move to our new commercial fab, and we’re confident it can be done. This means within nine months we’ll be ready for a qualification cycle, and at that point, we’ll have customers coming in with their requirements.

Do you have any technology partners that you are working with to bring this to market?

Crossbar has a complete technology, so in terms of partnerships, we are set. In fact, we already have about 45 issued patents for our technology, with another 55 pending. From a manufacturing standpoint and implementing at specific nodes, we’ll be partnering with large, commercial fab partners.

What is Crossbar’s roadmap for competing with Flash technologies?

There are several requirements to put our technology into a competitive position versus Flash. For our one terabyte devices, we will need a 20nm fab, and we will have had to perfect the stacking of MLC. Once we transfer the cells and arrays to a commercial fab, we can move from an embedded design to a standalone design. To do this, we will need to change the controller because standalone designs need to have different interfaces to the outside devices. Switching the controller from an embedded architecture, which uses someone else’s controller logic, to a standalone controller, is the biggest hurdle.

The other requirement is moving from a 40nm node to a 28nm node. Currently, we plan to use a 55nm node, based on the toolsets available right now. After that, we will convert to a smaller node. Most of the 28nm fabs are in various stages of availability, but are ready to be introduced to a new technology. We expect it to take about two years to completely transfer our technology to a 28nm fab.

Could you talk about some of the non-volatile aspects of this memory?

When it comes to Flash, there are several parameters that people use. The most critical parameters are voltage, current, endurance and temperature. For us, the voltages are all well below 5V because we are 100 percent CMOS and very close to 2V for read and write. The power has to be extremely low, and in our case, we are 20 times lower than anyone else. In terms of endurance, we are pushing 100,000 cycles.

What is Crossbar most excited about with this technology?

We are a highly technical group, so we tend to get excited about the technology. But, we get even more excited about it when we meet with potential customers and we see they’re excited. We feel we are at the leading edge of a new memory landscape, and we strongly believe our technology will bring great improvements to current systems and will enable a new era of applications and products.

“THE PHYSICS ACTUALLY WORKS IN OUR FAVOR; AS YOU SHRINK THE DEVICE, THE PERFORMANCE GETS BETTER.”

CROSSBAR RRAM TECHNOLOGY: SIMPLE CMOS INTEGRATIONIts simplicity, stackability, and CMOS compatibility enables logic and memory to be easily integrated onto a single chip at the latest technology node.

Page 40: EEWeb Pulse 122 - Crossbar

PULSE

40 Visit: eeweb.com

Altera Cyclone V GXStarter Kit

The Altera Cyclone V GX Starter Kit is built around the Altera Cyclone V GX FPGA. The FPGA has 75,000 logic elements, 4.4Mb of on-chip memory, 150 hard block elements, as well as transceiver support of up to 3.125Gb per second. The Starter kit offers a wide range of I/O and expansion options, allowing for greater flexibility for the designer.

Overview of the

Page 41: EEWeb Pulse 122 - Crossbar

41Visit: eeweb.com

PRODUCT OVERVIEW

Altera Cyclone V GXStarter Kit

The Altera Cyclone V GX Starter Kit is built around the Altera Cyclone V GX FPGA. The FPGA has 75,000 logic elements, 4.4Mb of on-chip memory, 150 hard block elements, as well as transceiver support of up to 3.125Gb per second. The Starter kit offers a wide range of I/O and expansion options, allowing for greater flexibility for the designer.

Overview of the

Page 42: EEWeb Pulse 122 - Crossbar

PULSE

42 Visit: eeweb.com

Included Hardware SpecificationsThe Altera Cyclone V GX Starter Kit allows for greater design flexibility by using a Cyclone V GX FPGA device. In terms of memory, the Starter Kit has 4Gb of LPDDR2, 256Mb of EPCQ flash, as well as a Micro SD slot for further expansion. The board implements Arduino’s standard expansion footprint, 2x20 GPIO, and an HSMC connector, resulting in over 300 possible expansion cards for shorten development time.

The board also has audio output and input, as well as an HDMI, which is compatible with DVI v1.0 and HDCP v1.4. These I/Os allow for a full, multi-media streaming experience. The Altera Cyclone V GX Starter Kit is the first to implement the Arduino footprint, allowing for a highly versatile and powerful FPGA development platform.

Watch Video

USBType-B Ports

HDMI TX

AudioJacks

Micro SD Card

To watch a video overview and demonstration of the Cyclone V Starter Kit, click the image below:

2x20GPIO

User Interface

HSMCConnector

Altera Cyclone V FPGA

Arduino Footprint

Page 43: EEWeb Pulse 122 - Crossbar

43Visit: eeweb.com

PRODUCT OVERVIEW

Included Hardware SpecificationsThe Altera Cyclone V GX Starter Kit allows for greater design flexibility by using a Cyclone V GX FPGA device. In terms of memory, the Starter Kit has 4Gb of LPDDR2, 256Mb of EPCQ flash, as well as a Micro SD slot for further expansion. The board implements Arduino’s standard expansion footprint, 2x20 GPIO, and an HSMC connector, resulting in over 300 possible expansion cards for shorten development time.

The board also has audio output and input, as well as an HDMI, which is compatible with DVI v1.0 and HDCP v1.4. These I/Os allow for a full, multi-media streaming experience. The Altera Cyclone V GX Starter Kit is the first to implement the Arduino footprint, allowing for a highly versatile and powerful FPGA development platform.

Watch Video

USBType-B Ports

HDMI TX

AudioJacks

Micro SD Card

To watch a video overview and demonstration of the Cyclone V Starter Kit, click the image below:

2x20GPIO

User Interface

HSMCConnector

Altera Cyclone V FPGA

Arduino Footprint

Page 44: EEWeb Pulse 122 - Crossbar

PULSE

44 Visit: eeweb.com

FAN23SV04T High Efficiency Buck Converter

The FAN23SV04T High Efficiency Synchronous Buck Converter from Fairchild is designed to power DDR termination rails in applications such as servers, telecom hardware, game consoles, and notebooks. The device is suitable for any application requiring an efficient tracking converter capable of sourcing and sinking multiple amps of current.

Overview of the

Page 45: EEWeb Pulse 122 - Crossbar

45Visit: eeweb.com

PRODUCT OVERVIEW

FAN23SV04T High Efficiency Buck Converter

The FAN23SV04T High Efficiency Synchronous Buck Converter from Fairchild is designed to power DDR termination rails in applications such as servers, telecom hardware, game consoles, and notebooks. The device is suitable for any application requiring an efficient tracking converter capable of sourcing and sinking multiple amps of current.

Overview of the

Page 46: EEWeb Pulse 122 - Crossbar

46 Visit: eeweb.com

PULSE

Included Hardware Tech SpecsDDR4 requires a termination supply (VTT) to be half of VDDQ. This has typically been accomplished by generating VDDQ from the 12V

supply and using a linear regulator, supplied by VDDQ, to generate VTT. This setup results in a VTT conversion efficiency of about 45%. The FAN23SV04T uses VDDQ as an input, but can be supplied from a separate rail; typically the same 12V rail that supplies the VDDQ regulator. This setup results in a VTT conversion efficiency of about 80%—a significant improvement over the linear regulator approach.

The FAN23SV04T also integrates the 50% resistive divider network in the IC. This results in very tightly-matched resistors and eliminates the added footprint and cost of providing external high-precision resistors. If your design requires tracking at something other than 50% of your reference supply, it is possible to add external resistors to adjust the tracking ratio.

Watch Video

FAN23SV04TBuck Converter

Enable & Test Pins

TransientLoad Switches

To watch a video overview and demonstration of the FAN23SV04T Buck Converter, click the image below:

Power In

Power Out

Page 47: EEWeb Pulse 122 - Crossbar

47Visit: eeweb.com

PRODUCT OVERVIEW

Included Hardware Tech SpecsDDR4 requires a termination supply (VTT) to be half of VDDQ. This has typically been accomplished by generating VDDQ from the 12V

supply and using a linear regulator, supplied by VDDQ, to generate VTT. This setup results in a VTT conversion efficiency of about 45%. The FAN23SV04T uses VDDQ as an input, but can be supplied from a separate rail; typically the same 12V rail that supplies the VDDQ regulator. This setup results in a VTT conversion efficiency of about 80%—a significant improvement over the linear regulator approach.

The FAN23SV04T also integrates the 50% resistive divider network in the IC. This results in very tightly-matched resistors and eliminates the added footprint and cost of providing external high-precision resistors. If your design requires tracking at something other than 50% of your reference supply, it is possible to add external resistors to adjust the tracking ratio.

Watch Video

FAN23SV04TBuck Converter

Enable & Test Pins

TransientLoad Switches

To watch a video overview and demonstration of the FAN23SV04T Buck Converter, click the image below:

Power In

Power Out

Page 48: EEWeb Pulse 122 - Crossbar

48 Visit: eeweb.com

PULSE

Intel GalileoDevelopment Board

The Intel Galileo board is the first ever development board to be available from the Arduino Certified program. The Galileo board features the Intel Quark SoC X1000 processor, making it a great development board for hobbyist, student, and even professional use. The combination of the Quark SoC and the extensibility of the Arduino environment make the Galileo a powerful, easy-to-use, board for taking designs to the next level.

Overview of the

Page 49: EEWeb Pulse 122 - Crossbar

49Visit: eeweb.com

PRODUCT OVERVIEW

Intel GalileoDevelopment Board

The Intel Galileo board is the first ever development board to be available from the Arduino Certified program. The Galileo board features the Intel Quark SoC X1000 processor, making it a great development board for hobbyist, student, and even professional use. The combination of the Quark SoC and the extensibility of the Arduino environment make the Galileo a powerful, easy-to-use, board for taking designs to the next level.

Overview of the

Page 50: EEWeb Pulse 122 - Crossbar

50 Visit: eeweb.com

PULSE

Included Hardware Tech Specs

The Galileo board features the Intel Quark SoC X1000, a 32-bit Pentium class application processor with a single

core operating at 400MHz at a max TDP of just 2.2 watts. The board also has 256MB of DDR3 memory, a microSD slot, 100Mbps ethernet, UART, USB, and a full length mini-PCIe slot on the back. One of the board’s key features, however, is the support for Arduino shields and the Arduino IDE.

Although the IDE is modeled after the Arduino IDE and includes the same standard libraries, it currently only supports the Galileo board. This means that the standard functionality should work without any issues.

Watch Video

Ethernet

Support for Arduino Shields and Arduino IDE

To watch a video overview and demonstration of the Intel Galileo Board, click the image below:

Intel Quark Processor

UART USB

JTAG

Micro SD Slot

Page 51: EEWeb Pulse 122 - Crossbar

51Visit: eeweb.com

PRODUCT OVERVIEW

Included Hardware Tech Specs

The Galileo board features the Intel Quark SoC X1000, a 32-bit Pentium class application processor with a single

core operating at 400MHz at a max TDP of just 2.2 watts. The board also has 256MB of DDR3 memory, a microSD slot, 100Mbps ethernet, UART, USB, and a full length mini-PCIe slot on the back. One of the board’s key features, however, is the support for Arduino shields and the Arduino IDE.

Although the IDE is modeled after the Arduino IDE and includes the same standard libraries, it currently only supports the Galileo board. This means that the standard functionality should work without any issues.

Watch Video

Ethernet

Support for Arduino Shields and Arduino IDE

To watch a video overview and demonstration of the Intel Galileo Board, click the image below:

Intel Quark Processor

UART USB

JTAG

Micro SD Slot

Page 52: EEWeb Pulse 122 - Crossbar

52 Visit: eeweb.com

PULSE

International Rectifier IR3846

Audio Amplifier

The IR4312M is a class-D amplifier that integrates the power MOSFETs with the audio controller IC into a single package, which reduces parasitics and helps improve sound quality. This amplifier is good for applications such as docking stations, power or active speakers, as well as portable musical instruments.

Overview of the

Page 53: EEWeb Pulse 122 - Crossbar

53Visit: eeweb.com

PRODUCT OVERVIEW

International Rectifier IR3846

Audio Amplifier

The IR4312M is a class-D amplifier that integrates the power MOSFETs with the audio controller IC into a single package, which reduces parasitics and helps improve sound quality. This amplifier is good for applications such as docking stations, power or active speakers, as well as portable musical instruments.

Overview of the

Page 54: EEWeb Pulse 122 - Crossbar

54 Visit: eeweb.com

PULSE

Included Hardware

SpecificationsThe IR4312M integrates the MOSFETs as well as the audio controller IC into a single 7x7mm PQFN package. There is no mechanical heatsink required over it, so you get a class-D audio amplifier into a very small, easy-to-use, and easy-to-implement package.

The combination of the audio controller IC with the audio MOSFETs results in improved efficiency, THD, and EMI. The IR4312M has high voltage ratings and noise immunity to ensure reliable operation and prevents audio clipping at high volume.

To read module specifications and reference designs, click here.

Watch Video

Audio Input Jack

PowIRaudioIC

Power Terminal

Stereo Outputs

To watch a video overview and demonstration of the IR4312M, click the image below:

Page 55: EEWeb Pulse 122 - Crossbar

55Visit: eeweb.com

PRODUCT OVERVIEW

Included Hardware

SpecificationsThe IR4312M integrates the MOSFETs as well as the audio controller IC into a single 7x7mm PQFN package. There is no mechanical heatsink required over it, so you get a class-D audio amplifier into a very small, easy-to-use, and easy-to-implement package.

The combination of the audio controller IC with the audio MOSFETs results in improved efficiency, THD, and EMI. The IR4312M has high voltage ratings and noise immunity to ensure reliable operation and prevents audio clipping at high volume.

To read module specifications and reference designs, click here.

Watch Video

Audio Input Jack

PowIRaudioIC

Power Terminal

Stereo Outputs

To watch a video overview and demonstration of the IR4312M, click the image below:

Page 56: EEWeb Pulse 122 - Crossbar

eeweb.com/register

Join Today

Printed Valentine

Profile Pic

Magnet Thief


Recommended