DIGITAL EBOOK

AN EEWORLD REFERENCE

AUTOMOTIVE ELECTRONICS
PUT PEDAL TO THE METAL
INSIDE:
13 Optimizing the Automotive Camera Development
S P O N S O R E D B Y:
02 Touch Screen EMI/ESD Protection Phase Using a Modular Design Approach

06 Evaluating Functional Safety in 15 Autonomous Driving – Getting it Right Before

Automotive Image Sensors Handing Over the Keys

09 TVS Diode Selection Guidelines for 17 Automotive Imaging Power Architecture

the CAN bus and Design
 AUTOMOTIVE ELECTRONICS PUT PEDAL TO THE METAL

TND6033/D

TOUCH SCREEN EMI/ESD PROTECTION

INTRODUCTION hardware, their interfaces can be integrated touch panels are often used for smaller
throughout the usage cycle. screen sizes than surface capacitive touch
The touch screen revolutionized the panels. The internal structure of these touch
computing industry user interfaces and user Projected capacitive touch screens like those panels consists of a substrate incorporating
experiences especially after the perfection first featured by leading smartphone and an IC chip for processing computations, over
of multi-touch technology. A major force tablet manufacturers is as high as 54% of the which is a layer of numerous transparent
behind the resurrection of the once sleepy touch market in 2011. They also will remain electrodes is positioned in specific
touch screen industry is the spread of touch the dominant implementation for the space patterns. The surface is covered with an
panels and the benefits they offer in the way in the years to come, ahead of other touch- insulating glass or plastic cover. When a
of intuitive operation. Since they can be used sensor technologies like infrared, optical, finger approaches the surface, electrostatic
for input through direct contact with icons resistive and surface acoustic wave. capacity among multiple electrodes changes
and buttons, they’re easy for unaccustomed simultaneously, and the position were
users to comprehend and the ease of usage. contact occurs can be identified precisely by
Touch panels also permit miniaturization and T R E N D S I N TO U C H S C R E E N S
simplification of devices by combining display
and input into a single piece hardware. There are five main touch screen
Since touch panel buttons are software, not technologies: resistive, surface capacitive,
projected capacitive, surface acoustic wave,
and infrared. In terms of cost and size, the
first three suit mobile products. In all cases,
the system consists of a sensing mechanism,
a control circuit, and an interface to the
control circuit. For the purpose of this paper
only Projected Capacitive and Resistive
Touch Screen EMI/ESD are discussed.

Projected Capacitive Touch Panels

Leading smartphone and tablet
manufacturers adopt this method to achieve
Figure 1. Worldwide Shipment Forecast of
Touch-Screen Controller (SOURCE: IHS iSuppli
high-precision multi-touch functionality and
Research, March 2012) high response speed. Projected capacitive Figure 2. Projected Capacitive Touch Screen

SPONSORED BY 2
 AUTOMOTIVE ELECTRONICS PUT PEDAL TO THE METAL

TOUCH SCREEN EMI/ESD PROTECTION

CONTINUED

measuring the ratios between Resistive Touch Panels The voltage measured is determined by the
these electrical currents as A resistive touch screen voltage divider developed at the point of
shown in Figure 2. works by applying a voltage touch. For this measurement, the horizontal
across a resistor network panel resistance in the X+ lead doesn’t
A unique characteristic of a and measuring the change affect the conversion due to the high input
projected capacitive touch in resistance at a given impedance of the A/D converter. Voltage is
panel is the fact that the point on the matrix where a then applied to the other axis, and the A/D
large number of electrodes screen is touched by an input converts the voltage representing the X
enables accurate detection stylus, pen, or finger. The position on the screen through the Y+ input.
of contact at multiple points Figure 3. 4-wire Touch change in the resistance ratio This provides the X and Y coordinates
(multi-touch). However, the Screen Construction marks the location on the to the associated processor as shown in
projected capacitive touch touch screen. The two most Figure 4.
panels featuring indium-tin- popular resistive architectures
oxide (ITO) found in smart use 4-wire or 5-wire
phones and similar devices configurations as shown in
are poorly suited for use Figure 3.
in large screens, since
increased screen size results A 4-wire touch screen is
in increased resistance constructed as shown in
(i.e., slower transmission of Figure 3. It consists of two
electrical current), increasing transparent resistive layers.
the amount of error and noise The 4-wire touch screen panel
in detecting the points touched. works by applying a voltage across the
vertical or horizontal resistive network.
Larger touch panels use center-wire projected The A/D converts the voltage measured
capacitive touch panels in which very thin at the point the panel is touched. A
electrical wires are laid out in a grid as a measurement of the Y position of the
transparent electrode layer. While lower pointing device is made by connecting
resistance makes center-wire projected the X+ input to a data converter chip,
capacitive touch panels highly sensitive, they are turning on the Y+ and Y− drivers, and Figure 4. Wire Simplified Schematic
less suited to mass production than ITO etching. digitizing the voltage seen at the X+ input.

SPONSORED BY 3
 AUTOMOTIVE ELECTRONICS PUT PEDAL TO THE METAL

TOUCH SCREEN EMI/ESD PROTECTION

CONTINUED

The benefits of this design is unparallel than the sensor through the mobile device during
other touch screen design; include the low- the touch event. It can be seen as degraded
cost manufacture, due to the simplicity of accuracy or linearity of touch, false or phantom
structure. This design consumes less power touches, or just simply erratic behavior. The
and is strongly resistant to external impurity culprits include marginal layouts, poor antenna
such as dust and water since the surface is designs, and poor peripherals. These prove
covered in film. In addition, input relies on to be significant sources of EMI/ESD. With
pressure being applied to the film. It can PCB real estate continuing to be at a premium
be used for input not just with bare fingers, within the mobile device, layouts and designs
but even when wearing gloves or using a can sometimes be compromised: components
stylus. These screens can also be used to are literally being placed on top of each other.
input handwritten text. Drawbacks include This sometimes leads to antennas being
Figure 5. Projected Capacitive Touch Screen
lower light transmittance (reduced display placed in extremely closes proximity to the
with External ESD Protection
quality) due to the film and two electrode touch sensors, and can create problems with
layers. Relatively lower durability and shock the radios falsely activating the touch sensor.
resistance. And reduced precision of detection Common mechanical countermeasures such as
with larger screen sizes. (Precision can be metal shields no longer are effective solutions
maintained in other ways − for example, for resolving the EMI problems.
splitting the screen into multiple areas for
detection.) On chip protection solutions for CMOS ICs
often include high-voltage suppression
The market trend continues to produce thinner circuits on input and output pins. These are
mobile devices. This means projected and active RC shunts which turn on whenever
capacitive touch screens are the designs the pad voltage goes outside of the normal
of choice, as they allow direct lamination operating range. However, the effectiveness
of capacitive touch sensors to the display, is limited by gate oxide thickness and size; as
Figure 6. Projected Capacitive Touch Screen
migration of the sensor inside the display, and withExternal LC Filer + Integrated ESD Protection they dissipate high voltage energy, some of
so comes with many other challenges of EMI/ that energy is converted to heat can cause a
ESD with antennas, charger and ground loading. thermal event within the chip itself. In addition, these internal ESD
Unwanted EMI/ESD is one of the biggest concerns related to circuits have a maximum allowable current density proportional to
capacitive touch screens. This is noise that is physically coupled into their size. In an ESD event, as maximum current density is reached,

SPONSORED BY 4
 AUTOMOTIVE ELECTRONICS PUT PEDAL TO THE METAL

TOUCH SCREEN EMI/ESD PROTECTION

CONTINUED

the forward voltage potential across the circuit CONCLUSION

begins to rise. A highly energetic ESD event can
cause this potential to rise above the circuit’s Projected capacitive touch screens like those first
maximum allowable forward voltage, causing featured by smartphones will remain the dominant
damage, the IC’s internal ESD protection circuits implementation for the space in the years to come,
melt or break down (primary failure), and high ahead of other touch-sensor technologies like
energy is passed without attenuation to the infrared, optical, resistive and surface acoustic
device’s internal circuits, damaging them as well wave. Design considered trade offs are a direct
(secondary failure). Because of the speed of the function of the die area (and thus product cost)
ESD event (most last less than a microsecond), devoted to the internal protection circuits versus
the damage from heating will tend to stay within the cost of adding extra protection devices outside
the local region of the IC’s ESD protection circuit. the capacitive sensing controller for those lines in
Also, the amount of energy any single protection the system which may be more vulnerable to ESD
Figure 8. 8-wired Touch Screenwith
circuit can carry is directly proportional to its events. Impedance of the integrated circuit’s pin is
External ESD Protection
size. ESD immunity tests introduce a single, a significant factor in the usefulness of external ESD
measured amount of energy to the IC. As a protection circuits.
result, if that energy can be spread between
several protection circuits both internally and External circuits can be designed to work
external ESD protection circuits the likelihood synergistically with the complex impedance of
of damage is greatly reduced. To do this, the IC’s pad circuitry and its packaging. The
additional external ESD protection devices are combination can provide an effective level of ESD
implemented shown in Figures 5 and 6. control that is hard to achieve by using only on-
chip, integrated protection circuitry. Levels of ESD
protection for CMOS circuits are based on a balance
between product cost and expected requirements
forprotection in production and end-use. For Figure 9. 8-wired Touch Screen
with LC Filter + Integrated ESD Protection
applications requiring high ESD immunity on some
lines, additional external protection can be provided
Figure 7. 4-wired
using inexpensive methods by employing single
Touch Screenwith channel ESD protection or low pass filer couple with
External ESD ESD circuit plastic package.
Protection

SPONSORED BY 5
 AUTOMOTIVE ELECTRONICS PUT PEDAL TO THE METAL

TND6233/D

EVALUATING FUNCTIONAL SAFETY

IN AUTOMOTIVE IMAGE SENSORS

ABSTRACT Underlying this trend is the fact that to ensure

a level of safety in ADAS and autonomous
Almost all Advanced Driver Assistance Systems (ADAS) both today and in the systems, the image sensor becomes a critical
foreseeable future are built primarily on machine vision to drive the decision component in the system’s overall functional
process. With the rapid proliferation of ADAS solutions and the introduction of the safety. With the introduction of ISO 26262, the
ISO 26262 safety standard for passenger vehicles, functional safety considerations concept of automotive safety integrity levels has
for those imaging systems becomes paramount. The manner in which safety been defined. ASIL levels range from the lowest,
measures are implemented and verified can have significant impact to the overall ASIL−A (lowest), to ASIL−D (highest). An ASIL
system design including cost, reliability, and complexity. This paper will examine level is determined by three factors, severity of
functional safety in the imaging subsystem and it implications to system design. a failure, the probability of a failure occurring,
and the ability for the effect of the failure to be
controlled. This paper will explore the issue of
INTRODUCTION functional safety as it relates to the image sensor
as well as to examine failure modes and safety
The first rear view cameras appeared features like lane keeping assist, mechanisms that can be implemented to detect,
in vehicles as early as 1991, primarily as adaptive cruise control, and automated protect, and/or correct image sensor failures. The
an aid to safe reversing. In 2004, ON braking for collision avoidance evolve key metrics that affect the safety performance of
Semiconductor introduced the first CMOS into true autonomy, additional cameras the system include detection, delay, efficiency,
sensor for automotive applications. In are making their way into production and effect.
the United States the National Highway vehicles. The primary sensor in almost
Transportation Safety Administration all ADAS systems is the image sensor. Faults in semiconductor devices arise from a
(NHTSA) mandated that by May 2018, all new As ADAS systems progress from number of causes including cosmic radiation,
passenger vehicles are required to include assistance to automation, the safe electromigration, early mortality, and a host
back-up cameras. Auto makers are now operation of the vehicle will depend of other reasons. It is not the objective of this
incorporating increasing levels of autonomy more and more on the reliability of the paper to examine the causes of faults in image
to further improve vehicle safety. As ADAS imaging subsystem. sensors, rather it will be to examine the nature of

SPONSORED BY 6
 AUTOMOTIVE ELECTRONICS PUT PEDAL TO THE METAL

EVALUATING FUNCTIONAL SAFETY IN AUTOMOTIVE IMAGE SENSORS

CONTINUED

faults, methods of fault detection i.e. safety functional safety, specifically, that safety to actually correcting a fault that has already
mechanisms, and the effectiveness of those starts at the source. occurred. A careful and balanced selection of
mechanisms. Here we will also discuss some system components can contribute to a more
factors that differentiate fault coverage In order to minimize the risk caused by optimized and efficient
claims and methodologies. a particular failure, the system designers implementation.
must, of course, identify possible failure
modes that could affect the safety of the In order to further consider the concept
FUNCTIONAL SAFETY vehicle and determine an appropriate of functional safety, a definition of failure
action to mitigate the risk. A key part of must be made. In the case of ADAS, we
To implement functional safety in an ADAS this process is the identification of all can generally accept that a failure is any
system requires that the system prevent components that could impact the safe condition that causes the system to make an
or mitigate any action or behavior that operation of the system. For each such incorrect or even less than optimal decision.
could cause harm. The assessment of the component, an analysis of every possible Examples of undesired decisions include
probability of harm and the severity of that failure mode must be made to determine late braking, over-steering, false object
harm caused by a failure in the system whether a given failure mode might identification, or unintended acceleration.
allows system designer to classify levels contribute to a failure in the system. Once
of risk the system and to take appropriate the failure modes have been identified,
measures to minimize the risk. mechanisms can be implemented to detect, IMAGE SENSORS AND
correct, and/or protect the system from, a T H E A DA S A P P L I C AT I O N
Often this requires fundamental changes given failure.
not only in the development process, but Image sensors are a core component of an
also in the corporate safety culture ranging The specific implementation of safety ADAS system and the primary source of all
from organization structure to safety roles/ mechanisms for a given system has a vision system data. They provide the raw
officers and safety documentation, manuals, tremendous impact on the cost, reliability data which the rest of the system uses to
and standards. Responsibility for functional and complexity of the solution, as well as analyze the environment and then make
safety compliance in a system involves its effectiveness at mitigating the risks operational decisions in the vehicle. In
not only the ADAS supplier, but the entire to the system. Various levels of safety effect, image sensors are the eyes of the
supply chain from OEM to ADAS supplier to mechanisms can be implemented that autonomous vehicle. Other sensors like radar
component providers. For robust functional range from simple fault detection and and lidar may also be used, but the primary
safety, all key safety relevant components reporting, to mechanisms that protect source of data are the image sensors. In
in the system must contribute to the overall from the occurrence of faults all the way addition to the sensors, other components

SPONSORED BY 7
 AUTOMOTIVE ELECTRONICS PUT PEDAL TO THE METAL

EVALUATING FUNCTIONAL SAFETY IN AUTOMOTIVE IMAGE SENSORS

CONTINUED

Figure 1. Block Diagram of a Typical Image Sensor

in the ADAS system include components that

perform the functions of image processing,
analysis and decision making.

As stated previously, the number of image

sensors in a typical ADAS system is rapidly
growing. From a single forward looking camera,
to full surround view systems, the number of
cameras in a vehicle can be anywhere from one
(1) to over ten (10). The effect of failures in the
sensor depend on the nature of the failure
and can range from insignificant to critical.
The ability for a system to detect, protect and complexity, these decisions have advanced from Implied are any problems in the
correct individual failures in the image sensor generating simple audible and visual warnings, to internal operation of the device, either
have significant ramifications to the overall much more complex decisions including braking, analog or digital, that could manifest
safety and reliability of the system. acceleration and steering, and in the future will themselves as pixel, row, column,
progress to completely autonomous driving. These or frame errors. Finally, errors in the
At its core, a CMOS image sensor is a advances in autonomous and semi-autonomous physical transmission of data from
rectangular array of photo-sensitive pixels vehicles places increasing reliance on the image the sensor to the rest of the system
organized in rows and columns. These pixels sensor and it safe operation. present another potential cause for
convert the incident light into voltage or current failure. Due to the dynamic nature of
with a per-pixel analog circuit. The current/ video, faults can be both static, e.g.
voltage is then converted into digital values, FA I L U R E S I N I M A G I N G A P P L I C AT I O N S permanent or fixed, and dynamic, both
typically in a row-by-row order. Additional digital spatially and temporally.
logic enables the data to be stored, processed, A very conservative view of a failure in an image
and transmitted to other devices in the system sensor would be to define an unsafe fault as any
for subsequent processing and analysis. output that differs from a “fault-free” model or
known-good device output as shown by the
CLICK for the
complete

HERE
The data captured by the image sensor in an diagram below. At a granular level, this would
technical
ADAS application is typically used by the system imply that errors even at the pixel level could note
to make decisions that affect the operation of constitute a failure. At higher levels, row, column
the vehicle. As ADAS systems have increased in and frame errors could also constitute a failure.

SPONSORED BY 8
 AUTOMOTIVE ELECTRONICS PUT PEDAL TO THE METAL

AND8181/D

TVS DIODE SELECTION

GUIDELINES FOR THE CAN BUS
INTRODUCTION area of a PN junction to provide robust as MOVs and choke filters. A TVS diode
protection against transient EMI surge provides a fast response time, low line
The Controller Area Network (CAN) is a voltage and ESD. Figure 1 provides a circuit capacitance and low clamping voltage. The
serial communication protocol designed diagram of the NUP2105L. NUP2105L has a time response time of less
for providing reliable high speed data than 1.0 ns and is able to clamp fast surge
transmission in harsh environments. This The NUP2105L has been tested to EMI and transient voltages before damage can occur.
document provides guidelines to select ESD levels that exceed the specifications of The silicon design has a line capacitance
Transient Voltage Suppression (TVS) diodes popular high speed CAN networks. Listed less than 30 pF, which is required for the
to protect CAN data bus lines. TVS diodes below is a summary of the NUP2105L’s EMI high 1.0 MHz data transmission rate. The
provide a low cost solution to conducted and ESD specifications: voltage clamping limit of the device, defined
and radiated Electromagnetic Interference • 350 W Peak Power Dissipation per line, by the 8 x 20 s exponential waveform, is
(EMI) and Electrostatic Discharge (ESD) noise (8 x 20 s) approximately equal to 42 V for a surge
problems. The noise immunity level and • Human Body Model ESD protection, current of 10 A. The low clamping voltage
reliability of CAN transceivers can be easily 16 kV ensures that the transient sure voltage will
increased by adding external TVS diodes to • IEC−61000−4−2 ESD level, 30 kV for not exceed the CAN transceiver’s maximum
prevent transient voltage failures. contact discharge voltage specification for the CAN_H and
• ISO 7637−1, nonrepetitive EMI surge CAN_L data lines.
pulse 2, 9.5 A (1 x 50 s)
UP2105L CAN BUS TVS DIODE • ISO 7637−3, repetitive Electrical Fast
A R R AY Transient (EFT) EMI surge pulses, 50 A
(5 x 50 ns)
The NUP2105L provides a transient • IEC 61000−4−5 lightning and load switch
voltage suppression solution for CAN data immunity, 10 A (8 x 20 s)
communication lines. The NUP2105L is a
The NUP2105L uses silicon semiconductor Figure 1. NUP2105L
dual bidirectional TVS device in a compact Bidirectional TVS/ESD
SOT−23 package. This device is based on technology to offer distinct advantages over Protection Device
Zener technology that optimizes the active alternative TVS protection devices such

SPONSORED BY 9
 AUTOMOTIVE ELECTRONICS PUT PEDAL TO THE METAL

TVS DIODE SELECTION GUIDELINES FOR THE CAN BUS

CONTINUED

Figure 2 provides an example of a typical CAN 2. Reverse Breakdown

bus protection circuit. The circuit provides Voltage (VBR) is the point
protection for the CAN_H and CAN_L data lines where the device conducts
by clamping the surge voltage to a level that in an avalanche mode and Figure 3. Bidirectional
will not damage the CAN transceiver. Further becomes a low impedance. TVS Characteristic

details on CAN protection circuits are provided The breakdown voltage Curve

in reference (1). is typically measured at a

current of 1.0 mA.
3. Maximum Clamping Voltage
(VC) is the maximum voltage
drop across the diode at the
maximum peak pulse current.
Figure 2. High− 4. Reverse Leakage Current (IR)
Speed and Fault is the current measured at the
Tolerant CAN TVS
Protection Circuit
reverse working voltage.
5. Test Current (IT) is the current
where the breakdown voltage
is measured.
TVS DIODE TERMINOLOGY 6. Peak Pulse Current (IPP) is the
maximum surge current specified for the 4. ESD
The first step in selecting a TVS diode device device. 5. EMI Immunity
is to define the device parameters. Figure a. Coupled Electrical Disturbance
3 provides a graphical definition of the on the Data Lines
bidirectional TVS diode parameters. T R A N S C E I V E R S P E C I F I C AT I O N S i. Nonrepetitive Surge
ii. Repetitive Surge / Electrical
The key TVS parameters are: There are several CAN transceiver Fast Transient (EFT)
1. Reverse Working Voltage (VRWM) is specifications that must be evaluated in order
defined as the maximum DC operating to select an appropriate TVS diode. The critical Table 1 provides a summary of
voltage. At this voltage the device is in a transceiver characteristics include: the system requirements for a
non−conducting state and functions as 1. Maximum supply voltage CAN transceiver. The ISO 11898−2
essentially a high impedance capacitor. 2. Common mode voltage physical layer specification forms
VRWM is also known as the stand−off voltage. 3. Maximum transmission speed the baseline for most CAN systems.

SPONSORED BY 10
 AUTOMOTIVE ELECTRONICS PUT PEDAL TO THE METAL

TVS DIODE SELECTION GUIDELINES FOR THE CAN BUS

CONTINUED

The transceiver requirements for the battery power lines and CAN signal lines. of the NUP2105L provides predictable
Honeywell Smart Distribution Systems (SDS) In addition, some applications often have device performance over potential system
and Rockwell (Allen−Bradley) DeviceNetE unique short duration maximum supply deviations. Figure 4 shows the VBR versus
high speed CAN networks are similar to ISO voltage specifications. For example, some IT characteristics of the NUP2105L over a
11898−2. The SDS and DeviceNet transceiver 12 V automotive systems have the provision temperature range of −55°C to +150°C.
requirements are similar to ISO 11898−2; of allowing a jump start from a 24 V battery.
however, they include minor modifications The minimum VRWM and VBR of the
required in an industrial environment. NUP2105L are specified at 24 V and 26.2 V,
respectively.

M A X I M U M S U P P LY V O LTA G E The NUP2105L has a nominal VBR of 27

V which is measured with a 1.0 mA, 1.0
The TVS diode VRWM and VBR should be ms pulse test current. The TVS’s Zener
greater than the maximum system supply technology produces a breakdown voltage
voltage because the transceiver must be characterized with a sharp knee and very
immune to an indefinite short between the low leakage current. The sharp knee

TA B L E 1 . T R A N S C E I V E R R E Q U I R E M E N T S F O R H I G H − S P E E D C A N N E T W O R K S Figure 4. VBR versus IT Characteristics of the NUP2105L

C O M M O N M O D E V O LTA G E
The common mode voltage specification is
required because there can be a significant
difference in the voltage potential between
the ground reference of the transmitting and
receiving CAN nodes. The CAN transceivers
must be able to function with a signal line
voltage that can be offset by as much as
2.0 V above or below the nominal voltage
level of the CAN_H and CAN_L signal lines.

SPONSORED BY 11
 AUTOMOTIVE ELECTRONICS PUT PEDAL TO THE METAL

TVS DIODE SELECTION GUIDELINES FOR THE CAN BUS

CONTINUED

A solution to the common mode problem is voltage of the high−speed and fault tolerant CAN EMI IMMUNITY TESTS
to use bidirectional TVS devices that will not CAN transceivers can be estimated to be
clamp if the voltage at the signal lines equal to the recessive voltage of 2.5 V. The Background
is offset. typical capacitance of the NUP2505L is Electromagnetic Compatibility (EMC) has
approximately 19 pF at 2.5 V. become a major design concern for network
products. Designers are being challenged to
MAXIMUM TRANSMISSION SPEED include EMC protection, without incurring a
size and cost penalty. CAN modules must be
The CAN data transmission rate determines must be compliant with strigent EMI standards
the maximum capacitance of the TVS device. and operate without either becoming effected
A large capacitance on the data lines causes by or adversely effecting the operation of
distortion in the signal waveforms. The neighboring units. CAN networks must have
distortion on the data lines is minimized by good noise immunity because the data lines
selecting a low capacitance TVS device. It is are a major source and entry point for both
recommended that the maximum capacitance conducted and radiated EMI and ESD.
of the protective network measured from
each signal line to ground should be less
Figure 5. IR versus Temperature
than 35 pF for 1.0 Mbits/s and 100 pF for 125 Characteristics of the NUP2105L
kbits/s.

The capacitance versus bias voltage

relationship of the NUP2105L is shown in CLICK
Figure 6. The capacitance between the
signal lines and ground was measured HERE
by varying the DC bias, at a frequency of for the
1.0 MHz and peak−to−peak amplitude of complete
technical note
60 mV. A diode’s data sheet specifies the
maximum capacitance at a bias voltage of 0
V; however, the average voltage of the data
lines will provide a more accurate estimation
of the capacitive loading. The average DC Figure 6.Capacitance versus VR
Characteristic of the NUP2105L

SPONSORED BY 12
 AUTOMOTIVE ELECTRONICS PUT PEDAL TO THE METAL

BLOG POST

OPTIMIZING THE AUTOMOTIVE

CAMERA DEVELOPMENT PHASE USING
A MODULAR DESIGN APPROACH
AS THE AUTOMOTIVE INDUSTRY
progresses rapidly toward fully autonomous
vehicles, the number and varieties of imaging
cameras are increasing. Various camera
applications include backup, surround view,
forward facing Advanced Driver Assistance
Systems (ADAS), Camera Mirror Replacement
Systems (CMS), and driver monitoring.
Amazingly, some vehicles are already being
equipped with more than twelve image
sensors.

Each individual camera has its own unique

features ranging from different resolutions,
optical formats, Color Filter Arrays (CFA), often leaves designers ill-equipped to develop Understanding this, ON Semiconductor
lens types, Automotive Safety Integrity Level in real-world scenarios. At present, automotive has created an innovative development
(ASIL) functions and transmission protocols. Original Equipment Manufacturers (OEM), Tier tool called the Modular Automotive
Dramatic market growth combined with 1’s and Tier 2’s, and Independent Software Reference System (MARS). MARS is
multiple camera types makes streamlining Vendors (ISV) spend tens to hundreds of capable of mixing and matching the
development imperative. thousands dollars in the camera design phase. various imaging components and
A modular compact development system transmission protocols using standard
While the industry with the flexibility to adapt to the required boards and connectors. It consists of
in general provides application and the ability to easily mount on set of miniature 25mm2 boards with
sufficient tools to the vehicle has the potential to dramatically each board containing one or more of
evaluate cameras in reduce cost, make resources more efficient and the major imaging components in the
a laboratory setting, it improve time-to-market. system.

SPONSORED BY 13
 AUTOMOTIVE ELECTRONICS PUT PEDAL TO THE METAL

OPTIMIZING THE AUTOMOTIVE CAMERA DEVELOPMENT

PHASE USING A MODULAR DESIGN APPROACH
CONTINUED

HELPFUL LINKS
The imaging industry almost universally FOR MORE INFORMATION:
recognizes ON Semiconductor’s sensor/
ISP evaluation tool, called Devware,
as best-in-class. Using a matching
deserializer, and adapter the designer
can use our world-class evaluation
tools in the real-world by mounting the
miniaturized MARS camera boards to the
Every single automotive qualified image sensor vehicle. Additionally, if the designer has
(or SoC) from ON Semiconductor has a MARS an existing ECU system they can connect
board, meaning that designers can interchange the MARS camera directly. Since MARS
this board nearly seamlessly to test all of our uses known vendors and transmission
watch the Introduction to MARS webcast
automotive sensor types. All of the boards standards, bring-up time for the camera
can be mixed and matched to the automotive requires little to no effort.
imaging system design team’s desire. The
component boards have consistent signal/
power interconnect definitions to enable users
MARS bridges the gap between
laboratory and real-world development. CLICK to configure your
product with the
to swap individual boards, creating a wide range
of options for experimenting, while eliminate the
It’s a complete development system
capable of truly optimizing the automotive
HERE MARS product
recommendation tool.
need for constructing custom boards. camera design.

SPONSORED BY 14
 AUTOMOTIVE ELECTRONICS PUT PEDAL TO THE METAL

BLOG POST

AUTONOMOUS DRIVING – GETTING

IT RIGHT BEFORE HANDING OVER
THE KEYS
RELINQUISHING HUMAN CONTROL of the car on
a mass scale is, for sure, a major step to take. There
are emotional/human acceptance as well as technical
barriers to overcome. Succeeding in the latter needs
to be based on a huge amount of data, proving and
validation. Accumulating all of this through physical,
‘on the highway’ testing alone will take a long time and
likely slow the realization of the widespread adoption
of autonomous cars and other vehicles. Interestingly,
large amounts of proving of the effectiveness and
safety of autonomous driving technology, with the
right publicity, should also help address the human
apprehension element – especially for those with many
years and a love of ‘hands-on’ driving.

The time it takes to gather real-world data from

autonomous vehicles on the road is why tools such
as NVIDIA’s DRIVE™ Constellation – an open, cloud-
based platform that can perform bit-accurate simulation
for large-scale, hardware-in-the-loop testing and
validation of autonomous vehicles – are so important.
The viability and success of such platforms in
supplementing and corroborating real-world testing
for the proven, safe and earlier realization of driverless
cars, is dependent upon collaboration with specialists
in specific technologies that are cornerstones to the
implementation of autonomous vehicles.

SPONSORED BY 15
 AUTOMOTIVE ELECTRONICS PUT PEDAL TO THE METAL

AUTONOMOUS DRIVING – GETTING IT RIGHT

BEFORE HANDING OVER THE KEYS
CONTINUED

ON Semiconductor, with its leading image processing. The complex sensor model
sensor knowhow and technologies, is one will utilize all critical parameters in the
such company. Its recent announcement path to provide an accurate output of a
of an agreement to work with NVIDIA real-world image sensor.

WANT
underlines its reputation and will see Check out the Avnet
the company’s CMOS image sensor The changes to our everyday lives Automotive Blog for more
modelling technology being used as a could be massive, as too will be the resources about autonomous
primary function to simulating real-world
sensor performance within the DRIVE
effect on the safety and environmental
impacts of getting from A to B for millions MORE? driving, sensor performance,
and more!

Constellation. across the globe. The ‘virtual proving’

provided by tools such as NVIDIA’s
The image sensor model receives both DRIVE Constellation and supported by
scene information and control signals collaborators such as ON Semiconductor,
from the DRIVE Constellation to calculate will benefit those engaged in the ground-
and output a real-time image based on breaking work to take autonomous
the inputs. It then transmits the simulated driving from concepts under test to
image back to the DRIVE Constellation for mainstream reality.

VIDEO: Solve Leak Problems in New Vehicles

Over 3% of all new vehicles have leaks. Here we show a CLICK HERE
simple, inexpensive solution for solving this problem.
WAT C H N OW > >

SPONSORED BY 16
 AUTOMOTIVE ELECTRONICS PUT PEDAL TO THE METAL

AND9708/D

AUTOMOTIVE IMAGING POWER

ARCHITECTURE AND DESIGN
INTRODUCTION
Imaging system input and intermediate 1. Conduction Losses
voltages play a critical role in total
system efficiency, performance, and
component selection. Proper selection
of these voltages depend on multiple
system variables, such as: total power 2. DC Voltage Drop
consumption of the image sensor/ISP,
regulator efficiency, and resistance of
the coax cable and filter. This document
discusses the trade−offs associated
with changing module input voltages, so Both of these concerns are factors of
that informed design−decisions can be the DC transfer current, DCR of the filter Figure 1. General Overview of an Automotive
made in optimizing automotive imaging inductors at each coax connector, and Imaging System Organized into Three Section

modules. coaxial cable resistance. Higher currents

contribute to higher power losses and
larger DC voltage drops. Moreover, a large
S Y S T E M P OW E R OV E R V I E W enough DC voltage drop could induce
under voltage−lockout in downstream
Power Transfer regulators and shut down the system.
In an automotive imaging system, power
and data are combined and transferred to Decreasing the transfer current (Iin) will
remote parts of the vehicle over a single minimize both conduction losses and the
coaxial cable. This implementation is DC voltage drop, however, in order to
known as “Power over Coax” (POC). From decrease the transfer current, the voltage
a power standpoint, the two main figures transmitted over the coaxial cable must be
of merit for this stage are: increased to maintain the same amount of

SPONSORED BY 17
 AUTOMOTIVE ELECTRONICS PUT PEDAL TO THE METAL

AUTOMOTIVE IMAGING POWER ARCHITECTURE AND DESIGN

CONTINUED

transferred power. This is not always ideal, input voltage that results in the lowest
as increased coaxial voltage will lead to more total power loss. Again, both transfer and
losses in the power conversion of the image conversion losses are factors of component
sensor module. parameters.

Power Conversion Conversion Loss Component Factors

Integrated circuits in automotive imaging • Regulator Efficiency at specific Vin/Vout
systems (Image Sensors, Image Processors, ratio
SERDES links) normally utilize the voltages Figure 2. Power • Voltage Conversion Topology
2.8 V, 1.8 V, and 1.2 V. Because of these Conversion
Diagram
multiple, low−voltage rails, it is necessary to
convert down from a single higher coaxial SYSTEM PERFORMANCE
voltage. The primary power concern in this
stage is voltage conversion losses. These Altering the system’s input voltage can
losses can be quantified as: pose performance risks to the system if the
voltages are defined by the image sensor, effects are not properly realized. A changing
this ratio can be decreased by lowering the input voltage alters the transfer current
input voltage to the power conversion stage. levels which affects voltage regulator
Unfortunately, decreasing the input voltage transients, filter performance, and dark
will increase the transfer current, resulting in current generation.
larger power transfer losses.
Where ηi is regulator efficiency, and Pi is the Transient Impact
regulator’s output power. Given a constant Transfer Loss Component factors Increased coaxial current consequently
output power, the loss is a factor of regulator • Inductor DCR increases voltage transients seen at the
efficiency. Furthermore, in both switching • Coaxial Series Resistance input of the Vmid Regulator. Large enough
and linear regulators, efficiency is a factor • Coaxial Connector Resistance transient events could impact the regulator’s
of input/output voltage ratio (Vin/Vout). The ability to maintain regulation, or even impact
larger the ratio, the worse the efficiency. The trade−off between transfer and the AC data traveling along the coaxial line.
conversion losses must be considered when To mitigate the effect of these increased
Lowering the VIN/VOUT conversion ratio choosing the proper input voltage. Optimally transients, sufficient capacitance should be
of each regulator will minimize power designed power architectures select an placed on the input of the Vmid Regulator.
conversion losses. Because the output

SPONSORED BY 18
 AUTOMOTIVE ELECTRONICS PUT PEDAL TO THE METAL

AUTOMOTIVE IMAGING POWER ARCHITECTURE AND DESIGN

CONTINUED

Signal Integrity
Another side−effect of larger coaxial current
is the need for larger POC filter inductors.
The POC filter is required to prevent the AC
coaxial data from reaching the input of the
Vmid regulator. If the inductors of the filter
are not properly sized, the higher currents
could saturate the inductors and lead to Figure 3. Efficiency
Graph of the
degradation of the AC signal integrity, or NCV890430 5−12Vin
even lose PLL (Phase Lock Loop) locking of Automotive Voltage
the serialized communication. Regulator. Note how the
input−to−output voltage
ratio drastically affects
Dark Current Temperature Variation the efficiency.
Dark current negatively impacts image
quality by altering the black level or
reference point of the image sensor.
Dark current is also highly temperature
dependent; larger temperatures near the
sensor lead to larger levels of dark
current. Minimizing the heating at or
around the image sensor is essential in

WANT
minimizing dark current levels, which in turn
ensures proper image quality. Dark Current

CLICK
generation is reduced by limiting power for the
dissipation in voltage regulators near the
image sensor. Ideally, dissipated system
power should be remote to the image sensor HERE
complete
technical
note
MORE?
Check out Avnet’s
such as the coaxial cable or filter. Within website for more
compact designs, lower conversion voltages informative articles about
Automotive Evolution.
and higher coaxial currents are preferred
as because it minimizes heating around the CLICK HERE >>
image sensor, which reduces image sensor
dark currents.

SPONSORED BY 19
 AUTOMOTIVE ELECTRONICS PUT PEDAL TO THE METAL

ON Semiconductor (Nasdaq: ON) is driving energy efficient

innovations, empowering customers to reduce global energy use.
The company is a leading supplier of semiconductor-based solutions,
offering a comprehensive portfolio of energy efficient power C O N TA C T U S
management, analog, sensors, logic, timing, connectivity, discrete,
SoC and custom devices. The company's products help engineers www.onsemi.com
solve their unique design challenges in automotive, communications,
www. avnet.com/transformon
computing, consumer, industrial, medical, aerospace and defense
applications. ON Semiconductor operates a responsive, reliable,
world-class supply chain and quality program, a robust compliance
ON SEMI SOCIAL MEDIA
and ethics program, and a network of manufacturing facilities, sales
offices and design centers in key markets throughout North America,
Europe and the Asia Pacific regions.

From idea to design and from prototype to production, Avnet supports AV N E T S O C I A L M E D I A

customers at each stage of a product’s lifecycle. A comprehensive
portfolio of design and supply chain services makes Avnet the go-to
guide for innovators who set the pace for technological change. For
nearly a century, Avnet has helped its customers and suppliers around
the world realize the transformative possibilities of technology.