Atmel Autocompilation Vol7 Dec2010

Automotive Compilation
Volume 7 December 2010
Contents
Table of Contents
Configurable RF Architecture Gives Engineers Greater Design Flexibility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Design and Security Considerations for Passive Immobilizer Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1 3
Automotive LIN Bus Driving Sensor Applications . . . . . . . . 12 Designing Next Generation Key Fobs . . . . . . . . . . . . . . . . . . 15 Capacitive Proximity Detection in the Automotive Industry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 Extending the Range of E-vehicles with Li-ion Batteries Using Active Balancing . . . . . . . . . . . . . . . . . . . . . . 24 Precision Battery Monitors for Standard Lead Acid Batteries Ensure Engine Start-up . . . . . . . . . . . . . . . . . . . . . . 27 DC and BLDC Motor Control ICs . . . . . . . . . . . . . . . . . . . . . . . 29 EMC Synonym for Exasperating, Magic, Confusing? . . . 36 EMC Pulse Immunity System-level Tests . . . . . . . . . . . . . . . 41
Configurable RF Architecture Gives Engineers Greater Design Flexibility

Ahmad Chaudhry and Jim Goings
In the automotive environment, RFenabled sub-systems continue to evolve and proliferate. Such systems include Tire Pressure Monitoring Systems (TPMS), which are mainly unidirectional on a single RF channel at relatively high data rates; Remote Start (RS) systems, which are generally bidirectional on single or multiple RF channels at relatively low data rates; Passive Keyless Entry (PKE) and Remote Keyless Entry (RKE) applications, which are unidirectional on single or multiple RF channels at moderate data rates. TPMS uses both On Off Keyed (OOK) and Frequency Shift Keyed (FSK) modulation, RS uses OOK only, and RKE uses either OOK or FSK. To accommodate such multiple systems and applications, automotive RF semiconductor devices must possess flexible as well as configurable architectures. The need for architectural flexibility, combined with customer expectations of high performance, improved range, and reliability, is driving the next generation of RF IC designs. Beginning with range and reliability, improvements here are easier to realize when the transmit path has programmable parameters. One can
1
usefully employ a Power Amplifier (PA) with reserve capacity, which can be trimmed in output power level to provide the maximum allowable output in accordance with local regulatory requirements. Further benefit can be obtained from a PA whose output impedance can be trimmed to optimize the antenna match. Range and reliability gains on the receive side can be enhanced through proper selection of sensitivity-related parameters, such as RF carrier frequency, subchannels, modulation, data rate, and IF bandwidth.
and RS. The devices also support all automotive bands: 310 to 318MHz, 418 to 477MHz, and 836 to 928MHz and use one device and a single crystal frequency. These two devices are also designed for architectural flexibility. A dualLNA architecture with two separate input pins natively supports multiband applications using one single IC, PCB, and bill of materials (BOM). Additionally, dual-parallel demodulation paths support simultaneous ASK and FSK sensing capability. These features
Atmel Has Introduced Next-Generation Transceiver And Receiver Devices With Configurable Options
Flexibility for the design engineer is also gained through access to a receiver in which these parameters are available as a programmable option. Atmel has introduced next-generation transceiver and receiver devices with these configurable options. For example, the Atmel ATA5830 transceiver and Atmel ATA5780 receiver can accommodate automotive applications such as RKE, PKE, TPMS, accommodate multiple polling schemes including TPMS, RS and up to three RKE channels, and can be configured to support RF protocols across multiple frequency bands, modulation schemes, and data rates. To implement the vast array of configurable content in both Atmel devices, the desired configuration settings are stored in the built-in EEPROM and automatically applied to
2010 / www.atmel.com
32 31 30 29 28 27 26 25
ATEST _ 101 TEST _ 102 AGND PB7 PB6 PB5 PB4
1 1 1 3 1 4 3 5 4 6 5 7 6 8 7 8
ATEST _ 101
TEST _ 102
AGND
PB7
PB6
PB5
PB4
RFIN _HB RFIN _LB
PB3
RFIN _LB
32 31 30 29 28 27 26 25
PB2
24
SPDT_RX RFIN _HB SPDT_ANT SPDT_RX ANT_TUNE SPDT_ANT SPDT_TX ANT_TUNE RF_OUT SPDT_TX VS_PA RF_OUT VS_PA XTAL1 TEST_ EN
(Optional) (Optional)
ATA5830 Atmel ATA5830

XTAL2 XTAL2 AVCC PC0 PC1 VS XTAL1
Atmel
23 24 22 23 DGND PB0 21 22 DVCC DGND 20 21 PC5 DVCC 19 20 PC4 PC5 18 19 PC3 PC4 17 18
PB0 PB1 PC3
PB1 PB2
Elements Inductor Elements Capacitor Inductor Capacitor Crystal Crystal
Number
Comments
9 9
10 11 12 13 14 15 16
AVCC PC0 PC1
17
10 11 12 13 14 15 16 VS = 3V VS = 3V
2Number (3) Matching Comments 7 3 (5) for matching 2(9) (3) Matching 4 Blocking 7 (9) 3for (5) for matching 1 1 1 crystal for 3 4single for Blocking different frequency ranges 1 single crystal for 3 different frequency ranges
Figure 1. Atmel ATA5830 Transceiver Application Circuit
accessible firmware, making it possible to develop an entire application using just one single IC. Both the ATA5830 transceiver and the Atmel ATA5780 receiver are also highly integrated, requiring very few external components. The application circuits (see Figures 1 and 2) show a standard implementation of each device. For a typical application, the ATA5830 transceiver only requires 10 external elements and the ATA5780 only six external elements. Both devices are packaged in a 5 x 5mm, 32-pin QFN package.
PB3
PC2
ATEST _ 101
TEST _ 102
AGND AGND RX_Active RX_Active LEDO LEDO NPWRONG NPWRONG IRQ IRQ
NSS
PWPON LED1
1 1 1 3 1 4 3 5 4 6 5 7 6 8 7 8
ATEST _ 101
TEST _ 102
NSS
PWPON LED1
RFIN _HB RFIN _LB
MISO
RFIN _LB
32 31 30 29 28 27 26 25
MOSI
MISO
SPDT_RX RFIN _HB SPDT_ANT SPDT_RX N.C SPDT_ANT N.C N.C N.C N.C N.C N.C N.C XTAL1 XTAL1 TEST_ EN TEST_ EN
TEST_ EN
PC2
32 31 30 29 28 27 26 25
Atmel ATA5780 Atmel

ATA5780
NRESET XTAL2 XTAL2 AVCC VS NRESET AVCC
9 9
10 11 12 13 14 15 16 NPWRON2 17 10 11 12 13 14 15 16 VS = 4.5v to 5.5v VS = 4.5v to 5.5v

VS TRPA
Figure 2. Atmel ATA5780 Receiver Application Circuit
Atmel Devices Also Have Enhancements Which Simplify Design and Reduce BOM Cost
the device on power-up. This enables autonomous (stand-alone) operation and polling for incoming signals from multiple RF systems with differing RF carrier frequency bands, modulation formats, and data rates. Stand-alone operation allows an external controller to sleep while the device polls, validates start of frame, and checks for proper transmitter ID data. The device only wakes the controller when a valid message is detected. This is critical to reducing Ignition Off Draw (IOD) in vehicle-mounted applications and to extending battery life in handheld fob applications. Atmel devices also have some other enhancements which simplify design and reduce BOM cost. The Atmel ATA5830 transceiver device, for example, has an embedded Atmel AVR microcontroller on the same silicon die. The microcontroller includes 6kByte of Flash and a 24kByte ROM library of user-
VS
24
NSS MISO MOSI NSS MISO SCK MOSI SCK
23 24 CLK_OUT SCK 22 23 DGND CLK_OUT 21 22 DVCC DGND 20 21 NPWRON5 DVCC TRPB 19 20 NPWRON4 NPWRON5 TRPB 18 19 NPWRON3 TMDO NPWRON4 NPWRON2 17 18 TRPA
NPWRON3 TMDO
SCK MOSI
Microcontroller Microcontroller
Elements Inductor Elements Capacitor Inductor Capacitor Crystal Crystal
Number
Comments
1Number Matching Comments 4 1 for matching 1 Matching 4 1 1 3 1for forBlocking matching 1 crystal for 3 3single for Blocking different frequency ranges 1 single crystal for 3 different frequency ranges
In conclusion, the newest generation of configurable Atmel RF semiconductor devices provides the design flexibility needed for rapidly evolving automotive RF-enabled subsystems .
NPWRON1 NPWRON1
Automotive Compilation Vol. 7
Design and Security Considerations for Passive Immobilizer Systems

Jim Goings, Toby Prescott, Michael Hahnen, Karl Militzer For years, consumers have come to rely on the convenience and added security that a passive vehicle immobilizer system offers . These systems consist of a key fob, carried by the driver, and a base station, mounted in the vehicle . They work together to determine if the driver is authorized to start the vehicle . Of equal or greater importance is the systems ability to prevent unauthorized sources from using the vehicle . While top-level functionality of a vehicle immobilizer is simple to describe, the underlying technology enabling it is intriguing and sophisticated . This article will explore both the hardware and software aspects of vehicle immobilizer systems as well as offer noteworthy comments on design and security considerations .
Communication
The prevailing method of communication between key fob and vehicle in passive vehicle immobilizer systems today is with a modulated magnetic field. This field is created by the vehicles immobilizer base station from a low frequency alternating current, typically 125kHz. The magnetic field serves three fundamental purposes: A) the power source for the key fob, hence the term passive, B) a carrier on which information from the base station to key fob is conveyed, e.g. downlink, C) a carrier on which information from the key fob to the base station is conveyed, e.g. uplink.
authentication process, which has a direct impact on response time. The sections that follow will address these topics in greater detail.
System Interfaces
The system architecture of a vehicle immobilizer has several layers of abstraction, each representing different system interfaces. Figure 1 provides a visual representation of these layers.
Crytographic Protocol
Characteristics of magnetic fields that are of particular appeal for a vehicle immobilizer system pertain most to the key fobs need to operate completely passively, e.g. without a battery. Downlink field detection and uplink field modulation can both be achieved using circuitry that consumes very little current. Furthermore, harnessing sufficient field energy to power these circuits in the key fob electronics can be achieved with relative ease. During the system design phase, care must be taken to carefully consider key performance parameters such as key fob energy requirements, which affect antenna coil geometries and drive levels, and the
Logical Physical
Figure 1. Vehicle Immobilizer Interface Layers
The Physical Layer

A t the lowest level of a vehicle immobilizer system is the physical layer. It consists of a vehicle-mounted antenna coil capable of creating a magnetic field of sufficient magnitude to enable its detection and modulation by an antenna coil mounted in the users key fob.
Field Generation and Modulation

Vehicle immobilizer systems can be classified in one of two different ways based on how the magnetic field is used to support the transfer data: half duplex or full duplex. In a half-duplex system, the vehicle-mounted antenna coil alternates between periods of energy transfer and data transfer. Uplink data (e.g., fob to vehicle) is modulated using Frequency Shift Keying (FSK). A graphical representation of this communication method is shown in Figure 2. Two points should be intuitively obvious from viewing Figure 2. First, the rate of data transfer suffers from a significant compromise
due to the recurring need to perform the energy transfer, e.g., charge up the key fob. Second, the modulated signal is very small compared to the field present during the energy transfer period, making it more susceptible to interference from ambient noise which results in reduced range. These characteristics have caused the popularity of half-duplex systems to wane.
The dominant system in use today is the full-duplex system in which the vehicle-mounted antenna coil performs energy transfer AND data transfer simultaneously. Uplink data is modulated using Amplitude Shift Keying (ASK). A graphical
Magnetic Field Charges Key Fob
FSK Key Fob Return Frequency (red)

Figure 2. Half-duplex Communication Using FSK
FSK Key Fob Return Frequency (blue)

0
Magnetic Field Charges Key Fob
ASK Key Fob Return Signal - Resistively Damped ( Green ) ASK Key Fob Return Signal - Undamped ( Black )
Figure 3. Full-duplex Communication Using ASK
representation of this communication method is shown in Figure 3. Clearly, the ability to simultaneously transfer data while keeping the key fob energized, or charged, provides the design engineer a significant data transfer rate advantage over half-duplex systems. Additionally, the constant carrier field tends to mask out interferences and enables robust communication during the data transfer. Furthermore, this approach can be realized using simple envelope detection circuitry. Because of the popularity of the full-duplex vehicle immobilizer system in the market place today, the rest of this document will focus on this type of system.
to predetermined times; T0 for a logic 0 and T1 for a logic 1. The advantage to this approach is that it embeds energy transfer from vehicle to key fob into the data encoding and ensures the key fob will be supplied enough energy to process the encoded data. However, a side effect of this encoding method is that the data transfer baud rate depends on the logical value of the data bit stream being sent since the transmission times for each binary state are different. See Figure 4 for a more detailed graphical depiction of this coding method.
System Interfaces: the Logical Layer

1-bit value BPLM bit frame example
Tgap
The next layer above the physical layer is the logical layer. This layer captures the characteristics and requirements for the coding and transfer of data across the magnetic field. It applies to the bi-directional data transfer that takes place from vehicle to key fob, commonly referred to as downlink, as well as key fob to vehicle, also known as uplink.
LF-Field
0
12 x CLKFC
Downlink
Downlink information is coded using Pulse Length Modulation; typically Binary Pulse Length Modulation (BPLM) or Quad Pulse Length Modulation (QPLM). This method is based on inserting a carrier field gap, Tgap, of fixed duration and setting the gap to gap timing intervals
LF-Field
Tgap
20 x CLKFC
Figure 4. BPLM Coding Method
QPLM is a variation of BPLM that is sometimes used. With this modulation, two bits are transmitted after one gap, and therefore more power is available on the transponder side. In addition, the average baud rate is higher compared to BPLM. The coding method follows the same basic implementation as BPLM, except the allowed number of states is extended from two to four and the predetermined gap to gap timing
Clock Data Manchester Bi-phase

Figure 6. Manchester and Bi-phase Coding
2-bit value
QPLM bits frame example
LF-Field
00
18 x CLKFC
Tgap
01
LF-Field
Tgap
Both encoding techniques enable clock extraction from the encoded data stream. This is possible because all time durations in the coded bit stream are quantized to one of two values; T or 2T, where T is what is referred to as a half bit. Data rate is fixed by the relationship 1/(2T). Clock extraction merely requires the detection of the minimum time duration element, T, and synchronizing its phase with the coded bit stream.
28 x CLKFC
LF-Field
Tgap
Protocol Layer
The protocol layer defines how individual data bits are grouped to enable communication between the vehicle base station and key fob. It defines how many bits and in which order they are transmitted between the reader and the transponder. As a simple analogy, this can be compared to the rules governing the formation of sentences in using words. The protocol layer would be like the sentence formed from the logical layer which would be like the words. It forms a fixed set of commands along with their allowable responses.
10
40 x CLKFC
11
LF-Field
Tgap
62 x CLKFC
Figure 5. QPLM Coding Method
Authentication
Authentication is the term used to describe the process of deciding whether the driver is authorized to start the vehicle. The simplest form of this is called unilateral authentication. In this case, the vehicle tests the key fob to determine if it has been associated/learned to the vehicle. When an additional step is added to this process, in which the key attempts to test the vehicle to determine if it has been associated with the key fob, it becomes bilateral or mutual authentication. Clearly, this added step increases security strength but comes at the expense of longer authentication time.
intervals are expanded to cover the additional states. See Figure 5 for a visual representation of QPLM.
Uplink
Information communicated from the user fob to the vehicle base station is typically encoded using Manchester or Biphase. These encoding methods share several characteristics that differ from the downlink: A) the encoded bit stream always has an average duty cycle of 50%, B) the time to send encoded data is based solely on the baud rate.
Unilateral Authentication Protocol

Key
LF-Field On
Detection Header (optional)
Car
Read UID Command
Key Memory ID Memory ID
4-bit Command + 4-bit CRC 8-bit Header + 32-bit ID + 8-bit CRC
Read UID ID
OK?
Key Memory
Stop
Y Random Number
Challenge
8-bit Command + N Challenge-bit + 8-bit CRC
Challenge
AES-128 (enc.)
AES-128 (enc.)
Response
8-bit Header + M Response-bits + 8-bit CRC
Response Y
=
Valid
Stop
OK, it is the right key
Start Authentication Command
Figure 7. Unilateral Authentication
Unilateral Authentication
Typically, the unilateral authentication protocol is initiated by the vehicle and consists of the following steps: 1. Vehicle reads the key fobs unique ID (not to be confused with the secret key) 2. Vehicle generates a random number challenge and sends it to the key fob 3. Key fob encrypts the challenge (using the secret key) and sends this response to the vehicle 4. Vehicle compares key fobs response with its calculated response (using same key and challenge) Note: The vehicle must posses the key fobs secret key to enable the success of this transaction. The process of sharing the secret key is called Key Learn and is described in the next section.
Key Learn: Open/ Secure

The Key Learn Protocol is the process that is used to allow the vehicle to establish a secret key and share it with the key fob. Depending on the restrictions and safeguards placed on the initiated Key Learn session by the vehicle, secret keys can be shared openly or securely. An open Key Learn process would typically consist of the following steps, also shown in Figure 8: 1. Vehicle generates a secret key based on a random number and proposes it to the key fob 2. Key fob accepts secret key, saves to memory, and responds with an acknowledgment 3. Vehicle saves secret key to memory after successful receipt of key fobs acknowledgement If the Key Learn Protocol cant be protected from eavesdroppers or unauthorized access to the vehicle, it may be desirable to utilize a Secure Key Learn Process.
Open Key Learn 1/2

Key
LF-Field On
Detection Header (optional) 8-bit Command + 128 bit Key + 8-bit CRC
Car
Random Number
Secret Key
Secret Key
Learn Secret Key (1 or 2) Cmd
Key Memory
Pass Y Response ( Pass/Fail )

8-bit Header +8 bit Status + 8-bit CRC
Stop
Key Memory
Figure 8. Open Key Learn
Bilateral or Quasi-mutual Authentication

A more complex form of an authentication process is the quasi-mutual or bilateral authentication. It is not a full mutual authentication that is implemented in the Atmel immobilizer system because it does not use random generators on both sides of the system, the car and the key. The implemented solution uses a MAC (Message Authentication Code) to authenticate the car vis--vis the key. Again, the authentication protocol is initiated by the vehicle and in case of a bilateral authentication consists of the following steps as shown in Figure 9: 1. Vehicle reads the key fobs unique ID 2. Vehicle generates a random number challenge and sends it to the key fob 3. Vehicle encrypts the random number and appends it to the challenge
4. Key fob encrypts the challenge (using secret key 1) and compares it with the received encrypted challenge (MAC) 5. If the result matches, the key fob encrypts it (using secret key 2) and sends this response to the vehicle 6. Vehicle compares key fobs response with its calculated response (using same key and challenge)
Cryptographic Layer
The cryptographic layer provides the highest level of encryption. It contains the mathematical function that transforms a plain text message into a secret message. Ideally, this function should have two properties: 1. Unique: For every plain text input, there must be a unique secret text output 2. Unpredictable: It must not be possible to predict a plain text to secret text pair, even if a large sample of known good plain text to secret text pairs was available for analysis
Bilateral Authentication Protocol

Key
LF-Field On
Detection Header (optional)
Car
Read UID ID
OK?
Read UID CMD
ID memory
ID
4-bit Command + 4-bit CRC 8-bit Header + 32-bit ID + 8-bit CRC
Stop
Y Random Number
Start Authenitcation Command
128-bit Key 1
Random Number
8-bit Command + N bit RandN + M bit (RandN) + 8-bit CRC
128-bit Key 1 AES-128 encryption 128-bit Key 2 AES-128 encryption

Hidden challenge (HCH) expanded to 128 bits
AES -128 encryption
Ok, is it the right car, continue
Stop
=
Y
Hidden challenge (HCH) expanded to 128 bits
AES -128 encryption 128-bit Key 2

8-bit Header + M bit {Resp}AES + 8-bit CRC
=
Y Valid
Stop
Ok, is it the right key, car match
Figure 9. Bilateral Authentication Protocol
Public vs . Private
For many years, private cryptographic algorithms were commonplace. However, private algorithms have drawbacks: A) uncertainty of algorithms strength, B) lack of being subjected to critical peer review, C) potential wide-scale security compromise should the algorithm be leaked to the public. In recent years, several high-profile examples can be cited that illustrate these weaknesses. Perhaps even more compelling is the lack of interoperability in systems that share the same physical and logical layers. This interferes with basic competitive market forces and in many cases drives higher system costs. In an effort to address these concerns, public opinion has shifted toward the acceptance of a public domain encryption algorithm the Advanced Encryption Standard (or AES, as it is more commonly referred to). Its origin comes from the 1997 initiative at the National Institute of Standards and Technology (NIST) to select a public-domain encryption algorithm. Within a year, fifteen candidate algorithms were identified and subjected to critical review by the cryptographic research community. This analysis included an
assessment of security and efficiency characteristics for each algorithm. After trimming the list of candidates from fifteen to four, NIST subjected them to a second round of public review before finally selecting the AES algorithm in 2000. AES, as we know it today, is a symmetrical block cipher that combines a 128-bit plain text input with a 128-bit secret key to create a 128-bit encrypted output. Due to its symmetrical characteristics, AES can also be used in reverse to combine the encrypted output with the secret key to find and extract the original plain text input.
System Security Considerations Attacks and Countermeasures

A common misconception held today is that the security of a vehicle immobilizer system is established by the strength of the encryption algorithm. While encryption algorithm strength is important, it alone does not define the overall systems resistance to attack. Each of the interface layers in the immobilizer system, algorithmic, protocol, logical and physical, contributes to the systems overall security and must be studied and fortified against attack.
Algorithmic Security and Countermeasures

As noted earlier, it is imperative that the encryption algorithm possess unique and unpredictable characteristics. In the case of AES, the details of how the algorithm operates is freely available to the public and as a result, it has been subject to critical review by the research community. This, by far, is the best countermeasure available. To date, scientific studies have confirmed the algorithms strength as it has withstood the test of time (over 10 years). However, in the case of private algorithms, scientific analysis by the research community was not possible, leaving the strength of these algorithms in question. In fact, many have failed to withstand the test of time and in recent years their weaknesses have been exposed.
Physical/Logical Security and Countermeasures

In recent years, attack methods have grown more sophisticated. Side-channel attacks such as Simple Power Analysis (SPA) and Differential Power Analysis (DPA) as well as other invasive attacks have been successfully applied to extract secret keys from key fobs. These socalled side-channel attacks measure and evaluate the power consumption of a cryptographic device and combine it with knowledge of the plain text or cipher text in order to extract an otherwise secret key. The theory underlying these methods is quite sophisticated and beyond the scope of this document. The strongest defense against the side-channel attacks noted above are: Randomization of clock frequency and operation Interleaving of digital control and the encryption operation Invasive attacks dwell on the physical implementation of the encryption-related circuitry on the silicon die itself. The best countermeasures are fairly simple to implement as long as they are considered early in the design process. The following are examples of steps that could be considered: Metal shielding of memory blocks Using non-standard synthesis libraries Scrambling the location of critical digital elements used during encryption Restricting memory access and automatic chip-erase function if attempted
Protocol Security and Countermeasures

In systems using unilateral authentication, attacks on the protocol layer are typically accomplished using scan or dictionary methods. In a scan attack, the attacker receives a challenge from the vehicle and returns random values in response. If the protocol consisted of a 56-bit response, then the bit security is 256 , i.e., it takes 256 trials to get one valid challenge-response pair. To resist this type of attack, the following measures can be considered:
Increasing the response bit length to add complexity Having the vehicle embed exponentially growing timeouts between consecutive unsuccessful trials Having the vehicle block trials after a fixed number of consecutive unsuccessful trials are attempted
System Performance Considerations

Current Consumption
System performance has different aspects. One is the power consumption of the key fob. This parameter relates directly to the achievable communication distance between key fob and vehicle base station. Car manufacturers and Tier1 suppliers tend to emphasize the importance of the coupling factor as a critical system parameter. However, it mainly represents the relationship of the mechanical dimensions between the key fobs antenna and the vehicle base stations antenna. This parameter is only valid for a given system configuration and depends on antenna inductance, Q-factor, driver current, reader sensitivity, and ignition lock cylinder material. Because of this, use of this parameter alone to compare different systems is inadequate. Of equal if not more importance than the coupling factor is current consumption, especially given that the key side current consumption is a limiting factor in a passive, batteryless environment where
In a dictionary attack, the attacker collects valid challenge (from the attacker) response (from the key fob) pairs by communicating directly with the transponder. These pairs are placed in a look-up table or dictionary for future reference. Equipped with this dictionary, the attacker then sequentially triggers the vehicle for a challenge, which can be checked in the dictionary for a valid response. If the protocol included a 100-bit response, one would need 251 trials to get one valid challenge-response pair. The birthday paradox states that after 2n/2 logged challenge-response pairs and 2n/2 trials, the probability of a valid result is 0.5. Using this, it can be shown that the overall complexity of this attack is 2n/2+1 = 251. Countermeasures to consider in this case are: Increasing the challenge bit length to add complexity Implementing a bilateral authentication protocol
10
the energy is harvested from a magnetic field and stored in a small capacitor. By selecting system components designed for extremely low power consumption and microcontrollers capable of being programmed with well-balanced software (putting the controller in sleep mode whenever possible), the engineer is able to overcome earlier system disadvantages requiring high coupling factors to compensate for high current consumption in the key fob.
Summary
By selecting system components that meet the security and performance expectations of the automotive market place, and support a highly configurable and open-source immobilizer software stack, the task of developing a robust vehicle immobilizer system can be greatly simplified. As a leader in automotive vehicle access solutions, Atmel offers such a complete system solution consisting of both hardware and software. Key fob designs can be realized with the Atmel ATA5580 and the Atmel ATA5795. These devices boast an LF front end, an AES hardware block to perform fast and efficient encryption calculations, coupled with an Atmel AVR microcontroller that has been optimized for extremely low current operation. Both include programmable flash memory that can be used to run the Atmel open immobilizer protocol or other customerspecific software and are capable of completely passive immobilizer operation. A base station can be realized with the Atmel ATA5272. This device integrates the LF base-station function with an AVR microcontroller with 8K of programmable flash memory. As a final complement to these devices, the open immobilizer protocol software from Atmel is available to users at no cost. It provides an unprecedented level of configurability including many user selectable features enabling the dynamic evaluation of system parametric tradeoffs and accelerates the development and optimization process: 1. A logical layer with uplink and downlink baud rate, bit encoding, and modulation depth 2. A protocol layer with challenge and response bit lengths, unilateral or bilateral authentication, data field CRC, two secret keys, secure or open Key Learn 3. A cryptographic layer with AES crypto clock speed from 125 kHz to 4 MHz on the fly
Authentication Response Time

Another important factor in immobilizer systems is the time it takes from turning the key fob inserted inside the lock until the engine starts. This time should be short enough to avoid the drivers perception of a delay. Depending on mechanical and electrical system design and how quick a person can introduce and turn the key, an overall timing budget in the range of 300ms to 500ms is typically available. A significant part of this budget is consumed through mechanics and overhead in the body control module. What remains is between 100ms and 200ms for the authentication process. A good compromise in terms of speed and security seems to be a bilateral authentication with a challenge length of 100 bits and a response length of 56 bits. In most systems this results in a response time of under 100ms.
Error Handling
In case an authentication failed for whatever reason, todays systems require the complete authentication cycle to restart from the beginning and allowing a maximum of three retries within a reasonable time. The retry strategy from Atmel looks a bit different and enables the system to recover from communication errors more quickly. All commands and optionally the data are protected by a Cyclic Redundancy Check (CRC). Both the key fob and the base station can make use of the CRC to detect errors and signal these conditions to their respective communication partner. This enables the base station to be selective about the amount of repeated info, the last action, the last response, or the last command. This feature enables quicker communication recovery and more attempts at communication recovery in the same amount of time (five-seven retries instead of three).
11
Automotive LIN Bus Driving Sensor Applications

Dr. Stephan Hartmann
Todays cars contain, on average, more than 50 different sensors to monitor various physical variables . This number is growing, driven mainly by the proliferation of actuators, which require sensors to deliver the relevant input values . In addition, requirements for signal systems are now increasing, and analog data transfer techniques are showing their limitations . The engineer now faces the challenge of transforming the sensor area into an efficient, high-performance digital subsystem .
LIN bus driving sensor technology enables efficient management of digital data, combining the benefits of existing voltage modulation and current modulation approaches . Atmel provides all necessary products independent of the integration level of LIN applications .
12
Sensor systems differ in many respects from other electronic components of the car. The most important difference is that sensors are mostly located outside the vehicle in harsh environments where they are subject to changes in humidity, temperature or pressure. In most cases, sensors also have to be mounted in areas with very limited space and are connected with a 2- or 3-wire harness.
pull-up resistor. For the connection to the sensor, a 2-wire connection is used. Disadvantages include thermal power dissipation in the sensor module, as well as a limited data rate dominated by the pull-up resistor. Other current-based transfer methods, such as Manchester-coded protocols, require dedicated transceiver ICs, driving system costs up.
The Applications for Sensors are as Diverse as the Application Areas Themselves:
In the comfort area: - Temperature sensors - Solar altitude sensors - Light sensors - Humidity sensors - Dew point sensors In the body control area: - Pressure sensors - Gyro sensors A typical sensor node contains the sensing element itself, a microcontroller for signal conditioning, and a transceiver for signal transmission. As the length of the data line is often more than 1m, the data transfer is dominated by analog signal conditioning which has a portion of about 90%. Analog signal conditioning does have some advantages. It is compatible with previously existing mechanical or electromechanical detection systems, and is also easy to use and to plug in. Analog data can be provided in a voltage range of, for example, 0 to 5V and the sensor can then be monitored by an ADC port on a microcontroller which converts the data into the digital domain. Generally, however, requirements for sensor systems are increasing, making analog signal conditioning less attractive. With ADC resolutions up to 10 Bit, and the ability to indicate two different types of failure modes by clamping the signal voltage either to the lowest or to the highest voltage levels, analog techniques have already reached the limits of their performance and will be replaced by different types of digital data transfer. Digital data transfer can be managed either by voltage modulation or current modulation. Both types have advantages and disadvantages. Simple current modulation allows a very cost-efficient design of the ECU as well as of the wiring harness. Inside the ECU, the different current levels can be transferred into voltage levels using a single In the powertrain area: - Position sensors - Speed sensors - Pressure sensors - Knock sensors
Figure 1. Basic Current Interface Set-up
Voltage modulation has the advantage of allowing a variety of protocols, beginning with simple PWM and moving to more complex versions like SENT, which have higher data rates than current modulation. Additionally, the ECU input can be designed as a capture compare unit on a timer basis. The main drawback of voltage modulation is that a wiring harness with a 3-wire connection is mandatory. Further issues may arise at EMC testing because most PWM drivers do not include a slope rate control. Additionally, ESD protection is low.
Figure 2. Basic Voltage Interface Set-up
The LIN bus protocol combines all advantages described above. As a two-wire interface, LIN helps save cost in the wiring harness. LINs slew rate control in the transceiver ICs supports excellent EMC performance, while LINs ESD protection features allow robust system designs suitable for harsh environments. Finally, the high production volumes
13
for LIN bus transceivers lead to very cost-efficient designs compared to other protocols requiring transceiver ICs (e.g., current-based transfer protocols).
and a microcontroller for signal conditioning and protocol handling. Atmel serves all integration levels with LIN transceivers, LIN system basis chips (SBCs), and LIN system-in-package (SiP) devices. With SiP, the customer benefits from the ultra low-power designs of Atmel AVR microcontrollers with Atmel picoPower technology. As a second step, the designs can be converted into single-chip, multi-die SiP designs. This saves PCB space and allows the engineer to include all electronics in the connector of the sensor element. Finally, by integrating the sensor element into the chip, and by implementing an intelligent state machine, the engineer takes a further step toward advanced, single-chip LIN sensor node designs.
Figure 3. LIN Interface Set-up Modified to Support the Current Supply of the LIN Sensor Slave
UC
The concept for the system design specifies a pull-up resistor at the ECU level. This is required to control the current supply for the LIN slave sensor through the data line. At the LIN slave sensor, only a buffer capacitor is required. Depending on the capacitor value, the data rate can be chosen up to 100 kbit/s. Supply voltage reduction provides an additional option for influencing the data transmission speed. As the dominant and recessive levels are referenced to the supply voltage, a reduction of the supply voltage directly reduces the gap between bus low and high level (the delta) to reach the correspondent levels. A 2V supply voltage reduction leads to an increase in the data transmission rate of roughly 15%. Unfortunately, the supply voltage cannot be lowered in all cases. The time portion of the bus-dominant level must also be considered, as this state discharges the buffer capacity of the LIN slave sensor Sensor system design using LIN can be viewed as a threestep process. First approaches to a discrete slave-node design include a sensing element, a LIN system basis chip,
Voltage 20 18 16 14 12 10 8 6 4 2 0 LINH VCC
99% 98% 95% 86%
63%
Figure 5. Capacity Discharge Depending on the Bus-dominant State Time
In summary, LIN not only enhances the driving of cars from the in-vehicle networking point of view, it also allows the rumble change of the sensor area to a cost-efficient and high-performing digital sub-system. Atmel offers all necessary products regardless of the integration level of LIN applications. To support low-power designs, the AVR microcontroller with picoPower technology is key. In addition, engineers can design the most robust and EMV-tolerant systems with the leading-edge EMC and ESD performance of LIN transceivers and SBCs from Atmel.
Special thanks to Daniel Yordanov and Keith Nicholson for supporting me during the writing of this article.
LINL
Figure 4. Bus Level States Bus Level States Depending on the Supply Voltage
14
Designing Next-Generation Key Fobs

Paul Lepek
Key Fobs Today
Todays key fobs can be generally subdivided into two different functional categories. The first includes Remote Keyless Entry (RKE) devices which require some sort of human intervention or a physical interface of the user to the key fob (e.g., key push) in order for the fob to produce the desired function effect such as unlocking a door or opening a sunroof. The second group of devices provides similar functionality but also features an Atmel has Introduced Next-Generation Transceiver added level of comfort by performing the same and Receiver Devices with Configurable Options function without physical intervention by the user. Instead of the push button or touch sensor interface, a Passive Entry (PE) identifies the user (and the Moreover, all key fobs support engine immobilizer system key fob) as a legitimate entity and automatically triggers authentication. To prevent theft every automobile uses an authentication or issues a request (e.g., passive door unlock, immobilizer system which authenticates engine starts. In trunk release, etc.). this case the key fob acts as a passive authentication tag Both systems are based on a preprogrammed key fob device similar to the RFID tag but with a larger feature set. Most ID and authentication protocols which include an encryption automotive key fobs use Near Field Communication (NFC) stage for authorizing the issue of key fob commands to the transponders which communicate with the engine controller. vehicle. In this way the key can be identified by the vehicle The transponder is integrated into the key and is a passive and vice versa before any action is executed. All RKE-based device. It does not need a battery for operation, but instead systems require key fobs to support RF links which fall into uses a magnetic field generated by the LF vehicle coil. It also Industrial, Scientific and Medical (ISM) frequency bands transmits the device ID and executes a special immobilizer (i.e., 0 - 135kHz, 13.56MHz, 315/433MHz, 869MHz, and
915MHz). However, for a PE system the LF downlink is used by the key fob to compute a Received Signal Strength Indicator (RSSI) value and thus the fobs physical coordinates in relation to the vehicle while the RF link is used to execute the authentication protocol with the vehicle. RKE and PE system fobs are designed to be powered by a small coin battery intended to last for the life of the vehicle.
15
Table 1 . Key Fob Features Today

Physical Interface 3 (Button, Touch, etc.) Remote Start Basic RKE (std.) RKE (ext.) PE (std.) PE (ext.) x x x x x x/ x/x x/x x/x x x x x Remote Entry Accessories
Class
Immobilizer1
Batttery/ Recharge2
Passive (Entry/Go) Accessories
Personalization Settings
Time/ Data Logging in Fob4
x x
x x x
x/x x/x
x x x x x
1 2 3 4
Immobilizer support includes secure fob and vehicle authentication via the LF field using an integrated LF transponder. The fob includes a built-in battery with the option of recharging the battery via the LF field. Remote Start and Remote Entry are controlled via RF uni- or bi-directional link. Control of accessories can either be done using RF or IrDA links. Data can be logged such as time stamp data, last device ID, last vehicle service date, and much more either via LF or RF links.
protocol for its authentication but all communication takes place via an LF field generated by the vehicle.
Expanding Fob Applications

Originally key fobs were designed for only one purpose: to unlock the door and start the engine with the metal key. Later, RKE devices were used to remotely (HF field) unlock the door. The integrated contactless passive transponder (LF field) then unlocked the steering column and enabled the engine start. Only more recently fobs began to penetrate convenience, general utilities, secure communications, and secure access ID applications (even extending to payment systems and e-ticketing). These functions were not developed before due to a lack of hardware and software resources, primarily because of the fobs physical size and power consumption. Recently, however, it has become possible to overcome these shortcomings by incorporating much larger user and program memories, and the use of faster, more compact, and ultra-low power processors without increasing the cost of production. Additionally, the integration of flexible, reconfigurable, and secure authentication peripherals can be made feasible. These types of peripherals include crypto units, secure key management features, and integration of smart cards useful for payment, user ID, and cipherbased authentication systems. An automotive key fob can therefore be used not only to interact with the vehicle but
also to gain entry to a park garage, ski lift or to purchase train tickets. This can deliver considerable benefits when goods are purchased using one of the major credit card networks. The secure user memory can also be used to store personal and secure information as personal data ID and provide transit information for e-ticketing.
System-in-fob Hardware Resources

Ultra-low-power 8-bit microcontroller Large Flash program and EEPROM data memory (memory segmentation with locks) RF communication interfaces - Infrared (IrDA) IF - Immobilizer IF at 125kHz - Passive entry IF at 125kHz (RX only) - Smart card IF at 13.5MHz - RKE IF at 315, 413, 868, 915MHz (frequency hopping) Power management (optional battery charge) Hardware cryptological unit (AES-128) Integrated proximity coupling smart card (ISO 14443) Cyclic Redundancy Check (CRC) block Serial interfaces (SPI, SSI) ISP/debug (dW) Analog comparator Flexible GP timers and WDT Oscillators: RTC, INTRC (125kHz, 4MHz)
16
The heart of a modern key is an ultra-low power microcontroller with sufficient program and data memories. Typical program memory can range from 8KB to 16KB and beyond with its data memory ranging from 1KB to 2KB depending on application requirements. Because of secure application support, the key fobs program and data memories must have provisions for memory segmentation and locks in both memory blocks. For example, the application firmware resident in program memory can be divided into separate memory sectors (e.g., application and immobilizer sections). Also data memory may have its own partitioning which can allow for soft and hard memory locks when it comes to releasing user-sensitive device-stored data (e.g., authorization password or secure key). While the microcontroller core executes application firmware, the secure user and key data is stored in the on-chip nonvolatile memories (EEPROM). The core uses various wireless communication peripheral interfaces to communicate with the infrared transceiver (IrDA), the LF transponder (125kHz), the 3D LF receiver (125kHz), the smart card (13.5MHz) and the RKE transceiver (315, 413, 868, 915MHz). Flexible serial interfaces can be shared such as SPI or Serial Synchronous Interface (SSI) to enable data exchange with every communication peripheral. A hardware data integrity check module, based on a Cyclic Redundancy Check (CRC) checksum algorithm, supports validation of received data. A unique feature of the immobilizer transponder interface is that it is closely bound to the power management unit which is used to provide power supply voltage, VDD, while exchanging LF data with the immobilizer in passive mode. In this mode all other communication with the key fob is disabled to support batteryless operation. Some key fobs may also support the battery charge feature which is integrated into the power management module for recharging the battery with the engine running. While in secure smart card mode the device can also operate in passive mode and exchange as well as encrypt and decrypt proprietary data using its own crypto module with the reader at FC = 13.5MHz. It can use its own device memory or the fobs internal nonvolatile memory. Integrated crypto modules can support many different cipher algorithmsthe most popular being the 256-bit block cipher known as the Advanced Encryption Standard (AES) based on the Rijndael algorithm which can be used with a 128-bit, 196-bit or 256-bit secret key.
The fob must also contain analog peripherals such as internal oscillators where F = 125 kHz and 4 MHz to generate its internal clock signals used for the transponder front end and the microcontroller core, respectively, with low frequency deviation across VDD and temperature. The supplied analog comparator can facilitate detection of VDD drops and prevent data corruption during nonvolatile data memory writes in passive mode.
Power Management Contactless Interface/ Transponder 3D LF Receiver
SPI EEPROM/ 2KB
FRC -Oscillator
ECIN
RTC
Clock Management & Monitoring
Contactless Interface/ SmartCard
SRC Oscillator Atmel AVR - Core Watchdog Timer
Volt. Monitor
Smart Card
IR - Driver SRAM 512B PM ROM 2kB PM - Flash 14kB debug WIRE POR / BOD & RESET RF TX/TRX Module
Timer Block
Crypto Unit
Data Mem. Secure EEPROM .
CRC
Serial IF IO-Ports
Figure 1. Fob Floor Plan
Wireless Data Communication Interfaces

An entry/immobilizer system consists of at least two communication partners where one side is on the vehicle and the other on the key fob. Depending on the link type, there are several possible communication interfaces available, including: 1. IrDA link for convenience and comfort applications 2. PE/PEG link to enter and start engine, including LF downlinks and UHF uplinks 3. Immobilizer link to start engine and emergency vehicle entry, including LF down- and uplinks 4. Chip card as user ID, authorization, authentication at pay stations, including HF down- and uplinks 5. RKE as entry authentication, including UHF down- and uplinks
17
Table 2 . Modern Key Fob Communication Links Overview

Application Standard Link Type FC Modulation Anticollision x Data Encoding PIE BR [Baud] Range Average Current 2 - 20A
PE/PEG
Custom
3D LF downlink UHF uplink
0130kHz 315, 433, 868, 915MHz 0130kHz
ASK
3.9k
3 - 5m
ASK/FSK
Manchester, Biphase
Up to 80k
30 - 120m
8 - 20mA
Immobilizer/ Emergency Entry
ISO14223 /Custom
LF downlink
ASK
PIE
3.1 -8.9k
2 - 10cm
40 - 260A
LF uplink Chip Card ISO14443 ISO15693 ISO18000 HF downlink HF uplink 13.56MHz
ASK ASK/PSK x
Manchester PIE, Miller, NRZ Manchester, NRZ x Manchester, Biphase
4.4k Up to 20k 5 - 20cm 20 - 120a
ASK/BPSK
RKE
Custom
UHF downlink UHF uplink
315, 433, 868, 915MHz
ASK/FSK/ PSK
Up to 80k
30 - 120m
8 - 20mA
VBAT=1 & VFLD=0
The summary of the communication channels and a brief overview is shown in table 2.
(Start-up)
Enter RKE Mode RKE
Secure and Reconfigurable Firmware

Application firmware which supports the complete functionality and feature set is the fundamental building block of the key fob. It may consist of many different modules and must encompass all functional and likely operating scenarios, including battery failure which comprises emergency or passive operation mode. To improve reliability it is a common practice to keep both application and immobilizer programs separate and in two distinct program spaces. While the immobilizer firmware supports distinct engine starts, the application software controls all other fob functionalities including RKE, convenience or user-ID applications. The immobilizer/emergency functionality is required to take priority over any other function, which is the equivalent of an override which suspends any function currently in progress when the LF field is detected at the transponder LF coil. Figure 2 depicts a flow diagram and interaction between application and immobilizer firmware.
APPLICATION SW MODULE
VFLD?
No
Yes
VBAT=0 & VFLD=1 (Start-up)
Enter IMMO. Mode
IMMOBILIZER APPLICATIONSW MODULE
VFLD?
No
Yes
Figure 2. Immobilizer and Application Firmware Interaction Flow Chart
All data communication is fully supported in the fobs firmware. Various communication protocols including unilateral, bilateral for immobilizer, PE/PEG and RKE systems can be fully configured by the application software. Based on
18
Coupling factor [%]
protocol topology, the application software controls dedicated peripherals by enabling them, and reading data during RX phases, and writing data during TX phases of the protocol as soft triggers (e.g., immobilizer and PE applications) or hard triggers (e.g., RKE or IrDA applications using a push button interface). A major advantage of a next-generation key fob is its in-field programmability, which can be very helpful in the event of a firmware or user data upgrade or programming. The fob can be initially configured using its dedicated general-purpose software via the LF field while the final test is performed at the factory. The user data can be added later by the Tier1 or OEM without modifying the original configuration. Even while in the field the fobs can be reprogrammed with new
Coupling Factors for Immobilizer System Coupling Factors for Immobilizer System
9.00% 8.00% 7.00% 6.00% 5.00% 4.00% 3.00% 2.00% 1.00% 0.00% 0 0.02 0.04 0.06 0.08 Lr=Rehfeld, Lt=2.45mH Lr=Rehfeld, Lt=5.1mH Lr=738uH, Lt=2.45mH Lr=738uH, Lt=5.1mH
Distance d [m]
Software Model
Application Layer
Figure 4. Immobilizer Coupling
Protocol/Session Layer
API, Configuration
Transport/Link Layer
Data I/O
Physical Drivers
Resources
Figure 3. Software Partitioning as a Bottom-up Approach
application and user data via one of the communication interfaces at a later date. Of course, in this case only a single functionality can be enabled while using memory locks to provide security. This is especially useful when used as a pay token in e-commerce or e-ticketing environments.
Figure 5 shows one complete challenge-response authentication protocol which could be used in a passive automotive immobilizer application. The LF field voltage (green) is enabled for 160ms. The field is damped (2.2V) during RX state and then switched to the undamped level (6V) during TX state. The charge storage capacitor voltage (yellow) which provides VDD to the transponder is immediately charged to 2.2V during the RX data stage. The transponder encrypts (AES-128) received plain text data (128-bit challenge) and transmits the response. In many immobilizer systems the system authentication time is a major concern. To minimize authentication time, the number of bits transmitted can be reduced without compromising system security. It is common for authentication time TAUT < 130ms at BR = ~3.9kBaud.
Abstraction Level
Transponder LF Field Coupling

Transponder-to-base-station coupling still remains the most challenging aspect for key fobs. The proper transponder coupling can be achieved when sufficient energy is transferred from the base station to the transponder for the transponder to communicate with the base station. During design, the L-C tank must be carefully selected for optimum energy and communication performance. Figure 4 shows a typical transponder coupling at 125kHz vs. operating distance from the base station coil5.
Figure 5. Immobilizer Protocol Execution Scope Shot/Power Analysis
5
Assumes the fob coil is placed in the center of the base station coil where the coupling is best
19
Figure 6 shows field voltage and VCC traces as the key fob is energized by the field and begins to receive a BPLM data stream. Field gaps are visible which separate field On interval which is decoded by the fob using a dedicated hardware peripheral. The fobs microcontroller is in sleep mode 95% of the time to save power consumption.
Table 3 . RF Transceiver Parameters

Antenna Gain -6dB Average Current ~6 - 8mA
Sensitivity Receiver -109dBm
POUT -
Transmitter
-18dB
10dBm
~9 - 10mA
Table 3 shows some typical RF transceiver parameters. A snapshot of the transmitter spectrum taken at 433MHz during transmission of an RKE message to the vehicle is shown in Figure 7. The transmitter carrier frequency, the span, power output, and device setting are configured using the serial interface by shifting configuration data (in this case the 32-bit configuration word) into the RF transmitter via the MCU when the user presses the open-door button.
Figure 6. Transponder at Power-up
RF Communication Links
Both RKE and PE/PEG systems utilize HF communications links. In comparison to LF links, HF links are superior in their operating range (up to several hundred meters) and baud rate (up to 80kBaud can be achieved). RF transceivers currently available on the market use N-fractional PLL frequency tuning techniques where the carrier frequency can be selected in firmware by the MCU. Some devices allow large tuning frequency variations, permitting more design flexibility. The transceivers operating range remains a key performance parameter. To extend operating range, it is common for the transmitter power to be as high as 12.5dBm and the receiver sensitivity to be less than -100dBm. Antenna design is also a determining factor providing additional performance gain. Although whip antennas add additional performance gain, small loop antennas printed on the PCB are usually chosen for use in automotive key fobs. Power consumption is another critical factor on the receiver as well as on the transmitter side. Power consumption can be reduced by selecting higher data baud rates. Choosing ASK modulation tends to lower operating current since the power amplifier is momentarily disabled during modulation.
Figure 7. RF TX Spectrum at 433MHz
References: 1. ISO 14223 Advanced Transponders Standard 2. ISO 10536 Close Coupling Smart Cards Standard 3. ISO 14443 Proximity Coupling Smart Cards Standard 4. ISO 15693 Vicinity Coupling Smart Cards Standard 5. ISO 18000 Item Management Standard
20
Capacitive Proximity Detection in the Automotive Industry

Luben Hristov
There has been a steady rise in demand for proximity detection sensors in automotive applications which reliably detect the presence of objects near the sensor surface without physical contact and the number of possible proximity detection applications is countless: Door entry control: detecting a hand approaching the door handle to initiate the car unlocking process Illuminating and waking up the touchscreen when a hand approaches the screen surface Switching interior car lights on/off when the hand is near the sensor Detection of simple spatial gestures to switch devices on/off Sensing the presence of large objects around the car during parking Many different proximity detection methods exist, for example, capacitive, infrared, ultrasonic, optical, etc. For the 5mm to 300mm proximity detection range, capacitive
sensing has many advantages compared to other methods: excellent reliability, simple mechanical design, low-power consumption, and low cost base. This article describes capacitive proximity detection technology from Atmel. Atmel is a leading touch solutions manufacturer with many years of experience in this field. The companys capacitive sensors are based on charge-transfer technologya method pioneered by Atmel where voltage is generated on the sampling capacitor during the repetition of a specific control sequence applied over the I/O pins. Atmel currently holds multiple patents in the area of charge-transfer technology for self-capacitance sensors (Atmel QTouch) and mutualcapacitance sensors (Atmel QMatrix). The Atmel chargetransfer technology delivers key benefits to the user and offers advantages compared to other capacitive measuring methods: increased flexibility, very high sensitivity, excellent moisture resistance, and noise immunity. QTouch and QMatrix technologies have been implemented in multiple touch controllers supporting touch buttons, sliders, wheels as well as touchscreens. Proximity detection support is also available with some of the standard products. Atmel is now developing and manufacturing new proximity algorithms
21
conductive surface as shown in Figure 1. Please refer to the Atmel application note about using active shields at
http://www.atmel.com/dyn/resources/prod_documents/ankd02_103-touch_secrets.pdf, page 7. A further advantage of
active shields is their neutralizing effect on water films. Moisture resistivity: Moisture-induced changes in the measured signals can be more significant than changes from approaching objects. Water film on the surface is one of the biggest problems for capacitive solutions. Water films are more or less conductive and create a change of the measured signals that is similar to normal touch events. There are mainly two ways to handle effects caused by water films: 1. Use of active shields (described above) 2. Shorter charge transfer time the water film could be utilized as a distributed RC circuit (as shown in Figure 2). Reduced charge transfer pulses will prevent full charging of the distributed capacitors C and hence reduce the impact of the water films. Best results can be obtained if the charge transfer time is in the range of 100ns to 250ns. A proper mechanical design of the sense area and the use of the appropriate materials prevent the emergence of thick water films on the sensing area.
Figure 1. Active Shielding of Proximity Sensor
to increase sensitivity to support finger or hand detection ranges of up to 200mm and more. The release of standard products and software library modules is being planned for 2011 and 2012. Capacitive proximity sensors measure the capacitance change between the single electrode and ground (selfcapacitance sensors) or between two electrodes (mutual capacitance sensors) as objects approach electrodes. While constant capacitance is between 10pF to 300pF, the capacitance changes are typically extremely small, ranging from a few fF to several pF. Since the electrical field lines around the self-capacitance sensors spread far away from the sensing electrode, self capacitance is the preferred proximity detection method over mutual capacitance where field lines are largely concentrated in the area between the transmitting and receiving electrodes.
Figure 2. A Water Film Acting as a Distributed RC Circuit
Characteristics of Capacitive Proximity Sensors For Automotive Applications:

High sensitivity: Detecting small changes in the measured capacitance requires increased and stable sensitivity. Special measures should be taken to reduce negative effects on sensitivity caused by capacitive loading, especially if the sensing electrode is placed on a conductive surface (metal plane, car body, etc.). An active shield layer is used to reduce the negative effect of capacitive loading between the electrode and the
Temperature stability: In automotive applications extreme and rapid temperature changes may occur at any time. Special care should be taken with regards to a stable mechanical design even the smallest gap changes near the conductive surfaces may cause false detection. Noise immunity: Due to the high sensitivity, noise interference could compromise normal operation of the proximity sensor. The electrical and mechanical design of the PCB should be carried out to avoid noise interference caused by adjacent cables or conductive surfaces. Fast response time: The expected response time is usually between 10ms and 100ms
22
The following sections provide more detailed scenarios of automotive capacitive proximity detection.
Door Entry System

One example of capacitive proximity detection is in car door entry systems (see Figure 3). The proximity sensor that detects hand approaches is located within the car door handle (1). Once object proximity has been detected, the main unit (2) sends a wake-up signal via the LF antenna (3) which activates the car key transmitter (4). The car key transmitter then exchanges information with the RFID receiver (5) and if the code matches the main control unit (2) the door is unlocked. The entire process of proximity detection and ID recognition takes a fraction of a second. This means when the hand pulls the door handle, the door is already unlocked. The advantage of using proximity detection rather than touch detection in door entry systems is the extended time to identify a person. As a result, the door lock state will always be resolved before the door handle is pulled.
causes the light to switch on, a wave of the hand in the opposite direction switches it off. The system is able to analyze the signals from the proximity sensors and to decide whether to switch the lamp on or off. There are many different options available for designing sensing electrodes inside a light from using thin copper wires to conductive polymers that can be applied directly over the plastic.
Figure 4. Spatial Gestures for Switching Devices On/Off Inside the Car
Conclusions
Implementation of capacitive proximity sensors in automotive applications paves the way to a broad range of comfort applications Moisture and rapid temperature changes are the main challenges for capacitive proximity sensors used in automotive systems. These challenges can be overcome by utilizing the most recent innovations in capacitive sensing technology from Atmel.
Signal
Figure 3. Door Entry System with Proximity Detection
Spatial Gestures to Switch Devices On/Off

The simultaneous use of two or more capacitive proximity sensors enables simple spatial gestures such as hand waving in front of the device to be detected. Figure 4 shows a simple example of such a system to switch lights on/off inside the car a wave of the hand in front of the light in one direction
Capacitive ProximitySensor
23
Extending the Range of E-vehicles with Li-ion Batteries Using Active Balancing
Claus Mochel
Global warming has now become the main talking point of our era, and electric or hybrid vehicles provide an excellent way of saving fuel and reducing CO2 emissions. However, electric and full hybrid vehicles have one particular Achilles heel, which is the capacity of the battery and therefore the limited range of these vehicles. The maximum battery size that can be installed in a vehicle is limited because of its volume and weight, making it all the more important to make optimum use of the battery capacity that is available. In order to provide the several hundred volts that are needed in modern high-performance batteries for electric vehicles, several individual battery cells are connected in series. Each cell in a battery pack such as this differs from the others, e.g. in terms of cell capacity, self-discharge rate, different temperature characteristics and cell impedance differences that are intensified by the ageing effect of the battery. When the cells are being charged, this leads to a situation whereby some cells have not yet received their maximum possible charge, but other cells are already fully charged. Unless additional measures are taken the charging procedure has to be aborted, since individual cells would otherwise be overcharged, which would damage or possibly even destroy the battery cells. The situation is similar with regard to discharging. Whereas individual cells are already discharged, others still contain enough energy to continue powering the vehicle (in theory). However, the vehicle must not continue to be driven, since the weaker cells would otherwise be overdischarged, which would lead to the destruction of these cells. In order to avoid all of this, active balancing between the individual battery cells is required.
Passive Methods Convert Valuable Energy Into Heat Loss

The method that is most widely used at present is passive balancing, in which cells that are already fully charged are discharged again using a resistor so that the other cells can continue to be charged. The disadvantages of this method are obvious: Cells can only be discharged for the purpose of balancing The discharge current causes power loss in the bypass resistor Valuable energy is converted into heat and is no longer available for powering the vehicle The range of the vehicle is reduced While the passive balancing method only converts the energy that is stored in the cells into heat loss, active balancing enables to transfer the charge from one cell to another. There are several ways of achieving this charge transfer, e.g. using switched capacitors or inductors. When the capacitive method is used, a capacitor is connected in parallel with the cell with the higher voltage. Once this has been charged, it is connected in parallel with the cell with the lower voltage and can charge it up. This procedure is repeated until the same voltage is present at both cells. The method that uses capacitors is extremely cost-effective, but has the disadvantage that the average balancing current is limited to less than 50mA. This limitation does not exist
24
MBAT7
DISCH6 MBAT6
PMOS Zelle n+1
RIN
MBATn+2 CIN Zelle n+1 DISCHn-+1 PMOS
RIN
MBATn+2 CIN DISCHn-+1
DISCH5 MBAT5
DISCH4 MBAT4
CSHUFFLE
NMOS
RPD MBATn+1
CSHUFFLE
NMOS
RPD MBATn+1
DISCH3 MBAT3
Zelle n
DISCH2 MBAT2
DISCHn
Zelle n
DISCHn
DISCH1 MBAT1
MBATn
MBATn
Figure 1. Passive Balancing Using Resistors
Figure 2. Active Balancing Capacitors
with the inductive method, and in this case it is easily possible to have balancing currents of 1A or more.
Active Balancing Method for Rapid and Almost Loss-free Charge Transfer
Active balancing is realized by connecting an inductive resistor in parallel with the cell from which the charge is to be taken. This results in a constant increase in the current flow in the coil. Once the coil has been decoupled from the cell that is being discharged via a transistor, the energy stored in the cell can charge the neighboring cell via a diode. It is therefore possible to move charges to and fro between the individual cells extremely efficiently and almost loss-free. This method has some decisive advantages: Balancing currents of 1A and more are possible Balancing is essentially loss-free Balancing is extremely rapid The efficiency and the capacity of the battery are increased The range of the vehicle is increased Compared to the other methods that have been mentioned, implementing an active balancing method using inductive resistors is not exactly low-cost because of the relatively high cost of the inductive resistor component. However, this
is not really a problem. A modern high-performance battery currently costs approximately USD $10,000 Even if only gaining an extra 10% of capacity using the inductive balancing method, this represents a value of USD $1,000 an amount for which a considerable number of inductive resistors can be purchased.
1 = Discharging a cell 2 = Charging a neighboring cell
Cell n
Cell n-1
Figure 3. Active Balancing Using Inductors
Individual cell monitoring is required for Li-ion batteries for safety reasons, since these cells burn when they are overloaded, and can even explode in extreme cases. As well as the overvoltage, undervoltage and temperature monitoring that is required, additional functions such as precise charge condition determination are needed. A component is now available on the semiconductor market with which both these functions and also the different balancing methods can be realized. With the Atmel ATA6870, each cell has individual electronic monitoring in order to provide functions such as state-of-charge determination, active/passive balancing or overvoltage, undervoltage and temperature monitoring.
25
Measurement Data Recording and Active Balancing for Li-ion Battery Systems
The heart of the Atmel ATA6870 consists of six highprecision 12-bit AD converters. Each cell is monitored by a separate AD converter, which has several advantages. On the one hand, all cells can be measured simultaneously, and on the other hand the cell voltage does not have to be transferred analog to ground, which would reduce the accuracy of the system. As well as providing efficient balancing, another prerequisite for making optimum use of the cell capacity is precise recording of the cell voltage because of the extremely flat characteristic curve of Li-ion battery cells. The ATA6870 therefore transfers the digital voltage value to ground without loss of accuracy after analog/digital conversion. As well as the six high-precision AD converters for measuring cell voltage, the ATA6870
is the worlds only device with the possibility of achieving both passive balancing and active balancing between the individual cells using capacitors or inductive resistors in high-performance batteries with up to 300 cells or more connected in series. In order to prevent the balancing from taking an extremely long time because of the large number of cells, balancing can be carried out simultaneously on any number of cells with the ATA6870. With the aid of this circuit it is now possible to develop low-cost, efficient and safe battery management systems that can squeeze the last Coulomb out of a battery due to extremely precise recording of the charge status of the individual cells and the efficiency of the balancing method that is used, and make it available for driving operation. This is another step that increases the range of electric vehicles and plug-in hybrid vehicles and therefore contributes to the triumphant march of these vehicles.
Cell 1 Reference ADC Charge Transfer
Digital Level Converter
Standby 3.3V Voltage Regulator 3.3V Internal Voltage Regulator
Digital Level Converter
Cell 6 Reference ADC Charge Transfer
Logic
Current Supply
Communications Bus Cell Temperature Measurement Digital Level Converter
Microcontroller Interface
Microcontroller
Figure 4. Block Diagram of the Atmel ATA6870 Li-ion Battery Management IC
26
Precision Battery Monitors for Standard Lead Acid Batteries Ensure Engine Start-up
Claus Mochel Today, one in three car breakdowns is related to the vehicles electronic equipment. Today, high-end cars incorporate up to 65 electronic control units (ECUs), compared to only 10 ECUs in 1995. Battery monitoring is one of the most important current ECU applications, since 60% of breakdowns caused by electronic devices are caused by defective or empty batteries. Batteries now face many challenges for energy management. Formerly, the batterys main task was to provide high-power output for a short time to enable engine start-up. This has changed dramatically, Today, the battery must provide continuous power to the large number of electronic devices inside the car. This function can be problematic in winter, when battery capability is reduced through cold temperatures and high energy consumption due to loads such as electrical seat heating. In addition, driving in slow traffic means that the car generator provides limited energy to charge the lead acid battery.
Concerns about global warming, along with resulting stringent legal provisions for reduction of vehicle CO2 emissions, are driving an additional challenge for energy management: micro-hybrids. These cars save fuel by switching off the combustion engine when stopping (for example, at a traffic light) and then automatically restarting the motor when the driver wants to continue. The primary issue with such an engine is that the vehicles starting capability must be guaranteed under all circumstances. Otherwise, traffic chaos is bound to develop, given the growing number of vehicles of this kind on city streets. With the advent of micro-hybrids, reliable battery diagnosis and modern vehicle energy management become all the more important. The core of such an energy management system is an intelligent battery sensor (IBS) to constantly monitor the charge and discharge currents, the battery voltage, and the temperature. The resulting data is used to determine the batterys load and state-of-charge and for distributing power within the car. Priority is given to critical functions such as engine start-up. Furthermore, the sensor calculates the batterys state-of-health (SoH) and alerts the driver when battery lifetime is expiring in order to avoid car malfunction or breakdown. With IBS information, engineers can also reduce fuel consumption and resulting CO2 emissions. This is done by tailoring the charging cycles depending on the required current for the various loads and the energy available from the generator. In cars without an IBS energy management system, the generator must provide a continuous electrical charge for the battery while the motor is running; with such a system, the generator can be disconnected from the battery when the battery charge is sufficient and the vehicles engine is under heavy load.
27 27
Atmel has provided initial models of a microcontroller perfectly matching the requirements of an advanced IBS system. The device is based on the well-known Atmel AVR microcontroller core in Atmel products, with additional mathematical extensions adapted to the requirements of battery management systems. The circuit has two 16-bit - analog-to-digital converters (ADCs) with synchronous sampling; the accuracy of these two ADCs is due to a very precise 5ppm/K reference voltage with special second-order temperature compensation. The first ADC is dedicated to current measurement and has a large dynamic range. Using a 100Ohm shunt resistor, currents from 1mA up to 1500A can be measured due to a programmable gain amplifier (PGA) at the input of the ADC. The second ADC is dedicated to system temperature determination and monitoring of the voltage of the lead acid battery. The temperature can be measured using either the ICs in-built temperature sensor or an external sensor. The measurement of battery voltage measurement is particularly interesting. This measurement requires an extremely precise resistor divider, which is integrated into the device. As a special feature, the divider can be connected directly to the battery pole to enable precise voltage measurement without failures caused by a reverse protection diode or other protection devices. This is very important because systems require reliable battery data regarding state-of-charge and state-of-health. Reverse polarity protection at this pin also poses no problems because the device is based on SOI (silicon-on-insulator) technology. SOI enables the design of circuits operating at voltage levels
lower than substrate voltage, since SOI devices are isolated by an oxide instead of P-N junctions, which avoids issues with parasitic diodes and transistors. The microcontroller device also has some additional features. First, due to its integrated voltage regulator, it is able to directly use the vehicles power net for operation. Second, it is available with 32kB and 64kB flash memory. Third, in addition to the in-built power supply, it incorporates a LIN transceiver based on the Atmel LIN transceiver IP with excellent electromagnetic compatibility (EMC) characteristics. Atmel products based on this LIN IP have received approval from nearly every OEM worldwide, ensuring smooth integration into a vehicle network. Fourth, Atmel provides a version of this microcontroller without a voltage regulator or LIN transceiver for systems that handle these functionalities externally. Finally, due to the restricted space and harsh environment of IBS applications, Atmel provides the ICs in a robust QFN 7 x 7mm package. This enables customers to design intelligent battery sensors the size of a postage stamp, which fit into the pole niche of almost any standard lead acid battery. In conclusion, the new microcontroller device from Atmel provides an advanced intelligent battery sensor system to ensure starting capability in todays vehicles, including microhybrids.
16-bit S D ADC
Voltage Regulator Oscillators
Watchdog
Timer/Counter
Temperature
Reference Low-power Atmel AVR CPU LIN Transceiver to LIN Bus
Shunt
16-bit S D ADC
Supervision and Diagnostics
28
DC and BLDC Motor Control ICs

Klaus Schweizer
Atmel has more than 25 years of in-depth automotive design expertise. The automotive specialist supplies drivers, predrivers, microcontrollers, and software for DC and brushless DC (BLDC) motor control solutions. This article describes Atmel system basis predriver ICs and emphasizes its most recently developed BLDC motor driver device. One of the key fields Atmel specializes in is application development for controlling brushed DC motors. There are a large number of suppliers offering a wide variety of automotive-qualified DC motors. Automotive applications continue to see a high demand for brushed DC motors. In most cases, these durable motors, based on proven technology, meet the customers application requirements and failure rates are low. But brushless DC motor applications are now emerging, and market analysts have assigned them the highest growth rate. The extended functionalities of BLDC motors are becoming more attractive.
Regardless of whether DC-brush or brushless DC motors are used, the challenge in electronic motor control is to design for new functionality while maintaining reliability and high-performance. Atmel smart drivers and bridges enable a myriad of electronic applications. Manufactured using 0.8m BCD-on-SOI process technology, they support operation up to 40V and can be optimized for motor control applications in harsh environments such as the engine compartment. The Atmel systems basis predriver ICs are of particular interest in managing the shift to brushless motors. To enable new, additional BLDC motor applications, a next generation of advanced system basis B6 predrivers has been developed. Together with Atmel microcontrollers and software, these predrivers will support successful design of future BLDC motor control ICs.
29
Brushed DC Motor Advantages

Brushed DC motors are a proven technology offering several advantages. In addition to low initial cost, brushed DC motors also stand out for their reliability, the high volume of production possible with this technology, as well as the ease with which brushed DC motor speed can be controlled. The low initial cost involved is perhaps the most important benefit. Customers still prefer brushed motor control in pricesensitive applications, which comprise almost all convenience electronics in small and mid-sized vehicles. In addition, many functions in this environment still run without semiconductor driver ICs, making them simpler and cheaper.
In summary, relay control is still the simplest method of controlling brushed DC motors, but perhaps performs better for certain applications with unidirectional operation, i.e., radiator fans, as well as for those with reversed operation, such as power windows.
DC Motor Control in H-bridge

Typically, half-bridge drivers are employed for brushed DC motor control, so it is important to utilize the right combination of microcontroller and driver. In reversed DC motor applications, for example, the DC motor is typically in H-bridge configuration with four power MOSFETs forming the bridge. The ATA6836, a fully protected hex half-bridge driver designed in the Atmel Smart Power SOI technology, can be used by a microcontroller to control up to six different loads. Up to a current limit of about 650mA, the Atmel ATA6836 can be used to drive up to 5DC motors directly in H-bridge configuration. This applies, for example, to flaps in air conditioning systems and side mirrors. In the case of DC motors with higher wattage (about 10W to 800W), Atmel recommends using an integrated gate driver (Atmel ATA6823 or Atmel ATA6824), an Atmel AVR microcontroller (e.g., Atmel ATmega32M1), and four discrete N-channel power MOSFET to be selected according to the DC motors wattage. The H-bridge driver ICs currently available are often simple predrivers that activate and deactivate the gates of discrete power MOSFETs. The approach taken by Atmel, however, is more complex. Based on its extensive experience with stand-alone watchdog ICs and system basis chips including LIN transceivers, low drop voltage regulators, and window
Brushed DC Motor Disadvantages

On the other hand, this technology has its drawbacks. One disadvantage is that carbon brush wear can reduce the motors service life. Also, brush fire of the DC motor may cause EMI. In addition, as indicated above, the low effort and cost of electronic control with brushed DC motors is associated with the use of relays, and a trend is underway to replace relays with transistors. While relays have sufficient reliability in most cases, particularly when motors are rarely switched on, they do have disadvantages. Transistors are increasingly preferred when the following variables are considered: Mechanical stress, vibration Switching frequency High coil current Speed control Size, space Clicking noise
VBAT
Voltage Regulator Atmel ATmega88 or Atmel ATtiny45
2 High-side Drivers
Watchdog Timer
Atmel ATA6823
LIN Transceiver
2 Low-side Drivers
LIN
Figure 1. Atmel ATA6823 Block Diagram
30
Battery CP VBATSW
VBAT
watchdogs in different partitioning Atmel has combined these IPs with push-pull drivers. The drivers are designed using a very robust SOI technology (Atmel SMART-I.S.). The numerous advantages of this technology have been described in great detail in several previous issues of Automotive Compilation and there is no need to detail them here. They include: Low leakage currents High-temperature and highvoltage capability Excellent radiation hardness Improved latch-up immunity High switching frequency
VCC VMODE
CPLO
VBAT
PBAT
CPHI
VINT
VG
VCC Regulator
VINT 5V Regulator
12v Regulator
Charge Pump
CP VRES
Bandgap
/RESET
VBG
S2
Oscillator
OTP 12bit
HS2 Driver
H2
Micro Controller
WD EN1
H1
HS1 Driver
S1
DIR
Logic Control
PWM mode
PWM
LS1 Driver
L1
L2
RX TX
LIN
WD timer
CC timer
VBAT
OV
UV
OT
LS2 Driver
Supervisor
PGND
VCC RWD GND DG1 DG2 LIN DG3 EN2 CC
1. Operating Current 2. Free wheeling current
Figure 2
LIN
In summary, smart H-bridge drivers from Atmel offer a more robust and flexible approach to controlling multiple brushed DC motors.
High-Temperature Applications
When designing brushed DC motor control systems, high-temperature is a second consideration. Taking advantage of Atmel SOI technologys high-temperature capability, the company has designed the Atmel ATA6824. Specifically for the high demands of engine compartment applications where electronic control units are specified for ambient temperatures of 150C or higher. However, a system basis predriver with an integrated 100mA, 3.3/5V linear regulator (designed for such an environment) needs to be qualified for a junction temperature of up to 200C.
solutions. With such as shift, the user expects that manufacturers guarantee full protection. Discrete power MOSFETs, for example, need to be protected against overcurrent conditions. This is typically done by monitoring the drain source voltage that is fed to a comparator integrated in the predriver. A high short-circuit current will flow if a high-side and a low-side power MOSFET are activated simultaneously within the same branch. As a countermeasure, Atmel system basis gate drivers feature implemented shoot-through protection. The dead time can be adjusted individually by choosing an R/C combination at the CC pin of the cross conduction timer. This allows the engineer to flexibly adapt the dead time to the switching characteristic of the power MOSFETs used.
functions into the driver IC. In this specific case, speed and direction control are made extremely easy. The microcontroller needs only two command lines to set the speed and direction of the DC motor: pins DIR (cw or ccw operation) and pin PWM. The moment the PWM signal is low, the system basis predriver activates both high-side drivers so that the free-wheeling current can flow without any additional microcontroller command. The Atmel ATA6823/ ATA6824s extensive control logic supports the microcontroller by taking over several tasks.
Shift to Brushless Motors

As stated above, BLDC motors are only now emerging in automotive applications, although they have been popular in disk drives, industrial applications and hobby electronics for several years. Automotive industry studies reveal that about 80% of DC motor applications is still equipped with brushed DC motors. Market analysis conclusively shows, however, that brushless DC motor applications are growing at the fastest rate. This
Protection
Protection is also an important consideration, particularly when customers are hesitant to adopt a new technology, as with migrating from relays to semiconductor
Smart 2-pin Motion Control

In choosing a motor driver, the user is also looking for good speed and direction control. Characteristically, the architecture of SBCs involves the integration of all necessary peripheral
31
VBAT
Voltage Regulator Atmel ATmega32M1 or Atmel ATtinyx61
3 High-side Drivers
Atmel ATA6833/34
Watchdog Timer
LIN Transceiver
3 Low-side Drivers
LIN
Figure 3. Atmel ATA6833/34 Block Diagram
means that BLDC motors will soon take the lions share of new motor control electronics R&D. The advantages of BLDC motors over brushed motors are obvious: Improved speed vs. torque characteristics High dynamic response High efficiency Noiseless and interference-free operation Extended speed ranges Long operation life Maintenance-free operation is a very important advantage for all systems that operate continuously while the engine is running, such as fuel pumps or variable vanes in turbo chargers. Also, the smaller size and reduced weight are plus factors for BLDC motors. On the other hand, the increased effort and cost of electronic control (including both hardware and software) could slow the replacement of proven DC motors with BLDC technology. Customer resistance to
BLDC can be overcome, however, if the motor control ECU cost is effectively managed by using a BLDC motor control system approach comprising a microcontroller, a system basis gate driver, and the necessary software.
The B6 Predrivers
Currently, vehicle applications mainly incorporate highly efficient 3-phase brushless DC motors. Such motors typically need a B6 bridge to control three high-side and three lowside power MOSFETs. As with the system basis H-bridge predrivers Atmel ATA6823/24, the B6 bridge predrivers Atmel ATA6833/34 include all the elements needed to form a complete system. The system consists of a pin-programmable linear voltage regulator (100mA, 3.3/5V); a LIN transceiver; and a window watchdog, in combination with six push-pull stages. The stages are needed to control the six discrete N channel power MOSFETs which operate 3-phase brushless DC motors.
These predrivers also offer outstanding space saving on the PC board, helping the designer minimize board size. And because the system basis predrivers are assembled in QFN packages with exposed pad, the board design can easily be optimized for perfect heat transfer. The QFN48 7x7 mm package features a thermal resistance of Rthjc = 5K/W, and thus a thermal resistance of Rthja = 20K/W can be achieved with an elaborate board layout.
Inputs/Outputs for all Kind of Commutations

Flexibility is also a must for motion control of power MOSFETs in 3-phase brushless DC motors. In a B6predriver, it makes no sense to use a 2-pin motion control (as with the Atmel ATA6823 or Atmel ATA6824) because flexibility in the control of the power MOSFETs is mandatory for different kinds of commutation. Therefore, power MOSFETs should be controlled from the C separately via the three high-side and three lowside inputs of the Atmel predrivers ATA6833/34.
32
U Nominal
incorporate a broad range of functionalities, some functions should be intentionally assigned to the microcontroller. In case of motor current measurement to detect overcurrent, the designer should feed the voltage drop via a shunt resistor to the analog comparators of one of the microcontrollers (e.g., the Atmel ATmega32M1). These comparators can be used to condition either the Hall sensor signals of a sensor-type motor or the BEMF signals of a sensorless motor.
Figure 4. Crank pulse = 12V; drop: -7V
Operation During Crank Pulses

In order to enable operation of BLDC motor systems during the crank pulse (as defined by ISO7637, see Figure 4), the minimum supply voltage must be as low as 5V and the predriver must be able to activate the high-side and low-side N-channel power MOSFETs. Hence the charge pump needs to be powerful enough to operate both the high-side and the low-side FETs via the internal voltage regulator. Because of this, the Atmel ATA6833/34 devices are ideal for systems which need to be in operation during engine start, e.g., fuel pumps.
The ATA6834 targets hightemperature applications with an ambient temperature of 150C or more. The device is designed and qualified for a continuous maximum junction temperature of 200C. Power dissipation calculations can be found in the application note Estimated junction temperature rise due to power dissipation during operation (see http://www.atmel.com/dyn/ resources/prod_documents/ doc9181.pdf) With temperature warning functionality and switch-off thresholds adapted to high junction temperature, and also with a special bond pad coating, the ATA6834 is perfectly suited for ECUs attached to coolant pumps, turbo chargers with variable geometry, or exhaust gas recirculation systems (EGR).
New Generation: Atmel ATA6843, Atmel ATA6844

To enable new, additional BLDC motor applications that take advantage of extended functionalities, a next generation of advanced system basis B6 predrivers has been developed. Like their predecessors, they consist of six push-pull stages combined with a 100mA, 3.3/5V pin-programmable linear voltage regulator; a window watchdog with a separate time base that is independent of the microcontroller; and a LIN transceiver. Since this LIN transceiver is based on the stand-alone LIN transceiver IP, Atmel ATA6663, the system basis gate drivers LIN section meets the high ESD protection requirements at the 8kV LIN pins and complies with all relevant conformance testing requirements. The increased functionalities of the next-generation devices Atmel ATA6843/44 include: Expanded overvoltage detection threshold (up to 30V). This is an important feature for all applications requiring full functionality during jump-start
B6 Driver Versions
As with as the H-bridge predrivers ATA6823/24, which are both available for the standard and the extended maximum temperature range, there are also two hightemperature B6 bridge driver types available for 3-phase BLDC motor control. The ATA6833 is rated for a maximum junction temperature of 150C. For all convenience applications, and also some powertrain applications, this temperature is sufficient.
Universal Use
Because BLDC motor applications involve multiple types of signal conditioning, universal use of system basis gate drivers is recommended. These drivers do not handle Hall sensors and BEMF signals, for example, but can be used universally to make stock-keeping easier. And although system basis devices
33
Atmel ATmega32M1
PSCOUT0A PSCOUT0B PSCOUT0C PSCOUT1A PSCOUT1B PSCOUT1C Over Current Current HallA HallB HallC PSC : Power Stage Controller ACMPi : Analog Comparator Positive Input (i = 0,1,2) AMPi+/- : Analog Differential Amplified Channel Positive/Negative Inputs (i = 0,1,2)
Figure 5.
can therefore be equipped with BLDC motors:

PH_A PH_B PH_C
PSC
Power Bridge + Shunt Resistor
Motor
PSCIN0 AMP1+;AMP1ACMP0 ACMP1 ACMP2
Electronic throttle control Exhaust gas recirculation Turbo charger with variable geometry Dual clutch Automated manual transmission Fuel pump Coolant pump Radiator fan Variable manifold Synchronous rectifier Electric power steering Unlike the standard-temperature version Atmel ATA6843, the warning threshold of the ATA6844 is set to 150C, and its excess temperature switch-off threshold is set to 200C. This makes this IC able to cope with the environment of electronic devices attached to parts of the engine, such as coolant pumps, EGR, or turbo chargers. Using NiAu plating on the Aluminum pads, the Atmel hightemperature ICs are able to withstand the challenges of high ambient temperatures over a products complete life cycle. The plating avoids dangerous gold-aluminum corrosion (purple plaque).
conditions, such as fuel pumps. In this case, the supply voltage is in the range of 5V to 30V. Adjustable and very low shortcircuit-detection threshold for increased compatibility with low-impedance power MOSFETs. The drain source monitoring can be adjusted by feeding a voltage in the range of 0.5V to 3.3V to pin SCREF. An internal voltage divider sets the detection threshold to 2.5V if the pin is left open. Note that the devices Atmel ATA6833/34 do not provide this pin. A digital input at pin COAST connected to the logic control. If this input is activated, all power MOSFETs are switched off, allowing the motor to coast. In some situations, the motor is asked to coast until it stops. Also, the coast function may be used in the case of overvoltage (e.g., load dump) or to reduce speed before reversing the BLDC motor. Note that this pin is not available in ATA6833/34.
providing freedom to control discrete power MOSFETs, e.g., for sinusoidal commutation. The new predrivers also allow simple and cost-effective control of MOSFETs via the microcontrollers three command lines: the gate driver ICs high-side and low-side inputs are designed with opposite input logic (ILx and /IHx).
High-temperature ICs for Hot Applications

These automotive applications are located within hot environments and
VBAT VBAT COAST 3.3/5V VCC Regulator Supervisor: short circuit openload over temperature under voltage
VMODE
VINT VG
CPHI CPLO VRES
PBAT H3 H2 H1 S1 S2 S3 L1 L2 L3
VCC DG1
13V Regulator 13V Regulator
CP
High Side Driver 3 High Side Driver 2 High Side Driver 1
Atmel ATmega32M1 or Atmel ATtiny157
DG2 DG3 /RESET WD IH1-3 IL1-3 Clear RX Hall A Hall B Hall C TX
VBG
Atmel ATA6843/44 Osc.
Logic Control
Low Side Driver 1 Low Side Driver 1
LIN LIN ENx
WD timer RWD
CC timer CC
Low Side Driver 3 SCREF GND
PGND
The new predrivers include 6-pin control (as with the ATA6833/34),
Figure 6. Atmel AT6843/44 Block Diagram
Hall A Hall B Hall C
34
Table 1 . System Basis B6 Predrivers Overview

Tj
ATA6833 ATA6834 ATA6843 ATA6844 150C 200C 150C 200C
Operating Range
5-20V 5-20V 5-30V 5-30V
Drain Source Monitoring

4V 4V 0.5-3.3V 0.5-3.3V
Coast Function
No No Yes Yes
3-pin Control
No No Yes Yes
Atmel AVR Microcontollers

To support the growing demand for BLDC motors, Atmel supplies not only systembasis gate drivers, but microcontrollers with software support as well. The automotive market for electronics is growing rapidly as the demand for comfort, safety and reduced fuel consumption increases. All of these new functions require local
intelligence and control, which can be optimized by the use of small, powerful microcontrollers. Taking advantage of its unsurpassed experience in embedded Flash memory microcontrollers, with a large number of devices of Atmel AVR devices from 8- to 32- bit microcontrollers. Atmel brings innovative solutions, whether for sensor or actuator control or more sophisticated networking applications. These microcontrollers are fullyengineered to fulfill OEMs quality
requirements towards zero defects. Several AVR microcontrollers are qualified for operation up to + 150C ambient temperature (AEC-Q100 Grade0). Designers can distribute intelligence and control functions directly into or near gearboxes, transfer cases, engine sensors actuators, turbo chargers and exhaust systems.
Links
Overview of Atmel Automotive Microcontrollers http://www.atmel. com/products/automcu/default. asp?source=overview_automotive Application note, AVR194, featuring BLDC motor basics, hardware implementation, and code example. http://atmel.com/dyn/resources/prod_ documents/doc8138.pdf System basis B6 predrivers: Atmel ATA6833 Data Sheet: http://www.atmel.com/dyn/resources/ prod_documents/doc9122.pdf Atmel ATA6834 Data Sheet: http://www.atmel.com/dyn/resources/ prod_documents/doc9122.pdf Application note: http://www.atmel. com/dyn/resources/prod_documents/ doc9143.pdf
35
EMC Synonym for Exasperating, Magic, Confusing?

Juergen Strohal
The acronym EMC stands for electromagnetic compatibility and means the ability of an electronic device (or a module, printed circuit board or integrated circuit) to operate in an electromagnetically distorted environment while keeping its own distortions below certain thresholds so that other devices do not suffer any serious adverse effects. For many people this field of expertise seems more like a form of black magic: No matter what work is done related to EMC, not only will something completely unexpected and unpredictable happen, worse than that it can be assumed that things will always take a turn in the wrong direction. It will become apparent through this article whether this viewpoint is accurate and what engineers can do to gain more facility in this field. Those involved in the field of electronics, especially in the automotive sector, will certainly have been confronted by EMC-related issues more than once. The phenomenon of radio interference is nearly as old as the invention of radio itself and at an early stage led to the definition of guidelines for noise suppression. The other part of EMC, the immunity against distortions, only began attracting attention around fifty years ago. While it is merely a nuisance to experience radio interference, it is definitely a serious matter if an ABS or airbag may also be affected by interference, e.g., when passing a TV tower. Not only is the number of electronic control units in cars on the rise, so too is the number of electronic devices frequently used inside the cars such as cellphones, portable navigation devices, wireless headsets which may also cause interference. Making matters even worse is that more and more of devices of this kind operating at higher and higher frequencies are constantly being introduced to the market. Higher frequencies imply that
smaller structures can behave like an antenna and crosscoupling needs to be considered even for relatively small coupling capacitances. It is therefore only natural that there has been a growing need to define certain rules of the game over the past decades. Today all car manufacturers are aware that EMC testing is an important part of car electronics development and understand that EMC issues become costlier the later they are discovered. That is the reason why they do not just rely on a final test inside the car but insist on tests of the electronic control unit (ECU) and even on test results of the integrated circuits used in the design before deployment in vehicles. All around the world a wide variety of test methods have been developed for both unwanted electromagnetic emissions as well as the susceptibility for electromagnetic distortions. In the meantime, all integration levels are covered and over the past 10 years the various standardization committees have devoted their time to the IC level. As a semiconductor manufacturer, Atmel is confronted primarily with IC-level and ECU-level tests. Unfortunately, not only have there been quite a large number of different standards established (perhaps a bane to testing specialists), at the same time, many OEMs apply these standards in slightly different ways. Two different types of measurements can be distinguished in the case of emission as well as susceptibility tests: The radiated measurements, involving an antenna, a coupling clamp, a magnetic or electric probe, a stripline or a TEM cell
36
Measurements carried out using galvanic coupling to certain ports to measure or inject RF signals Semiconductor companies have do deal mostly with testing at the IC level, however certain applications such as automotive networking systems also require the proof that some module level tests are passed by those ICs that include integrated bus line transceivers. The classic EMC tests at the IC level measure the emissions in varying frequency bands at defined frequency steps, dwell times, measurement bandwidths and detector types as well as measuring immunity by applying an unmodulated or AM-modulated RF signal of defined amplitude, again while varying the frequency at defined steps. These tests are specified in the international standards IEC 61967 for emission and IEC 62132 for immunity. Currently these standards are supplemented by new test standards for pulse measurements. Most ports at the IC level are designed to be connected internally on the PCB only while just a few ports, such as supply pins, bus lines or antenna pins, are connected externally. While EMC requirements are less stringent for local (internal), the global (external) ports are far more critical because the cable lengths involved increase the cross-coupling between different lines or may behave as an undesired antenna for RF signals. Certainly a few ports such as those for bus lines are external ports by definition, but for most ports it is the application that determines whether they need to be treated as local or global ports. The good news is that designing circuits and board layouts which are robust will not only ensure an assembly works better and is more dependable but in most cases also help to achieve the required performance regarding both electromagnetic emissions and immunity. Having said this, how can circuit design be achieved which is that robust? Of course there are a few general rules which help to improve the EMC behavior of the circuit: Careful consideration should be given to what clock frequency is really needed for the development application. The lowest clock frequency possible should be chosen because this is the first measure which will reduce electromagnetic emissions. High-impedant ports are susceptible to RF distortions; therefore, impedance should be kept as low as acceptable or a low-impedance path to GND for the RF disturbance should be provided. If an integrated
circuit indicates that some GND pins are related to certain VCC pins or ports, where decoupling caps are placed should take this knowledge into account. Ports that are connected on the outside of the ECU require special attention plan for decoupling capacitors to GND and series resistors if possible 10 to 100 are often acceptable, whereas higher values will form a more efficient filter but also cause a higher voltage drop for DC signals. If emission of a particular port is the problem, one end of the resistor is connected to the port while the capacitor is on the other side. To protect a port against RF distortions the components are arranged inversely. For frequencies >10MHz ferrite beads may be more efficient than only utilizing small resistors and as a further benefit their DC resistance is negligible, meaning the circuit will not suffer from a drop in voltage. For higher frequencies a capacitor functions not only as a capacitor but also has some inherent, built-in parasitic components such as series inductance and resistance known as equivalent series resistance (ESR) to name only the most important components with adverse effects. Because the correct choice and placement of decoupling capacitors is of vital importance, this will be discussed in more detail below. In addition, resistors need to be regarded as a more complex component (highly depending on the type of construction and also on the resistance value). Fortunately, for a typical low-ohmic thin film resistor as they are used for building EMC filters, the contribution of its parasitic components is largely negligible up to 1GHz. When developing the PCB layout, the different circuit blocks should be arranged in such way that sufficient space around sensitive inputs is provided towards switched signals of high amplitude and/or frequency because of the possible interferences these signals may cause. Having parallel tracks on a board provides good coupling between the signals on these tracks if this kind of coupling is not desired, insert some GND area between the tracks; should these tracks cross each other at different layers, have the tracks cross at right angles to minimize the coupling area. Keep tracks as short as possible, especially those carrying RF or switched signals with fast slopes. Critical parts of the circuit such as tuners may require shielding. The highest frequency the circuit produces or is exposed to should be considered and the track length of critical connections kept below 1/10 of the wavelength. Two considerations must be kept in mind here: First, the
37
wavelength on the PCB is shortened due to r of the board material, for FR4 this is typically around 4.5; however the effective r will be somewhat lower because part of the electric field of a micro strip line is in free space. For a frequency of 3GHz the formula
Capacitor
ESR
Parasitic Inductance
Cp
Rp
reveals that a track length of about 50mm already equals /10. Secondly, the highest frequency in the circuit is Capacitor ESR soParasitic Inductance determined by the fastest slopes if some parts of the design operate at 1MHz but with slopes of 1ns, there will be frequencies of at least 500MHz on the Cp PCB. Differential signals must be routed close to each other with the same track length for both lines. Avoid generating large loops and keep the path for the return Rp current in mind. The larger the area of a loop, the higher the susceptibility and the lower the frequencies which may impact the circuit. Analogously, this is valid for emission, too any tracks forming a loop with an RF current flow can behave like a loop antenna. The ground plane should be designed as solidly as possible, preferably using a multilayer PCB with dedicated layers for GND and power planes. Typically, the signal layers will be on the top and bottom side of the PCB and the GND/power planes on the inner layers. It is advantageous to keep the distance between signal layer and adjacent GND/power plane to a minimum. Doing so helps achieve relatively low track impedance even for fairly thin connections. Slots in the GND plane should be avoided to prevent creating unwanted slot antennas. In addition, small islands should be avoided, different GND areas need to be connected using a sufficient number of vias (one via every 3 to 5mm is sufficient for most designs). Hopping between the layers should be kept to a minimum during board development. Every via, especially long ones from top to bottom layer, involve some inductance; as a rule of thumb this is in the range of 0.5 to 1nH. Particular care needs to be taken about GND connections of decoupling capacitors. Atmel highly recommends placing several vias in parallel close to the respective capacitor.
Figure 1. Equivalent Circuit of a Capacitor
Resistor
Parasitic Inductance
Cp Shunt Capacitance
Figure 2. Equivalent Circuit of a Resistor
Equivalent Circuit Model of the Capacitor

In the capacitors equivalent circuit model, the simplest model comprises just a serial connection of the nominal capacitor, an equivalent series resistance and a parasitic series inductance. The ESR determines the lowest impedance reached at the capacitors series resonance. Above this series resonance the capacitors impedance will increase with frequency, thus behaving like an inductor. A more sophisticated model would also include the components Cp and Rp, connected in gray in Figure 1. Modified equivalent circuits are also found in literature which show Cp and Rp in parallel to the whole serial connection of (desired) capacitor, ESR and parasitic inductance; it is merely a question of transforming the values of the respective inherent components. The parasitic inductance together with Cp leads to a parallel resonance that is frequently neglected, because such parallel resonance of typical SMD ceramic capacitors will only appear at several GHz. The series resonance of the capacitor is determined by its type (electrolytic, foil, ceramic), mechanical dimensions (axial, radial, SMD, size) and of course its value. The higher the capacitance of a certain capacitor type, the lower the series resonance frequency. Therefore, it is advisable not just to place a single capacitor for decoupling purpose, but combine two or several caps to achieve broadband decoupling. For example, it is often recommended to pair a
38
Figure 3. Attenuation of two Ceramic Capacitors of 10nF and 100pF in Parallel in a 50 System
10nF capacitor for lower frequencies with a 100pF cap for higher frequencies. The following will explore whether this is advisable. A very basic linear RF simulation tool is sufficient for demonstrating this; there are even freeware tools available for this purpose. Many manufacturers of ceramic capacitors supply S-Parameter files for their products and it is advisable to use them. Figure 3 shows the attenuation of the above two capacitors when placed in parallel from a 50 track to GND. This looks quite acceptable. Attenuation of at least 30dB was achieved for frequencies between 20MHz and well above 1GHz. If a higher reference impedance than 50 had been used, it would look even better. In a rather ideal world, it would be possible to stop here. But has something been overlooked? In reality, it is not possible to connect the capacitors perfectly to GND or to the track or pad which needs to be decoupled. Every track on the PCB above behaves like a transmission line and its impedance is determined primarily by track width, the thickness of the PCB or, in case that a multilayer PCB is used, the distance between signal and GND layer, the distance to adjacent GND areas and the dielectric constant r of the PCB material. Again, there are special books and free calculation tools available for guidance on this issue. With track widths of 0.2mm, GND area >0.5mm away from the track and r = 4.7, impedance of well
above 100 for 2-layer boards (1.6mm standard thickness) will result and close to 50 will result for a multilayer board with 150m distance between signal layer and GND plane. When considering the red curve in Figure 4, it is obvious it looks quite different from the previous graph. Figure 4 shows how decoupling performance changes if the board layout is not done with care. The assumptions for the red curve were as follows: standard 2-layer PCB, distance between caps and to their GND vias: 10mm with just one GND via per cap. Now, quite unexpectedly, there is a highly undesirable resonance around 130MHz with attenuation of only 6dB. The green graph in Figure 4 shows the performance for an improved board layout. Now a multilayer board is used, the two caps are closer to each other and each one has two GND vias only 1mm away from the respective cap. The resulting decoupling performance is significantly improved, but there still seems to be room for improvement. The lesson learned from this example is, first, the decoupling caps need to be as close as possible to each other and to the component which needs to be decoupled. Secondly, using a multilayer PCB with a GND plane just below the signal layer is beneficial too. And, finally, it appears to be a good idea to do some simulation with real capacitors rather than just
39
Red: Distance between caps and their GND vias: 10mm, one GND via per cap, track width: 0.2mm, board thickness top to GND: 1.6mm.
Green: Optimized, distance between caps: 5mm, between caps and their GND vias: 1mm, two GND vias per cap, track width: 0.2mm, board thickness top to inner GND plane: 0.15mm
Figure 4. Attenuation of two Ceramic Capacitors of 10nF and 100pF in Parallel in a 50 System with Realistic Parasitic Components in Connecting Tracks
selecting them based on instinct. Innovative layout designs even place only a single centralized group of capacitors to decouple a larger area but that is something which should not be attempted without careful simulation. As mentioned previously in this article, car manufacturers are aware that EMC issues generally become costlier the later they are discovered. Engineers may benefit from this insight within their own development work. Giving some thought to EMC behavior already when designing a circuit definitely helps avoid unpleasant surprises during EMC approval testing. Having said that much what needs to be done if a design fails EMC testing despite the care exercised prior to testing? Or, just in case it was not possible to include sufficient time, budget or experience appropriate for EMC assessment in a project, what can be done to improve matters? The truth is that no standard procedure exists. If emission is the problem, a probe across the circuit can be attempted with a field probe to detect any potential hot spots. Or, if fast enough, it is possible to re-perform the particular failed emission measurement while connecting a short isolated wire to some suspicious spots on the PCB. If a critical one shows up, the number of spurs will increase;
this becomes immediately noticeable on the connected receiving instrument. If the design performs weakly in terms of susceptibility, consideration must be given to what parts of the circuit are affected (this can often be deduced from the malfunction occurring during the immunity measurement), and the coupling path must be located. Once the critical parts of the circuit have been pinpointed, the techniques described above can be used to improve EMC performance. A key consideration in this regard is that effective decoupling requires a solid GND area. If this is lacking, it may be easier to redesign the board first; alternatively some copper foil could be added to facilitate further optimization measures in the lab. Hopefully it has become apparent that there is nothing magical at all about EMC and that it is simply applied physics. Naturally, our knowledge about coupling mechanisms and particularly their parameters tends to be inaccurate and sometimes incomplete, even if highly sophisticated electromagnetic simulation tools are used. So there is a trace of magic (or uncertainty) left in the process after all. But these imponderables are also what challenge us daily in this specialized discipline.
40
EMC Pulse Immunity System-level Tests

Stephan Gerlach System-level tests for EMC pulse immunity have been established in order to test the immunity of an electronic device or module against strong electromagnetic fields (e.g., the near field of an AM or FM transmitter station or the short but intense shock from a flash that can damage or at least disable vehicle electronics). Many car manufacturers and Tier1 suppliers have tried to simplify test procedures and to standardize both the test set-up as well as workbenches. By using mature but still valid test cases they aim to reduce the use of wiring harness, interacting modules and the cars body to only the basic functional minimum. Unfortunately, this led to a large number of experimental set-ups that were similar but incompatible for comparing technical parameters. Two examples developed by car OEMs will be explained in this article. Figure 2 shows an example that is comparable to the set-up of Figure 1 regarding functionality, but leads to differing results.
Connector Pin
Connector
Wire Communitcation Bus or Remot I/O Harness
BATTERY
Default Connection for Bleed-Off Resistor
Test Fixture
DUT
Insulator Insulator
Table with Ground Plane
Bleed Off Resistor
ESD Simulator
ESO SImulator GND
Chamber of Facility Floor
Figure 1. Principle of Car Wiring Harness Pulse Immunity Test Using an ESD Simulator (ESD Pulse Gun)
41
EST
Electrostatic Discharge Generator
Actuators Power Supply Sensors
Figure 2. Workbench of a Major French Car Manufacturer
1 2 4 6 7
+ -
8 9
10
80mm 100mm >=200 mm
5
10mm 1500mm >=100mm
11
Figure 3. EMC Pulse Immunity System-level Test According to ISO 10605 Annex F
(1 = metal ground plane, 2 = ESD resistors, 3 = reference ground, 4 = optional ground line, 5 = DUT1 (master), 6 = wiring harness, 7 = car battery, 8 = DUT2 (slave), 9 = artificial mains network, 10 = battery ground, 11 = ESD pulse generator with gun and ground cable)
GR
OU
Battery
ND
50mm
PL
AN
50mm
42
To overcome this obstacle, a major German car manufacturer started an initiative that was accepted by all major car OEMs. The proposed test set-up simulates a wiring harness that is exposed to ESD events. Since 2007 this setup has been part of the international standard ISO 10605 as Annex F.
Functionality of EMC Pulse Immunity System-level Testing

By using an ESD gun, different pulses with positive and negative polarity in the range of 1 to 30kV are applied to three test pads. The so-called conducted discharge is utilized to guarantee the highest possible repeatability which is a vitally important characteristic of any test procedure. Nevertheless, it cannot be ruled out that a single pulse delivers the same reaction on the device under test. The standard procedure therefore prescribes a repetition rate of 10 ESD pulses for each polarity. To guarantee 100% test reliability, this procedure has to be repeated on the test fixture for all three pads. The reason for this is quite simple: To a certain extent, the shape and amplitude of the stimulated waveform within the wiring harness is arbitrary. Moreover, because the ESD pulses are applied to a living system (i.e., ongoing LIN communication between master and slave) they have a different impact on the DUTs failure behavior until final destruction. Figure 4 shows that current spikes of up to 37A are created with a burst frequency of several hundred MHz. As explained above, neither the pulse shapes nor the effects are identical. A DUT can withstand five or seven pulses, but the 8th pulse will cause final destruction.
Figure 4. Stimulated Transient Signal Measured with Current Probe (Clamp-on Ammeter)
Figure 5. Workbench System-level EMC Pulse Immunity Test in Application Lab
Figure 6. Workbench System-level EMC Pulse Immunity Test Pulse Measurement
43
Atmel Corporation 2325 Orchard Parkway San Jose, CA 95131 USA Tel: (+1)(408) 441-0311 Fax: (+1)(408) 487-2600 www.atmel.com
Atmel Asia Limited Unit 01-5 & 16, 19F BEA Tower, Millennium City 5 418 Kwun Tong Road Kwun Tong, Kowloon HONG KONG Tel: (+852) 2245-6100 Fax: (+852) 2722-1369
Atmel Munich GmbH Business Campus Parkring 4 D-85748 Garching b. Munich GERMANY Tel: (+49) 89-31970-0 Fax: (+49) 89-3194621
Atmel Japan 9F, Tonetsu Shinkawa Bldg. 1-24-8 Shinkawa Chuo-ku, Tokyo 104-0033 JAPAN Tel: (+81)(3) 3523-3551 Fax: (+81)(3) 3523-7581
2010 Atmel Corporation. All Rights Reserved. Atmel, Atmel logo and combinations thereof, AVR, QTouch, picoPower, SMART-I.S. and others are registered trademarks, or trademarks of Atmel Corporation or its subsidiaries. Other terms and product names may be trademarks of others.
Disclaimer: The information in this document is provided in connection with Atmel products. No license, express or implied, by estoppel or otherwise, to any intellectual property right is granted by this document or in connection with the sale of Atmel products. EXCEPT AS SET FORTH IN THE ATMEL TERMS AND CONDITIONS OF SALES LOCATED ON THE ATMEL WEBSITE, ATMEL ASSUMES NO LIABILITY WHATSOEVER AND DISCLAIMS ANY EXPRESS, IMPLIED OR STATUTORY WARRANTY RELATING TO ITS PRODUCTS INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTY OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, OR NON-INFRINGEMENT. IN NO EVENT SHALL ATMEL BE LIABLE FOR ANY DIRECT, INDIRECT, CONSEQUENTIAL, PUNITIVE, SPECIAL OR INCIDENTAL DAMAGES (INCLUDING, WITHOUT LIMITATION, DAMAGES FOR LOSS AND PROFITS, BUSINESS INTERRUPTION, OR LOSS OF INFORMATION) ARISING OUT OF THE USE OR INABILITY TO USE THIS DOCUMENT, EVEN IF ATMEL HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. Atmel makes no representations or warranties with respect to the accuracy or completeness of the contents of this document and reserves the right to make changes to specifications and products descriptions at any time without notice. Atmel does not make any commitment to update the information contained herein. Unless specifically provided otherwise, Atmel products are not suitable for, and shall not be used in, automotive applications. Atmel products are not intended, authorized, or warranted for use as components in applications intended to support or sustain life.

Atmel Autocompilation Vol7 Dec2010

Enviado por

Dados do documento

Direitos autorais

Formatos disponíveis

Compartilhar este documento

Compartilhar ou incorporar documento

Opções de compartilhamento

Você considera este documento útil?

Este conteúdo é inapropriado?

Direitos autorais:

Formatos disponíveis

Atmel Autocompilation Vol7 Dec2010

Enviado por

Direitos autorais:

Formatos disponíveis

Automotive Compilation

Volume 7 December 2010

Configurable RF Architecture Gives Engineers Greater Design Flexibility

RFIN _HB RFIN _LB

ATA5830 Atmel ATA5830

Elements Inductor Elements Capacitor Inductor Capacitor Crystal Crystal

Figure 1. Atmel ATA5830 Transceiver Application Circuit

RFIN _HB RFIN _LB

Atmel ATA5780 Atmel

10 11 12 13 14 15 16 NPWRON2 17 10 11 12 13 14 15 16 VS = 4.5v to 5.5v VS = 4.5v to 5.5v

Figure 2. Atmel ATA5780 Receiver Application Circuit

Elements Inductor Elements Capacitor Inductor Capacitor Crystal Crystal

Automotive Compilation Vol. 7

Design and Security Considerations for Passive Immobilizer Systems

The Physical Layer

Field Generation and Modulation

Magnetic Field Charges Key Fob

FSK Key Fob Return Frequency (red)

FSK Key Fob Return Frequency (blue)

Automotive Compilation Vol. 7

Magnetic Field Charges Key Fob

System Interfaces: the Logical Layer

Figure 4. BPLM Coding Method

Clock Data Manchester Bi-phase

QPLM bits frame example

Figure 5. QPLM Coding Method

Automotive Compilation Vol. 7

Unilateral Authentication Protocol

Read UID Command

Key Memory ID Memory ID

4-bit Command + 4-bit CRC 8-bit Header + 32-bit ID + 8-bit CRC

8-bit Command + N Challenge-bit + 8-bit CRC

8-bit Header + M Response-bits + 8-bit CRC

Start Authentication Command

Figure 7. Unilateral Authentication

Key Learn: Open/ Secure

Open Key Learn 1/2

Pass Y Response ( Pass/Fail )

Figure 8. Open Key Learn

Bilateral or Quasi-mutual Authentication

Automotive Compilation Vol. 7

Bilateral Authentication Protocol

Read UID CMD

4-bit Command + 4-bit CRC 8-bit Header + 32-bit ID + 8-bit CRC

Start Authenitcation Command

128-bit Key 1 AES-128 encryption 128-bit Key 2 AES-128 encryption

AES -128 encryption

Ok, is it the right car, continue

Hidden challenge (HCH) expanded to 128 bits

AES -128 encryption 128-bit Key 2

Ok, is it the right key, car match

Figure 9. Bilateral Authentication Protocol

System Security Considerations Attacks and Countermeasures

Algorithmic Security and Countermeasures

Physical/Logical Security and Countermeasures

Protocol Security and Countermeasures

System Performance Considerations

Automotive Compilation Vol. 7

Authentication Response Time

Automotive LIN Bus Driving Sensor Applications

Automotive Compilation Vol. 7

Figure 2. Basic Voltage Interface Set-up

99% 98% 95% 86%

Figure 5. Capacity Discharge Depending on the Bus-dominant State Time