• Tidak ada hasil yang ditemukan

CAN-FD Transceiver Design & Implementation with Improved Matching on Switching Behavior

N/A
N/A
Protected

Academic year: 2023

Membagikan "CAN-FD Transceiver Design & Implementation with Improved Matching on Switching Behavior"

Copied!
52
0
0

Teks penuh

In the automotive industry, automotive integrated chips (ICs) require high levels of electromagnetic immunity (EMI) and low electromagnetic emissions (EME). EME is more important for the Flexible Data Rate Controller Area Network (CAN-FD) because the CAN bus acts as an antenna and interferes with the other automotive ICs. In this thesis, a new architecture of the CAN-FD transmitter with improved tuning of the CAN bus to switching behavior is proposed, resulting in the reduction of EME.

The proposed transceiver presents a new CAN bus drive transceiver architecture using 30-level cascaded current sources and a cross-control method, which can increase CAN bus matching in switching behavior. The measurement results show that the result is better for CAN bus matching than other CAN-FD receivers and transmitters.

Introduction

Classical CAN and CAN-FD

EMC Issues of the Automotive ICs

Design Considerations for CAN-FD Transceiver

  • Design Considerations for Automotive Environment
  • Specific Requirements for Classical CAN
  • Specific Requirements for CAN-FD
  • Additional Design Considerations for CAN-FD Transceiver

In ISO 11898 there are specifications for the physical CAN layer, which refers to the CAN transceiver. This is because the CAN bus can be shorted to the supply lines even in dual battery and inverted dual battery mode. The transmitter sends the signal to the CAN bus and converts a digital signal to an analog signal.

The additional design considerations of the CAN-FD transceiver are to extend IC life, reduce EME, and increase RF immunity, which can lead to EMI. By lowering the on resistors relative to the load, the CAN bus current is load dependent. Second, it is difficult to match symmetries, including matching the CAN bus and the on resistors of both switches.

Defined as VCAN_H+VCAN_L in ISO 11898-5, driver symmetry is the matching parameter of the CAN bus. When applying differential signals to the CAN bus, transients and noise can occur in the common-mode signal. Thus, matching the CAN bus of the two switches is necessary for canceling each other's bus line EM fields.

But this structure cannot resolve the CAN bus matching in switching behavior when transitions occur on the CAN bus. Also, circuit stability is difficult to achieve as the bus topology changes. The split termination is between the termination resistors and the CAN bus capacitor.

If the receiver can detect a small differential signal in a noisy common-mode signal, high immunity of the CAN-FD transceiver can be achieved [22]. There are some approaches that reduce bus noise by adding bypass capacitors to the receiver inputs [7]. At the end of the phases, the hysteresis comparator can be used to generate a signal that is robust against the disturbances of the CAN bus.

Table 2.2 Maximum ratings of CANH, CANL, and VSPLIT [2].
Table 2.2 Maximum ratings of CANH, CANL, and VSPLIT [2].

Principle of Proposed CAN-FD Transceiver

  • Features of the Proposed Architecture of CAN-FD Transceiver
  • Structure of Transmitter
  • Structure of Receiver
  • Concept of Cross-control Method
  • Application of Cross-control Method in the Transmitter
  • Circuit Implementations of the Cross-control Method
  • Simulation Results
    • CAN-FD Transceiver Version I: Core-structure Version
    • CAN-FD Transceiver Version II: Stand-alone Version
    • Simulation Results of Proposed CAN-FD Transceiver
  • Layout Information

PD is separated by PD_H and PD_L which drive both sides of the power stage. RxFEOfs block is the block that makes the offset to recognize the recessive state of the CAN bus signal. It converts the CAN bus signal to the digital signal for the Digital block of the CAN transceiver.

This method matches the average operating time of both switches regardless of process variations. The crossover control method controls both sides of the current source, known as PSW, which drives the CAN bus. 2.2, each driver independently turns on and off each side of the switch in the power phase, resulting in an inconsistency in the switching behavior of the CAN bus.

The PSW unit consists of 4 pairs of high and low side current sources driving the CAN bus, 3 phases of cross control mirrors (#1 ~ #3) to match the driving strength of the high side current sources and low. Each of the current sources is checked by each stage of the cross-check mirrors. The cross-check mirror controls the incoming current sources located on the opposite side of the cross-check mirror.

At this moment, the current bias is activated by turning on the first current sources about 270 μA of current generation. 3.12 (b) shows that the first phase of the cross-check mirrors which are connected to the other side of the current sources turn on the second current sources. By turning on the current sources, the second stage of the cross-check mirrors is simultaneously activated, as in Fig.

The PD block is a pre-driver of the transmitter connected to the PSW block of the device. OPDH and OPDL are output pins connected to the unit's PSW block, which is a single power stage in the CAN bus. It is the core part of the proposed transmitter which improves the matching of the CAN bus on switch behavior presented as the cross control method.

3.16 (a) shows a power stage with an IFSW block, which is a power stage protection block. They are controlled by comparing the reference voltage with the voltage level of the CAN bus.

Fig. 3.1 Block diagram of the transmitter part.
Fig. 3.1 Block diagram of the transmitter part.

Measurement Results

Experimental Set-up

Measurement Results Compared with Other Products

VCANH+VCANL parameter shows 700mV variation while TCAN1041GVDQ1 shows 1.63V variation. These results show that the proposed receiver is better in matching the CAN bus in switching behavior.

Fig. 4.5 shows the comparison between the proposed transceiver and TCAN1041GVDQ1 transceiver  at 5Mbps with 200ns of the bit time
Fig. 4.5 shows the comparison between the proposed transceiver and TCAN1041GVDQ1 transceiver at 5Mbps with 200ns of the bit time

Future Works

Furthermore, when applying CAN-FD to different network topologies, such as star topologies or hybrid topologies, as shown in Fig.

Fig. 5.2 Generation mechanism of ringing on the CAN-FD bus [28].
Fig. 5.2 Generation mechanism of ringing on the CAN-FD bus [28].

Conclusions

4] Atmel, CAN-FD and Ethernet Create Fast Reliable Automotive Data Buses for the Next Decade, 2013. 7] Won-Hee Jo, “A CAN Transceiver for an Intelligent Output ASIC of Automotive Electronic Control Units: Design, Implementation and Measurement”, 2016. Catrysse, “Towards an EMC Guide for Integrated Circuits Asia-Pacific Symposium on Electromagnetic Compatibility and 19th Zurich International Symposium on Electromagnetic Compatibility, Singapore, 2008, p.

Steyaert, "An EMI Resisting LIN Driver in 0.35-micron High-Voltage CMOS," in IEEE Journal of Solid-State Circuits, vol. Steyaert, "EMI Resistive Slimkrag-Integrated LIN Drywer with Reduced Ramp Pump IEEE Custom Integrated Circuits Conference, San Jose, CA, 2008, pp. Janschitz, 'n EMI Robust LIN Drywer with Low Electromagnetic Emission th International Workshop on the Electromagnetic Compatibility of Integrated Circuits (EMC Compo), Edinburgh, 2015, pp.

Nascimento, "EMC-EMI Optimized High Speed ​​CAN Line Driver th Symposium on Integrated Circuits and Systems Design, Florianopolis, 2005, pp.

Gambar

Fig. 1.1 A signal definition of the CAN protocol at CAN bus [3].
Fig. 1.2 Data frames compared with classical CAN and CAN-FD [6].
Fig. 2.2 is the most common structure of a transmitter which drives the CAN bus. The power stage  consists of two pairs of Lateral Diffused MOS (LDMOS) operating as switches and diodes protecting  supply rails from the reverse polarity voltage of bus pins
Fig. 2.4 CAN transmitter using current sources at the power stage of transmitter [22]
+7

Referensi

Dokumen terkait

The 30° head-up position is suggested according to previous studies, which can decrease ICT and increase the pressure of cerebral perfusion compared to the supine position.8 A 30°