ARCNET .

The universal, realtime capable fieldbus solution

Take the lead with ARCNET !

AUG - ARCNET user group e. V.

AUG - ARCNET user group e. V.

CONTENTS 1 ARCNET, the universal, realtime capable Fieldbus Solution ____3 2 History________________________________ _________________ 4 3 Characteristics of modern Fieldbusses _____________________ 4
3.1 Topology ____________________________________________________ 5
Bus __________________________________________________________ 5 Star __________________________________________________________ 5 Tree__________________________________________________________ 6 3.1.1 3.1.2 3.1.3

3.2 3.3 3.4 3.5 3.6

Bus Access Management _______________________________________ 7 Transmission Protocol __________________________________________ 8 Transmission Integrity __________________________________________ 8 Physical Interface _____________________________________________ 8 Implementations ______________________________________________ 9

4 ARCNET________________________________ _______________ 10
4.1 4.2 4.3 4.4 Topology ___________________________________________________ 10 Bus Access Management ______________________________________ 10 Protocol Components _________________________________________ 10 Network Access______________________________________________ 13
Passing on the Token____________________________________________ 13 Data Transfer _________________________________________________ 13 Broadcast Message_____________________________________________ 13

4.4.1 4.4.2 4.4.3

4.5

Configuration Mechanisms _____________________________________ 14

Participant joins the Net __________________________________________ 14 Participant leaves the Net_________________________________________ 14

4.5.1 4.5.2

4.6 4.7 4.8

Security Mechanisms _________________________________________ 15 Overhead ___________________________________________________ 15 Timing Behaviour and Data Throughput ___________________________ 15
Example for a Data Transmission___________________________________ 16

4.8.1

4.9

Physical Properties ___________________________________________ 18

Coax Cable and Twisted Pair with HIT - Interface _______________________ 18 Twisted Pair with RS 485 Interface__________________________________ 19 Optical Waveguide ______________________________________________ 19

4.9.1 4.9.2 4.9.3

4.10 Implementation ______________________________________________ 20

Take the lead with ARCNET !

Page 1

AUG - ARCNET user group e. V.

5 Summary ________________________________ ______________ 21 6 Classification of the different Bus Systems_________________ 21 7 ARCNET as a Systembus ________________________________ 22 8 ARCNET in Comparison ________________________________ _23 9 Circuit and Layout Proposals _____________________________ 26
9.1 Coax Interface _______________________________________________ 26
Physical Transmission Layer ______________________________________ 26 Important Wiring Hints ___________________________________________ 28 Circuit _______________________________________________________ 29 Layout _______________________________________________________ 32 Transceiver ___________________________________________________ 35 9.1.1 9.1.1 9.1.2 9.1.3 9.1.4

9.2

TWP_______________________________________________________ 35
Physical Transmission Layer ______________________________________ 35 Circuit _______________________________________________________ 38 Layout _______________________________________________________ 40

9.2.1 9.2.2 9.2.3

9.3

OWG ______________________________________________________ 41
Physical Transmission Layer ______________________________________ 41 Circuit _______________________________________________________ 42 Layout_______________________________________________________ 42 Network Extension for OWG Interfaces ______________________________ 44

9.3.1 9.3.2 9.3.3 9.3.4

9.4

Low Cost OWG ______________________________________________ 46

Overview _____________________________________________________ 46 Functional Specifications _________________________________________ 46 Circuit _______________________________________________________ 47

9.4.1 9.4.2 9.4.3

9.5

RS485 _____________________________________________________ 48
Pysical Transmission Layer _______________________________________ 48 Circuits ______________________________________________________ 54

9.5.1 9.5.2

10

Appendix ________________________________ ____________61

10.1 Definitions __________________________________________________ 61 10.2 Literature ___________________________________________________ 62 10.3 Referenced Documents________________________________________ 62

Take the lead with ARCNET !

Page 2

AUG - ARCNET user group e. V.

1 ARCNET, the universal, realtime capable Fieldbus Solution

With ARCNET the market offers a fieldbus solution with numerous benefits which supply the user with powerful and inexpensive solutions for fieldbus applications. These benefits include the following features: Realtime Capability by Token Passing technology, this results in computable response times Safety due to checksum in the data packet and handshake protocol between transmitter and receiver Protocol is implemented within a single chip for low cost applications Simplicity, because OSI layers 1 and 2 are implemented in hardware Flexible network topology, bus-, tree- and star topologies are possible and may be randomly combined Flexible use of media, coax-, twisted pair- and fibre optics can be combined Huge network extent at standard baud rates, at reduced the baud rate even more extendable Automatical logon and logoff of participants, the network automatically integrates new stations into the ring or logs them off Variable packet length between 1 and 508 data bytes Master master access, the node which has the right to send is master Variable transfer rates between 30 bps and 10Mbps High efficiency, up to 71% of datarate as data channel

Take the lead with ARCNET !

Page 3

AUG - ARCNET user group e. V.

2 History
ARCNET has been developed in the USA by Datapoint and successfully used in the office LAN field. With an originally fixed transfer rate of 2.5 Mbps, ARCNET was pushed out of office LAN applications by the faster competitor Ethernet. For office applications the fast transfer of a large amount of data is the primary goal and not the realtime capabilities of the network. Just this ARCNET property is the major interest for industrial applications. SMC has introduced a family of ARCNET controllers to the market which is designed for the use in fieldbus applications and for internal networking within units especially. These circuits are used in large quantities in very different fields of application. In the meantime ARCNET became one of the most popular network solutions for industrial purposes.

3 Characteristics of modern Fieldbusses

Modern manufacturing processes require machinery and installations of increasing capabilities and complexity. The number of system components in use is increasing. For the communication between the components a fieldbus is used as a network for the industrial environment. Primary technical properties of fieldbusses are: Topology Bus access management Physical properties Transmission protocol Bus access management within multimaster systems Transmission security Overhead Real Time behaviour Economical and strategical aspects These items are first discussed in general and then detailled for ARCNET.

Take the lead with ARCNET !

Page 4

AUG - ARCNET user group e. V.

3.1 Topology
There are three different basic topologies: 3.1.1 Bus An example for the bus topology is shown below. Fieldbus nodes are connected to every bus segment in parallel. In every fieldbus the segment length depends on the transmission media and the number of the connected clients. The total length can be extended by using repeaters. A repeater (amplifier with in- and outputs) regenerates and amplifies the signal on the bus. The bus segment is terminated at both ends with resistors.

3.1.2

Star

In the star topology every segment only supports one node. Central point is a HUB (amplifier with up to 10 in- and outputs) which also contains the termination resistors.

Take the lead with ARCNET !

Page 5

AUG - ARCNET user group e. V. 3.1.3 Tree The tree topology is a mixture between bus and star. These can be randomly combined via HUBs.

Take the lead with ARCNET !

Page 6

AUG - ARCNET user group e. V.

3.2 Bus Access Management

Systems with centralized Bus Access Management One participant works as a master and controls the arbitration of the bus. This makes sense for applications, which are based on a system manager with high performance centrally distributing tasks and managing the information flow. The guaranteed access time can be configured individually for each participant. The availability of the hole system depends on the master. Token Passing Method A participant is allowed to access the bus as soon as it receives the token. The token is passed to the next participant after the action has been completed. Every participant of the system takes part in the communication process as a master according to a predefined timing grid. There is no preset information flow. A maximum access time can be calculated (deterministic system). The hole system does not depend on single participants. Free Bus Access There is no preset order for bus access. Every participant can try to access the bus at any time. Collisions that may occur must be avoided by the "Carrier Sense Multiple Access" method or corrected by the "Collision Detect" (CD) method. Compared to deterministic systems the free bus access method allows higher data throughput rates for applications with low bus load. However access times cannot be guaranteed. Such a system has no predictable realtime capabilities. At higher bus loads the bandwidth for data transmission decreases rapidly due to collisions.

Take the lead with ARCNET !

Page 7

AUG - ARCNET user group e. V.

3.3 Transmission Protocol

The different layers in industrial automation lead to different requirements for the fieldbus. Operation Layer, Control Layer Few participants (operation terminal, control terminal) communicate via LAN at high throughput rates. File transfer is emphasized. The LAN has to be optimized for high data rates and large data packets. System Layer, Process Layer Cell computers and micro programmable controllers (PLC s) are connected. The number of participants is greater than in the higher layers. Both, file transfer and byte oriented data transfer is needed. Often deterministic realtime behaviour is required. Sensor- /Actor Layer A large number of sensors / actors is connected to the. Only small numbers of bytes are transmitted.

3.4 Transmission Integrity

In fieldbus systems, there are high demands regarding data integrity. The common security mechanisms are CRC (cyclic redundancy checksum) and parity checking.

3.5 Physical Interface

Coax Cable supports huge network extension, is easy to install and meets the electromagnetic compatibility requirements. Twisted Pair Cable is the most common wiring type used for fieldbusses. It meets the electromagnetic compatibility requirements, often based on RS 485 Transceivers. Optical Waveguide (OWG) ensures high data integrity and data rates, however, limitations in terms of topology (mostly star) and handling have to be accepted. The electromagnetic

Take the lead with ARCNET !

Page 8

AUG - ARCNET user group e. V. behavior is excellent.

3.6 Implementations
In fieldbus solutions usually only OSI layers 1 and 2 are implemented. Layer 7 directly communicates with layer 2 There are various possibilities for the implementation of fieldbus solutions: Software Implementation Parts of layer 1 and 2 are implemented as software. Microcontrollers are used. With fieldbusses of higher performance this solution exhausts the available microcontroller resources or the use of expensive microcontrollers is required. ASIC Implemention Layer 1 and 2 are implemented as an ASIC. The computing power of the CPU can be exclusively used by the application. The user does not have to deal with fieldbus protocol details.

Take the lead with ARCNET !

Page 9

AUG - ARCNET user group e. V.

4 ARCNET
Below the specific ARCNET features are discussed.

4.1 Topology
ARCNET is one of the few networks that can be used with every topology. Bus-, star- or tree topologies as well as combinations of these can be used. Therefore ARCNET supports a wide range of applications.

4.2 Bus Access Management

ARCNET uses the Token Passing method. Every participant is assigned a unique number (ID) between 1 and 255, no ID may be used twice. The token is passed to the node with the next higher number independently from the network topology. The token allows the participant to access the network for transmitting one data packet of variable length (1-508 bytes). After a completed transmission attempt the token has to be passed to the next node whether the data block has been sent successfully or not. At any time it is possible for a participant to leave or to enter the logical ring of the token passing network. The access method ensures equal availability of the network for all connected nodes.

4.3 Protocol Components

It is important for the understanding of the following chapter (network access sequence) to explain the basic elements of the low level ARCNET protocol, which are used for the physical layer of data transmission. These protocol components are invisible for the user because they are part of layer 1 which is implemented in hardware.

Take the lead with ARCNET !

Page 10

AUG - ARCNET user group e. V.

There are five different protocol elements. The Alert Burst, which consists of six consecutive bits with the value 1, precedes all ARCNET packets. The header for every transmitted byte of data is represented by a 110 bit combination. ITT - Invitation to Transmit ITT is the token which is passed from one node to another node. Alert Burst ITT DID DID

FBE - Free Buffer Enquiry FBE serves as query to initiate a data transmission to another participant. Alert Burst ENQ DID DID

PAK - Data Packet PAK performs the transmission of a data packet.

Alert Burst PAK SID DID DID COUNT SC DATA // DATA CRC CRC

ACK - Acknowledgement ACK confirms a successful action. It is used as an answer to a FBE as well as for the termination of a successful data transmission. Alert Burst ACK

NAK - Negative Acknowledgement NAK is used as a negative answer to a FBE. A failed data transmission (CRC error) is recognized by the transmitter based on a timeout. Alert Burst NAK

Take the lead with ARCNET !

Page 11

AUG - ARCNET user group e. V.

Definitions: DID SID COUNT DATA CRC ACK NAK Destination ID Source ID Number of data bytes Data (1 - 507 byte) Cyclic Redundancy Check Acknowledgement Negative Acknowledgement

Take the lead with ARCNET !

Page 12

AUG - ARCNET user group e. V.

4.4 Network Access

A node can only actively access the physical network if it has received the token. There are several possibilities for a node accessing the network. 4.4.1 Passing on the Token A participant receives the token but has no data packet for transmission. In this case the token will be passed on to the next NID (next ID) immediately. If the NID does not become active within a given period of time, the owner of the token assumes that this node has left the network. Then it will start searching for a new neighbour by incrementing the Next ID register and repeating the token passing procedure. This is being continued until an active node has been found. 4.4.2 Data Transfer A node receives the token and wants to transmit data to node IDn. By transmitting FBE to the receiver IDn the owner of the token requests IDn to signal its ready-to-receive state. The ready-to-receive state is acknowledged by ACK. Now data are transmitted with checksum. If IDn is not prepared to receive data (for example: receive buffer full), it responds with a NAK, the token is passed on and the data transfer is retried at the next token reception. If a data packet has been sent with a correct CRC it will be acknowledged by the receiver, which responds an ACK. A bad CRC leads to a timeout because the receiving node does not acknowledge this packet anyway. After the timeout the token will be passed on. If the transmitter node receives the token again, it will retry to transmit the data packet. The number of retries for this procedure can be configured. 4.4.3 Broadcast Message During a broadcast message the data is sent to every participant without acknowledgement. Broadcast messages are transmitted with a CRC without a prior FBE and are addressed to DID = 0. The token is passed on without the need for awaiting an ACK. This type of transmission is not equipped with a security mechanism.

Take the lead with ARCNET !

Page 13

AUG - ARCNET user group e. V.

4.5 Configuration Mechanisms

ARCNET allows to add or remove participants to or from the network during operation. The user is not involved in this action, however, he receives a status message about this reconfiguration procedure by the ARCNET controller. 4.5.1 Participant joins the Net If a new node is added, it will destroy the token by a reconfiguration signal and enforces a complete reconfiguration of the network. After the token destruction a timeout will occur at every node. The timeout length depends on the ID of a node. Therefore the node with the highest ID times out first and tries to pass on the token to the node with the next ID. This procedure is repeated with an incremented address according to a given timeout period until activity is detected on the net and the next ID was located. The ID of the found participant is stored into the Next-ID register and the token is passed on. This procedure continues until every participant has investigated his NID. 4.5.2 Participant leaves the Net If a participant leaves the net only the neighbouring participant updates his Next-ID register. It is recognized that the neighbouring node has left the net by the absence of bus activity for the token passing procedure. The NID register of the searching node is incremented according to the reconfiguration procedure until a new a new neighbouring node has been found.

Take the lead with ARCNET !

Page 14

AUG - ARCNET user group e. V.

4.6 Security Mechanisms

The data transmission within ARCNET networks is secured in several ways based on low level mechanisms. prior to data transfer the ready-to-receive state of the target node is checked by a FBE. a 16 bit CRC is added to each data packet which is checked by the receiving node. In case of a lost token a reconfiguration is initiated and the network will be automatically reestablished. additionally the ARCNET controllers provide registers for diagnostic purposes.

4.7 Overhead
A fixed overhead cannot be defined for the ARCNET protocol due to the variable size of the data packets. The efficiency of the data transmission for ARCNET has a maximum of 71% of the selected baud rate (for long packets).

4.8 Timing Behaviour and Data Throughput

The timing behaviour can be exactly computed because of the token passing method. It is deterministic. Also during high bus loads the maximum access time and the token turnaround time can be calculated for each node, according to the equation below:

t Token = (145.4 + 4.4 x nBytes) nNodes s

145.4 s is the time required by the protocol for the transmission of each data packet. This value as well as the 4.4 s per byte is not fixed but depends on the data rate which is used within the network. For a network with 32 participants this results in a maximum access time of about 8 ms for a data packet length of 32 bytes and a data transfer rate of 2.5 Mbps. The maximum data throughput with ARCNET is reached for the maximum

Take the lead with ARCNET !

Page 15

AUG - ARCNET user group e. V. packet length of 508 bytes. The following example demonstrates this in detail. 4.8.1 Example for a Data Transmission In this example the total time for the transmission of a short data packet of 253 bytes shall be calculated. Node 1 has the token and transmits to node 2.

Node 1 asks with the aid of a FBE whether a data transmission to node 2 is possible. On every data transmission a turnaround time (Tta) of 12.6s is inserted before the addressed node answers. Node 2 waits for this period of time until it signals its ready-to-receive state with an ACK. Node 1 waits again for the Tta before it transmits a short data packet. Node 2 acknowledges the successful transmission after Tta with an ACK. After Tta node 1 passes the token to node 2, the transmission cycle is finished. After another Tta node 2 can start its activity. This leads to following total time consumption for the protocol: 2 * 15.6 s + 2 * 6.8 s + 37.6 s + 5 * 12.6 s = 145.4 s (at 2.5 Mbps) The related data transmission requires: 4.4 s * 253 = 1113.2 s This leads to a total time of 1258.6 s and a usable data rate of 88 %.

Take the lead with ARCNET !

Page 16

AUG - ARCNET user group e. V. For detailed information on times and data rates refer to [3].

Take the lead with ARCNET !

Page 17

AUG - ARCNET user group e. V.

4.9 Physical Properties

As transmission media coax cabling, twisted pair (shielded or unshielded) or (glas-, plastic-) optical waveguides can be choosen. The possible cable length without the use of amplifiers depends on the media, on the selected baud rate and on the number of participants. The ranges per segment vary from about 120 m with the ordinary twisted pair bus up to 3000 m with optical waveguide, at 2.5 Mbps. By the use of appropriate HUB's (amplifier boards for one or more cable types) different topologies and transmission media can be combined and the transmission range increased. 4.9.1 Coax Cable and Twisted Pair with HIT - Interface With coax and twisted pair cables a so called HIT (High Impedance Transceiver) is used as bus interface which converts the TTL signal of the ARCNET-controllers into a sinusodial signal of 15.4VPP up to max. 25.0VPP. Furthermore it performs a potential segregation. For coax applications a cable type with a characteristic impedance of 93 Ohms, e. g. RG 62 is used. The recommended twisted pair cables are specified inIEEE 802.3i-1990. At 2.5 Mbps, for coax cables, eight participants and without a HUB the maximum length of a bus segment is 300 m. Under the same conditions with twisted pair cables a maximum length of 120 m can be achieved. The maximum transmission length depends also on the number of participants. For a decreasing number of participants the usable length of a network segment is increasing due to the reduced attenuation of the signal. The total expansion of a network can be increased by active amplifiers (HUBs). Depending on the technical implementation of these amplifiers there is a limit for the maximum number of cascaded HUBs related to the propagation delay of the HUBs. For coax cabeling and at a transmission rate of 2.5 Mbps a maximum of 16 km network extent can be reached, for twisted pair cabeling the limit is about 6 km.

Take the lead with ARCNET !

Page 18

AUG - ARCNET user group e. V.

4.9.2 Twisted Pair with RS 485 Interface RS 485 has been established as a standard for fieldbus systems. There is no fixed default data rate, the number of participants is limited to 32 units per segment using standard RS 485 transceivers. Today some types of transceivers on the market are able to overcome these traditional limitations. For detailed information refer to the RS-485 Cabling Guidelinesby SMC. The network extent can also be increased by using HUBs, there are the same limitations as for coax applications. 4.9.3 Optical Waveguide Optical waveguide connections are available for applications, which require extremely high noise immunity of the cabeling technology and huge distances for single network segments. There is a difference between glass- and plastic optical waveguide being used, further the transmission range depends on the type of used transmitters and receivers and their optical transmit power and reception sensitivity. At 2.5 Mbps and with glass fibre a range of 3000 m can be reached, with plastic OWG only short distances up to 100 m are possible, although plastic OWG is less expensive. The baud rate does not play an important role with OWG transmissions. The range can also be increased with the aid of HUBs, there are the same limitations as with coax cables. A disadvantage of OWG is that either the star topology is used - which means that a hub is the center of the network - or that every participant contains the hub functionality. This leads to the fact that the operation of the whole system depends on the operation of every single participant. Nevertheless there are solutions - especially for the automobile industry - with passive optical hubs which avoid these limitations.

Take the lead with ARCNET !

Page 19

AUG - ARCNET user group e. V.

4.10 Implementation
For own applications inexpensive ARCNET controllers are available which support datarates of up to 10 Mbps on the bus. (COM 20020 bis 5 Mbps, COM 20022 bis 10Mbps) The lower two OSI layers are implemented on the chips. The user communicates with the chip via a register interface. Registers for configuration and monitoring purposes are available. The received data or data to be sent are exchanged via an integrated Dual Port Ram. A 8051 compatible microprocessor with built in ARCNET controller is also available. (COM 20051) For the coax interface connection a HIT is available. (HYC 9088A) The connection to RS 485 and optical waveguide networks is possible with inexpensive standard components, There are driver libraries for the INTEL 8051 and 80x86 processors for software development. (ControLink 51 and 86) The standardization of the transmission protocols, of layer 7 and of an OPC interface is promoted by the newly founded ARCNET USER GROUP (AUG).

Take the lead with ARCNET !

Page 20

AUG - ARCNET user group e. V.

5 Summary
ARCNET, due to its fast and computable realtime behavior and its flexible data packet size, can be used in the sensor / actor field as well as on process- and cellbus level. Inexpensive node implementations and the possibility to setup networks of large dimensions for multiple nodes are further ARCNET advantages Field devices can be connected to and disconnected from the bus without interferences. A classification and a comparison with other bus systems follows below. Due to the complete implementation of the network protocol on chip level the use of ARCNET as systembus solution within stand-alone units is also very interesting.

6 Classification of the different Bus Systems

The presently available fieldbusses can be classified into the following groups: event controlled systems, like CAN and LON systems which can be used for explosion protected applications , like FIP, PROFIBUS-PA and P-NET fast, deterministic systems, like ASI, INTERBUS-S/DP, SERCOS, ARCNET and IEEE-Bus deterministic systems, like BITBUS, PROFIBUS-FMS ARCNET is to be placed into the group of the fast, deterministic systems. However, also for other applications the use of ARCNET would make sense due to its specific benefits.

Take the lead with ARCNET !

Page 21

AUG - ARCNET user group e. V.

7 ARCNET as a Systembus
Due to the high integration of the ARCNET controllers and a special transmission mode, ARCNET is also used as a networking bus for single components within stand-alone units. Multimaster capabilities and the simple processor interface support architectures with distributed intelligence. This results in advantages concerning device maintenance and diagnostics.

Take the lead with ARCNET !

Page 22

AUG - ARCNET user group e. V.

8 ARCNET in Comparison
Bus type Comment No. of Mastersl Topology ASI only binary I/O 1 line INTERBUS-S SERCOS optical waveguide 1 Ring -60m(3) -250m(4) 245 4 Mbit/s 1 LWL NRZI according HDLC 16 Bit CRC 1 bis 16 Connection oriented
PROFIBUSDP

ARCNET

1 ring -12,8km(1) -10m(2) 256(1) 7(2) 500 kbit/s 2(6) 8(2) RS-485 (NRZ) 16 Bit CRC 4 to 64 Connection oriented

1 to 32 Line 200m bei 1,5Mbit 124/32 1,5 Mbit/s 2 RS-485 (NRZ) Parity (2 dim.) 1 to 246

1 to 255
line, star, tree

Cable Length 100m No. of Participants

Max. Datarate

31 167 kbit/s 2(6) ALP Manchester (RZ) Parity (1 dim.) 5 Bit

Refer to table 2 256 10 Mbit/s 1 2 (7) RS-485 (NRZ) 16 Bit CRC 1 to 508

No. of Wires Coding

Error Recognition No. of data bytes

Communication Connection Structure oriented

Connection Connection oriented and oriented

Global
Master/Master TokenRing Master/Slave Polling

Access Method

Master/ Slave Polling Service controlled

Master/Slave Master/ sum frame Slave Time slots Service controlled ID-numbers

Master/ Master Token Ring

User Access

User Interface DDLM Chip DIN E 19245

Data blocks and Commands Chip

ATA/ANSI 878.1, 2 and 3

Layer 2 Standardizati on

Chip no

Chip DIN E 19258

Chip IEC DIS 44

(1) far distance bus (2) peripheral bus (6) incl. auxiliary power supply

(3) plastic OWG

(7) 1 with coax, 2 with twisted pair and OWG

Take the lead with ARCNET !

Page 23

AUG - ARCNET user group e. V.

Bus type Comment No. of Mastersl Topology Cable Length No. of Participants
Max. Datarate

CAN

LON Comprises layer 1 to 7 Multimaster line / tree 500m at 1.25Mbit/s 64 1.25 Mbit/s 2 RS-485 / Manchester (RZ) 16 Bit CRC

multimaster line 40m at 1 Mbit/s 256 1 Mbit/s 2 RS-485 / ISO11898 (NRZ) 15 Bit CRC

No. of Wires Coding

Error Recognition Max. 228 (1) No. of data bytes 1 to 8 Communication Object oriented Object oriented
Structure

Access Method

CSMA/CA

CSMA/CA

User Access

CAN Application Layer Chip ISO11898 ISO11519-1

Netzwork Variables, Explicit Messaging Chip (2) No

Layer 2 Standardization

(1) LON is optimized for small packet sizes (2) Protocol has not been publicated
Source: Robert Busse, Feldbussysteme im Vergleich, Pflaum Verlag, Mnchen

Take the lead with ARCNET !

Page 24

AUG - ARCNET user group e. V.

Transmission Media

Segment length with 2 Participants

300 m about 120 m 3000 m 200 m

Network Dimension

Coax TWP LWL RS 485

16 km 8 km 20 km 16 km

Table 2: Cable length for ARCNET at 2.5 Mbps

Take the lead with ARCNET !

Page 25

AUG - ARCNET user group e. V.

9 Circuit and Layout Proposals

This part of the ARCNET description contains circuit proposals for the physical ARCNET bus interface. Beside circuit schematics also design proposals and further hints are included to simplify the implementation of user specific interfaces.

9.1 Coax Interface

9.1.1 Physical Transmission Layer The physical transmission layer consists of transceivers (HIT), the transmission media coax cable with an impedance of 93 Ohms (RG62) and the corresponding bus termination networks (resistors). 9.1.1.1 Data Rate

The coax interface supports a fixed datarate of 2.5 Mbps. 9.1.1.2 Connectors

Allowed are BNC female connectors for the nodes with a characteristic impedance of 50 or 75 Ohm and the corresponding male connector for the cables. 9.1.1.3 Transmission Media and Termination

Coax cable with an impedance of 93 Ohm e.g. type RG 62 U The cable has to be terminated on both sides with a resistor of 93 Ohm. The termination can either be integrated in the device or supplied by an external resistors (integrated in a BNC male connector). It has to be possible to activate / deactivate internal termination by switches or jumpers.

Take the lead with ARCNET !

Page 26

AUG - ARCNET user group e. V.

9.1.1.4

Signal Shape and Level

For the transmission of the information a dipulse signal is used as shown in the figure below. The signal level has to be between 15.4V and 25.0V peakpeak.

A "1bit is transmitted as a 200 ns dipulse followed by 200 ns without a signal, a "0" as 400 ns without a signal. For further information refer to [ 5 ]. Signal shape for dipulse transmission

Take the lead with ARCNET !

Page 27

AUG - ARCNET user group e. V.

9.1.1 Important Wiring Hints The table below shows the most important wiring parameters. Minimum distance between two nodes Node impedance Termiantion resistance Maximum no. of nodes per segment Maximum segment length Bus 1m > 1,2 kOhm 93 Ohm 8 300 m Star 1m > 1,2 kOhm 93 Ohm 2 610 m

The maximum segment length depends on the number of connected participants. It can be calculated based on the following equation: Segment length lk = (Po dt nt) / dk Po = transmit power above receiver threshhold [dB] (typical value 11dB) dt = attenuation per participant [dB] (typical value 0,45dB) nt = no. of participants dk = cable attenuation per m [dB/m]

Take the lead with ARCNET !

Page 28

AUG - ARCNET user group e. V.

9.1.2 Circuit The following design proposal describes an ARCNET node for coax cabling. Hereby electromagnetic compatibility aspects are especially emphasized. 9.1.2.1 Circuit Schematic The following circuit diagram shows a typical coax interface containing a hybride module (type HYC9088(SMC), a BNC connector and the protection circuit. The termination is not included here because it is usually implemented as external termination resistor at both ends of the coax cable. The approved design consequently separates electronic ground (GND_E) and chassis ground (GND_A). The protection circuit consists of R1, R2, V1 and C1. By this configuration any noise power, which is coupled into the coax cable in a noisy environment, is carried down to the reference potential and therefore kept away from the network interface hybrid module. As reference potential a metal cover (housing) for the unit offers high noise immunity especially for industrial applications. The metal cover needs to be designed for high impedance segregation related to the remaining electronic circuit within the housing. R1 and R2 serve as leakage resistors to avoid high static potential differences at the network interface. C1 is a high frequency short circuit. The varistor limits voltage peaks to protect C1 against overload. L1 and L2 are ferrite beads which can be optionally used to increase the immunity against high frequency noise. As an alternative to the varistor a supressor diode can also be used. A coax connector with integrated bleed off capacitors should be used.

Take the lead with ARCNET !

Page 29

AUG - ARCNET user group e. V.

9.1.2.2 Components Components for the protection circuit: Number Reference 1 U1 1 C1 Designator HIT Metallized Film Capacitor 400V Varistor Metalfilm resistor 1/2W Type HCY9088 AS FKC3 Value

10nF

1 2

V1 R1, 2

V275LA20 PR01 5k6

Take the lead with ARCNET !

Page 30

AUG - ARCNET user group e. V.

Figure: Circuit schematic of a coax interface

Take the lead with ARCNET !

Page 31

AUG - ARCNET user group e. V.

9.1.3 Layout For the described circuit a layout has been developed which emphasizes electromagnetic compatibility aspects. 9.1.3.1 Ground Layout The layout of the ground signal is the most important factor for the noise immunity. The connection between protection circuit and chassis should be as short as possible and/or connected by a wide track. The following example layout for a PC card clarifies this. The slot panel with two lugs is the low impedance connection between protection circuit and chassis.

Take the lead with ARCNET !

Page 32

AUG - ARCNET user group e. V.

9.1.3.2 Component Placement and Pads The placement of the components refers to the track layout which is shown below.

Take the lead with ARCNET !

Page 33

AUG - ARCNET user group e. V.

9.1.3.3 Layout Important is the wide grounded area which ensures the connection between protection circuit and front panel. The grounded area is segregated from the PC ground on the ARCNET card.

Take the lead with ARCNET !

Page 34

AUG - ARCNET user group e. V.

9.1.4 Transceiver The following table lists different transceiver modules currently available on the market. Type HYC9088A JTL7958R MIT P/N A60152-10 MIT P/N A60151-10 Vendor SMC JASMINE MIT Supply Impedance Voltages +5V, -5V high impedance +5V +5V -5V or -12 V +5V -5V or -12 V high impedance high impedance Isolation Voltage n.A. n.A n.A Datarate 2.5 MBit/s 2.5 MBit/s 2.5 MBit/s 2.5 MBit/s

MIT

93 Ohm

n.A.

This table does not intend to be complete.

9.2 TWP
9.2.1 Physical Transmission Layer This physical transmission layer consists of the transceivers, the transmission media twisted pair cable and the corresponding bus terminators. The signal lines are twisted to achieve a defined impedance and a high noise immunity. 9.2.1.1 Bit Rate The coax interface supports a fixed datarate of 2.5 Mbps. 9.2.1.2 Connectors For the nodes a 9-pin SUB-D male connector is used. The cables are equipped with 9-pin SUB-D female connectors. The housing of the connectors at the cable should consist of a conducting material (metal or conducting plastic cover) and has to be connected to the twisted pair cable shield over a wide area for low inductance. We additionally recommend to connect the shield to pin 1 of the connector.

Take the lead with ARCNET !

Page 35

AUG - ARCNET user group e. V.

9.2.1.3 Connector Pin Assignment Pin No. 1 and chassis 3 8 2,4-7, 9 Signal Name Shield TWPTWP+ not used Comment Chassis and front panel potential "shield" I/O of the HIT "center" I/O of the HIT

9.2.1.4 Transmission Media and Termination TWP uses shielded twisted pair cables. The two wires of the pair are twisted to achieve a defined impedance and a high noise immunity. Attenuation and recommended cable types meet the specifications described in chapter "RS485". The cable has to be terminated at both ends with a resistor of 120 Ohm/0.5W. The termination can either be integrated in the device or supplied by an external plugable resistor. It has to be possible to activate / deactivate internal termination by switches or jumpers.

Take the lead with ARCNET !

Page 36

AUG - ARCNET user group e. V.

9.2.1.5 Signal Shape and Level For the transmission of the information a dipulse signal is used as shown in the figure. The signal level has to be between 15.4V and 25.0V peak-peak. A "1bit is transmitted as a 200 ns dipulse followed by 200 ns without a signal, a "0" as 400 ns without a signal. For further information refer to [ 5 ]. The interface has to be designed in such a way that the ARCNET controller interprets the transition from logical 0 to logical 1. 9.2.1.6 Important Wiring Hints The table below defines the most important wiring parameters. Minimum distance between two nodes Node impedance Termination resistor Maximum no. of nodes per segment Maximum segment length Bus 2m > 1.2 kOhm 120 Ohm 8 about 120 m Star 2m > 1.2 kOhm 120 Ohm 2 about 280 m

Take the lead with ARCNET !

Page 37

AUG - ARCNET user group e. V.

9.2.2 Circuit In the following a design proposal for an ARCNET node for twisted pair cabling is presented. Hereby electromagnetic compatibility aspects are especially emphasized. TWP is based on a HIT device driving the twisted pair cable. In contrast to RS485 systems the electrical segregation is done by the transformer within the HIT device. The setting of a defined inactive level by pullup and pulldown resistors as it is used for RS485 is not necessary. 9.2.1.7 Circuit Diagram The following circuit diagram describes a typical twisted pair interface based on a hybride of the type HYC9088, a 9-pin SUB-D male connector and the corresponding protection circuit. The approved design consequently segregates electronic ground (GND_E) and chassis ground (GND_A). The protection circuit consists of R1, R2, V1 and C1. By this configuration any noise power, which is coupled into the coax cable in a noisy environment, is carried down to the reference potential and therefore kept away from the network interface hybrid module. As reference potential a metal cover (housing) for the unit offers high noise immunity especially for industrial applications. The metal cover needs to be designed for high impedance segregation related to the remaining electronic circuit within the housing. R1 and R2 serve as leakage resistors to avoid high static potential differences at the network interface. C1 is a high frequency short circuit. The varistor limits voltage peaks to protect C1 against overload. L1 and L2 are ferrite beads which can be optionally used to increase the immunity against high frequency noise. As an alternative to the varistor a supressor diode can also be used. A coax connector with integrated bleed off capacitors should be used. We recommend the use of a shielded twisted pair cable which meets the specifications described in the chapter "Design Proposal for a RS485 Interface". The housing of the connectors at the cable should consist of a conducting material (metal or conducting plastic cover) and has to be connected to the twisted pair cable shield over a wide area for low inductance. We additionally recommend to connect the shield to pin 1 of the connector.

Take the lead with ARCNET !

Page 38

AUG - ARCNET user group e. V.

Take the lead with ARCNET !

Page 39

AUG - ARCNET user group e. V. 9.2.1.8 Termination The cable has to be terminated with the characteristic impedance of the cable on both ends of the bus. This is performed by jumper X2 and resistor R5. 9.2.1.9 Components Number Reference 1 U1 2 C1, C2 4 1 R1-R4 R5 Designation HIT ceramic capacitor, > 1000V resistor, 0.5 W coal or metal film resistor, 0.5 W coal or metal film Type HCY9088AS DE0910SL220J6K Value 22pF 5k6 120

9.2.3 Layout The recommendations describes for the coax design proposal are also valid for twisted pair applications.

Take the lead with ARCNET !

Page 40

AUG - ARCNET user group e. V.

9.3 OWG
9.3.1 Physical Transmission Layer 9.3.1.1 Overview The transmission via optical waveguide is recommended especially for long distances and/or high requirements concerning noise immunity. 9.3.1.2 Datarate All datarates supplied by ARCNET are supported. 9.3.1.3 Connectors Due to the numerous available OWG systems and fields of application no standardization will be offered. 9.3.1.4 Signal Shape and Level A logical 0 corresponds to no light. A logical 1 corresponds to a pulse (=light) of the length T_On = 0.25 to 0.5 bit. The interface has to be designed in such a way that the ARCNET controller interprets the transition from "no light" = logical 0 to "light" = logical 1.

Figure 1: OWG signal shape

A careful design of the OWG circuit is recommended. It has to be considered that the bit jitter should be kept as small as possible. Bit jitter occurs if the optical transmission power is too high (photo diode overdrive) as well as if it is too low (noise).

Take the lead with ARCNET !

Page 41

AUG - ARCNET user group e. V.

9.3.1.5 Transmission Media Due to the numerous available OWG systems and fields of application no standardization will be offered. 9.3.2 Circuit In the following a design proposal for an ARCNET node with an optical waveguide interface is presented. 9.3.1.6 Circuit Diagram The following circuit diagram shows a typical interface between an ARCNET controller and an optical waveguide (SMA connectors). RX_DATA is the received signal which has been converted into a TTL signal which is connected to the data input of a COM20020 device. _TX_DATA is the transmitted TTL signal. A TTL high level corresponds to a logical 0 (no light) and is defined as the idle state. The _PULSE1 output of the COM 20020 device supports this type of signal. It has to be considered that the transmitted data are also returned to the input via the gates U3B and U3C. This is necessary because the ARCNET controller continuously monitors the bus for timeouts and initiates a reconfiguration cycle if a timeout occurs. 9.3.1.7 Power Setting We strongly recommend to design the OWG interface for an adjustable transmission power. In the circuit diagram the resistors R6, 8, 9 and jumper X3 are used for that adjustment of the required transmission power. Most of the OWG receivers at the market are sensitive against overdrive, which is caused by receiving a high amount of light power at the input. This results in an increased bit jitter for misaligned input power which also causes a considerable increase of the bit error rate. For a calculation example on the correct OWG interface design refer to [4]. 9.3.3 Layout

OWG receivers are usually very sensitive against noisy supply voltages. Therefore a proper decoupling of the power supply has to be performed by the use of bypass capacitors.

Take the lead with ARCNET !

Page 42

AUG - ARCNET user group e. V.

Figure: circuit diagram of an OWG interface

Take the lead with ARCNET !

Page 43

AUG - ARCNET user group e. V.

9.3.4 Network Extension for OWG Interfaces The extension of an ARCNET network mainly depends on optical transmission power, receiver sensivity, fiber attenuation and transition losses at connectors etc. An additional amount of reserved power has always to be included. The range (in m) is calculated according to the equation below: r = (Po - Pi - m - Ls + Gdia)/ L *1000 Po = optical transmission power of the transmission diode [dBmW] Pi = minimum receive power required for the receiver module [dBmW] m = margin = reserve [dB] L = attenuation of the OWG [dB/km] per km Ls = all connector attenuation and splice losses Gdia = adaption factor [db] which depends on the cable type r = range [m] 9.3.1.8 Example 1 Example for an OWG interface based on a TOSLINK module TODX296 and a plastic fibre of the type TOCP200. Bit rate = 2.5 Mbit/s. All values are worst case. Po = -20 dBm, Pi = -31 dBm, m = 3 dB, L = 240dB/km Ls = 0, because the cable calculation already includes the connector losses --> r = 33 m

Take the lead with ARCNET !

Page 44

AUG - ARCNET user group e. V.

9.3.1.9 Example 2 Example for an OWG interface based on a Honeywell transmission diode HFE 4401-014, receiver HFD 3403-002 and a glas fibre of the type HOCZ 0052. Bit rate = 2.5 Mbit/s. All values are worst case. Po = -17 dBm, Pi = -25.5 dBm, m = 3 dB, L = 3.5 dB/km Ls = 0, because the cable calculation already includes the connector losses --> r = 1570 m Note: Gdia is set to 0 dB in the examples.

Take the lead with ARCNET !

Page 45

AUG - ARCNET user group e. V.

9.4 Low Cost OWG

9.4.1 Overview For short distances (up to about 50 m) at high noise immunity requirements, plastic optical waveguides are especially suitable as transmission media. However, the wavelength of the transmission diode has to fit to the spectral attenuation of the plastic OWG. At a wavelength of 660 nm transmission distances of up to 30 m can be achieved without a special treatment of the plastic OWG ends. 9.4.2 Functional Specifications 9.4.1.1 Datarate All datarates supplied by ARCNET are supported. 9.4.1.2 Signal Shape and Level Signal shape and level meet the specifications above for standard OWG interfaces. 9.4.1.3 Mechanichal Specifications The connection of the optical waveguide to the transmitter and receiver devices is performed either directly (Siemens) or by using a SMA-connector (Honeywell). The devices of Siemens and Honeywell can be combined within a network. For shorter distances (about 20 m) also standard OWG components can be mixed up with low cost OWG devices. Siemens: Transmitter unit SFH 752V, receiver unit SFH 551 V The OWG is cut to length, inserted into the OWG module without further treatment and mechanically fixed with a nut. For long transmission distances the end of the OWG fibre can be treated with grinding paper. Honeywell: Transmitter unit HFX6015 xxx, receiver unit HFD3403-2xx Connection by SMA-connectors.

Take the lead with ARCNET !

Page 46

AUG - ARCNET user group e. V.

9.4.1.4 Transmission Media Plastic fibre OWG with a core of 1 mm diameter and a black cover, total diameter of 2.2 mm. Installation as single and duplex cable. Duplex OWG Honeywell HOCD0202 or Hirschman OKD 1000-B Typical technical data: attenuation at 660 nm about 200 dB/km evironmental temperature -20 to +80 C Isolation voltage 110 kV/m longitudinal strength 100 N 9.4.3 Circuit For the electrical circuit, layout and power settings, the same specifications as for OWG are valid.

Take the lead with ARCNET !

Page 47

AUG - ARCNET user group e. V.

9.5 RS485
9.5.1 Pysical Transmission Layer The transmission technology is based on the EIA RS-485 stansard. 9.5.1.1 Datarate

All datarates supplied by ARCNET are supported. 9.5.1.2 Signal Shape and Level

A logical 0 corresponds to a positive voltage difference between the signals DATA+ und DATA- of at least 0.2 V. A logical 1 corresponds to a negative voltage difference between the signals DATA+ und DATA- of at least 0.2 V. T_On equals 0.5 * bit length +- 10%. The following figure shows the logical shape during the transmission. The interface has to be designed in such a way that the ARCNET controller interprets the transition from logical 0 to logical 1.

Figure 2: logical RS485 signal shape

Take the lead with ARCNET !

Page 48

AUG - ARCNET user group e. V.

9.5.1.1 Mechanical Specifications For the connection of the bus cable to the bus participant a 9-pin SUB-D connector is used. The female connector is used at the bus cable, the male connector is used at the bus interface of the participant. The connectors should always be equipped with metal housings. The mechanical fixture between connectors should be based on conducting screws. For the connection of bus participants to the bus cable a so called T-connector part should be used which supports an easy removal or exchange of the bus participant without the need of breaking up the cable segment. 9.5.1.2 Connector Pin Assignment Pin No.
1 2

Designator
Connector housing Shield RP-P

Description
Wide area connection to shield Shield and protective ground Optional Reserved for auxiliary voltage supply + , for the supply of bus interface units and repeaters without private power supply Optional Data line B (=RS485-B), in idle state more negative than DATA-A Reserved for signals for direction control of repeaters with RS485 technology (=RS485-A), in idle state more positive than CNTR-B Optional Data reference potential, supply of the termination networks Supply voltage +, supply of the termination networks, only necessary for participants at end of the cable Reserved for auxiliary voltage supply-GND, for the supply of bus interface units and repeaters without private power supply Optional Data line A (=RS485-A), in idle state more positive than DATA-B Reserved for signals for direction control of repeaters with RS485 technology (=RS485-B), in idle state more positive than CNTR-A Optional

3 4

DATA-B CNTR-A

5 6 7

DGND VP RP-N

8 9

DATA-A CNTR-B

Take the lead with ARCNET !

Page 49

AUG - ARCNET user group e. V.

9.5.1.3 Transmission Media As bus cable a shielded twisted pair cable is used. The shield is necessary for the improvement of the electromagnetic compatibility. The cable used should meet following specifications:
9.5.1.5.1 Attenuation

at 16MHz at 4MHz at 38.4kHz at 9.6 KHz

9.5.1.5.2

<45dB/km <22dB/km <5dB/km <3dB/km

Characteristic Impedance

at 3MHz bis 20MHz at 38.4kHz at 9.5kHz rated value:

9.5.1.5.3 Operation Voltage

15015 18518,5 27027 150

100V
9.5.1.5.4 Capacity

9.5.1.5.5

30pF/m (for transmission rates from 3 to 20MHz) 60pF/m (for transmission rates below 3MHz)
Wire Diameter
2

0,34mm (for transmission rates from 3 to 20MHz) 0,22mm (for transmission rates below 3MHz)
2

Tee lines to participants should be kept shorter than 0.3 m.

Take the lead with ARCNET !

Page 50

AUG - ARCNET user group e. V.

9.5.1.6

Circuit Diagram for Cable bewteen two Nodes

For the circuit shown above, the potential difference between all the data reference potentials (DGND) of all interfaces must not exceed 7V. If this cannot be ensured a separate potential compensation has to be used. At both connectors the shield should be wired to the protective ground with a connection of low inductance in order to achieve optimum electromagnetic compatibility. This connection is preferably established by the metal chassis and the screws of the SUB-D plug. If this is not possible the connection should be established by pin 1 of the connector. 9.5.1.4 Termination Both ends of the bus cable have to be equipped with a termination resistor Rt = 220 . To ensure a defined signal level at the line for inactive states an additional resistor Rd = 390 has to be connected against GND and another resistor Ru = 390 against the supply voltage +. The participants which act as line termination have to provide VP and DGND (recommendations are valid for a supply voltage of +5V 5%). The line termination can be performed within the nodes or within the housing of the external connectors.

Take the lead with ARCNET !

Page 51

AUG - ARCNET user group e. V.

Vp Ru = 390 RS485-A Rt = 220 RS485-B Rd = 390 GND

Rt = 220 2%, min W Rd = 390 2%, min W Ru = 390 2%, min W The total high frequency impedance RT calculates as follows: with Rd = Ru 1/RT = 1/Rt + 2/Rd = 1/Zcable
9.5.1.5

Wiring Example for the minimum Configuration

Take the lead with ARCNET !

Page 52

AUG - ARCNET user group e. V.

9.5.1.6 Wiring and Network Extension Different factors determine the maximum extension of a network segment: data rate technical specifications of the cable (e.g. attenuation) attenuation by the participants attenuation by protection circuits power of the transceiver units

Therefore the extension of an RS485 network has to be evaluated in details for each application and is not outlined here.

Take the lead with ARCNET !

Page 53

AUG - ARCNET user group e. V.

9.5.2 Circuits 9.5.2.1 Protection Circuit To protect the RS485 driver against fast transients the data lines A and B are connected to suppressor diodes. In order to minimize the load on the bus lines low capacity suppressor diodes are used (suppressor diodes with a standard diode in series). The load by the transient protection is hereby limited to a value of 10 pF per wire. As a protection againt lightning discharges gas discharge elements should be used against the ground wire. 9.5.2.2 Factors for the Improvement of the Electromagnetic Compatibility Tee lines between transceiver and SUB-D connector as short as possible Placement of bus termination and transient protection as close to the SUBD connector as possible Shield connected to protective earth (at both sides, connection with low inductance over a large area) Shield soldered to the metal cover of the connector (better results than for connection at strain relief) Ground areas Separation of electrically segregated parts Ferrite tube or 6-hole cores over bus cable (subsequent installation possible) protective ground area at the front panel is recommended Y-capacitors (4.7nF, 100V) for voltage supply and auxiliary energy

Take the lead with ARCNET !

Page 54

AUG - ARCNET user group e. V.

9.5.2.3 Circuit Layout without electrical Segregation Due to a limited noise immunity the use of this circuit is only recommended small systems (e.g. within a wiring cabinet)
1 2 3 4 5 6 7 8

uC-Interface

/RESET /INTR /CS /WR /RD GND IC2 13 A[0..2] 1 2 3 4 5 6 8 9 10 11 12 14 22 7 NC RD/DS DIR/WR A0/MUX CS A1 INTR A2/ALE RESET AD0 AD1 TXEN AD2 RXIN D3 D4 PULSE2 D5 PULSE1 D6 D7 XTAL2 XTAL1 Vss Vss VDD Vss VDD COM20020 +5V C2 100nF GND Q1 20MHz C3 27pF C4 27pF GND Transientenschutz (optional) GND Suppressor diodes, Array LCDA05 GND +5V +5V 27 26 25 24 23 21 20 19 18 GND 17 16 28 15 R7 100K R8 100K R6 680 IC1 1 2 3 4 RO VCC RE DE DI A B VSS 8 6 ST1 7 5 C1 100nF J1 LTBR J4 LTBR J2 J3 LTBR LTBR R5 220 R3 390 R4 390 1 2 3 4 GND 5 +5V 6 7 DATA-A 8 9 DATA-B D-Sub 9pol. Stecker +5V +5V R2 100K GND R1 100K GND

D[0..7]

75ALS176D

GND

Busabschlu (optional)

Title Size A3 Date: File: 1 2 3 4 5 6

ARCNET Interface ohne galv. Trennung

Number S351_2 15-Jul-1997 Sheet of Maier I:\PRJ\ARCNET\S351_2.SCH Drawn By: 7 8 Revision

Take the lead with ARCNET !

Page 55

AUG - ARCNET user group e. V.

9.5.2.3.1 Components

Number 1 1 1 1 1

Reference IC1 IC2 Q1 ST1

Designation RS485 transceiver ARCNET controller Crystal oscillator Connector Suppressor diodes

Type SN75ALS176D COM20020

Value

20MHz SUB-D 9-pin connector Array LCDA05

9.5.2.3.2

Wiring

1 x 2: twisted pair cable with two unshielded wires for data signals A and B

Take the lead with ARCNET !

Page 56

AUG - ARCNET user group e. V.

9.5.2.4 Optically decoupled Circuit Layout

1 2 3 4 5 6 7 8

Galvanische Trennung

D +5V 1 DC 2 C10 27F 35V C12 100nF DC 5 C9 27F 35V C11 100nF U1 6 +5Vext

UMS 05S05 ACAL

GND

GNDext

uC-Interface +5Vext /RESET /INTR /CS /WR /RD IC4 13 A[0..2] 1 2 3 4 5 6 8 9 10 11 12 14 22 7 NC RD/DS DIR/WR A0/MUX CS A1 INTR A2/ALE RESET AD0 AD1 AD2 D3 D4 D5 D6 D7 Vss Vss Vss TXEN RXIN PULSE2 PULSE1 XTAL2 XTAL1 VDD VDD 27 26 25 24 23 21 20 19 18 +5Vext 17 16 28 15 R5 680 IC2 8 +Vcc A1 7 Vo1 K1 6 C6 100nF GND Q1 20MHz C7 27pF C8 27pF C3 100nF GND 5 Vo2 A2 K2 0V HCPL2631 GNDext 1 2 4 3 C5 100nF R2 300 Transientenschutz (optional) Busabschlu (optional) GNDext GND +5V +5V Suppressor diodes, Array LCDA05 GNDext +5Vext R1 300 R3 300 IC3 +Vcc 1 A1 2 K1 Vo1 4 A2 Vo2 3 K2 0V HCPL2631 C1 100nF C2 100nF R4 680 8 7 6 5 1 2 3 4 +5V R6 680 GNDext +5Vext R13 680 IC1 RO VCC RE DE DI A B VSS 8 6 ST1 7 5 C4 100nF J1 LTBR J4 LTBR J2 LTBR R12 220 GNDext GND GNDext +5Vext R10 390 R11 390 J3 LTBR 1 2 3 4 GNDext 5 +5Vext 6 7 DATA-A 8 9 DATA-B D-Sub 9pol. Stecker B R9 100K R8 100K +5Vext GNDext

+5Vext

75ALS176D

D[0..7]

COM20020 +5V

A galvanisch getrennt

GND

Title Size A3 Date: File: 1 2 3 4 5 6

ARCNET Interface mit galv. Trennung

Number S351_1 Revision

15-Jul-1997 Sheet of Maier I:\PRJ\ARCNET\S351_1.SCH Drawn By: 7 8

Take the lead with ARCNET !

Page 57

AUG - ARCNET user group e. V.

9.5.2.4.1

Components

Number 1 2 1 1 1 1 1

Reference IC1 IC2, IC3 IC4 Q1 ST1 U1

Designation RS485 transceiver opto coupler ARCNET controller Crystal oscillator Connector DC/DC converter Suppressor diodes

Type SN75ALS176D HCPL2631 COM20020

Value

20MHz SUB-D 9-pin connector UMS 05S05 array LCDA05

9.5.2.4.2

Wiring

2x2:

1 twisted pair for data signals A, B and 1 twisted pair for repater control signals A, CNTR-B

Take the lead with ARCNET !

Page 58

AUG - ARCNET user group e. V.

9.5.2.5 Circuit Layout of the Repeater Control based on RS485 It has to be considered that repeater control based on RS 485 is only possible if all bus participants are equipped for this purpose.
1 2 3 4 5 6 7 8

Galvanische Trennung

+5V 1

U1 6 DC

+5Vext +5Vext 5 C10 C12 27F 100nF 35V +5Vext +5Vext GNDext R5 680 R21 680 IC1 1 RO VCC RE DE DI A B VSS 8 6 ST1 7 5 C4 100nF J1 LTBR J4 LTBR J2 J3 LTBR LTBR R19 220 R15 390 R16 390 1 2 DATA-B 3 CNTR-A 4 GNDext 5 +5Vext 6 7 DATA-A 8 CNTR-B 9 D-Sub 9pol. Stecker R12 100K R11 100K GNDext

D 2 C11 C13 27F 100nF 35V DC UMS 05S05 ACAL

GND

GNDext

2 3 4

uC-Interface

R2 680

75ALS176D

/RESET /INTR /CS /WR C /RD IC5 13 A[0..2] 1 2 3 4 5 6 8 9 10 11 12 14 22 7 NC RD/DS DIR/WR A0/MUX CS A1 INTR A2/ALE RESET AD0 AD1 TXEN AD2 RXIN D3 D4 PULSE2 D5 PULSE1 D6 D7 XTAL2 XTAL1 Vss Vss VDD Vss VDD COM20020 +5V C6 100nF GND Q1 27 26 25 24 23 21 20 GND 19 18 17 16 28 15 R1 300 R9 300

+5V +5Vext +5Vext IC4 +Vcc 1 A1 2 K1Vo1 4 A2Vo2 3 K2 0V HCPL2631 8 7 6

GNDext

C Suppressor diodes, Array LCDA05 GNDext +5Vext Busabschlu (optional) GNDext C2 100nF R8 300 Transientenschutz (optional) +5Vext +5Vext +5Vext GNDext R22 680 IC2 R14 100K 8 6 7 5 C7 100nF J5 LTBR J8 LTBR J6 J7 LTBR LTBR R20 220 GNDext R17 390 R18 390 B R13 100K

C1 100nF

D[0..7]

GND +5V +5V R3 680 IC3 8 +Vcc A1 7 Vo1K1 6 C3 100nF GND Vo2A2 K2 5 0V HCPL2631

GNDext +5Vext

1 1 2 4 3 C5 100nF GNDext IC6A 1 74HC14 2 2 3 4

RO VCC RE DE DI A B VSS

75ALS176D

20MHz C8 C9 27pF 27pF

GND

Suppressor diodes, Array LCDA05

GNDext +5Vext Busabschlu (optional)

GNDext Transientenschutz (optional)

galvanisch getrennt A A

Title Size A3

ARCNET Interface mit Repeater-Richtungssteuerung

Number S351_3 15-Jul-1997 Sheet of I:\PRJ\ARCNET\S351_3.SCHDrawn By: Maier 7 8 Revision

Date: File: 1 2 3 4 5 6

Take the lead with ARCNET !

Page 59

AUG - ARCNET user group e. V.

9.5.2.5.1 Components

Number 2 2 1 1 1 1 1 2

Reference IC1, IC2 IC3, IC4 IC5 IC6A Q1 ST1 U1

Designation RS485 transceiver opto coupler ARCNET controller Hex Schmitttrigger inverter Crystal oscillator Connector DC/DC converter Suppressor diodes

Type SN75ALS176D HCPL2631 COM20020 MC74HC14

Value

20MHz SUB-D 9-pin connector UMS 05S05 array LCDA05

9.5.2.5.2

Cabling

2 x 2 +3 :

2 x 2 wires, twisted pairs for data signals A, B, and control signals CNTR-A, CNTR-B and 1 additional wires for data reference potential DGND, 2 additional wires for auxiliary voltage supply RP-P, RP-N (without galvanic isolation for the bus participants compensation current has to be considered)

Take the lead with ARCNET !

Page 60

AUG - ARCNET user group e. V.

10 Appendix
10.1 Definitions

Term TTL - H

Definition voltage level > 2.4 V

Comment

TTL - L

voltage level < 0.8 V

Logical 0

signal level at the ARCNET bus if there is no activity [5]

_pulse1 output of the COM20020 = TTL-H [1]

Logical 1

signal level at the ARCNET bus when activity begins [5].

_pulse1 output of the COM20020= TTL- L [1]

Bit length

the total time for the transmission of a bit

400 ns at 2.5 Mbit/s

Mbit/s kBit/s

number of transmitted bits per second. "M" means 1E06, "k" means 1000.

Bitjitter

The time deviation of the real signal ralted to the basic time grid for bits. It is defined as a fraction of a bit length.

Take the lead with ARCNET !

Page 61

AUG - ARCNET user group e. V.

10.2 Literature
Siemens "Optokoppler Lichtschranken Lichtleiter-Bauelemente". Order number B349-B6039 Honeywell OPTOELECTRONICS data sheet "HFX6015 xxx 660 nm LED for Industrial Bus Systems" Honeywell OPTOELECTRONICS data sheet "HFD3403-2xx 5Mb/s Receiver for Industrial Bus Systems"

10.3 Referenced Documents

Document Data sheet COM20020 by SMC Data sheet HY 9088A by SMC ARCNET Designers Handbook, Dok Nr. 61610, by DATAPOINT Cabling Alternatives for the COM20020, COM90C66, AND COM90C165, SMC

Reference [1] [2] [3]

Version 2 9/4/92 Edition 02

Date 7/20/92 9/4/92 1988

[4]

1/92

ARCNET Local Area Network Standard, by ATA Standards Commitee Proposal for a RS 485 Interface Guide to Configuring an ARCNET Network by Contemporary Control Systems, Inc. RS-485 Cabling Guidelines for the COM20020 by SMC

[5]

1.6

5.1.1991

[6]

00.02

8.10.96

May 1994

Take the lead with ARCNET !

Page 62

Take the lead with ARCNET!

AUG - ARCNET user group e.V. Bussardstr. 19 90766 Frth Germany Tel. ++ 49 911 97341 24 Fax. ++ 49 911 97341 10

email: [email protected] Internet: WWW.ARCNET.DE

Take the lead with ARCNET!