1. Industrial Automation Automation Industrielle Industrielle Automation 3.2Field bus operation Buses de terreno: modo de trabajo Bus de terrain: mode de travail Feldbus: Arbeitsweise 3 Industrial Communication Systems

2. Field Bus Operation 3.2 - 2 Industrial Automation 2013 Fieldbus - Operation 3.1 Field bus types 3.2 Field bus operation Data distribution Cyclic Operation Event Driven Operation Real-time communication model Time distribution Networking 3.3 Standard field busses

3. Field Bus Operation 3.2 - 3 Industrial Automation 2013 Objective of the field bus Distribute process variables to all interested parties: source identification: requires a naming scheme accurate process value and units quality indication: {good, bad, substituted} time indication: how long ago was the value produced (description) timequalityvaluesource description

4. Field Bus Operation 3.2 - 4 Industrial Automation 2013 Data format In principle, the bus could transmit the process variable in clear text (even using XML..) However, this is quite expensive and only considered when the communication network offers some 100 Mbit/s and a powerful processor is available to parse the message More compact ways such as ASN.1 have been used in the past with 10 Mbit/s Ethernet Field busses are slower (50kbit/s..12 Mbits/s) and thus more compact encodings are used. timequalityvaluesource description valuelengthtypeASN.1: (TLV) minimum

5. Field Bus Operation 3.2 - 5 Industrial Automation 2013 Datasets wheel speed air pressure line voltage time stamp analog variables binary variables all door closed lights on heat on air condition on bit offset 163248064 6670 size Field busses devices have a low data rate and transmit always the same variables. It is economical to group variables of a device in the same frame as a dataset. A dataset is treated as a whole for communication and access. A variable is identified within a dataset by its offset and its size Variables may be of different types, types can be mixed. dataset identifier dataset

6. Field Bus Operation 3.2 - 6 Industrial Automation 2013 Dataset extension and quality To allow later extension, room is left in the datasets for additional variables. Since the type of these future data is unknown, unused fields are filled with '1". To signal that a variable is invalid, the producer overwrites the variable with "0". Since both an "all 1" and an "all 0" word can be a meaningful combination, each variable can be supervised by a check variable, of type ANTIVALENT2: A variable and its check variable are treated indivisibly when reading or writing The check variable may be located anywhere in the same data set. dataset

7. Field Bus Operation 3.2 - 7 Industrial Automation 2013 hierarchical or peer-to-peer communication AP all traffic passes by the master (PLC); adding an alternate master is difficult (it must be both master and slave) inputoutput PLCs may exchange data, share inputs and outputs allows redundancy and “distributed intelligence” devices talk directly to each other separate bus master from application master ! inputoutput PLC central master / slave: hierarchical peer-to-peer: distributed “slaves” “master” “slaves” “masters” alternate master AP Application

8. Field Bus Operation 3.2 - 8 Industrial Automation 2013 application processor application processor Broadcasts A variable is read in 1..3 different places on avergage Broadcasting messages identified by their source (or contents) increases efficiency. … instances … = variable application processor plant image plant image plant image = distributed data base The bus refreshes plant image in the background Each station snoops the bus and reads the variables it is interested in. Each device is subscribed as source or as sink for a number of process variables Only one device is source of a certain process variable (otherwise collision) The replicated traffic memories can be considered as "caches" of the plant state (similar to caches in a multiprocessor system), representing part of the plant image. bus plant image

9. Field Bus Operation 3.2 - 9 Industrial Automation 2013 Decoupling Application and Bus traffic sending: application writes data into memory receiving: application reads data from memory the bus controller decides when to transmit bus and application are not synchronized application processor bus controller traffic memory (buffer) decoupled (asynchronous): sending: application inserts data into queue and triggers transmission, bus controller fetches data from queue receiving: bus controller inserts data into queue and interrupts application to fetch them, application retrieves data application processor bus controller queues coupled (event-driven): events (interrupts)

10. Field Bus Operation 3.2 - 10 Industrial Automation 2013 Traffic Memory: implementation Bus and Application are decoupled by shared memory, the Traffic Memory, (content addressed memory, CAM, also known as communication memory); process variables are directly accessible by application. Ports (holding a dataset) Application Processor Bus Controller Traffic Memory Associative memory two pages ensure that read and write can occur at the same time (no semaphores !) bus an associative memory decodes the addresses of the subscribed datasets

11. Field Bus Operation 3.2 - 11 Industrial Automation 2013 Freshness supervision Applications tolerate an occasional loss of data, but no stale data, which are at best useless and at worst dangerous. Data must be checked if they are up-to-date, independently of a time-stamp (simple devices do not have time-stamping) To protect the application from using obsolete data, each port in the traffic memory contains a freshness counter. This counter is reset by the bus or the application writing to that port. It is incremented regularly, either by the application processor or by the bus controller. The application should always read the value of the counter before using the port data and compare it with its tolerance level. The freshness supervision is evaluated by each reader independently, some readers may be more tolerant than others. Bus error interrupts in case of severe disturbances are not directed to the application, but to the device management.

12. Field Bus Operation 3.2 - 12 Industrial Automation 2013 Example of Process Variable application interface (program) Access of the application to variables in a traffic memory is simple: ap_get (variable_name, variable value, variable_status, variable_freshness) ap_put (variable_name, variable value) Rather than fetch and store individual variables, access is done by clusters (predefined groups of variables): ap_get (cluster_name) ap_put_cluster (cluster_name) The cluster is a table containing the names and values of several variables. Note: Usually, only one variable is allowed to raise an interrupt when received: the one carrying the current time (sent by the common clock) The clusters can correspond to "segments" in the function block programming.

13. Field Bus Operation 3.2 - 13 Industrial Automation 2013 Transmission principle The previous operation modes made no assumption, how data are transmitted. The actual network can transmit data cyclically (time-driven) or on demand (event-driven), or a combination of both.

14. Field Bus Operation 3.2 - 14 Industrial Automation 2013 Cyclic versus Event-Driven transmission event-driven: send when value change by more than x% of range limit update frequency !, limit hysteresis cyclic: send value strictly every xx milliseconds nevertheless transmit: - every xx as “I’m alive” sign - when data is internally updated - upon quality change (failure) misses the peak (Shannon-Nyquist!) always the same, why transmit ? how much hysteresis ? - coarse (bad accuracy) - fine (high frequency) time individual period hysteresis

15. Field Bus Operation 3.2 - 15 Industrial Automation 2013 Fieldbus: Cyclic Operation mode 3.1 Field bus types Classes Physical layer 3.2 Field bus operation Data distribution Cyclic Operation Event Driven Operation Real-time communication model Time distribution Networking 3.3 Standard field busses

16. Field Bus Operation 3.2 - 16 Industrial Automation 2013 Cyclic Data Transmission address devices (slaves) Bus Master Individual period RTD N polls time [µs] read transfer time [ms] The duration of each poll is the sum of the transmission time of address and data (bit-rate dependent) and of the reply delayof the signals (independent of bit-rate). plant The master polls the addresses in a fixed sequence, according to its poll list. 123456 address (i) data (i) address (i+1) 10 µs/km Poll List 123456123456123456 Individual period 44 µs.. 296 µs

17. Field Bus Operation 3.2 - 17 Industrial Automation 2013 Round-trip delay of master-slave exchange The round-trip delay limits the extension of the bus master most remote data source repeater closest data sink Master Frame access delay propagation delay (t_pd = 6 µs/km) t_source distance next Master Frame t_ms Slave Frame T_m T_s T_m t_repeat (t_repeat < 3 µs) t_repeat t_sm t_mm

18. Field Bus Operation 3.2 - 18 Industrial Automation 2013 Cyclic operation principle The delivery delay (refresh rate) is deterministic and constant. No explicit error recovery needed since a fresh value will be transmitted in the next cycle. Only states may be transmitted, not state changes. To keep the poll time low, only small data items may be transmitted (< 256 bits) Cyclic operation is used to transmit the state variables of the process. These are called Process Data (or Periodic Data) The bus is under control of a central master (or distributed time-triggered algorithm). Data are transmitted at fixed intervals, whether they changed or not. Cycle time is limited by the product of the number of data transmitted and the duration of each poll (e.g. 100 µs / point x 100 points => 10 ms) The bus capacity must be configured beforehand. Determinism gets lost if the cycles are modified at run-time.

19. Field Bus Operation 3.2 - 19 Industrial Automation 2013 Source-Addressed Broadcast The bus master broadcasts the identifier of a variable to be transmitted: Phase1: Process Data are transmitted by source-addressed broadcast. The device that sources that variable responds with a slave frame containing the value, all devices subscribed as sink receive that frame. Phase 2: bus. master bus subscribed devices subscribed device subscribed device source sink variable value bus variable identifier bus master devices (slaves) source sink subscribed devices sink device devices (slaves)

20. Field Bus Operation 3.2 - 20 Industrial Automation 2013 Optimizing Cyclic Operation Problem: Cyclic operation uses a fixed portion of the bus's time The poll period increases with the number of polled items The response time slows down accordingly Solution: introduce sub-cycles for less urgent periodic variables that are a power of 2 multiple of the base period: The poll cycles should not be modified at run-time (non-determinism) A device exports many process data (state variables) with different priorities. If there is only one poll type per device, a device must be polled at the frequency required by its highest-priority data. To reduce bus load, the master polls the process data, not the devices group with period 1 ms time 4a81614b643 1 ms period (basic period) 2 ms period 24a 4 ms period 1 ms 1112

21. Field Bus Operation 3.2 - 21 Industrial Automation 2013 Cyclic Transmission and Application Bus and applications are decoupled by a shared memory, the traffic memory, which acts as distributed database updated by the network. The bus master scans the identifiers at its own pace. The bus traffic and the application cycles are asynchronous to each other. Traffic Memory cyclic algorithms cyclic algorithms cyclic algorithms cyclic algorithms port address application 1 Ports application 2 4 source port sink port port data sink port cyclic poll bus controller bus master application 3 bus Periodic List Ports bus controller bus controller bus controller bus controller

22. Field Bus Operation 3.2 - 22 Industrial Automation 2013 Application Of Cyclic Bus The principle of cyclic operation, combined with source-addressed broadcast, has been adopted by most modern field busses Deterministic behavior, at expense of reduced bandwidth and geographical extension. It is currently used for power plant control, rail vehicles, aircrafts, etc... The poll scan list located in the central master (which may be duplicated for availability purposes) determines the behavior of the bus. It is configured for a specific project by a single tool, which takes into account the applications’ requirements. This guarantees that no application can occupy more than its share of the bus bandwidth and gives control to the project leader.

23. Field Bus Operation 3.2 - 23 Industrial Automation 2013 Example: delay requirement Worst-case delay for transmitting all time critical variables is the sum of: Source application cycle time Individual period of the variable Sink application cycle time 8 ms 16 ms 8 ms = 32 ms subscribers application instances device publisher application instance bus instance device applications bus

24. Field Bus Operation 3.2 - 24 Industrial Automation 2013 Example: traffic pattern in a locomotive number of devices: 37 ( including 2 bus administrators) 37 of 16 bits 16 ms32 ms64 ms128256 49 frames of 256 bits 30 frames of 128 bits 1024 65 frames of 64 bits 18 of 32 period % periodic time occupancy is proportional to surface total = 92%

25. Field Bus Operation 3.2 - 25 Industrial Automation 2013 Fieldbus: Event-driven operation 3.1 Field bus types 3.2 Field bus operation Data distribution Cyclic Operation Event Driven Operation Real-time communication model Time distribution Networking 3.3 Standard field busses

26. Field Bus Operation 3.2 - 26 Industrial Automation 2013 Event-driven Operation Detection of an event is an intelligent process: Not every change of a variable is an event, even for binary variables. Often, a combination of changes builds an event. Only the application can decide what is an event, since only the application programmer knows the meaning of the variables. Events cause a transmission only when an state change takes place. Bus load is very low on average, but peaks under exceptional situations since transmissions are correlated by the process (christmas-tree effect). event- reporting station event- reporting station event- reporting station plant Multi-master bus: uses write-only transfers intelligent stations sensors/ actors

27. Field Bus Operation 3.2 - 27 Industrial Automation 2013 Bus interface for event-driven operation Application Processor Bus Controller message (circular) queues bus driver filter application Each transmission on bus causes an interrupt. Bus controller checks address and stores data in message queues. Driver is responsible for removing messages of queue memory and prevent overflow. Filter decides if message can be processed. interrupt

28. Field Bus Operation 3.2 - 28 Industrial Automation 2013 Response of Event-driven operation Interruption of server device at any instant can disrupt a time-critical task. Buffering of events can cause unbounded delays Gateways introduce additional uncertainties Since events can occur anytime on any device, stations communicate by spontaneous transmission, leading to possible collisions Caller Application Called Application Transport software Transport software interrupt request indication confirm Bus time

29. Field Bus Operation 3.2 - 29 Industrial Automation 2013 Determinism and Medium Access In Busses Although the moment an event occurs is not predictable, the bus should transmit the event in a finite time to guarantee the reaction delay. Events are necessarily announced spontaneously The time required to transmit the event depends on the medium access (arbitration) procedure of the bus. Medium access control methods are either deterministic or not. Non-deterministic Collision (CSMA/CA) Deterministic Central master, Token-passing (round-robin), Binary bisection (collision with winner)

30. Field Bus Operation 3.2 - 30 Industrial Automation 2013 Events and Determinism Deterministic medium access is necessary to guarantee delivery time bound but it is not sufficient since events messages are queued in the devices. The average delivery time depends on the length of the queues, on the bus traffic and on the processing time at the destination. Often, the applications influence the event delay much more than the bus does. Real-time Control = Measurement + Transmission + Processing + Acting bus data packets acknowledgements input and output queues events producers & consumers

31. Field Bus Operation 3.2 - 31 Industrial Automation 2013 Events Pros and Cons In an event-driven control system, there is only a transmission or an operation when an event occurs. Advantages: Drawbacks: Can treat a large number of events – but not all at the same time Supports a large number of stations System idle under steady - state conditions Better use of resources Uses write-only transfers, suitable for LANs with long propagation delays Suitable for standard (interrupt-driven) operating systems (Unix, Windows) Requires intelligent stations (event building) Needs shared access to resources (arbitration) No upper limit to access time if some component is not deterministic Response time difficult to estimate, requires analysis Limited by congestion effects: process correlated events A background cyclic operation is needed to check liveliness

32. Field Bus Operation 3.2 - 32 Industrial Automation 2013 Fieldbus: real-time communication model 3.1 Field bus types 3.2 Field bus operation Centralized - Decentralized Cyclic Operation Event Driven Operation Real-time communication model Time distribution Networking 3.3 Standard field busses

33. Field Bus Operation 3.2 - 33 Industrial Automation 2013 Mixed Data Traffic represent the state of the plantrepresent state changes of the plant -> Periodic Transmission of Process Variables short and urgent data items Since variables are refreshed periodically, no retransmission protocol is needed to recover from transmission error. -> Sporadic Transmission of Process Variables and Messages infrequent, sometimes long messages reporting events, for: System: initialisation, down-loading,... Since messages represent state changes, a protocol must recover lost data in case of transmission errors Users: set points, diagnostics, status Process DataMessage Data... motor current, axle speed, operator's commands, emergency stops,... periodic phase event sporadic phase time basic period sporadic phase

34. Field Bus Operation 3.2 - 34 Industrial Automation 2013 Mixing Traffic is a configuration issue Cyclic broadcast of source-addressed variables is the standard solution in field busses for process control. Cyclic transmission takes a large share of the bus bandwidth and should be reserved for really critical variables. The decision to declare a variable as cyclic or event-driven can be taken late in a project, but cannot be changed on-the-fly in an operating device. A message transmission scheme must exist alongside the cyclic transmission to carry not-critical variables and long messages such as diagnostics or network management An industrial communication system should provide both transmission kinds.

35. Field Bus Operation 3.2 - 35 Industrial Automation 2013 Real-Time communication stack The real-time communication model uses two stacks, one for time-critical variables and one for messages Logical Link Control time-critical process variables Management Interface time-benign messages Physical Link (Medium Access) Network (connectionless) Transport (connection-oriented) Session Presentation Application 7 6 Remote Procedure Call 5 4 3 2' 1 connectionless connection-oriented medium access implicit Logical Link Control 2" media common

36. Field Bus Operation 3.2 - 36 Industrial Automation 2013 Application View Of Communication R4 Traffic Memory Periodic Tasks R3R2R1 Message Data (unicast) Process Data (Broadcast) E3E2E1 Event-driven Tasks bus Supervisory Data bus controller Message ServicesVariables Services Queues node

37. Field Bus Operation 3.2 - 37 Industrial Automation 2013 Cyclic or Event-driven Operation For Real-time ? Data are transmitted at fixed intervals, whether they changed or not. Data are only transmitted when they change or upon explicit demand. cyclic operationevent-driven operation (aperiodic, demand-driven, sporadic)(periodic, round-robin) Worst Case is normal caseTypical Case works most of the time Non-deterministic: delivery time vary widelyDeterministic: delivery time is bound All resources are pre-allocatedBest use of resources message-oriented busobject-oriented bus Fieldbus Foundation, MVB, FIP,..Profibus, CAN, LON, ARCnet The operation mode of the communication exposes the main approach to conciliate real-time constrains and efficiency in a control systems.

38. Field Bus Operation 3.2 - 38 Industrial Automation 2013 Fieldbus: Time distribution 3.1 Field bus types 3.2 Field bus operation Data distribution Cyclic Operation Event Driven Operation Real-time communication model Time distribution Networking 3.3 Standard field busses

39. Field Bus Operation 3.2 - 39 Industrial Automation 2013 Time: TAI and UTC TAI (Temps Atomique International) is the scientific time scale. It is continuously incrementing and will never be reset or discontinued. It is the base of all other scales. UTC (Universal time coordinated) is the legal time. It is the base for the clocks of all countries. It indicates approximately 12:00:00 at solar noon in Greenwich at the Spring equinox (it was formerly called Greenwich Mean Time). The Bureau International des Poids et Mesures (BIPM), Paris, is responsible for the definition of UTC. Rate: UTC and TAI proceed at exactly the same rate; both were identical back in 1961. Leap Seconds: Since 1961, the earth rotation slightly slowed down, days became longer. When the difference between UTC and solar noon exceed 0,9 s, which happens after some years, the BIMP adjusts UTC by letting all clocks insert a leap second, so the last minute of a day lasts 61 seconds (the reverse case is also possible, but very unlikely). Leap seconds cannot be anticipated, since irregularities of the earth’s rotation are unpredictable. In 2011, UTC lagged behind TAI by 34 seconds: when TAI was 2011-02-04 12:00:00, UTC was 2011-02-04 11:59:26. One cannot deduce TAI time from UTC time without a table of all elapsed leap seconds, and UTC cannot be predicted for a given TAI, since the introduction of a leap second is a decision of the BIPM. The system of leap seconds is still subject to discussion and could be revised in the future – it could be abolished in 2015.

40. Field Bus Operation 3.2 - 40 Industrial Automation 2013 Time-stamping and synchronisation In many applications, such as disturbance logging and sequence-of-events, the exact sampling time of a variable must be transmitted together with its value. For this purpose, the devices are equipped with a clock that records the creation date of a value (not the transmission time). To reconstruct events coming from several devices, clocks must be synchronized. considering transmission delays over the field bus (and in repeaters,....) A field bus provides means to synchronize clocks despite propagation delays and failures of individual nodes. bus input processing t1t1 t2t2 t3t3 t4t4 t1t1 val 1

41. Field Bus Operation 3.2 - 41 Industrial Automation 2013 Syntonization vs Synchronization distinguish: 1) frequency synchronization (syntonization) 2) relative synchronization 3) absolute synchronization

42. Field Bus Operation 3.2 - 42 Industrial Automation 2013 Example: Phasor information Phasor transmission over the European grid: a phase error of 0,01 radian is allowed, corresponding to +/- 26 µs in a 60 Hz grid or 31 µs in a 50 Hz grid.

43. Field Bus Operation 3.2 - 43 Industrial Automation 2013 Time distribution In master-slave busses, the master distribute the time as a bus frame. The slave can compensate for the path delays. Time is relative to the master In demanding systems, time is distributed over separate lines as relative time (e.g. PPS = one pulse per second) or absolute time (IRIG-B), with accuracy of 1 µs. In data networks, a reference clock (e.g. GPS or atomic clock) distributes the time. A protocol evaluates the path delays to compensate them. NTP (Network Time Protocol): about 1 ms is usually achieved. IEEE 1588, all network devices collaborate to estimate the delays, an accuracy below 1 µs can be achieved without need for separate cables (but hardware support for time stamping required). (Telecom networks typically do not distribute time, they only distribute frequency)

44. Field Bus Operation 3.2 - 44 Industrial Automation 2013 NTP (Network Time Protocol) principle time request time response t1t1 t2t2 t3t3 t4t4 time request time response t’ 1 t’ 2 t’ 3 t’ 4 distance time  network delay servernetworkclient network delay  Measures delay end-to-end over the network (one calculation) Problem: asymmetry in the network delays, long network delays

45. Field Bus Operation 3.2 - 45 Industrial Automation 2013 IEEE 1588 principle (PTP, Precision Time Protocol) Grand Master Clock Two calculations: residence time and peer delay All nodes measure delay to peer TC correct for residence time (HW support)

46. Field Bus Operation 3.2 - 46 Industrial Automation 2013 IEEE 1588 – 1 step clocks Sync (contains all  + ) residence time Pdelay_Resp (contains t 3 – t 2 ) Pdelay_Req ordinary (slave) clock distance time Sync 1-step transparent clock grand master clock 1-step transparent clock residence time bridge  link delay t2t2 t3t3 Pdelay_Resp t1t1 t4t4 Pdelay_Req t2t2 t3t3 t1t1 t4t4 t2t2 t3t3 t1t1 t4t4 Sync Pdelay_Resp Pdelay_Req  t5t5  t5t5 t6t6 Grandmaster sends the time spontaneously. Each device computes the path delay to its neighbour and its residence time and corrects the time message before forwarding it residence time calculation peer delay calculation

47. Field Bus Operation 3.2 - 47 Industrial Automation 2013 References To probe further http://www.ines.zhaw.ch/fileadmin/user_upload/engineering/_Institute_und_Zentr en/INES/IEEE1588/Dokumente/IEEE_1588_Tutorial_engl_250705.pdfhttp://www.ines.zhaw.ch/fileadmin/user_upload/engineering/_Institute_und_Zentr en/INES/IEEE1588/Dokumente/IEEE_1588_Tutorial_engl_250705.pdf http://blog.meinbergglobal.com/2013/11/22/ntp-vs-ptp-network-timing- smackdown/http://blog.meinbergglobal.com/2013/11/22/ntp-vs-ptp-network-timing- smackdown/ http://blog.meinbergglobal.com/2013/09/14/ieee-1588-accurate/

48. Field Bus Operation 3.2 - 48 Industrial Automation 2013 Fieldbus: Networking 3.1 Field bus types 3.2 Field bus operation Data distribution Cyclic Operation Event Driven Operation Real-time communication model Time distribution Networking 3.3 Standard field busses

49. Field Bus Operation 3.2 - 49 Industrial Automation 2013 Networking field busses Networking field busses is not done through bridges or routers, because normally, transition from one bus to another is associated with: - data reduction (processing, sum building, alarm building, multiplexing) - data marshalling (different position in the frames) - data transformation (different formats on different busses) Only system management messages could be threaded through from end to end, but due to lack of standardization, data conversion is not avoidable today.

50. Field Bus Operation 3.2 - 50 Industrial Automation 2013 Networking: Printing Example B C D E PMLS PMLS PMLS PMLS MPS Section Control Line bus (AF100) Section Busses (AF100) Console, Section Supervision Reelstand bus (Arcnet) Reelstand-Gateways Operator bus (Ethernet) Plant-bus (Ethernet) Production Reelstands Printing Towers RPERPDRPCRPB SSCSSDSSESSB multiplicity of field busses with different tasks, often associated with units. main task of controllers: gateway, routing, filtering, processing data. most of the processing power of the controllers is used to route data MPS = Master Printing System LS = Leitstand (section supervision) PM = Print Master SS =Section Steuerung (section control)

51. Field Bus Operation 3.2 - 51 Industrial Automation 2013 Assessment What is the difference between a centralized and a decentralized industrial bus ? What is the principle of source-addressed broadcast ? What is the difference between a time-stamp and a freshness counter ? Why is an associative memory used for source-addressed broadcast ? What are the advantages / disadvantages of event-driven communication ? What are the advantages / disadvantages of cyclic communication ? How is time transmitted ? How are field busses networked ?