Messaging application layer (MODBUS) over ethernet to transport layer (TCP) communications method and apparatus for a modular terminal input/output system

Abstract
A communications adapter is provided for interfacing between a master device and an I/O device (body) having an output and/or an input. In the case of the I/O body having an output, the adapter has a TCP port for coupling to the master device via a transmission path for receiving a request message. The adapter also has a connector for operable coupling to the I/O device for receiving the output of the I/O device. The adapter further has an interface circuit operably connected to the TCP port and the connector for transmitting a response message over the transmission path in response to the request message, the response message correlating to the output received from the I/O device. The request message and the response message is limited to a length that is less than a TCP transaction length and/or a maximum transmission unit limit, or both, depending on the embodiment of the present invention.
Description


TECHNICAL FIELD

[0002] The present invention relates to input and output (I/O) terminal interfaces for communication between input and output field devices, and programmable logic controllers (PLCs), and for communication between input and output field devices, and other field master devices, such as a host device. More specifically, the present invention relates to a modular I/O terminal interface for communication over MODBUS encoded Ethernet to TCP.



BACKGROUND OF THE INVENTION

[0003] Within industrial automation systems market, there are various types of communications network protocols which were developed for products, such as PLCs, to run on the products to be networked together, and for the field devices to be monitored and controlled from various locations within the particular automation systems. Thus, various types of input and output communications devices have been produced to communicate within the various types of communications protocols for the various types of communications networks for the automation systems. For example, FIG. 2 shows various types of communications protocols, such as Interbus-S, Profibus DP, Modbus Plus, Echelon, Seriplex, CAN DeviceNet, CAN SDS, and CANCAL, to name only a few. An additional protocol is FIPIO. Each of these various communications network types require specific input and output devices for communication with the input and output field devices based on the various types of communications protocols each having specific and different communications requirements.


[0004] In addition, not only does each network protocol require different input and output communication devices for communication with the various above protocols, but there is also the need of having input and output communication devices for communication directly with the PLCs. Communication between the PLCs and the input and output communication devices can have yet another type of communication protocol which may require different types of input and output communication devices for each different type or brand of PLC.


[0005] The present invention is provided to solve these and other problems.



SUMMARY OF THE INVENTION

[0006] The present invention is a communications adapter for interfacing between MODBUS over Ethernet to TCP for the communication of information between field devices and a field master using these types of protocols. Field devices can include devices such as digital or binary inputs, digital or binary outputs, analog inputs, analog outputs, QPR units or other special units, and INTIO devices to name only a few. Field masters can include programmable logic controllers (PLCs) (sometimes referred to as process control devices or PCDs), application specific controllers, and host computers/devices such as a personal computer with industrial automation software running thereon. For example, U.S. Pat. No. 5,611,059 discloses various types of controllers within a control structure for interfacing with the field devices, as well as user interface of a personal computer having industrial automation software running thereon.


[0007] The communications adapter is for providing an interface between a master device and an I/O device (body) having an output and/or an input. In the case of the I/O body having an output, the adapter has a TCP port for coupling to the master device via a transmission path for receiving a request message. The adapter also has a connector for operable coupling to the I/O device for receiving the output of the I/O device. The adapter further has an interface circuit operably connected to the TCP port and the connector for transmitting a response message over the transmission path in response to the request message, the response message correlating to the output received from the I/O device. The request message and the response message is limited to a length that is less than a TCP transaction length and/or a maximum transmission unit limit, or both, depending on the embodiment of the present invention.


[0008] The present invention also includes a method for providing a connection between a master device and an I/O device having an output and/or and input. In the case of the method for providing a connection between a master device and an I/O device having an output, the method includes receiving over a transmission path a request message on a preregistered TCP port selected from a plurality of TCP ports. The method also includes receiving the output from the I/O device. The method further includes transmitting a response message over the transmission path in response to the request message, the response message correlating to the output of the I/O device. The request message and/or the response message is limited to a length that is less than both a TCP transaction length and/or a maximum transmission unit limit, or both, depending on the embodiment of the present invention.


[0009] In the case of the method for providing a connection between a master device and an I/O device having an input, the method includes receiving over a transmission path a request message on a preregistered TCP port selected from a plurality of TCP ports. The method also includes transmitting a response message over the transmission path in response to the request message. The method further includes transmitting data to the input of the I/O device in response to the request message. The request message and/or the response message is limited to a length that is less than a TCP transaction length and/or a maximum transmission unit limit, or both, depending on the embodiment of the present invention.


[0010] The communications adapter (adapter or COM-adapter) attaches to an input/output body as is described U.S. patent application Ser. No. 09/036,565 and/or German Patent No. DE 196 15 093, which are hereby incorporated by reference.


[0011] Specifically, the communications adapter is configured to directly attach to and communicate through at least an in-data port, the out-data port, and the identification port of the input/output body. In addition, the communications adapter is also configured to communicate with MODBUS over Ethernet type field masters. The communications adapter can have an input multiplexer for accepting data from the in-data port and from the identification port, an output multiplexer for providing data to the out-data port, and a processor for communicating with the input multiplexer and the output multiplexer. The processor is also provided for converting the data, received from the input multiplexer and the output multiplexer. Field bus circuitry is also connected between the processor and the field bus, within the communications adapter, for allowing the processor to communicate with the field master on the field bus.


[0012] Alternatively, the communications adapter has at least one application specific integrated circuit (ASIC) for accepting data from the in-data port and from the identification port, and for providing data to the out-data port. The ASIC converts the data to and from a MODBUS over Ethernet communications protocol of a PLC type of field master.


[0013] The present invention can take the form of a communications adapter adapted to only an input body or only an output body having the same advantages and the full input/output body, as is understood with reference to the incorporated references above. Thus, the invention allows inexpensive standard network components to be used in place of specialized real time field bus components in communicating with industrial sensor and actuator devices. This enables major savings in cost and complexity when connecting simple devices to a network solution involving programmable controllers or other industrial computer systems, since the same networking infrastructure components can be shared. Other advantages and aspects of the present invention will become apparent upon reading the following description of the drawings and detailed description of the invention.







BRIEF DESCRIPTION OF THE DRAWINGS

[0014]
FIG. 1 is a perspective view of the input/output body, with the communications adapter of the present invention attached thereto.


[0015]
FIG. 2 is a modularity chart showing the flexibility of the present invention.


[0016]
FIG. 3 is a block diagram of the internal structure of the present invention.







DETAILED DESCRIPTION

[0017] While this invention is susceptible of embodiment in many different forms, there is shown in the drawings and will herein be described in detail a preferred embodiment of the invention with the understanding that the present disclosure is to be considered as an exemplification of the principles of the invention and is not intended to limit the broad aspect of the invention to the embodiments illustrated.


[0018] With reference to FIG. 1, the present invention is a communications adapter (COM-adapter) for connecting to and communicating with an input/output (IO) body 2. The IO body 2 interfaces between field devices (not shown in FIG. 1) and a field master (not shown in FIG. 1). The communication adapter will allow an IO base (body) to reside on a Ethernet network and communicate using Modbus messages over the TCP/IP protocol.


[0019] Some abbreviations for this specification can be defined as follows:


[0020] MIO: Momentum Terminal Input Output


[0021] ARP: Address Resolution Protocol used to get Ethernet physical address, given the IP address


[0022] MI-Interface: Momentum IO Internal Interface


[0023] BOOTP: Protocol used at power up to associate a MAC address with an IP address


[0024] COM-adapter: Communication Adapter


[0025] ICMP: Error indication and control protocol


[0026] I/O-body: Input/Output body (base unit)


[0027] IP: Internet network layer providing worldwide addressing capability


[0028] Modbus: Messaging application layer (Read/Write services)


[0029] TCP: transport layer for point to point


[0030] STP: Shielded Twisted Pair


[0031] With reference to FIGS. 1 and 3, the COM-adapter 10 is for providing an interface between a master device 12 and an I/O device (body) 14 having an output and/or an input. In the case of the I/O device 12 having an output, the adapter 10 has a TCP port 16 for coupling to the master device 12 via a transmission path 18 for receiving a request message. The adapter 10 also has a connector 20 for operable coupling to the I/O device 14 for receiving the output of the I/O device 14. The adapter 10 further has an interface circuit 22 operably connected to the TCP port 16 and the connector 20 for transmitting a response message over the transmission path 18 in response to the request message, the response message correlating to the output received from the I/O device 14.


[0032] The request message and the response message is limited to a length that is less than a TCP transaction length and/or a maximum transmission unit limit, or both, depending on the embodiment of the present invention.


[0033] The selected TCP port in the COM-adapter should be TCP port number 502 according to the design assumptions for one embodiment of the present invention, which are provided further below. In addition, the assumptions also provide that the interface circuitry ignores a connection request received over the transmission path by another TCP port. In addition, the interface circuitry discards a message of unknown meaning received by the TCP port. The interface circuitry closes the transaction path in response to an error received by the TCP port. In addition, the interface circuitry transmits an Address Resolution Protocol Response over the transmission path if an Internet Protocol address encoded with the request message matches an Internet Protocol address associated with the I/O device 14. The interface circuitry also transmits a connection confirmation in response to a TCP connection request.


[0034] The present invention also includes a method for providing a connection between a master device 12 and an I/O device 14 having an output and/or an input. In the case of the method for providing a connection between a master device and an I/O device 14 having an output, the method includes receiving over a transmission path 18 a request message on a preregistered TCP port 16 selected from a plurality of TCP ports. The method also includes receiving the output from the I/O device 14. The method further includes transmitting a response message over the transmission path in response to the request message, the response message correlating to the output of the I/O device 14. The request message and/or the response message is limited to a length that is less than both a TCP transaction length and/or a maximum transmission unit limit, or both, depending on the embodiment of the present invention.


[0035] In the case of the method for providing a connection between a master device 12 and an I/O device 14 having an input, the method includes receiving over a transmission path 18 a request message on a preregistered TCP port 16 selected from a plurality of TCP ports. The method also includes transmitting a response message over the transmission path in response to the request message. The method further includes transmitting data to the input of the I/O device 14 in response to the request message. The request message and/or the response message is limited to a length that is less than a TCP transaction length and/or a maximum transmission unit limit, or both, depending on the embodiment of the present invention.


[0036] MODBUS is an industrial control protocol that is a widely-implemented standard where each transaction is independent and comprises a request message and response message pair, each of bounded length. The length of the request and response messages is such that the encapsulated message, when sent as part of a standard TCP connection, is smaller than both the TCP ‘window’ and Maximum Transmission Unit (MTU) limits. In addition, because the response from a previous transaction must be received before the request for the next transaction may be sent, there is no reason to break up, or ‘fragment’ a MODBUS message when transmitted over TCP. The result is that there is a direct relationship between the MODBUS message encoding and the TCP frame encoding on any given network.


[0037] By making a number of simplifying assumptions about the relationship between the target ‘slave’ device and its interrogating >master= device, the obligations of the receiving software (See Appendix A) can be reduced from the traditional ‘network protocol stack’ consisting of a number of interacting software (See Appendix A) components to simpler ‘state machine’ where the correct response to an incoming request can be rapidly determined from the content of the start of the request message. For one embodiment of the present invention, the following are the significant assumptions:


[0038] All requests will be initiated at the ‘master’. This means the slave has no need to initiate connections or resolve symbolic Domain Name Service (DNS) names.


[0039] All requests will appear on registered TCP port number 502, which is registered already with the Internet Assigned Numbers Authority (IANA) for the purpose of carrying MODBUS traffic. This means that attempted connection requests on other ports can be ignored.


[0040] The slave may assume that a valid return path to the master is either ‘direct’ or can use the network bridge device which last processed the request message on its way to the device. This means that it is no necessary to maintain IP routing tables or understand ICMP redirect messages.


[0041] It is appropriate if presented with a message of unknown meaning to discard the message as if the device is hiding behind a ‘network firewall’ device.


[0042] The appropriate response to any MODBUS protocol or encoding error is to unilaterally close the TCP connection.


[0043] As a result, the MODBUS/TCP/Ethernet implementation, as an example, can implement largely pre-calculated responses to the following messages:


[0044] Address Resolution Protocol (ARP) Request—send ARP response if IP address matches.


[0045] Internet Control Management Protocol Echo (ICMP PING) request—send PING response (to aid in standard network setup and troubleshooting).


[0046] TCP connection request (SYN)—send connection confirm (SYN ACK) if the requested port number is 502, ignore otherwise.


[0047] TCP disconnect request (FIN)—send disconnect confirm (FIN ACK).


[0048] MODBUS request as TCP data frame—send MODBUS response as TCP data frame, generating most of the protocol prefix information from the equivalent information in the request.


[0049] As a result, performance is enhanced in at least two significant respects as compared to a traditional implementation:


[0050] The amount of network traffic is reduced because each MODBUS transaction involves typically 2 messages only (the encoded request and response). A traditional protocol stack would generate 4 messages (adding an ACK message for each data unit transmitted). This results in a saving of 20% to 50% of the network traffic, allowing the same network components to handle an additional 25% to 100% throughput with no additional cost.


[0051] The calculation time at the slave is drastically reduced, allowing simpler and less expensive microprocessors to substitute for expensive ones and yet achieve the same effective response performance. At the same time, compatibility with conventional protocol stacks on large computers is maintained, because adherence to the standard encoding for MODBUS over TCP has not been violated.


[0052] In the present invention, the Ethernet (MAC) address for the COM-adapter can be stored in flash memory. The COM-adapter can operate on a standard Ethernet network that should include a BOOTP server. The IP address will be obtained from a network server using the BOOTP protocol. A commercially available BOOTP server can be used for this purpose. The COM-Adapter communicates using Modbus messages over TCP/IP to an NOE2×1 module, a host computer, or any device using the Modbus Protocol over Ethernet. The COM-adapter will be compatible with at least, host programs running over Windows 95, Windows NT, and UNIX TCP/IP stacks. The COM-adapter should meet the main international standards: U. L., C.S.A.., F.M., and C.E. The COM-adapter enables the connection of all MIO Boards to the Ethernet. Used with the COM-adapter, the MIO will provide at least the following functions: communication using a limited set of Modbus commands over the TCP/IP protocol; input and output data exchanges; parameter management (at initialization time and at run time); diagnostic information (LEDs, communications statistics); and downloading of new operating software (See Appendix A) over the Ethernet connection.


[0053] The external accesses to the Ethernet Comm Adapter are:


[0054] RJ45 female connector for 1OBaseT network connection.


[0055] 2 LEDs:—Run: Green LED. Indicates the module's operating state.—Communication: Green LED: Indicates network activity.


[0056] The mode of operation is as follows:


[0057] Initialization: at power up, a kernel firmware (see Appendix A, which is hereby incorporated herein by reference) will perform an internal initialization and self-tests. If the self tests fail, a run LED will flash, if possible, indicating a failure reason. In this state, the COM-adapter does not attempt to communicate over the network. If initialization is successful, the COM-adapter then requests its IP parameters (IP address, default gateway, and sub-net mask) using the BOOTP protocol over the Ethernet network and the MAC address stored in a nonvolatile memory. The COM-adapter will wait ten seconds for a BOOTP server to respond before trying again. The COM-adapter will retry six times, three times using the Ethernet II framing type, and three times using the 802.2 SNAP header. If no server responds, the COM-adapter will use the last valid IP stored in flash memory. If no valid IP data exists, the COM-adapter will flash the run LED in a specified pattern and retry the BOOTP request every thirty seconds. If the IP parameters are successfully obtained, the kernel (firmware) performs a checksum test on the executable image. If the image is invalid, the kernel will put the I/O base into a safe mode, flash the run LED with a pattern indicating its condition and wait for a download command sequence on the Ethernet port.


[0058] After receiving its network parameters, the COM-adapter runs an identification procedure with the I/O-body. If the Identification procedure fails, the Run LED flashes a failure code. When the initialization phase is successfully completed, the COM-adapter is ready to communicate using the Modbus protocol over TCP/IP. The Run LED will be on steady.


[0059] The COM-adapter will enter the failed state if the I/OERR signal is pulled down by a complex I/O body for more than one second. If the I/OERR signal is pulled down by a simple I/O-Board, the COM-adapter remains in the ready state. The Run LED will indicate the failure condition and the I/O base will be forced into a safe state.


[0060] With respect to downloading, at any time after successfully obtaining its IP information, the COM-adapter will accept Modbus commands with function code 125 for downloading executive code. Downloading executive code should be understood in the art. When entering kernel software (See Appendix A), the unit will place the I/O in a safe state and enter the kernel. The Run LED will flash the code indicating that a download is in progress. During kernel downloading, only Modbus 125 commands will be accepted. When the download is complete, the COM-adapter will restart the initialization sequence. If the download fails, the unit will flash the LED sequence indicating that a download is required.


[0061] The following describes the I/O operating modes. The COM-adapter will not support peer cop or global data. The COM-adapter has three groups of internal registers: module data, configuration, and status. All three register groups can be accessed over the Ethernet network by standard Modbus commands to ensure compatibility with existing devices (i.e., user logic MSTR block). The COM-adapter will restrict write access to the first node that communicates with it. The COM-adapter will maintain this lockout until communication with the master times out. The COM-adapter will allow the master to specify up to three other Aowners”60 in order to facilitate the efficient implementation of hot standby systems. The user can access various registers to obtain I/O module information via the Ethernet network. These internal registers are mapped to emulate 4xxxx registers allowing read/write 4xxxx register commands to be used (i.e. by using a MSTR block).


[0062] Below is a Table to show data flow between the Ethernet network and the COM-adapter internal registers:
1Data GroupOffset in HexSize of FieldRead OnlyInput Data0001Module dependentWrite OnlyOutput Data0001Module dependentConfiguration GroupOffset in HexSize of FieldRead/WriteTime-out RegisterF0011 wordRead/WriteReserved for DistributedF201-F3FF(See programIOlisting).Read/WriteOwnership RegisterF4016 wordsRead/WriteIP Parameter SaveF4101 wordRegisterStatus GroupOffset in HexSize of FieldRead OnlyStatus registerF801word 1 of status registerRead OnlyModule ASCII headerFC01word 6 of status register


[0063] Data group information is as follows: The input buffer scheme captures a snap shot of all input data The output buffer scheme insures that the newest copy of output data (only one buffer) is written to the output modules. A special algorithm is also used to insure old data is not lost, during a single word update of a multiple word output buffer field.


[0064] Configuration group registers are as follows: The configuration group contains three registers that are used by the COM-adapter: a module hold-up time register, a write privilege register, and a save IP parameter register. A block of registers in this area is reserved for use by distributed I/O.


[0065] Module hold-up: Module hold-up time-out is the amount of time the output modules are held in their current state without being updated by Modbus write commands. The module time-out is one word at offset F001. This register can be read from and written to using Modbus commands, and the default value is 100 (1 sec.). The module time-out word is in increments of 10 msec, with a minimum value of 30 (300 msec) and a maximum value of 6,000 (60 sec.). All values outside this range will be logged as illegal data address errors. Write access is restricted to the current Amaster.”


[0066] Another timer is the reservation time-out. The COM-adapter is dedicated to one Ethernet device. Reservation time-out is the amount of time (60 seconds) that the output module will be dedicated to an Ethernet device that is no longer communicating with it. If the time-out expires the COM-adapter will be dedicated to the next Ethernet device that writes to it. If two Ethernet devices wanted to write data to the same COM-adapter, one Ethernet device would have to wait for the reservation time-out to expire before it could write its data. This time-out is a fixed value, preferably sixty seconds, and is not user accessible (not changeable).


[0067] Ownership register: the Ownership register is used so that more than one Modbus device can have write access to the COM-adapter. Up to three remote Ethernet devices can have write access at the same time. This special case overrides the reservation time limit. The ownership register is six words starting at location F401, two words for the IP address of each Ethernet device. The default setting for each ownership register is zero (no owner). Register F401 contains the first owner's IP address, register F403 the second owner's IP address, and F405 contains the third owners IP address. All three owners have the same write privileges. A fourth controller (Ethernet device) could write to the COM-adapter if the three known owners (Ethernet devices) had ceased communication for more than the reservation time-out of sixty seconds.


[0068] IP Parameter Save Register: this Boolean register is located at offset F410 and determines the behavior of the COM-adapter if a BOOTP server is not found at initialization time. If a one is written to the register, the current values of the IP parameters will be written to non-volatile storage (memory). If a BOOTP server cannot be found during the next initialization, these values will be used. If a zero is written to this register, any saved IP parameters will be erased. A change of state of this register will cause a reset of the COM-adapter. Writing the IP Parameter Save Register is restricted to owner or owners (Ethernet devices).


[0069] MBP Status Group registers: there are two registers in the Status Group: The internal status register starts at offset F801, and the ASCII header register starts at offset FC01.


[0070] Module (COM-Adapter) Status register:


[0071] The following table shows the contents of the Module status register:
2Words ofinformationDescriptionValueWord 1Number of status wordsMax. of 13 wordsWord 2number of module input bytesModule dependentWord 3number of module output bytesModule dependentWord 4Module ID numberModule dependentWord 5Module revision numberX RWord 6ASCII header length in wordsModule dependentWord 7last node to communicate lowIP AddresswordWord 8remaining reservation timeXWord 9remaining module holdup timeXWord 10Module health8000 hex = good healthWord 11Last 1/O module errorXWord 12I/O module error counterXWord 13last node to communicateIP Address


[0072] For the purpose of this chart:


[0073] X=upper four bits reserved for station management commands, the upper 4 bits in this word will always be zero; R=(module revision number) REV 1.00=100hex.


[0074] ASCII header block: there is an ASCII header block starting at offset FC01. This header is used to give a brief description of the module. The length of this block can be between one to sixty-four bytes. The length is contained in word six of the status register. This area is read only.


[0075] Compatibility: the COM- adapter is compatible to the ATI-interface and will function with all I/O bodies which operate according to the ATI-interface as described in U.S. patent application Ser. No. 09/036,565.


[0076] MTBF quality specification: the mean time between failure (MTBF) reliability calculation model is based on MIL_HDBK-217 (a military standard). MTBF=1/ failure rate. The specified value is calculated at 30 degrees C., GB (ground benign). The MTBF goal is 200,000 hours.


[0077] Performance: the communication stack is optimized to get the best possible performance for Modbus response time (time to issue a response after receiving a Modbus TCP request). The COM-adapter will support a MODBUS/TCP transaction rate of one per millisecond, providing a response to function code 23 for a simple 32 bit (2 word) in/out module in 500 micro-seconds. A master can recover from a TCP disruption by closing and reopening socket connection. This sequence will take no longer than 5 milliseconds due to delays at the COM-adapter.


[0078] Electrical specifications:


[0079] For the ATI-interface:


[0080] Logic supply Vcc :5V/+−5%/500 ma Max supplied from the I/O body to the interface., Levels, load, and timing will be according to other ATI-interface specifications.


[0081] For the Ethernet-interface:


[0082] Compliant with the STP 100 ohm connection.
35 V Tolerance+/−5%5 V Current Consumption200 MA max.; @ <100 uF Capacitive LoadProcessorAMD186ERMemory128 K byte EPROM, 32 K byte SRAMEthernet ControllerCrystal CS8900


[0083] EMC Requirements: The COM-adapter should meet EMC tests described in the applicable standards. The COM-adapter is considered as open equipment, which means it should be within an enclosure. The following tests can be performed with shielded cable:
4TOP HAT (COM-adapter) OPENEQUIPMENTStandardTest DescriptionApplicable PortParameters/LimitsEN 55011RadiatedEnclosureClass AInterferenceEN50140/Radiated RFEnclosure80-1000 MHzIEC1000immunity10 V/m4-3EN50140Radiated RFEnclosure900 Mhbimpulse10 V/mimmunityIEC 1000-4-2ElectrostaticEnclosure8 kV AirNoteDischarge4 kV contactENV50141/Conducted RFCOM Port.15-80 MHZ1EC 1000-4-6immunity10 VrmsNote 1IEC 1000-4-4Fast TransientComm. lines1 kV cap. ClampBurstENVSurgesEarth Port2 kV CM50142/(shield)IEC1000-4-5IEC 1131Protective EarthConnector to30 Apar 4.7.2ContinuityEarth<0.1 Ohms


[0084] Communication ports pass/ fail criteria B is acceptable


[0085] The COM-adapter should meet the following agency standards:


[0086] U.L. 508, 746C, 94.


[0087] IEC 1131-2 (where applicable)


[0088] CSA22.2 No. 142


[0089] CE Mark


[0090] FM Class 1 Div. 2


[0091] The COM-adapter in operation should be kept within the following ranges:
5Temperature0-60 degrees C. operating−40-+85 degrees C. storageHumidity5-95% RH (non-condensing)Vibration10-57 Hz @ 0.075 mmd.a57-150 Hz @ 19Shock+/−15 G peak, 11 ms, half sine-wave


[0092] The COM-adapter can have 2 LEDs, and use Ethernet shielded or unshielded RJ45 female connector. Shielding should be provided on 360°, and good contacts should be provided with the external metallic parts of the RJ45 male connector. The Ethernet RJ45 pin out is:


[0093] Pin 1: TX+


[0094] Pin 2: TX−


[0095] Pin 3: RD+


[0096] Pin 4: RD−


[0097] Modbus: the COM-adapter will accept MODBUS messages over TCP/IP using the MBAP protocol to communicate with certain boards. Modbus function codes 9 (read registers), 16 (write multiple registers) and 23 (read/write) will be processed by the software (See Appendix A), which is attached hereto, and which is incorporated by reference herein, and passed to the ATI interface. Message 8, sub-function 21 (get/clear statistics), will return Ethernet statistics similar to the NOE2×1. Modbus 125 commands will be processed by the kernel for executive download. The COM-adapter will respond to all other Modbus messages with exception code 01 (illegal function).


[0098] TCP/IP: the COM-adapter will run an optimized communication stack. This stack will enable the COM-adapter to respond to Modbus messages with minimum turn around time. It must also handle other network traffic, such as ARP requests and ICMP echo requests, in a manner consistent with the associated protocols. The COM-adapter will receive its network parameters from a BOOTP type server or use those retained in nonvolatile storage, if available.


[0099] The Modbus handler will field requests from the network and either respond directly or pass the request to the ATI interface. The handler will maintain the internal configuration s and status registers, and arbitrate write access to the COM-adapter.


[0100] The TCP/IP communication stack should be optimized for performance. However, these optimizations should not impair its ability to function as a standard TCP/IP node on an integrated network.


[0101] The kernel will provide basic services for the operation of the unit. These include timer services, resource management, interrupt handling and drivers for peripherals such as Ethernet controller. Initialization and fault handling will also be handled by this code.


[0102] The present application is being filed contemporaneously with a U.S. patent application Attorney Docket No. 401 P 129, both of which are, or will be, assigned to Schneider Automation, which other application is incorporated herein by reference to the extent necessary for the understanding of the present invention.


[0103] While the specific embodiments have been illustrated and described, numerous modifications come to mind without significantly departing from the spirit of the invention and the scope of protection is only limited by the scope of the accompanying Claims.


Claims
  • 1. A method for optimizing communication on a network comprising a master device and a slave device, each device utilizing TCP and IP protocols, the method comprising the steps of: transmitting a request message from the master device, the request message having a first part; transmitting a response message from the slave device, the response message being responsive to the first part of the request message; and, limiting the request message and the response message to a length that is less than both a TCP transaction length and a maximum transmission unit.
  • 2. The method of claim 1 wherein MODBUS is utilized as an application layer protocol.
  • 3. The method of claim 2 further including: designating a set of predetermined response messages comprising at least one predetermined response message, each predetermined response message being distinguishable by the first part of the request message; and, selecting a predetermined response message in response to the first part of the request message wherein the predetermined response message is rapidly determined from the content of the first part of the request message for quickly responding to the request message.
  • 4. The method of claim 3 wherein the set of predetermined response messages comprises a response message to an address resolution protocol request message.
  • 5. The method of claim 3 wherein the set of predetermined response messages comprises a response message to an Internet control management protocol request message.
  • 6. The method of claim 3 wherein the set of predetermined response messages comprises a response message to a TCP connection request message.
  • 7. The method of claim 3 wherein the set of predetermined response messages comprises a response message to a TCP disconnect request message.
  • 8. The method of claim 3 wherein the set of predetermined response messages comprises a response message to a MODBUS request message as a TCP data frame.
  • 9. The method of claim 1 wherein the message is received exclusively on TCP port number 502.
  • 10. The method of claim 9 further including ignoring any message that is not transmitted via a TCP port number 502.
  • 11. A network communication system comprising: a master device for initiating a request message; a slave device being exclusively responsive to the request message of the master device; and, an optimal protocol utilized to communicate the request message and the response message between the master and the slave devices, the optimal protocol comprising: an IP protocol; a TCP protocol; and, an application layer protocol wherein the building and parsing of the response message is responsive to a first part of the request message.
  • 12. The network communication system of claim 11 wherein the application layer protocol is MODBUS.
  • 13. The network communication system of claim 11 wherein the response message is responsive to the content of the first part of the request message. 14. The network communication system of claim 11 wherein the master device exclusively initiates the request message.
  • 14. The network communication system of claim 11 wherein the master device exclusively initiates the request message.
  • 15. The network communication system of claim 11 further comprising a set of predetermined response messages including at least one predetermined response message, each predetermined response message being distinguishable by the first part of the request message wherein the predetermined response message is determined from the content of the first part of the request message and rapidly selected for quickly responding to the request message.
  • 16. The network communication system of claim 15 wherein the set of predetermined response messages comprises a response message to an address resolution protocol request message.
  • 17. The network communication system of claim 15 wherein the set of predetermined response messages comprises a response message to an Internet control management protocol request message.
  • 18. The network communication system of claim 15 wherein the set of predetermined response messages comprises a response message to a TCP connection request message.
  • 19. The network communication system of claim 15 wherein the set of predetermined response messages comprises a response message to a TCP disconnect request message.
  • 20. The network communication system of claim 15 wherein the set of predetermined response messages comprises a response message to a MODBUS request message as a TCP data frame.
  • 21. The network communication system of claim 11 wherein each device limits its message to a length that is less than both a TCP transaction length and a maximum transmission unit.
  • 22. The network communication system of claim 11 wherein the optimal protocol exclusively utilizes a TCP port number 502.
  • 23. The network communication system of claim 22 wherein any message not transmitted via the TCP port number 502 is ignored.
  • 24. A high performance Ethernet module comprising: an Ethernet controller operably coupled to a network connection; a control processing unit operably coupled to the Ethernet controller; and, an optimal communication stack that executes on the control processing unit, the optimal communication stack being capable of simultaneously processing a TCP protocol, an IP protocol and an application layer protocol, the simultaneous processing further including building and parsing a communication message dependent upon a predetermined index of the message.
  • 25. The Ethernet module of claim 24 wherein the application layer protocol is MODBUS.
  • 26. The Ethernet module of claim 25 wherein the communication message is limited to a length that is less than both a TCP transaction length and a maximum transmission unit.
  • 27. The Ethernet module of claim 24 wherein the optimal communication stack is configured to quickly provide a response message responsive to a request message.
  • 28. The Ethernet module of claim 27 wherein the communication message further comprises the request message having a first portion and the response message being responsive to the first portion of the request message wherein the response message is determined from the content of the first portion of the request message and rapidly selected for responding to the request message.
  • 29. The Ethernet module of claim 27 wherein the communication message is limited to a length that is less than both a TCP transaction length and a maximum transmission unit.
  • 30. The Ethernet module of claim 24 wherein the communication protocol exclusively utilizes a TCP port number 502.
  • 31. The Ethernet module of claim 24 wherein the control processing unit is operably coupled to a factory automation device.
RELATED APPLICATIONS

[0001] This patent application is a divisional application claiming priority to U.S. Patent Application entitled, “Messaging Application Layer Over Ethernet to Transport Layer (TCP) Communications Method and Apparatus for a Modular Terminal Input/Output System, application Ser. No. 09/804,362, filed Mar. 12, 2001, currently pending, which is a continuation of application Ser. No. 09/166,870, filed Oct. 6, 1998, and Granted May 15, 2001 as U.S. Pat. No. 6,223,626 in the U.S. Patent and Trademark Office.

Divisions (1)
Number Date Country
Parent 09804362 Mar 2001 US
Child 10003123 Nov 2001 US
Continuations (1)
Number Date Country
Parent 09166870 Oct 1998 US
Child 09804362 Mar 2001 US