CN102026050A - Optical fiber CAN bus hub and networking method thereof - Google Patents

Abstract

The networking method of the optical fiber CAN bus hub is to combine the main hub and the main hub of the optical fiber CAN bus with n optical ports and a local twisted pair electric port expansion circuit with n optical nodes or n slave hubs or n remote ends The n twisted-pair CAN subnets connected to the electrical port expansion device and one local twisted-pair CAN subnet are connected to form a network. A fiber-optic CAN bus hub with twisted-pair electric port expansion circuit and dual CAN transceivers and two pairs of twisted-pair wires to separate signal transmission and reception, avoiding self-excited blocking communication caused by signal transmission loops; fiber optic CAN bus hub, fiber optic pair and The remote electrical port expansion device connects the optical node, the local twisted pair CAN bus subnet and the remote twisted pair CAN bus subnet together to form a high-speed wide-area networking solution.

Description

Optical fiber CAN bus hub and its networking method

technical field

The invention belongs to a CAN bus network, in particular to a CAN bus hub whose transmission medium is an optical fiber and a wide-area high-speed networking method based on the hub for connecting an optical fiber CAN node and a twisted-pair CAN bus subnet together.

Background technique

CAN (Control Area Network) bus is a field bus network that uses a non-destructive bit contention mechanism to realize serial multi-master communication. CAN bus has the advantages of strong anti-interference ability, short message, good real-time performance and low networking cost. It is especially suitable for fields such as industrial automation sites and auto parts control with harsh environments, large electromagnetic radiation, and high reliability requirements. The most commonly used transmission medium of the CAN bus is a twisted pair, and the highest rate achievable on this medium is 1Mbps. In actual use, the available communication rate of the twisted-pair CAN bus is lower than 1Mbps. The main reasons for the decrease in communication rate are as follows: (1) There are too many nodes connected to the bus, resulting in bus parameter mismatch; (2) The overvoltage and overcurrent protection circuit installed on the node causes bus parameter mismatch; (3) The geographical distribution of the nodes is too large, and the distribution parameters of the too long cable cause the mismatch of the bus parameters.

When the working environment of the bus is too harsh, such as when the common-mode potential difference between nodes exceeds the maximum value of the physical layer transceiver device, the bus will not work normally. Therefore, it is of great significance to study the use of optical fiber as the transmission medium, take advantage of its high transmission rate and strong anti-interference ability, study its networking mode, and solve the large-capacity, high-speed and long-distance networking problems of the CAN bus.

The twisted-pair CAN bus network is composed of CAN controllers in each node connected to twisted-pair wires through CAN transceivers. When multiple nodes compete for the bus, the "recessive" or "dominant" bit signal sent by a certain node, and the signals sent by other nodes realize logic "wired AND" on the bus, and send the result back to the sending node and Each receiving node implements non-destructive bit contention. Since the twisted pair does not limit the transmission direction of the signal, as long as every bit of data appearing on the bus can be correctly received by all nodes on the bus (including the sending node) within a specified time, the communication can proceed normally.

Optical fiber is a medium for one-way transmission of optical signals. At present, there is no one that can achieve "line-AND" superposition of optical signals when one or more nodes send out (or not send out) optical signals, so that all nodes are equal to each other. An optical transceiver device capable of receiving superimposed optical signals. Therefore, optical fiber cannot be directly used for CAN bus networking. When using optical fiber networking, the original CAN controller is generally retained, the CAN transceiver and twisted pair are discarded, and the physical layer of the network is redesigned. How to connect the optical fiber physical layer with the traditional twisted pair physical layer CAN network is also a problem to be solved.

At present, several implementation schemes have been proposed about the networking of optical fiber CAN bus. Among them: (1) the ring optical fiber CAN bus network scheme allows a very limited number of nodes (only optical fiber nodes) in the access ring, and with the increase of the number of access nodes, the communication rate drops sharply; (2 ) The optical fiber CAN bus self-healing ring network scheme also has similar problems; (3) The dual-star optical fiber CAN bus scheme allows more nodes to be connected, but like the previous two schemes, there is a signal transmission loop self-excitation that causes network congestion and inconvenient interfacing with twisted pair CAN bus subnetworks.

Contents of the invention

The purpose of the present invention is to provide an optical fiber CAN bus hub and a networking method thereof on the basis of the prior art.

Technical scheme of the present invention is:

A networking method of a fiber optic CAN bus hub is provided, which is characterized in that: the fiber optic CAN bus hub with n optical ports and a local twisted pair electrical port extension circuit and other n-1 optical nodes, local dual Twisted wire subnet, remote electrical port expansion device and remote twisted pair subnet are connected and communicated to form a network.

For example, the optical fiber CAN bus hub 1 with n optical ports and a local twisted pair electric port expansion circuit shown in Figure 1 and its connection with each optical node (comprising optical node 2), local twisted pair subnet 4 and remote A networking method of the twisted pair subnet 7.

Each optical port on the hub 1 has a pair of photoelectric conversion module PIN and electro-optical conversion module LED. Taking optical port #1 as an example, PIN12 and LED13 are respectively connected to LED23 and PIN22 in optical node 2 through optical fiber pair 3. The local electrical port expansion circuit 14 of the hub 1 can be connected to the local twisted pair subnet 4 through two pairs of twisted pairs. Each optical port (take n# optical port as an example) can also be connected to the far-end electrical port expansion device 6 through the long optical fiber pair 5, and then connected to the remote twisted-pair wire through two pairs of twisted pairs 7 on the net.

The optical node 2 as shown in FIG. 1 includes a CAN bus controller 21, its input pin RXD is connected to the output end of PIN22, and its output pin TXD is connected to the input end of LED23.

A schematic block diagram of a fiber optic CAN bus hub 1 provided is shown in Figure 2. The hub 1 has a built-in complex programmable device CPLD11, n photoelectric conversion modules PINs forming n optical ports, n electro-optical conversion modules LEDs and local electronics. Port expansion circuit 14. The outputs of the n PINs are connected to the input RX(1)...RX(n) of the CPLD11; the output of the local electric port expansion circuit 14 is connected to the input pin RX(n+1) of the CPLD11. The output TX(1)...TX(n) of the CPLD11 is connected to the input of each LED; the output TX(n+1) is connected to the input of the local electrical port expansion circuit 14 .

The local electrical port expansion circuit 14 is shown in FIG. 2 , which includes CAN transceivers 143 , 144 , and optical compartments 141 and 142 . The bus pins CANH1 and CANL1 of the transceiver 143 and the bus pins CANH2 and CANL2 of the transceiver 144 are respectively connected to the twisted pair 41 and the twisted pair 42 in the local twisted pair subnet 4 through two pairs of twisted pairs. The input pin TXD1 of the transceiver 143 is connected to a high level "1"; the output pin RXD1 is connected to the input pin of the optical isolation 141, and the output of the optical isolation 141 is connected to the input pin RX(n+1) of the CPLD11. The output pin RXD2 of the transceiver 144 is suspended; the input pin TXD2 is connected to the output pin of the optical barrier 142, and the input of the optical barrier 142 is connected to the output pin TX(n+1) of the CPLD11.

Described local twisted pair subnet 4 is shown in Figure 3, and it comprises node 43 and other multiple electrical port CAN nodes, two pairs of twisted pairs 41 and 42 and four 120Ω termination resistors. Taking the electrical node 43 as an example, it includes a CAN controller 435 , two CAN transceivers 431 and 432 , and two optical barriers 433 and 434 . The input pin TXD1 of the transceiver 431 is connected to the output pin TXD of the CAN controller 435 through the optical barrier 433 , the output pin RXD1 is suspended, and the CAN bus pins CANH1 and CANL1 are connected to the twisted pair 41 . The output pin RXD2 of the transceiver 432 is connected to the input pin RXD of the CAN controller 435 through the optical barrier 434, the input pin TXD2 is connected to a high level "1", and the CAN bus pins CANH2 and CANL2 are connected to the twisted pair 42. The twisted pairs 41 and 42 are respectively connected to two 120Ω terminal resistors. The methods for connecting other electrical nodes to the twisted pairs 41 and 42 are the same.

The remote electrical port extension device 6 is shown in FIG. 4 , which includes CAN transceivers 63 and 64 , a photoelectric conversion module 61 and an electro-optical conversion module 62 . The input pin TXD2 of the transceiver 64 is connected to the long fiber pair 5 through PIN61, the output pin RXD2 is suspended, and the CAN bus pins CANH2 and CANL2 are connected to the twisted pair 72. The output pin RXD1 of the transceiver 63 is connected to the long optical fiber pair 5 through the LED 62 , the input pin TXD1 is connected to a high level “1”, and the CAN bus pins CANH1 and CANL1 are connected to the twisted pair 71 .

The remote twisted-pair subnet 7 is shown in FIG. 4 , and its circuit structure and connection method are exactly the same as those of the local twisted-pair subnet.

The present invention also proposes a cascading networking method of an optical fiber hub as shown in Figure 6, which consists of a master hub 81 with n optical ports working in a master mode and a slave working in a slave mode with n optical ports The hubs 82 are cascaded. Each optical port of the main hub 81 can be connected to an optical node or a slave hub, while only one optical port of the slave hub 82 can be connected to the other optical ports of the main hub to connect to an optical node. For example, the n# optical port of the master hub 81 is connected to the 1# optical port of the slave hub 82 through the optical fiber pair 85 . Other optical ports of the main hub 81 and the slave hub 82 are respectively connected to n-1 optical nodes and connected to a local twisted-pair CAN subnet 83 and 84 respectively.

Technical effect of the present invention is embodied in:

1. The optical fiber CAN bus hub that uses programmable logic device CPLD to realize "wire-and" logic connects multiple optical CAN nodes together through optical fiber pairs, and can realize a high-speed CAN network that fully complies with the CAN bus (above the physical layer) specification.

2. Two sets of CAN transceivers and two pairs of twisted-pair wires separated by sending and receiving are used to solve the problem of self-excitation and blocking communication in the signal transmission loop existing in the optical fiber network in the past.

3. Using fiber optic CAN bus hubs, long fiber pairs, remote electrical port expansion devices and hub cascades to connect multiple optical nodes, local twisted pair CAN bus subnets and remote twisted pair CAN bus subnets together, Realize the wide-area CAN bus high-speed network that fully complies with the CAN bus specification.

Description of drawings

Figure 1 is based on the networking diagram of the fiber optic CAN bus hub;

Fig. 2 block diagram of optical fiber CAN bus hub;

Fig. 3 local twisted pair CAN bus subnet principle block diagram;

Figure 4 The functional block diagram of the remote electrical port expansion device and the remote twisted pair CAN bus subnet;

Figure 5 PCA82C250 CAN bus transceiver internal circuit block diagram;

The cascade schematic diagram of Fig. 6 optical fiber CAN bus hub;

Detailed ways

The present invention will be described in detail below in conjunction with accompanying drawings and examples.

In the wide-area CAN bus network composed of optical nodes, local electrical nodes and remote electrical nodes shown in Figure 1, the "recessive" bit and "dominant" bit sent by the node are represented as high level and high level respectively on the circuit. low level. When transmitting in the optical fiber medium, it should be designed so that there is optical transmission corresponding to the "dominant" bit, and no optical transmission (or when not connected to the optical fiber) corresponds to the "recessive" bit.

The hub 1 shown in Fig. 2 works in the main mode, and the logic in the programmable logic device CPLD11 wherein is designed according to following n+1 equation groups:

TX(1), TX(2), ..., TX(n), TX(n+1) =

RX(1)&RX(2)&...&RX(n)&RX(n+1) (1)

Its meaning is to "AND" all the n+1 signals RX(1), RX(2)...RX(n), RX(n+1) input to CPLD11, and then send them back to n+ 1 output TX(1), TX(2)...TX(n), TX(n+1). The "wire AND" function of the twisted pair is realized by using CPLD logic "AND". When the i-th optical port on the hub is unused, the corresponding PIN has no optical signal input, and the input RX(i)=1 of CPLD11 is "recessive" level, so it will not affect the "and" of equation group (1). "Effect.

In the local twisted-pair CAN bus subnet 4 shown in FIG. 3 , each node adopts two sets of CAN transceivers and two pairs of twisted-pair wires separated for sending and receiving. The electrical node 43 is taken as an example to illustrate the interconnection relationship between the electrical node and the optical node. The bit signal ("recessive" bit or "dominant" bit) that CAN controller 435 sends from TXD pin passes through optical barrier 433, and CAN transceiver 431 arrives and sends twisted pair 41, first with the signal that other nodes send in this subnet After "line and", through the CAN transceiver 143, the optical barrier 141 enters the CPLD11 from the RX (n+1) pin, and after "ANDing" with the bit signals of other optical nodes according to the equation group (1), from the TX (n +1) output, the CAN transceiver 144 reaches the receiving twisted pair 42 through the optical barrier 142 . The bit signal returns to the RXD input pin of the CAN receiver 435 through the CAN transceiver 432 and the optical isolation 434; the bit signal is also received by other nodes in the subnet at the same time.

Similarly, in the far-end twisted-pair CAN bus subnet 7 shown in Figure 4, the bit signals sent by each electrical node in the subnet first realize the "wire AND" on the sending twisted-pair 71, and then pass through the far-end The electrical port extension device 6 and the long fiber pair 5 are coupled and enter the CPLD11 from the RX(n) pin, and output from the TX(n) after "ANDing" with other signals, and then couple through the remote electrical port expansion device 6 and the long fiber pair 5 Arriving at the receiving twisted pair 72, it is received by all nodes in the subnet.

From the above analysis, it can be seen that the "recessive" bit or "dominant" bit signal sent from the physical layer by both the optical node and the electrical node is all processed in the CPLD11 after the "AND" logic operation, and then sent back to each node, which conforms to the CAN bus standard Due to the requirements of the physical layer signal transmission characteristics, it can ensure that each node in the network realizes the unique multi-master of CAN bus, and the communication without destructive bit competition.

When the local electrical port expansion circuit 14 shown in Fig. 2 was not connected to the local twisted pair subnet 4, the output RXD1 of the CAN transceiver 143 was "recessive" high level, and the CPLD11 of the RX (n+1) pin input hub Participate in the "and" operation, so it will not affect the "and" effect of the equation group (1).

The purpose of using two sets of CAN bus transceivers and two pairs of twisted pairs in the electrical port is to eliminate the signal transmission loop and avoid self-excited blocking communication. Now take the local electrical port expansion circuit 14 in FIG. 2 as an example for illustration. Fig. 5 is the internal structure diagram of CAN bus transceiver PCA82C250, it can be seen that the transceiver has the function of self-sending and self-receiving, if only one CAN transceiver 143 or 144 is used, the signal sent by CPLD11 from TX(n+1) will pass through the optical Interval 142, CAN transceiver 144 or 143, optical isolation 141 returns from RX(n+1) to CPLD11. TX(n+1) and RX(n+1) appear at both ends of equation group (1), so a signal loop is generated.

As shown in FIG. 2, this problem can be solved by adopting dual CAN transceivers 143 and 144 and two pairs of twisted pairs. If only one twisted pair is used, a signal loop will still be formed. If two pairs of twisted-pair wires are used, two sets of independent CAN bus transceivers must be used in each electrical node.

In the cascading mode shown in FIG. 6 , the main hub 81 works in the main mode, and the CPLD therein is still designed and implemented according to n+1 equation group (1). The slave hub 82 works in a slave mode, wherein the CPLD is designed and implemented according to the following single equation (2) and n equation groups (3).

TX(1)＝RX(2)&...&RX(n)&RX(n+1) (2)

TX(2),...,TX(n), TX(n+1)=RX(1) (3)

Since the 1# optical port is an upward cascading port from the hub 82, the meaning of the above equation is to input n signals RX(2)...RX(n), RX(n+1) from the hub 82 (not including RX (1)) after all phases "AND", send it to the main hub 81 by TX (1), and return to the slave hub 82 from RX (1) after all signals in 81 are phased "AND". Then forward to n output TX(2)...TX(n), TX(n+1). The purpose of this design is to avoid signal transmission loops between the master and slave hubs.

As long as the logic of the master and slave hubs is implemented according to (1), (2) and (3), the cascading of hubs can have many stages. With the increase of the number of cascades, the number of nodes that can be accessed is very large, but the delay of signal transmission increases, and the highest achievable CAN bus rate decreases accordingly.

In the wide-area CAN bus network formed by this method, the scale of each sub-network composed of electrical nodes should not be too large, so there is no problem of communication rate drop caused by too many bus nodes and too long cables, so the maximum communication rate of the network is basically It is only limited by the transmission delay of each device and the delay of the signal in the long fiber pair. This maximum rate is much higher than the maximum rate of the wide-area CAN bus network composed of fully twisted pairs.

The optical fiber pair 3 and the long optical fiber pair 5 etc. can adopt single-mode or multi-mode optical fiber.

The photoelectric conversion modules PIN12, PIN22, PIN15 and PIN61 can use any conversion module with a communication rate that meets the requirements, such as HFBR-2412 of HP Company.

The electro-optical conversion modules LED13, LED123, LED116 and LED162 can use any conversion module with a communication rate that meets the requirements, such as HFBR-1414 of HP Company.

The programmable logic device CPLD11 adopts a CPLD with sufficient logic resources produced by any company, such as XC9572.

The CAN bus controllers 21, 435 and 731 etc. adopt any chip conforming to the CAN bus specification, such as SJA1000.

The CAN bus transceivers 143, 144, 341, 342, 734 and 735 etc. adopt any chip conforming to the CAN bus specification, such as PCA82C250.

The optical isolators 141, 142, 433, 434, 732 and 733 etc. adopt any optical isolator chip with sufficiently small transmission delay, such as 6N137.

Claims (7)

1. the network-building method of optical fiber CAN bus hub is characterized in that: will have n twisted-pair feeder CAN subnet that the primary hub of optical fiber CAN bus of a n Guang Kou and a local twisted-pair feeder electricity port expansion circuit and primary hub and n optical node or n be connected from hub or n the electric mouth of far-end expanding unit and be connected with 1 local twisted-pair feeder CAN subnet and constitute network.

2. the network-building method of optical fiber CAN bus hub as claimed in claim 1 is characterized in that: each the light mouth on the described hub all has a pair of photoelectric conversion module PIN and electrooptic conversion module LED; The local electric port expansion circuit of hub links to each other with local twisted-pair feeder subnet by two pairs of twisted-pair feeders; Described each light mouth also can be connected on the far-end twisted-pair feeder subnet by two pairs of twisted-pair feeders by long optical fibers pair with after a far-end electricity mouthful expanding unit that is positioned at a distant place links to each other again.

3. the network-building method of optical fiber CAN bus hub as claimed in claim 1 or 2, it is characterized in that: described optical node, comprise a CAN bus control unit, CAN bus control unit input pin RXD connects the output of the photoelectric conversion module PIN that is associated, and CAN bus control unit output pin TXD connects the input of the electrooptic conversion module LED that is associated.

4. optical fiber CAN bus hub is characterized in that: its built-in complex programmable device CPLD11, constitute n photoelectric conversion module PIN, n the electrooptic conversion module LED and the local electric port expansion circuit of n light mouth.The input pin RX (1) of the output termination CPLD11 of n photoelectric conversion module PIN ... RX (n); The input pin RX (n+1) of the output termination CPLD11 of local electric port expansion circuit 14, the output pin TX (1) of CPLD11 ... TX (n) connects the input of each electrooptic conversion module LED; The output pin TX (n+1) of CPLD11 connects the input of local electric port expansion circuit.

5. the network-building method of optical fiber CAN bus hub as claimed in claim 1 is characterized in that: the electric port expansion circuit in described this locality, and it comprises that two CAN transceivers splitting and corresponding two light are every linking to each other;

6. the network-building method of optical fiber CAN bus hub as claimed in claim 1, it is characterized in that: described far-end electricity mouthful expanding unit, it comprises two CAN transceivers, photoelectric conversion module PIN and electrooptic conversion module LED, the input pin of one of them transceiver by photoelectric conversion module and long optical fibers to linking to each other, the output pin of transceiver is unsettled, and two pin of CAN bus connect on the twisted-pair feeder.To linking to each other, the input pin of another transceiver connects high level " 1 " to the output pin of another transceiver by electrooptic conversion module and long optical fibers, and two pin of CAN bus connect on another twisted-pair feeder.

7. the network-building method of optical fiber CAN bus hub as claimed in claim 1 is characterized in that: described primary hub by optical fiber pair with link to each other from hub, from hub with link to each other from hub and constitute network.

Images