US 4,964,120 A · Method of detecting a cable fault and switching to a redundant cable in a universal local area network

Abstract

A method by which the each module of a token passing local area network complying with IEEE standard 802.4 determines if the communication cable from which it is receiving signals is faulty and switches to a second, or redundant cable. Each module transmits the same signals over both cables, but can receive signals from only one, its selected cable. In normal operations, all modules receive signals from the same cable. Each of the modules has the capability of detecting faults in its selected cable, and of switching cables so that its second cable becomes its selected cable. Each module also has the capability of initiating the process or re-establising the network whenever a module switches cables in response to detecting a fault. Each module in response to the process of initiating re-establishing the network, switching cables so that all modules are listening to, or receiving signals from the same cable. Each module periodically checks to determine if low level noise is present on its selected cable, and if it is, the module switches cables. A module in the listen only mode of operation will periodically check to determine if the signals it is receiving or valid. If not, it will switch cables. A module whose redundant cable has been identified as being silent, will periodically check to determine if its redundant cable remains silent.

Claims15

Dependent claims

The method by which each active module of a token passing local area network which satisfies IEEE Standard 802.4 communicates over a pair of cables, with each module transmitting the same signals over both cables and receiving signals transmitted over the cables from only one of the cables at a given time, the cable from which a module receives signals at any given time being designated as that modules selected cable and the second cable being its redundant cable, with the selected cable of all modules being the same cable when the network is operating properly;

each module has the capability of detecting a fault in its selected cable, of switching to its second cable and of initiating the re-establishment the network with the second cable as the selected cable of the modules of the network;
said method comprising the steps of:
checking to determine if a module's selected cable has a fault;
and
a module detecting a fault in its selected cable switching cables to receive signals from its second cable;
the other modules in response to a module of the network initiating re-establishing the network;
switching to their second cable, so that the second cable becomes the selected cable of the modules of the network.

4 more elements

In a token passing local area network having a plurality of modules which network and modules satisfy IEEE Standard 802.4 and which modules communicate over a pair of cables, with each module transmitting the same signals over both cables and receiving signals transmitted over the cables from only one of the cables at a given time, the cable from which a module receives signals at any given time being designated as that modules selected cable and the other, or second, cable being the redundant cable, with each module including a token bus controller (TBC) that produces interrupts when certain conditions are detected;

the improvements comprising the method by which interrupts produced by a TBC are serviced;
said method comprising the steps of:
1. in response to a TBC producing a bus idle time-out interrupt;
a, checking to see if the redundant cable has a fault;
b, if not, switching cables and checking to see if valid signals have been received from the second cable, and
c, if valid signals have been received, returning from the interrupt;
2. in response to a TBC producing a noise/frame fragment counter interrupt;
a, switching to the redundant cable ;
and
b, returning from the interrupt;
3, in response to a TBC producing a successor change interrupt, a claim token interrupt, a win address sort interrupt, or a lost address sort interrupt;
a, switching to its redundant cable;
and
b, returning from the interrupt.

11 more elements

A method by which each module of a token passing LAN detects a fault in a first cable from which each module receives signals, each modules selected cable, and switches to a second redundant cable from which to receive signals, and reestablishes the LAN with all modules receiving signals from the redundant cable, an active module both transmitting and receiving signals and a passive module only receiving signals;

by executing a plurality of subroutines comprising the steps of:
subroutine 1,
(a) switching to the second cable if no signals are received from the first cable for a first period of time;
(b) testing the second cable to determine if valid sets of signals are being received from the second cable for a second period of time;
(c) terminating subroutine 1 if valid sets of signals are received in step b;
(d) repeating steps a and b if no valid sets of signals are received in step b until terminated by step c or until a third period of time has elapsed;
(e) terminating subroutine 1 when the third period of time has elapsed;
(f) testing to determine if any signals have been received during said first period of time after step e is completed;
(g) if valid sets of signals are received in step f, continuing to receive signals from the same cable for a fifth period of time and terminating subroutine 1;
(h) if no signals are received in step f, testing to determine if any signals have been received during the most recent fifth period;
(i) if signals were received during the most recent fifth period, terminating subroutine 1;
(j) if no signals were received during the most recent fifth period, switching cables and terminating subroutine 1;
subroutine 2,
(k) determining if the cable from which a module is receiving signals is noisy;
(l) switching cables if the cable is determined to be noisy;
and
(m) terminating subroutine 2 after completing step 1;
subroutine 3,
(n) switching to the second cable any time a module begins the process of re-establishing the LAN;
and
(o) terminating subroutine 3;
and
subroutine 4
(p) each active module periodically checking the cable from which it is receiving signals for low level noise;
(q) terminating subroutine 4 if the noise level of its selected cable is acceptable;
(r) switching to the redundant cable if the noise level of the received signals exceeds a predetermined threshold;
and
(s) terminating subroutine 4.

25 more elements

Dependencies are inferred from the claim text.

Method of detecting a cable fault and switching to a redundant cable in a universal local area network

Life of this patent

Method of detecting a cable fault and switching to a redundant cable in a universal local area network

Life of this patent

Abstract

Claims15