GB2288709A - Fault correction - Google Patents

Abstract

A fault correction arrangement for shelf based cards 2 in a data transmission system is disclosed in which each card includes a non volatile memory within which faults can be recorded. The non volatile memory is isolated from the remaining card circuitry so as to allow interrogation of the NVM if the remaining card circuitry is faulty. <IMAGE>

Description

FAULT CORRECTION SYSTEM FOR SHELF BASED CARDS This invention relates to a fault correction system for shelf based cards and is particularly applicable to systems operating with digital data transmission equipment, such as within a telecommunication system. The invention need, however, not be limited to such equipment and can be applied to any equipment which uses removable printed circuit boards. In electrical communication equipment and the like the operation of the equipment usually relies on a stack of printed circuit boards which are contained in shelves of a frame or rack. The printed circuit boards themselves are able to be slid in and out of the shelves and when in position, contacts on the boards make contact with corresponding connections at the back of the shelves. By interconnecting the various contacts the correct assembly of functions required can be achieved.

As with any equipment, faults do from time to time occur and it is necessary to be able to recognise and quickly correct for these faults. Normally once a fault has been identified, the card carrying the fault is removed and a new card is inserted. The old card can then be repaired remote from the equipment and used as a spare or reinserted in the equipment at a convenient time. In tie specification of our British patent 2245801 there is described a fault location system in which a separate data recording medium is included for recording any faults which occur.

This means that the medium can be removed from the equipment, the fault identified and the correct PCB or other faulty piece of equipment replaced.

It is an object of the present invention to simplify such a system and to give a reliable fault correction system which correctly identifies in respect of each printed circuit board the fault that occurs with it.

In accordance with the present invention, a fault correction arrangement for shelf based cards in a data transmission system is provided, in which each card includes a non volatile memory (NVM) within which faults can be recorded, the NVM being isolated from the remaining card circuitry so as to allow interrogation of the NVM if the remaining circuitry is faulty.

The NVM is to be construed as including a battery backed random access memory (RAM) or similar device, such as a smart card. The fact that the NVM is isolated from the remaining card circuitry allows independent operation from the rest of the card, and can allow interrogation of the memory to reveal the fault even in the event of severe card failure.

Preferably an independent interface is provided for connecting the NVM to an external controller, such as a control card within the shelf based system identified above.

In order that the invention may be well understood, an embodiment thereof will now be described by way of example with reference to the accompanying figure which shows schematically a shelf within a digital telecommunication network.

Referring to the figure the shelf 1 includes a plurality of equipment cards 2, only one of which is shown. Each equipment card 2 may be operative to perform any one of a number of different tasks and includes card control circuitry 3 and card function circuitry 4 interconnected by a link 5. In addition and according to the invention, a non volatile memory NVM 6 is provided within which faults occuring on the respective card 2 can be recorded. The NVM 6 is isolated from the remaining circuitry on the card to the extent that it performs no other function other than that discussed above. In a specific example the NVM 6 is an EEPROM, based on the Philips 12C bus. The memory 6 has its own independent access port 7 by which data can be read into and out of the memory from an external controller, such as a shelf control card 8.Another link 9 connects the shelf control card 8 to the card control circuitry 3 while a link 10 connects the shelf control card 8 to a management system (not shown).

In use fault data derived as a result of the equipments self diagnostic functions is passed from the card control circuitry 3 across the link 9 to the shelf control card 8. Card 8 writes fault data across the link 7 into its respective NVM 6. This data can include the time of the fault, based on the control functions real time clock.

By using the memory on the card itself, there is no risk that when the card is removed it is incorrectly indexed or tagged to show the fault since the fault can always be checked by interrogating the memory. The fault can then be quickly repaired and the card replaced.

It is preferable that the memory is used to log any previous faults which have been located in the card and corrected. This rqn tZ incorporated with the date, the identity of the repairer and any other relevant information. Furthermore the memory device in the card can be used to identify the site from which the card was removed so that it is returned to the correct location and there is no problem arising by the loss or elegibility of address labels and the like.

As shown the memory device is supplied with power common to the whole equipment.

In an alternative embodiment a local power source may be provided which would allow faults caused by failure of power supplies to be recorded in the memory.

It will be appreciated that by incorporating as much information as possible on the card as soon as a fault occurs, then the chance of inaccurate repair or the use of unsuitable cards will be minimised. Furthermore since each card carries its own history a respective card can be compared with other cards with similar histories so that any trend of faults may be analysed and recognised.

The memory 6 may also be used for the storage of card data or "inventory". This may include the build date, site of manufacture, build state, card type, serial number and the like. This allows remote interrogation from the external management system in order to establish the status of card build state etc in the field. It may also allow tracking down of a lost card based on its serial number, as well as identification of particular cards that may require a software or hardware update.

Claims (6)

1. A fault correction arrangement for shelf based cards in a data transmission system, in which each card includes a non volatile memory (NVM) within which faults can be recorded, the NVM being isolated from the remaining card circuitry so as to allow interrogation of the NVM even if the remaining circuitry is faulty.

2. A fault correction arrangement according to claim 1 in which an independent link is provided to allow the NVM to be accessed.

3. A fault correction arrangement according to any preceding claim comprising an external controller, a card being operative to transmit data indicative of detected faults on the respective card to the external controller, the external controller being operative to write data indicative of the fault into the NVM.

4. A fault correction arrangement according to any preceding claim for shelf based cards in a digital telecommunication system.

5. A fault correction arrangement according to any preceding claim in which the NVM is connected to the cards power supply.

6. A fault correction arrangement substantially as described with reference to the accompanying figure.