GE 90-30 Hot Backup System

[cool341 0]

Hi All, Can anybody here tell me about the switch over time between two CPU's in a GE 90-30 Hot Backup System. What is the exact relationship of this with the Update time of the Genius Bus? According to my knowledge the Switch Over Time = 3 times the Genius Bus Update time. Can anybody validate this? Is the Switch Over between the CPU's is really bumpless? And without using any extra hardware how they implement a Hot Backup system? Any manuals supporting this will be great...... thanks

[rammin48 1]

Yes the switch over time is quite fast and indeed it is bumpless, if the Genius I/O is programmed and capable of handling a redundant system. Actuall switchover time will depend on the amount of trafic on the Genius lan and the amount of data you set up to synchronize. So there will not be any published data on this as it is system dependant GE uses a software called Max on to perform the hotback up transfer logic and data mapping. You have to purchase one copy for each project as you authorize it to the hardware. Edited 6 Jul 2006 by rammin48

[cool341 0]

Thanks Rammin for replying..... I got the answer to my question while browsing and downloading some manuals from GE Support Website. There is no such manual about 90-30 Hot backup system.... but they have on online help file, which gives a fair idea about the system..... In that help file the Failovertime is described as: Failover Time: There are two considerations for failover time. 1. Token Rotation Time: This time interval varies somewhat from one bus to another…. it depends upon the update period for the I/O LAN. Also, it is a function of the number of devices on the LAN, the quantity of input/output data associated with each device, the LAN's baudrate, and the quantity of global data being transferred. Generally the time period ranges from a few milliseconds to perhaps 100 milliseconds. 2. Mastership Time: This is the time interval for the Backup PLC to recognize that the Master PLC has failed. It takes one or two PLC scans to determine that the Master has failed. Then it takes an additional scan to activate the output data stream in the Backup PLC. Lacking output data from the current Master's GBC, each output circuit on each device on the I/O LANs will hold its last state for up to three Token Rotation Times before it assumes the Default State unless there is output data from the Backup GBC. Then the output device will begin using output data from the other GBC. The actual failover time is the longer of either…. 3 Token Rotation Times, or 1 Mastership Time. I am attaching the On line Help File also...... Thanks Rohit Singhal MAXON.zip

[semilogical 0]

this is kine of an adder to the topic: I have not used the GE 90-30. I did attend a two week school in 1995 but have not touched a system since then. We have a project from 2004 that is in Qutar. The customer is asking questions regarding our layout of the system. We have a hot backup type layout. We show two CPU's in the in the Redundant CPU Racks adn then a thrid CPUin the Remote I/O Main Rack. In a system like this what is the function of the CPU in teh Remote I/O Main Rack?

[RussB 22]

I can only guess. It was common back then to use a CPU rack with I/O in it and a Genius Bus module as a 'remote I/O' rack. In this form there would be minimal logic, if any in the CPU. It was only needed to configure the rack. Quite often these were low end CPU311, CPU313 or CPU323 racks, which would give you a full 5 or 10 I/O slots as the CPU is built into the rack.

[co_chonx 0]

Hi guys, My name is chonx from Indonesia. I want to build a redundant system using CPU374 (90-30 GEFANUC PLC series). Could I get the part number or id number for max-on software? i 've search it, in ge fanuc catalogue, but i can't find it. Thanks before.. Hopelessly, Candra

[RussB 22]

http://gefanuc.bnsi.net/eforum/showpost.ph...amp;postcount=4