PDA

View Full Version : CPU Fault SLC 5/03


techimnot
July 15th, 2009, 08:59 PM
We are running an AB SLC 5/03. The machine will run for awhile and then stops due to a CPU Fault. I can usually clear it right away with the RSLogix, but it eventually comes back. When I "go to error" from the drop down the error message reads
"Remote power failure of an expansion I/O occurred"
Would this be coming from an I/O card, or as I suspect, the power supply or CPU itself? Any help is appreciated.

geniusintraining
July 15th, 2009, 09:22 PM
From AB if it is a 21h error

Error code 21H indicates that one of the modules in the slc 500 chassis is faulting.
The exact description is "A remote power failure of an expansion I/O chassis has occurred.
The customer should download a blank program and force the program into run mode.
Adding one module at a time to the I/O and placing the program in run mode.

I would say its a card... but just a guess, the above would take a long time to figure out in your case, I would power down and re-seat all of them just to make sure its not a bad connection

Ken Roach
July 15th, 2009, 10:12 PM
The controller is detecting a power failure in an extended I/O chassis.

The main chassis in which the CPU resides can be cabled to up to two additional chassis. It's the power supply in one of those remote chassis that's failing.

It could be a bad power supply, but it's more likely a loose power connection.

techimnot
July 15th, 2009, 10:46 PM
There is no remote chasis. What I have is the power supply, CPU, Triac AC output, DC source output, 115v input, and a DC sink input. And empty slots.

Would these cards be directly replaceable with a like card. Or would I need to get into the configuration. Been about ten years since I've done this.


Any help GREATLY appreciated.

Ken Roach
July 15th, 2009, 11:07 PM
OK, then the SLC is mis-identifying a major backplane failure.

I've seen this caused by spikes that come in from inductive loads (usually relays, sometimes ordinary AC outputs) and from damaged backplanes. There's an active chip on the backplane that can get damaged and fail progressively. It's not common, but it happens.

Another cause can be contamination on the PC board of the backplane bus itself, or in the connector that would plug into an expansion chassis.

The last two most common things are a failing power supply or a failing CPU.

It is unlikely to be a failing I/O module. The SLC can usually identify a failing I/O module by slot number, and if it can't identify the slot number it blames the last configured slot. But an individual module failure shouldn't be misinterpreted as a chassis failure. I can't say its impossible, but I have not seen it.

techimnot
July 16th, 2009, 09:20 AM
Thanks Ken and Genius. It is a 21h error and I will take your facts and ideas with me to work later. I will clean up the slots and re-seat the cards again. I have a used power supply on the shelf and will start there. Also have another 5/03 CPU. Hopefully won't have to go there. Will repost after applying your comments and suggestions.
Thanks,
Jim

techimnot
July 18th, 2009, 05:36 AM
OK, cleaned up the 7-slot(dusty), faulted--replaced Power Supply, had a dark almost burned area on a board, no joy. Replaced the 5/03 and have been running for 30 hours. I believe it's reasonably safe to say that was it. Thanks for all the help. Next up, MicroLogix 1000, but that's a new thread later today. Thanks again!:dance: