SLC500 Power Up Fault

Hi all
At work we have had some problems with SLC500's faulting out on power up, after being isolated for 6 or more hours. To get round this we decided to fit permanent live PLC supplies to a group of 3 control panels containing 5 PLCs between them (all on one ethernet network). Two of the panels containing two PLCs each have been done without any problems.
When the 3rd panel with one SLC500 in it was done we found that when the panel isolator was switched on the PLC would fault out even though it had not been powered down with the isolator.
Details of the PLC are

Slot0 5/05 CPU - 64K. OS501 /C
Slot1 Empty
Slot2 1746-IB32
Slot3 1746-IB32
Slot4 1746-OB32E
Slot5 1746-NI04V
Slot6 Empty

This PLC was NOT one of those that had been faulting out after power down, but it is part of the same system.
Unfortunately I do not have a fault code for the problem but I could recreate the issue to get one.
Has anyone come across this issue before?
any help is very welcome.
cheers
Ian

ian.smith7 said:

Hi all
At work we have had some problems with SLC500's faulting out on power up, after being isolated for 6 or more hours. To get round this we decided to fit permanent live PLC supplies to a group of 3 control panels containing 5 PLCs between them (all on one ethernet network). Two of the panels containing two PLCs each have been done without any problems.
When the 3rd panel with one SLC500 in it was done we found that when the panel isolator was switched on the PLC would fault out even though it had not been powered down with the isolator.
Details of the PLC are

Slot0 5/05 CPU - 64K. OS501 /C
Slot1 Empty
Slot2 1746-IB32
Slot3 1746-IB32
Slot4 1746-OB32E
Slot5 1746-NI04V
Slot6 Empty

This PLC was NOT one of those that had been faulting out after power down, but it is part of the same system.
Unfortunately I do not have a fault code for the problem but I could recreate the issue to get one.
Has anyone come across this issue before?
any help is very welcome.
cheers
Ian

Click to expand...

Get the fault code, but if I had to guess, you have a problem with your logic and are getting a divide by zero or math overflow on the first scan of the processor. Then, perhaps conditions are such that subsequent scans, the math error no longer occurs.

Hi Robert
I was on night shift after my post yesterday and during a short shutdown I had a chance to recreate the fault.
It was a Math overflow error so I have programmed an overflow trap just before the end statement. This seems to have cured the problem.
Thanks for your reply.
Ian

Technically, no that hasn't cured the problem. It has only masked the problem so you don't see it.

You really need to find and fix the cause of the overflow.

I am posting this here because it is relavant to the math overflow bit, and it might even being directly related to the issue related by ian.smith7.

After installing a 1746-NI8 input card in a 5/03, and scaling the input, 4000 - 20000 to 0 to 25000, (this is for conductivity), I had the same issue.

Rockwells knowledgebase stated that it is fine to unlatch the overflow bit if you are certain doing so is not an issue. Further research into programs,(not mine), using this card and scaling, I see that in virtually every case that S:5/0 is unlatched.

So I would ask...

1) What is it about the scaling that causes this issue? (I am assuming it is the scaling, and has nothing to do with the card itself.)

2) Is there a better solution?

I am absolutely certain the issue is from the scaling, and/or the input card.

This was a while ago, but if memory serves me, the sensor is powered from an analyser, and the input is in series with a passive chart input as well. Is it possible that the problem arises when the SLC powers up and connects what was an incomplete circuit, that it sees a spike, or dip outside the allowable range? I do not remember if I tried unlatching the bit manually after power up or not. This seems as though it might be a plausable explanation.