Modicon/Unity register Monitoring

Hello all,

Last week our main PLC had a system halt (EC0E) and since then, one internal register(mapped to an address) used in our CIP cycle has been acting strange. it's being written to somewhere in the program but I cannot find it anywhere.

The controller is a quantum (140 CPU 652 60).

I was wondering if there's a way to monitor a register in such a way it shows what/where it's being written from. I can watch it change in an animation table and it's almost predictable.

It's a word register that is used only once, it's in an array of words(think drum sequencer).

I've tried looking it up by every address I could think of as well as it's array and even it's index in the array. Has anybody come across this or does anybody have an idea for a solution?

Thank you in advance as always with any reply.
Phil

Do you have the Overlapping addressing warning set under build messages in the project settings? that should give you an idea if you have any overlapping addresses.

also double check the addresses aren't being used in the IO Scanner, io scanner health block and io scanner device control block. they aren't listed in the elementary variables window so its a common mistake.

could it be your scada/hmi writing a value to it? try changing the address?

were any changes made just before the halt? or was it random?

Hi Brendonh,

Thank you for the reply, I've triple checked the HMI and Scada systems that are connected to it. nothing there

Nothing in the IO Scanner

I did make a change just prior to the halt, I had imported a DFB, built changes. Then attempted to implement DFB (the DFB didn't output anywhere, I was testing some scaling from an analog input) and it had the communications error.

Upon bringing it back up after clearing RAM everything else seemed just dandy.
I wish I had more to go off of or for your information. Again thank you for the reply.

Phil

Did you back up your project before importing the DFB? I would try reverting to your last known good backup and see if the problem persists. If you can see the value changing then it must be being written by a process that is occurring less frequently than every cycle, i.e. It's coming from external to the PLC or from some function that is not resolving or executing a MOVE each cycle.

I've had that EC0E error before and it's nasty on the Quantum CPU. I'd call support - in my case, they blamed it on the CPU hardware.

http://www.schneider-electric.us/si...ang=en&locale=en_US&id=FA199158&redirect=true

I suspect if it is repeatable, the support team will want to know how you can make it happen. I could not - we just saw it randomly and I could never tie any type of specific trigger event to it.