What could cause a ML110 to error out for watchdog timer fault all of a sudden [Twice today]? There hasn't been a change to the program [8 rungs] and the only thing connected to it RSView32 trend graph and this isn't displayed most of the time.

We were/are have some network problems today but I don't have and faults set for communication errors.

I am getting to walk out the door from work and will think about this on the drive home.

Attached is the basic program minus some logic that does counting [The counter ACC was at 7 during the first fault].

Any ideas?



Thanks,

Bob

What is F8:0s value when it faults?

Don't ask me why but try changing Source B in your ADD statement to F8:2 and set it's value to 1. I am thinking of a problem I once had that was cause by something like this but I can't place it. If anything it wouldn't hurt to try it

Unlike other PLCs which have a seperate communication processor, I think the ML1100 handles it ethernet communications as a part of the scan. So network problems very possibly could be a part of the problem.

There may be a technote on this on the AB knowledgbase. You probably aren't the first person to deal with it. I suggest you look there.

An easy way to test that is to unplug the Ethernet power and cycle power. The Ethernet port does not function until it sees something plugged into the port.

But I tested faulting the network quite a bit on the Micrologix 1100 before we started using them and didn't run into any problems like this.

But anything is possible considering it is a simple 8 rung program

Thanke guys

I by-passed the ML1100 and then killed power to it before I left since it is just to monitor water flow and so they could run production.

While driving home I decided to call the guys and have them power the ML110 back up so I could pull the program out when I got home, but there was no program in it when I connected.

So do these units clear themselves when faulted on a power cycle?

Edit: I just looked at the error code and this time it is S:6 = 2h

Unexpected hardware watchdog timeout or an ASIC reset occurred and this one is different from the first one.

There is no ladder program in the processor? Is the internal 24VDC power supply being used?

No program and an external power supply.
Attached is a zip shot of the screen

One other thing the Micrologix 1100 does have a battery but I don't know what happens when the battery goes dead. Actually I can't imagine the battery going dead this early anyway it goes.

I put your program in a 1100 and speed it up to where it should be after running for about 12 hours at the maximum of the flowmeter input and it did not fault so I don't think there is anything wrong with the program. I did use F8:2 instead of 1.0 but I don't think that mattered

I think I would look at the internal 24VDC and grounds next

Edit - I posted this before I saw the previous post

I could not find a technote in the knowledge base about this issue with the ML1100.

I could put this EQU S:6 2 OTU S:1/13

at the end of Lad 2 but I would like to find out what is going wrong.






Thanks,
Bob

Can you call Rockwell about this issue? I would like to know what is causing it myself

I did and they said it is either noise or it is starting to fail. I asked about ethernet coms and he had not heard that the coms itself causing an issue with scan times.
I just reloaded a program and I am going to watch to see if it faults.
I will let you know what I find out.

this is interesting. it could be number of reasons including hardware malfunction.
but I also had similar problem with different system a while ago. it was working for years
when customer installed some new equipment (with some big motors and drives). from there on,
there was so much interference that all data collection PCs and nutrunner controllers ware locking up
every 15-20 minutes. checked all cables, shields, surge suppressors, nothing helped...

even plcs got reset every day or two (had line filters but ware on a ethernet network).
laptops would run as long as not connected to network and not using mouse (or anything with cable).
built in pointing device would work but the cursor was still jumping wildly...
replacing PCs didn't work, tried few diffetnt ones always with same result (but they all worked fine anywhere else).

that new equipment worked somehow and we ware not allowed to investigate so all we could do is
test on our end. after two long days we ended up changing ground point for our equipment and adding
opto isolation to all comm channels... this worked. ("Enterprise, shields up..." ).

i would start by stripping system down (remove un-necessary connections, specially comms) and
checking for power fluctuations, proper grounding and power quality. i would try to running plc from
UPS or battery powered inverter to see if the problem happens again.

anyway, it's just an idea.