L32E ethernet problem

Hi all and hopefully Ken Roach.

I have an unusual problem that I can't find a reference to in the knowlegebase and tech support have not heard of.
We have 2 near identical jobs using a compactlogix L32E connected to a PV+ via an ethernet switch.
Unfortunately these 2 jobs are 900 and 2000 kms from where we are based.

Job A runs on a generator which is switched on and off as the plant is required. (maybe 20 times a day)
Everything ran OK for 2 weeks then on a startup the PV will not communicate. I had a local electrician browse with RSWho on the ethernet network. He can see the PV+ and the L32E ethernet port. He was unable to expand the ethernet port to see the backplane or the processor. (he did not have a 1747-CP3) to go in through the serial port.

Job B ran for 2 months on mains power then today they ran it on the generator power and shut down the generator while the plant was connected. Same problem. The PV+ loses comms.
At this job I do not have access to someone with the programming software yet.
All the LEDs on the L32E are OK except for the link led which is flashing. This indicates that there is no CIP connection.

I am surmising the problem is developing when the control voltage on the system is falling from 240V to zero as the generator winds down.

Is it possible for the processor to lockup or lose comms between the ENet port and the internals of the processor.

Regards Alan Case

I have seen unusual behavior in systems that experience brownouts or slow declines in control power. It's conceivable that the switch itself might be part of the broken connection.

I suspect that the onsite guy's failure to browse was more about EDS files than about daughterboard/mainboard comms inside the CompactLogix. If there was a problem between those boards, or between the controller and the Compact I/O subsystem, you'd see a connection failure (the I/O LED would blink green) or experience a major fault.

My wild guess is that it's the PanelView that is failing to handle the brownout gracefully. My approach would be to fit a voltage sensing relay to drop out control power entirely to the control system once mains voltage gets below about 90 VAC.

Hi Guys...What about putting a UPS in to carry the PLC et al?...I had a similar problem at some block plants, and after we put in the UPS-a small Din rail mounted unit-we didn't have anymore problems.

David

The notebook he was using was the one I set up for him when I was on site and the EDS for the L32E was present and he could browse through to the processor. Since I left site I talked him through a small programming change and that was handled OK. RSWho can browse through the switch to see the L32E port but no further. The controller does not have any IO connected, just a PV+ that is not in the tree so I assume that the IO connection throught the Enet is OK because it does not expect to see an IO connection. The panelview is visible in a RSWho browse and the path in the error messge is correct. The link light indicates that the ENet port is OK but does not have any CIP connections established.

We will be installing a UPS and a undervoltage relay but it will require a bit of smart wiring as we only want the UPS to hold up the power for a max of a minute on loss of mains or generator voltage then shut down.

Also another SI I spoke to experienced the same problem on a L35E on a generator that was turned off. The ethernet port gave up communicating. It seems strange that all 3 give up at the same time, when on a generator supply that is turned off.

Ubnfortunately it looks like we may have to go to site.

Regards Alan

Hi David. Were you using L3?E processors and did the processors come good or were they replaced.
Regards Alan Case

Hi, Just an update.
I received the processor back from 1 job today.
No program in it (lost/dumped corrupted), but I can see where the mice have been pi*s*ng in the top of the unit. So it is not always what you think it is.
Regards Alan Case

Hum...I wonder if that could be trapped by some type of fault routine:book: