Rockwell POINT Guard I/O Modules dropping out.

[Struggling_forward]

Hi,
I've got an intermittant fault on a machine.

I have 5 identical machines, 4 work, one has a fault that appears every 5/10 mins.

THe fault is the safety system drops out, all the guards report being open, E/stops pushed, yet no one has touched them.

All the affected guards/stops are on a remote guard i/o connected over ethernet.

The machine runs well for a few mins so the fault is unlikely to be the program. I've put in a new cable and connectors from the main panel to the remote i/o but it made no difference.

Now I'm stuck.

Any ideas?

 

[dmbartender]

Try reseating the module into its PointIO backplane. I have had the case that the safety module wasn't seated 100%, although it looked to be. The vibrations caused by the machine moving may be creating a faulty contact between the card and the backplane.

Otherwise, couild be a faulty card. Try swapping the card out to another machine maybe, to see if the fault follows with the card. Of course you have to change the safety network identification associated with the card, so I would advise using a new virgin card if you have one.

 

[Struggling_forward]

Thanks, I've found a new card in the store, how do i configure it to the correct IP address?

When I run BootP I get the message: Cannot find valid ip address, check network settings.

 

[dmbartender]

Well the safety card doesn't possess an IP address. It has a slot number on the PointIO ethernet node. When you put the new card in, go to your hardware tree in RSLogix, and take a look at the card. If it is faulted, then go to the safety tab of its properties, and see if resetting ownership helps the syncing process.

 

[Paullys50]

Contact Rockwell, a colleague of mine had a similar issue with regular PointIO, it was a firmware issue in the 1734-ANET card. Check your firmware version, you may need to update it.

 

[The Plc Kid]

Contact Rockwell, a colleague of mine had a similar issue with regular PointIO, it was a firmware issue in the 1734-ANET card. Check your firmware version, you may need to update it.

Yes there is a issue that will cause dropped packets on major firmware rev 1. If this is your issue it can not be flashed thought . The adaptor has to be replaced.

 

[Ken Roach]

You're assuming it's a 1734 POINT I/O module in a 1734-AENT(R) adapter; this might be a 'brick' style module.

The fact that four out of five machines that are 'identical' work correctly suggests that the problem is not a configuration or firmware issue.

You should be able to diagnose the safety connection status and distinguish it from an actual module failure or wiring failure.

Start with the embedded web pages on the Guard I/O module itself, and on the 1756-EN2T module (or whatever interface you have on the GuardLogix). Look for errors on the wire first, then move up to see if there are lost packets or other errors at the IP level.

 

[Paullys50]

I was merely pointing out that we have experienced instabilities in the firmware lately, and that it is something to consider. If you look through the 3.x release notes you'll see the anomalies. Updating from 3.7 to 3.10 corrected the anomaly my colleague had seen (under the direction from Rockwell).

 

[Ken Roach]

Aah, I see where he said it was a POINT module in the title, but not in the body of the post. I blame jetlag.

I still think that a machine that is 'identical' to four others is unlikely to have a firmware problem.

Firmware revisions are easy to determine, using the embedded web pages or RSLinx Classic. The OP's first step if he thinks there is a firmware problem should be to do an inventory of his machines and compare them.

Visual inspection of Ethernet cabling is insufficient. You need a proper cable analyzer, and you need to look at each device's physical media diagnostics, to conclude that your media is in good condition.