1734-IB8S / OB8S getting 16#0203 fault. Please Help

Hi all,

I've contacted Rockwell numerous times about the PointGuard card faulting with the 16#0203 fault. I've implemented ALL of their suggestions but still consistently see the fault. Allow me to explain what is going on.

I have a 1768-L45S with 1768-ENBT communicating to 4 1734-AENTRs. Each rack has various standard point I/O and safety point I/O. All but one rack does not exceed the allowed CIP limit of 20. All but one rack does not exceed the power limitations of the AENTRs (added 1734-EP24DCs where necessary).

Every now and then (say 5 - 10 minutes), ONE of the safety cards on the network will fault with the 16#0203 and soon after it will return to running status, but the fault causes the safeties wired to that card (Estops, Guard switches, etc) to drop out or rather the processor looses the ability to read the card.

Rockwell has told me to increase the RPI, to increase the network timeout, to make sure my power limitations aren't being exceeded, to update firmware, to make sure CIP connection is below 20, to change rack connection to "none"...

I've done all of this with the exception of one AENTR rack, but the faults keep coming. Does anyone know a possible solution? Is the one rack that hasn't been changed a problem? I mean, would it cause the other safety cards on other racks to fault with the 16#0203 code?

Let me know if you need more details.

Thanks,

Eric

Well, trying to have more than 20 CIP connections to a single AENTR won't work for sure.

In the past, I don't remember seeing it cause problems with other than the AENTR that had too many connections, I wouldn't necessarily think it would cause the other things on the network to fail with 16#0203 but I can't be sure.

You know about having too many CIP connections already. You will have to tackle that problem at some point. I would attack known problems before trying to solve unknown problems.

BTW what is the network topology? I assume a device-level ring or linear network since you're using the AENTR's.

The network topology is linear/star. Thanks for your reply. I would like to fix the rack that has too many CIP connections; however, since the other two racks on the network haven't been working (even with the fix) there's no guarantee that that was the problem. I'm still having problems with the 1734-IB8S/OB8S faulting with the error 16#0203.

Strangely enough I've noticed it's almost always just 1 card at a time, never multiple cards. Not sure if that means anything.

This is possibly a firmware issue on the 1734-AENTRs and possibly on the 1768-ENBT as well...

68626 - PointGuard: Error 16#0203 In RSLogix 5000
Access Level: TechConnect

Regards,
George

Firmware is at 3.012 (the highest that rockwell is offering right now). Rockwell had me do this first. I've updated the Ethernet card and processor as well.

Thank you, though.

Well, if you have ticked all their suggestion boxes, then I also think you should start looking into resolving the known problems as suggested.

G.

Thank you, I'm still skeptical that that's the problem. Forgive me. I only think this because we have the EXACT same layout and program running on another line with none of the mentioned problems.

Have a look at the diagnostic counters for the AENTRs. Here's some guidance on what to look for, especially regarding Alignment and FCS errors...

29328 - Ethernet Alignment and FCS Errors
Access Level: TechConnect

G.

On one of the AENTRs I'm seeing "MAC Transmit Errors". Same with on the 1768-ENBT. I've look this up before but couldn't find a solution.

What do you think?

Issued solved!

Not sure if this was specific to me or not, since I've never had this problem before. I tried this solution out because of the "MAC Receive errors" I was seeing in the ENBT. I changed every managed switch, AENTR, and ENBT to "auto-negotiate" from forced full duplex and 100 Mbps. I've always used the latter configuration with EtherNet/IP and never had any problems until this line.

Anyway, thanks to all who replied.

Hmm, that should have been one of the first things RA should have advised you to do? Auto-negotiate should be used as an industry standard. It enables more flexibility between devices. I think some people think that forcing modality high ensures maximum speed at all times.

Well done anyway!

G.

Hi All

I have a similar problem to kurtzberger

I have done all the recommendations that you and AB said, but the problem continues.

My system has a:

1 CompactLogix 1768-L43S
6 1734-AENT POINT I/O
1 Powerflex 700
1 Ethernet Module (Resolver)

The problem is just in a 1734-IB8S, sometimes the machine stops, operator push the reset button and continue working, but sometimes the adapter goes to fault and we need to shutdown the machine. This fault appears 2 o 3 times a day.

All the firmware is updated, rpi changed to 20 ms, 1734-IB8S and 1734-AENT has been changed for another, auto-negotiate port speed and duplex

I have attached 2 images of the status of the ethernet card and the adapter that is failing. Somebody could help?

http://www.4shared.com/rar/3Nx_hfxhba/1734-1768.html

Thanks in advance.

Asteroid,

Did you ever resolve this issue? I am currently having the same problem.

Thanks

t.medina, how many modules do you have in your 1734-AENTs?

I remember finding out 1734-AENTs only support 20 CIP connections. And since POINT Guard I/O safety modules do not support rack optimization, each safety card counts as 2 connections, and standard I/O card as 1.

See this post here: http://www.plctalk.net/qanda/showthread.php?t=81438

What I ended up having to do, was split any Point I/O rack (by adding another AENT) to make sure each rack was under 20 CIP connections.

Let me know if this helps!