Ethernet/IP to Devicenet

tsgrab

Member
Join Date
Oct 2005
Location
Texas
Posts
4
Hi everybody. I have spent a lot of time lurking on this website and have found many of the posts quite interesting and useful.

A customer that my company works for uses a 1756-L63 ControlLogix that interfaces to an Ethernet to Devicenet linking device 1788-EN2DN through a 1756-ENBT/A ethernet module. The module also services other I/O devices on remote ethernet I/O racks.


The devicenet scanner controls an MCC devicenet network controlling E3Plus overload devices, Powerflex700 drives, and simple DSA modules. The total number of nodes is about 20. The baud rate is 500kbaud. I've noticed in the Devicenet for Logix user manual that it recommends that devices that have in excess of 15 bytes of input and output data be set with low addresses, however this is not the case in this system. The scanner interscan delay is 25ms. The ethernet RPI for the scanner is 100ms.



We have a problem of the scanners input and output data registers (GP1_MCC2_DN:I and GP1_MCC2_DN:O as viewed from Logix5000) locking up and not updating. The status registers (GP1_MCC2_DN:S) continue to update, eg scan counter, device status, active node registers. When we attempt to start and stop the scanner by turning on and off the GP1_MCC2_DN:O.CommandRegister.Run bit, the GP1_MCC2_DN:I.StatusRegister.Run continues to stay on. Any motors that were running when this anomaly occurs continue to stay running, even though the outputs in the output registers GP1_MCC2_DN:O.Data are off. As you can imagine this can cause major problems due to ignored interlocks not functioning.



We can communicate using explicit commands such as the CIP message instruction in Logix5000 and monitor parameters using Networx for Devicenet, but the I/O connections will not work until we cycle power on the scanner.



This problem doesn't occur on a continuous basic. The system may run for weeks with no problem, shut down once a day, or shut down several times per day. It requires continuous monitoring from an operator, even though the system was designed to operate unsupervised.



Have you heard of this happening before? I have not seen anything on the knowledgebase about this. I would like to continue using ethernet/devicenet enabled MCC's, but if these problems persist, we will have to think of something else.

Thank you for your support.
 
No help with your specific problem, sorry.

But I'm curious why one would use this device instead of a 1756-DNB...
 
Does the I/O connection fail (the yellow triangle in the I/O tree) ? You can use a GSV instruction to get the Module object's EntryStatus to check on this in ladder logic, too.

I've never heard of the Input and Output connections failing but the Status connection staying functional.

Does anything else on the EtherNet/IP network have this sort of problem ?

It's likely you have already done so, but this is something that should definitely be brought to the attention of Rockwell Automation Technical Support.
 
Ken,
If the I/O connection fails, it does so when I am not viewing it. I have entered some GSV blocks to see if I can catch and record some faults on the EN2DN, the ENBT, or on the CPU itself. Nothing else is haveing any problems on the ethernet network, and the whole I/O structure is centered around Ethernet/IP. We even have another MCC controlled by another PLC on the same network that hasn't had any problems. This is about to drive me crazy being that it only happens on occasion and the only way to get out of this anomaly is to cycle power on the module.

Gerry,
The PLC cabinet is located some distance away from the Devicenet MCC and A-B sells these MCC's with the EN2DN or CN2DN pre-installed. Seemed like the logical choice at the time.
 
I'll bet you have encountered something new in the wild. Get as much troubleshooting info as you can ready for Technical Support.

Since none of the DeviceNet products fault when this happens, and the scan counter keeps running, I'll bet the DeviceNet traffic has stayed steady. You could verify this with a NetMeter or a Traffic Analyzer. Although all A-B motor controls can be made to "hold last state" on a DeviceNet polling failure, most people don't configure that mode and so the behavior of the DeviceNet is a good assumption we can make.

I wonder if the HTTP server on the 1788EN2DN gives any useful information in such a circumstance.
 

Similar Topics

Hello, This will be my first time converting powerflex 40's and 400's from devicenet to ethernet. I did some research, and it seems I will need...
Replies
4
Views
699
A customer has a stamping press that has 2 DeviceNet nodes going to a DeviceNet control card in a ControlLogix rack. Older ControlLogix, L55...
Replies
41
Views
14,959
Hello Team, Is there a way to explore DH+ Networks like there is for EIP and Dnet? Thanks! -PreLC
Replies
3
Views
1,687
So I have a system that is running on DeviceNet. We have been having nothing but problems with it for the last 3 years. I am trying to switch...
Replies
2
Views
1,258
Hello Experts, I would like to post a question about migrating DeviceNet to Ethernet. No change in PLC and number of IOs. Just network migration...
Replies
7
Views
2,709
Back
Top Bottom