GE RX3i Modbus TCP/IP Client Commreq

[CodyB84]

I'm trying to talk Modbus TCP/IP from a redundant system using Lan 2 inthe CPE330's to a CPU310 using ETM001. I've had the system talking before I added the redundant chassis. Once I added the redundant chassis, I now get a value in the first Commreq Status Word after trying to open channel 2.

The value equals error:

C090H: (Redundancy only) COMMREQs commands are not allowed when the Redundant IP address is not active at this Ethernet interface.

I have a Wonderware PC talking to the redundant system via the Redundant IP address. I've tried this with the Wonderware talking and with it disconnected.

Anybody have any idea what's going?

 

[Steve Bailey]

Have you tried making the COMREQ command execute only when the CPU is the active one?

 

[CodyB84]

I'm not sure I know how to do that. Attached is the file I'm using to test.

 

[Steve Bailey]

There are %S bits associated with the redundancy status. There is a bit that indicates if the the CPU is the primary or the secondary and another that indicates whether or not it is currently controlling the outputs. I'll have to fire up a different computer to get more specific.

 

[CodyB84]

I understand and know the %S bits you’re talking about. I’ve used them before in another system. I just don’t know how to utilize them in the Commreq or where to use them to make this work.

 

[Steve Bailey]

Only permit the one-shot that triggers the COMREQ when the CPU is the active one. I'm not 100% sure that's what is causing the error, but the text of the error message seems to suggest that you can't execute the COMREQ from the secondary PLC unless it's active.

 

[CodyB84]

I’ll look into it but that’s not how I read it. I read it as the redundant ip address isn’t active but I know it is.

 

[Steve Bailey]

I know that you can choose to assign the same IP address to both CPUs in a RX3i redundant system, so there must be some mechanism in place to deactivate the port in the CPU that isn't controlling the outputs so you don't show a duplicate IP address. I suspect the error message is related to that, although I admit I don't know how it applies to non-CPU ports.

 

[CodyB84]

Primary CPU is xxx.xxx.xxx.10
Secondary CPU is xxx.xxx.xxx.11
Redundant IP is xxx.xxx.xxx.12

The HMI and other devices look at xxx.xxx.xxx.12 so there isn’t a list of view when the CPU’s fail over.

From PME, you use .10 or .11 to connect to the CPU you want to talk to.

 

[Steve Bailey]

Which CPU are you monitoring with PME when you get the error message and what is its role at the time? Do you get the same message on the other CPU?
My take on the error code is that you're triggering the COMMREQ on the CPU that isn't currently the one controlling the outputs. Assuming the same program is running on both CPUs you will be triggering the COMMREQ at the same time on both CPUs unless you use one or more of the %S bits related to primary/backup operations as permissives for the COMMREQ.

 

[CodyB84]

Which CPU are you monitoring with PME when you get the error message and what is its role at the time? Do you get the same message on the other CPU?
My take on the error code is that you're triggering the COMMREQ on the CPU that isn't currently the one controlling the outputs. Assuming the same program is running on both CPUs you will be triggering the COMMREQ at the same time on both CPUs unless you use one or more of the %S bits related to primary/backup operations as permissives for the COMMREQ.

I get it on both. If I’m on the primary and the primary is active, I get it. If I’m on the primary and the secondary is active, I get the error. The same thing happens if I’m on the secondary and look at the primary or secondary.

 

[Steve Bailey]

I guess that blows my theory out of the water. For some reason it doesn't seem to want to let you use port 2. Have you tried opening a case with tech support? How well-versed in redundancy is the automation specialist at your local distributor?
Are planning to read, to write, or both between the redundant CPE330s and the CPU310? In the redundant system I supported before I retired, we used the HMI for communications with PLCs outside the redundant system. If there was data from the outside PLCs we needed for control in the redundant system, we used a script in the HMI to write it to the redundant PLCs.

 

[CodyB84]

We opened a case with tech support yesterday. Waiting to hear back from them. Our local guy is excellent, but he has been tied up the past couple of days. He helped me get the system talking in a simplex setup. Hopefully him and I can sit down tomorrow and investigate the issue. I just plan to read Modbus data into the CPE330s.

Here is the plan and the backstory:

We have a redundant system with CPE330s in-service. We are adding an Ultrasonic Flowmeter and want to read some of the flowmeter's data via Modbus TCP. I currently do not have the flowmeter. So, I setup a CPU310 with an ETM001 to mimic the flowmeter until I get it here.

The first thing we did was setup a simplex system with a CPE330 talking to the CPU310/ETM001. We were able to read Modbus data from the CPU310/EM001 to the CPE330. Then we converted the simplex system to a redundant system with the CPE330s. This is when we started getting the error C090.

 

[Steve Bailey]

Sounds like if all else fails, you could let the HMI take care of reading from the flowmeter, especially if all you need to do is display the flowmeter data. If you need to make control decisions based on the flowmeter data, the HMI could write it to the PLC in a script (assuming your HMI supports that).

 

[CodyB84]

That’s definitely a possibility. It’s a PC running Wonderware HMI. It should do that. Thanks for the recommendation.