Hello ,

I have written a program that is uses the MB_CLIENT FB to communicate with Modbus TCP Servers.

There are 10 separate MB TCP/IP servers and I have 3 separate polls (Read Multiple Holding Registers) for information from each.

I therefore have in total 30 x MB_CLIENT FB's being used.

Each MB server has its own IP address on the subnet and they also have unique UNIT_ID’s / Slave Addresses.

The PLC can poll units 1 - 9 without any problem, all results are returned correctly and all MB_CLIENTS appear to disconnect correctly.

When polling UNIT 10, the first read works successfully, however the following 2 do not and return error 16#80C8. I have checked to ensure that all MB_CLIENT FB's are using unique CONNECT_ID's and they are. If I remove 2 previous reads of any other units then the queries for UNIT 10 are successful, therefore I know that the UNIT_ID is correct i.e. 16#000A = 10 and the MB server is responding to the queries.

I can see that within the DB for the 2 MB_CLIENT queries that don not seem to be working that the TRCV_SFB.EN_R value is FALSE, for all other queries that work this is TRUE.

It would appear that once I exceed 28 x MB_CLIENT FB Calls then I start to receive these errors, could this be a device limitation?

My PLC also provides 2 x MB Server connections on different ports, I have disabled these connections, however this makes no difference.

I have attached an example of a MB Query.

Any help is greatly appreciated.

Thanks

Kaulquappe

Hello NeilRu,
80C8 is also an error message of TCON. It means that the connection ID is already in use.

I would also check how many connections are supported by the used hardware.

Regards,
Kaulquappe

Hi Kaulquappe,

I did read this previously in the documentation, however I have checked and the connection ID is not in use anywhere in my project.

Each query is to dissconnect after it has either been flagged as DONE or ERROR, therefore it should not be using more than one sumultaneous connection at any time accross all queries?

Regards

Kaulquappe

Hello NeilRu,
TDISCON may need more than one cycle to finish. Due to the manual STATUS=0003 is reported when the connection is terminated. I recommend to check that before starting the next job.

Regards,
Kaulquappe

Hi Kaulquappe

Apologies, however I am not too sure how I can acomplish this.

I have used the TDISCON between each job, and I have waited for it to complete, however this does not resolve the issue. I get a status of 16#7000 from each TDISCON, is this correct?

Regards

Kaulquappe

Hello NeilRu,
when you set DISCONNECT=TRUE the MB_CLIENT calls TDISCON internally. The STATUS of TDISCON is transferred to STATUS of MB_CLIENT accordingly. You mus not call the T-blocks for the connection ID which is used with MB_CLIENT.

Regards,
Kaulquappe

 Hello Kaulquappe,

The connection IDs are not being used for any other T-blocks. I have removed all MB_SERVER instances and all other functionality within my program and I am still experiencing the same issue. I only have my MB_CLIENT queries and each query has its own unique connection ID.

Regards