Technical Forum

Communication / Networks

OPC random write problem

Erstellt von Kris1976 am 04.05.2012 10:05 (3 Antworten)

« Vorheriges Thema Nächstes Thema »
Seite 1 von 1 (4 Objekte)
  04.05.2012, 10:05
Beigetreten am: 20.09.2011
Letzter Besuch: 04.05.2012
Beiträge: 5
Bewertung:
Bewertung: Herausragend [4 von 5 / 1 mal bewertet]. (1) 
Bewertung: keine [0 von 5 / 0 mal bewertet]. (0) 
We have a 315-2 PN/DP v2.6 CPU, where a MES system (coolplex) is communicating with via the Simatic net OPC server (v6.4) through the ethernet interface. OPC server is running on Windows server 2003 R2.

The MES system is reading/writing to a DB of 56 bytes, where a command made of 12 bytes is written to the PLC.

For the last 2 years, it worked fine, but since a few weeks, now and then, some values of the command are no longer written to the CPU, although the writing of the command hasn't changed.

It happens up to 20 times a day that suddenly 1 or 2 values are not written. When the MES system rewrites the command, everything comes through again...

Anybody any idea what may cause this or how i can log the OPC server what is happening?

Thx

Kris
Top
  04.05.2012, 16:28
Beigetreten am: 17.06.2011
Letzter Besuch: 21.05.2013
Beiträge: 1277
Bewertung:
Bewertung: Exzellent [4,04 von 5 / 80 mal bewertet]. (80) 
Benutzer mit speziellen Fachkenntnissen, die einzelne Konferenzen betreuen
Bewertung: keine [0 von 5 / 0 mal bewertet]. (0) 

Hi Kris1976,

for OPC you can create OPC trace to see the OPC error code.
The easiest way is as following:
1. Open the Program "Configuration console" in SIMAITC NET Start Menu folder. Enter Application -- OPC -- Trace and activate the Trace DATA. Over there you can set the trace level to error first.

 

2. If the OPC server returns an error code you will find it in this trace. It starts normally with 800***. Check the error code with the manual or easier you will have a look to the file OPC_ERROR.H in Simatic net installation path (make a search about this file, because I don't know the exact location of this file). Search about the error code and in the comments of this message you will find a description about the reason.

 

But on general, if you don’t change anything it would not change the result. So check what is happened. Additional connection to PLC added? Changes in the OPC application made? Changes in PLC Program made? Try to figure out what was happened.




=== Edited by Rated X @ 5/4/2012 4:28 PM [GMT ] ===



Gruß / best regards
Rated X
Top
  04.05.2012, 16:57
Beigetreten am: 20.09.2011
Letzter Besuch: 04.05.2012
Beiträge: 5
Bewertung:
Bewertung: Herausragend [4 von 5 / 1 mal bewertet]. (1) 
Bewertung: keine [0 von 5 / 0 mal bewertet]. (0) 
We already tried the OPC trace, but it impacted the communicationspeed to much and it generated to much comm errors in the MES system.

This system is vital in a running plant (24h/7d), so I had to stop the tracing, no luck there
Top
  04.05.2012, 17:33
Beigetreten am: 07.07.2010
Letzter Besuch: 21.05.2013
Beiträge: 1978
Bewertung:
Bewertung: Exzellent [4,01 von 5 / 200 mal bewertet]. (200) 
Bewertung: keine [0 von 5 / 0 mal bewertet]. (0) 
Kris1976 wrote:
We have a 315-2 PN/DP v2.6 CPU, where a MES system (coolplex) is communicating with via the Simatic net OPC server (v6.4) through the ethernet interface. OPC server is running on Windows server 2003 R2.

The MES system is reading/writing to a DB of 56 bytes, where a command made of 12 bytes is written to the PLC.

For the last 2 years, it worked fine, but since a few weeks, now and then, some values of the command are no longer written to the CPU, although the writing of the command hasn't changed.

It happens up to 20 times a day that suddenly 1 or 2 values are not written. When the MES system rewrites the command, everything comes through again...

Anybody any idea what may cause this or how i can log the OPC server what is happening?

Thx

Kris

Having worked in a manufacturing plant for many years, I know how seemingly unrelated events can often be highly entangled.  I say this to make you think about what has happened within the past few weeks to the following:

1) Your plant electrical & network distribution in the vicinity of the MES computer and the PLC OPC is exchanging data with.
2) I.T. network changes that may have increased traffic or rerouted how the network functions between the PC/PLC.
3) Possibly an ethernet a connector starting to lose strong connection intermittently, say due to a prolonged vibration, whether it be mechanically, or acoustically induced.
4) And strangely enough, is the PC hard drive approaching capacity or in need of serious defragmenting?  We had a similar OPC problem that would result in sporadic loss of data acquisition and control.  Turns out it was an overly full hard drive resulting in time-outs without any errors showing up.

Please post your results!

Helpful? Rate it.
Solved? Post it.
Top
Seite 1 von 1 (4 Objekte)
Aktionen