mrtweaver
Member
Well it still does not answer my question directly. A previous post says that you need to go in and organize the tags yourself for optimum efficiency. So with that bit of information lets say you have 42 tags per IO device. In the DBF file as well as if you look at the tag listing within the CITECT software the first 42 records are for the first IO device, however they are not in sequential order. Now in my IO device, Horner PLC, they are in sequential order. So with all that has been said here, does the tag layout in the DBF, which the CITECT product uses for your tags, have to be in sequential layout as well for optimum efficiency?
Ex: In the PLC I have the information that I want to gather and setup in registers 362 thru 450. However in my DBF file for CITECT it jumps around, records 1 - 6 = R362 thru R372, then records 7 - 9 = R406, R413, and R423, from this point the rest of the records jump around are are not in consecutive order. But they are all still in the same grouping. Like I said records 1 - 42 are for machine center1, records 43 - 84 are for machine center 2, and so on. But if I get better efficiency by putting the records in sequential format then I will do that.
Just out of curiousity, do all SCADA software packages require that you put the tags in sequential format within the server portion?
Does this rule also hold true if you are using an OPC server such as KepServer?
Finally some people on this forum swear off of using an OPC server they prefer a direct connect from the SCADA software they are using to the device. What is gained by going directly to the device from the software?
Ie: In the line of say CITECT. They have a modbus driver which communicates nicely with my horner devices. However the system can also be setup with using a product such as KepServer. So what are the benefits downfalls of each method?
Ex: In the PLC I have the information that I want to gather and setup in registers 362 thru 450. However in my DBF file for CITECT it jumps around, records 1 - 6 = R362 thru R372, then records 7 - 9 = R406, R413, and R423, from this point the rest of the records jump around are are not in consecutive order. But they are all still in the same grouping. Like I said records 1 - 42 are for machine center1, records 43 - 84 are for machine center 2, and so on. But if I get better efficiency by putting the records in sequential format then I will do that.
Just out of curiousity, do all SCADA software packages require that you put the tags in sequential format within the server portion?
Does this rule also hold true if you are using an OPC server such as KepServer?
Finally some people on this forum swear off of using an OPC server they prefer a direct connect from the SCADA software they are using to the device. What is gained by going directly to the device from the software?
Ie: In the line of say CITECT. They have a modbus driver which communicates nicely with my horner devices. However the system can also be setup with using a product such as KepServer. So what are the benefits downfalls of each method?
PhilipW said:KnowledgeBase Article Q3414:
The Citect I/O Server (and compiler) will set up request blocks to optimise data requests and thus improve performance. Blocking is basically set-up on address and type but with OPC we are not using addresses but record numbers. Therefore, remembering that blocks are created using individual types, tags should be set-up per I/O device, per type in the variable tags database.
This means you have to manually sort the database.