1646 valKind=Conf in DAI element
Created: 04 Feb 2019
Status: Discussion (red)
Part: Part 6 (2009-12; Edition 2)
Paragraph: Table 20
Category: Issue for edition 2 of this part
Assigning valKind=Conf in instance (DAI) section makes no sense. This would allow different instances of an LNodeType to have different online data models. If valKind=Conf is desired then that fact should be exposed in the DataTypeTemplates section.
1. Modify text in Table 20 (DAI attributes) row valKind to note that "Conf" is not allowed here
2. Modify SCL_Enums.xsd to add a new type "tValKindDaiEnum" with same enumerations as tValKindEnum but excluding "Conf"
3. Modify SCL_IED.xsd complexType "tDAI" attribute "valKind" to use tValKindDaiEnum instead of tValKindEnum
The SISC I2111 is wrong. Support changed valKind shall be limited to SetToRo - the ICT/IED capability is described in the service capability.
This is not an SCT responsability to change valKind to Conf.
|07 Feb 19
It is a possibility since edition 1 to set valKind="Conf" in DAI and this is the responsibility of the manufacturer of the device to correctly implement it if he decides to support it.
For backward compatibility I propose to keep this capability.
The problem is when the system tool change the visibility of an existing datamodel.
We have introduced in edition 2.1 the attribute ValueHandling to indicate if an IED tool support changes of a valKind by a system tool.
Here we have a discrepancy between definition of the valueHandling and the SICS. Value handling is supposed to allow only change from Set to RO (to restrict edition of the value) and in SICS, the capability to change to Conf is also indicated.
My proposal is to update SICS to not allow change to Conf and maybe to add a note to valKind definition for DAI to warn about the impact of changing to Conf for an IED and restrict this capability to the IED (and his tool) only
|06 Feb 19
Privacy | Contact | Disclaimer
Tissue DB v. 184.108.40.206