1725   Can an IED publish GOOSE only with fixed-offset? Can the SCT modify GSEControl.fixedOffs?

Created: 20 Oct 2020

Status: Not Applicable

Part: Part 6 (2019; Edition 2.1)

Links:

Page: 81

Clause: 9.3.2

Paragraph: Table 11

Issue

IED.Services.GOOSE.fixedOffs indicates that "fixed offsets are supported and can be set individually per GocB".

Can a vendor declare that all their GOOSE messages are encoded using fixedOffset? In that case IED.Services.GOOSE.fixedOffs shall be set to true or false?

If a vendor declares a GSEControl.fixedOffs = true in their ICD, can the SCT modify this GSEControl.fixedOffs to false?

Can a vendor set IED.Services.GOOSE.fixedOffs = false and configure a GSEControl.fixedOffs = true?

Proposal

Thank you for your response.


Discussion Created Status
Approved not Applicable 03 Jun 21 Not Applicable
Seems that current service description is align with 8-1 and current behavior seems correct with expectation. If a tissue is really needed to update the behavior this has first to be resolved in 8-1.

Today implementation is as follow:
- publishers and subscribers shall support BER encoding for interoperability
- publisher may support additionally FixedOffset and declare this capability with service GOOSE.fixedOffs
- this capability allows any SCT to set the encoding in each GSEControl to FixedOffest as per 8-1 with attribute GSEControl.fixedOffs
- a subscriber supporting FixedOffset will be able to decode more efficiently such GSEControl without taking care of the size of each data of the GOOSE frame
- a subscriber not supporting FixedOffset will decode the GOOSE frame as a classic BER encoded GOOSE frame by reading size of each data, as the size is still indicated even if it is the maximum size of the data

If the request is to support only FixedOffset and not variable length on publisher side, then part 8-1 ed2.1 A.3 have to be reformulated to indicate that the Fixed-Length property do not occurs for each GOOSE Control. End then SCL will be adapted accordingly by deprecating GSEControl.fixedOffs which is not relevant to indicate a global statement for the whole publisher

22 Feb 21 Approval (N/A)
Since 8-1, Ed2. the support of BER encoding rules for the publisher is mandatory (8-1 Annex A.4 Ed2 and Ed2.1) garantees the interoperability.
The support of fixed-length encoding is optional and belongs to PICS and serviceCapability.
The support of Fixed-length decoding is mandated since Ed2.
Changing the behaviour is therefore an 8-1 issue. Part 6 currently is aligned with 8-1 definition.


23 Nov 20 Discussion (red)
Do we agree that the behavior as explained is the intended behavior?

FixedOffs has been introduced, to enable GOOSE implementation in HW. If now a device has implemented GOOSE sending in HW, how can it support BOTH?
03 Nov 20 Discussion (red)
According to IEC 61850-8-1 Ed2.1 Normative Annex A.4 - Conformance:

Publishers claiming conformance to this clause shall support BER encoding.

As soon as an IED declares the IED.Services.GOOSE.fixedOffs=true is supported, then it does support BOTH encoding, and the SCT is entitled to change the encoding of the GOOSE message setting GSEControl.fixedOffs attribute.


Can a vendor declare that all their GOOSE messages are encoded using fixedOffset?
RE: A vendor is mandated to support at least BER encoding. If the service capability indicates that fixed offset encoding is also supported, then the preconfigured GSEControl may be fixedOffs=true, but the SCT is entitled to change it to FALSE.

In that case IED.Services.GOOSE.fixedOffs shall be set to true or false?
It shall be set to true, as the vendor obviously supports fixed offset encoding

If a vendor declares a GSEControl.fixedOffs = true in their ICD, can the SCT modify this GSEControl.fixedOffs to false?
RE: Of course, as both encoding are mandated to be supported.

Can a vendor set IED.Services.GOOSE.fixedOffs = false and configure a GSEControl.fixedOffs = true?
RE: No, this is an inconsistent configuration.
21 Oct 20 Approval (N/A)

 

Privacy | Contact | Disclaimer

Tissue DB v. 23.12.13.1