The former (proven and widely implemented) recommendation (in note 2 of Fig. 10) that the publisher should send GOOSE-retransmissions when about half of the TAL has expired has been removed.
The new/remaining wording is ambiguous and opens room for discussions if it was also OK to send re-transmission exactly when TAL has expired, thus making them arrive at the subscribers too late and forcing the subscribers to cater for that.
This would break backward compatibility with existing subscribers and contradict some concepts in IEC 61850-90-28.
Proposal
Provide proper and unambiguous wording, stating that it is in the responsibility of the publisher to publish the GOOSE-retransmissions in a timely manner so that they arrive at the subscribers before TTL expires.
This will re-instate the clause of ED2 (under Fig. 8) that correctly says:
"Each message in the retransmission sequence carries a timeAllowedToLive parameter that informs the receiver of the maximum time to wait for the next re-transmission. If a new message is not received within that time interval, the receiver shall assume that the association is lost."
The row in Table 160, stating a formula "2*TAL + Offset ms" is contradicting and shall be removed.
Discussion
Created
Status
Editor Meeting:
Table 160 has been added on purpose in Ed2.1 to force subscriber to allow missing at least one retransmission.
In Previous editions, it was unclear how subscribers set their timer at all. It is not contradicting because this is describing the subscriber behaviour, and not the publisher behaviour.
Figure 10 is stating that the publisher shall set its retransmission timer to a value smaller than the Time Allowed to Live exposed in the GOOSE message (Restranmitting to half of the Time Allowed to Live is smaller than the Time Allowed to Live). It is a local issue to set a smaller value, as it always was: in Ed2, using half was a suggestion not a mandated behaviour.