User Manual

e1152.doc

Data and design subject to change without notice. Supply subject to availability

07.93 9

CDDL Protocol - Data link level

4.6 Time-out control and recovery procedure

These procedures are different in case of master or slave stations.

Below both cases are separately described.

4.6.1 Slave side

Two time counts are managed.

The first time-out, T2 (see the available range in Annex A), is used to verify the conti-

nuity of the master station polling activity. The T2 timer is reset and restarted each

time a polling/calling/waiting selection is received or when a [DLE, ACK] sequence is

waited after a transmission.

The second control, TC, is used during a packet reception, starting from the [DLE;

SOP] sequence up to the checksum characters, to verify that the transmission is not

interrupted. The maximum tolerable delay is 0.8 seconds between two consecutive

characters.

When either T2 or TC expires, there are two results:

• the slave station gets back in a polling/calling wait status

• the error counter is incremented

This counter is reset after a successful data exchange; when it reaches the value 10,

a line fault status is activated, for slave station internal use.

After a line fault, the communication gets back into a "no fault" status when a data

packet sending toward the master station is successful (i.e. when an acknowledge-

ment message is received).

4.6.2 Master side

Two time counts are managed.

The first time-out, T1 (see the available range in Annex A), is used to verify that

meaningful answers are received from the slave station after a polling/calling/waiting

sequence.

The T1 timer is reset and restarted each time a polling/calling/waiting selection is re-

ceived or when a [DLE, ACK] sequence is waited after a transmission.

The second control, TC, is used during a packet reception, starting from the [DLE;

SOP] sequence up to the checksum characters, to verify that the transmission is not

interrupted.

During this phase, the maximum tolerable delay is 0.8 seconds between two consecu-

tive characters.

The fault line treatment is similar to that described in Section 4.6.1. The communica-

tion gets back into a "no fault" status when an acknowledgement message is re-

ceived.