Application Note Flashing Rate and Clock Generator in SIPROTEC 4 Units

Flashing Rate and Clock Generator in SIPROTEC 4

Relays
Implementation in the CFC

Foreword
In many cases, the customer requires flashing LEDs, for example, to draw the operator’s attention to a
particular status. This feature is mostly necessitated by safety-critical requirements. For example, it
can be used to indicate a change from a safe to a critical state. This change could be caused by the
operator:
ΠKeyswitch turned to LOCAL
ΠControl command output
or also by the process:
ΠMeasured values are at limit value
ΠProtection function has been blocked
Applications are also possible in which a clock generator (analog flashing rate generator) outputs a
pulse at regular intervals that can be used, for example, to synchronize other relays.
In the SIPROTEC 4 relays, no such flashing mechanism is available. The LEDs and binary outputs
can only be routed to "Latched" or "Unlatched", but not to "Flashing".
However, this missing functionality can be compensated for using the programmable logic, a highlight
feature of the SIPROTEC 4 family. This paper describes two solutions; you must use one or the other
depending on the type of triggering signal:
ΠShort pulse (fleeting indication): e.g. pressing a function key on the relay
ΠSignal pending for a long time: e.g. "keyswitch to LOCAL" or "measured values above limit value"

CFC and timers

The flashing mechanism uses a signal that switches from "low" to "high" at regular intervals:

thigh tlow Time

50ms 250ms

The time block, which is the only block that processes the time, can be used in this case. Two are
available for this with device firmware 4.30 and higher. But for typical flashing the normal timer block is
sufficient. It actually consists of two timers.
Inputs:
With S, the "timer 1" is started and runs until the
time parameterized in T1 has elapsed. If the
starting signal S is reset, "timer 2", whose runtime
is set by T2, is started,. On a signal to R, all
running timers are stopped.
Outputs:
As long as "timer 1" is running, QT1 is "high". The
1
same goes for "timer 2" and QT2.

1
As a logic combination (Q=(S+T2) * T1 * R), Q is not required very often and is not relevant to this task.

PTD PA 13 -1-
Application Note Flashing Rate and Clock Generator in SIPROTEC 4 Units

CFC and priority class

As is stated in the help for the block, the timer only runs in certain priority classes. Why?
There are 4 priority classes:
1. Fast PLC (PLC_Bearb)
ΠPriority higher than the protection function
ΠEvent-controlled: Start of the class on change in input signal
ΠSuitable for blocking protection functions
ΠNo more than about 12 blocks can be placed
2. Slow PLC (PLC1_Bearb)
ΠPriority lower than the protection function
ΠEvent-controlled: Start of the class on change in input signal
ΠPreferred class for event-controlled applications (e.g. flashing mechanism)
ΠNo more than about 30 blocks can be placed
3. Switchgear interlocking (SFS_Bearb)
ΠPriority lower than the protection function
ΠEvent-controlled: Start of the class on output of a (switching) command
ΠSuitable for interlock conditions
ΠNo more than about 100 blocks can be placed
4. Measured value processing (MW_Bearb)
ΠPriority lower than the protection function
ΠCyclically controlled: Start of the class at regular intervals or if the processor is not off-loaded
ΠSuitable for processing measured values (e.g. limit value comparison, implementation of
protection functions such as "Reverse Power ANSI 32" or "Power Factor ANSI 55")
ΠNo more than about 100 blocks can be placed
Because the timer can be started at any point in time, it can only be placed in the PLC classes. In the
other levels, its execution would not be guaranteed.
The CFC chart that must be created for the flashing mechanism must therefore be set accordingly:
1 Create a new CFC chart.
2 Open that CFC chart.
3 Select in the menu "Edit" the entry "Run Sequence..." (a window "Run Sequence" opens).
4 Select "PLC1_Bearb" (i.e. the slow PLC class).
5 Select the entry "Predecessor for Installation" in menu "Edit" of this window.
6 Close this window.
The new CFC chart will now definitely run in the correct priority class.

PTD PA 13 -2-
Application Note Flashing Rate and Clock Generator in SIPROTEC 4 Units

Flashing rate generator – triggered by a short pulse

In many cases, pressing a function key causes switchover to a safety-critical state. A flashing LED
indicates that. The flashing mechanism is started by a short fleeting indication and continues to run
until it is stopped (e.g. by pressing a further function key).

Preparation in the routing matrix

For these two function keys you require two new internally formed information items (signals) that are
created in the routing matrix and routed to destination "C" (CFC):
1 Insert a new routing matrix into a new group, e.g. "Flashing LED".
2 Select type "Internal single-point indication" in the information catalog of the routing matrix (under
Information) and drag and drop the entry into the required group. Repeat this action.
3 Name the information items created in the group, e.g. "F1 Flash" and "F2 NoFlash".
4 Route the two information items in such a way that they are triggered by function keys 1 and 2
(source) and can be used for processing in the CFC (target "C").
In the CFC processing, a signal must be generated that makes the LED flash. Here again, an internal
single-point indication (selection) is necessary:
5 Drag an information item of type "Internal single-point indication" out of the information catalog into
the required group.
6 Name the new information item, e.g. "LED Flash".
7 Route the information as the result of the CFC processing: Source "C" (CFC).
8 Route the information to an LED "Unlatched", e.g. LED1.

Implementation in the CFC

Create a CFC chart in the correct priority class, as shown in Section "CFC and priority class".
Here is the solution to the problem:
The time sequence of the flashing rhythm can be described by the change of the times for "low"
(tlow=250ms) and "high" (thigh=50ms), as shown in the figure in Section "CFC and timer". However, this
sequence cannot be triggered with only one timer block because the second timer of the block is
already started when the start signal S (short pulse!) is falling again.
For that reason, two serially connected timer blocks are required:
The first is started by the signal "F1 Flash". Immediately after starting, the internal timer T1 runs for
250ms (tlow) and sets output QT1 high for that time.
This output signal of the first timer is used as the starting signal for the second timer block. However,
in this case, the internal timer T2 is used which is only started when the start signal at the start input
falls again, i.e. precisely when time tlow=250ms has elapsed. It runs for 50ms (thigh) and sets the
associated output QT2 high for this time. This signal is combined with the LED ("LED Flash"), which
lights up for precisely this time. After this timer has elapsed, it goes off again.
For the LED to flash now, the sequence must be repeated. For that reason, the same output QT2 is
combined with the starting input of the first timer block by means of a OR gate connected in front of it,
a permitted feedback loop which causes an irrelevant warning on compilation in his case. But careful:
If the OR gate is inserted after the two timer blocks, its sequence number is incorrect. You can change
it with menu item "Run Sequence" in menu "Edit". The important thing is that the OR gate is assigned
a sequence number that is greater than of the second time block.
Now only the mechanism that permits switch-off of the flashing rate generator is missing (in this
example using the F2 key): For that purpose, the reset inputs of the two timer blocks are connected
with the signal "F2 NoFlash".

PTD PA 13 -3-
Application Note Flashing Rate and Clock Generator in SIPROTEC 4 Units

Flashing rate generator – triggered by longer pending signal

To solve this problem, it is first necessary to understand why the above solution is not effective in this
case: Due to the feedback between the second and first timer block, the first timer is started again.
However, that only happens if the preceding start signal has already fallen. This is the case if you
press a function key but not for a much longer signal:
Example: You want to set the LED flashing if the user sets the switching authority to "LOCAL" on the
relay, the first time would only be started once – on triggering. The feedback would only run once but
do nothing.
This circumstance therefore requires conversion of the longer signal to a short pulse that then starts
the flashing mechanism as shown above. The timer block also performs that function:
At the start input of this additional timer block, the
signal to be converted is combined. T1 and T2 are
also assigned 10ms. That means: The rising start
signal – however long it may be – causes a
fleeting QT1 edge, the falling a fleeting QT2 edge.

These two signals can be used in the above solution instead of "F1 Flash" and "F2 NoFlash".
Here is the solution for the above example:

If the switching authority is implemented as a double-point indication, as it is for example on the

7SJ63, (keyswitch!), the conversion "Conv to impulse" must be place before conversion of the double-
point indication to a single-point indication (DI_TO_BOOL).

PTD PA 13 -4-