Use alarms to notify users of abnormal conditions for a given shared variable.

You can specify whether to enable alarms, under what circumstances a shared variable is in alarm, the priority level of an alarm, and how alarms are acknowledged. Each alarm limit has a priority ranging between 1 and 1,000.

The LabVIEW Datalogging and Supervisory Control (DSC) Module generates alarms depending on the value or status of a shared variable. The alarms based on value vary with the shared variable data type. For any shared variable, if the status is bad, the DSC Module can generate a bad status alarm.

Defining an Alarm Area

Use an alarm area to define a subset of shared variables in the system. You also can use an alarm area to perform tasks on a set of shared variables at the same time.

You can specify the area to which an alarm belongs on the Alarming page of the Shared Variable Properties dialog box. You also can use the Alarming.AlarmName.Area SharedVariableIO property, where AlarmName is HI_HI, HI, LO_LO, LO, RoC, U32 Bit Field, Bad Status, or Boolean, to specify an alarm area programmatically.

Configuring Alarm Acknowledgement Mode

You can use one of two modes to handle shared variables that were previously in alarm but have returned to normal: Auto Acknowledge and User Must Acknowledge. Configure these modes on the Alarming page for each shared variable alarm. If you configure an alarm for automatic acknowledgement, the acknowledgment status automatically changes from unacknowledged to acknowledged when the shared variable returns to normal. If you configure an alarm for user acknowledgement, the status remains unacknowledged until the operator acknowledges the alarm.

Setting Alarms for Numeric Shared Variables

You can configure numeric shared variables to have the following alarms: HI_HI, HI, LO, LO_LO, Bad Status, and RoC (Rate of Change). The DSC Module calculates alarms after performing scaling. The DSC Module lists alarm levels in engineering units.

Setting Alarm Deadband on Numeric Shared Variables

Alarm deadband defines the percentage a shared variable value must change from the alarm limit before the change is considered a significant change. For example, if a shared variable that represents a temperature value hovers near an alarm limit of 40°C, the shared variable might trigger an alarm many times in a relatively short period of time.

The following table shows examples of events with alarm deadband set to 0.0%:

Time Value Change Alarm Condition
9:15:05 41.1 Yes HI
9:15:10 39.9 Yes None
9:15:15 40.1 Yes HI
9:15:20 38.5 Yes None

This type of situation clogs event files with redundant information and can frustrate operators who have to acknowledge alarms even when the change is insignificant. Increase the alarm deadband to alleviate this problem.

For the shared variable to trigger an alarm, the shared variable value must exceed the exact alarm value (in the previous example, 40). However, to be considered normal again, it must deviate from the alarm value by an amount greater than the alarm deadband. For example, if the range is 0 to 40 °C, an alarm deadband of 1.0%, or one degree Celsius, eliminates unnecessary events.

The following table shows examples of events with alarm deadband set to 1.0%:

Time Value Change Alarm Condition
9:15:05 41.1 Yes HI
9:15:10 39.9 No HI
9:15:15 40.1 No HI
9:15:20 38.5 Yes None

Setting Alarms for Boolean Shared Variables

You can configure Boolean shared variables to have two alarm states—either the shared variable is in alarm or it is not. You can determine whether a Boolean shared variable is in alarm when it is ON (high) or OFF (low).

You also can specify a bad status alarm.

Setting Alarms for U32 Bit Field Shared Variables

You can configure U32 bit field shared variables to have one of two types of alarms — either any or all of its bits are in the alarm state.

You also can specify a bad status alarm.

Setting Alarms for All Other Shared Variable Data Types

You can specify a bad status alarm for all other types of shared variables.