What is OPC A&E?

Before OPC A&E, accessing alarm data from a DCS or PLC required two things that vary completely between vendors: proprietary tools or APIs for accessing the alarm system, and proprietary message formats defining how alarm information was structured. A third-party alarm management application, sequence-of-events recorder, or compliance audit tool had to be custom-written against each control system vendor's implementation. OPC A&E eliminated that by defining a common alarm message format and a common subscription mechanism.

OPC A&E is the third member of the OPC Classic family. OPC DA handles real-time process values. OPC HDA handles archived historical data. OPC A&E handles alarms and events as they happen, plus the ability to query recent event history and acknowledge outstanding conditions. All three share the same COM/DCOM foundation. TOP Server implements all three on a single installation alongside OPC UA.

Engineers familiar with OPC DA sometimes try to use tag subscriptions to implement alarm detection: subscribe to a tag, watch for values that exceed a threshold, and treat that as an alarm. This works in simple cases but misses what makes OPC A&E genuinely useful.

OPC DA delivers values. OPC A&E delivers events, and an event is more than a value crossing a threshold. An event has a type, a source, an area, a severity, a timestamp, a category, and for condition-based alarms, a state machine with acknowledgement requirements. An operator looking at an alarm display needs to know not just that a value is high, but whether the alarm is currently active, whether it has been acknowledged, by whom and when, and what the sub-condition is (HIGH vs HIGH-HIGH, for example). None of that state management is in OPC DA.

More importantly, OPC A&E receives alarms that originate inside the DCS or PLC itself, not inferred by an external application watching tag values. A DCS-generated alarm carries the DCS's alarm logic, deadbanding, shelving state, and priority classification. An alarm inferred by polling a DA tag does not.

OPC A&E defines three distinct event types. Each serves a different purpose and carries different fields in the event message. Understanding the distinction matters when configuring subscriptions and when evaluating what a particular DCS or SCADA system exposes through its OPC A&E server.

Condition events are the foundation of alarm management in OPC A&E. Unlike simple events, a condition has state that persists and changes over time. The OPC A&E specification defines this state machine precisely, and each state change generates a new event notification to all subscribers. This is what allows an alarm management application to display exactly which alarms are currently active, which are unacknowledged, and what the history of each alarm looks like.

To receive alarm and event notifications, an OPC A&E client creates a subscription with the server. The subscription defines which events the client wants to receive and how they should be delivered. Once the subscription is active, the server pushes notifications to the client via asynchronous callbacks as events occur, with no polling required.

OPC A&E organizes the alarm namespace into three hierarchical concepts that define where an event comes from and how clients can filter it.

An area is a logical grouping of equipment within the OPC A&E server's namespace. Areas map to physical or functional plant areas: a reactor building, a distillation column train, a cooling system. Areas are hierarchical and can be nested. A client subscribing to a specific area receives events only from the equipment within that area. This is how a control room operator display for the west production line shows only west production line alarms, while a plant-wide historian captures all areas simultaneously.

A source is a specific piece of equipment or system within an area that generates events. Sources are identified by name and sit within an area hierarchy. When an event fires, it carries the source identifier so the client always knows exactly which piece of equipment the alarm originated from. In TOP Server's OPC A&E plug-in, sources map directly to the tags and devices already configured in TOP Server's driver configuration.

A category is a server-defined grouping of related event types with a common set of attributes. For example, a "temperature alarms" category might include all high and low temperature conditions across the plant and define that temperature-specific attributes (engineering units, setpoint value) are carried in every event of that category. A "batch tracking" category might define batch ID and recipe version as standard attributes. Categories are specific to each OPC A&E server's configuration.

The TOP Server OPC A&E plug-in enables engineers to define custom alarm conditions directly against any tag already configured in TOP Server, without needing a separate alarm management system. Because TOP Server connects to 140+ device types, the A&E plug-in effectively brings standardized alarm delivery to any device that TOP Server can communicate with, including legacy hardware that has no native alarm export capability.

OPC UA carries alarm and event capability in its Alarms and Conditions (A&C) specification, defined in OPC UA Part 9. A&C is the functional successor to OPC A&E and carries the same core concepts forward: condition state machines, acknowledgement, subscriptions, filtering. The differences are architectural.