Alarm trips: the ups and downs - Part 2

Other types of alarm

Apart from high and low limit alarms, there are other functions that alarm trips can be used for:

• Rate-of-change alarm: Used to detect changes in the measured value in units per minute or second, a rate of change alarm monitors an input for a change in value with respect to time (Figure 1). The alarm is set to trip when the input rate-of-change exceeds a user-selected rate (delta) based on a change in value (ΔT) over a user-selected time period (Δt).

• Input fault alarm: On some alarm trips, you can set one or more of the relays to trip when an input is interrupted, such as in the instance of a sensor break. This provides an alert of a non-critical sensor break without causing a costly false shutdown.

• Self-diagnostic alarm: Some limit alarm trips continuously monitor their own status during operation and trip if they are not operating properly.

• Average and differential alarms: An average limit alarm trips when the average of two or three input signals exceeds a pre-selected high or low trip point (Figure 2). A differential alarm trips when the difference between two input signals, such as two RTD temperature sensors, exceeds a specific value.

Alarm trip relay responses

Normally open and normally closed

Figure 1: Rate-of-change alarm trip.

Figure 2: Averaging alarm trip.

There are three common types of alarm relay configurations: single-pole/single-throw; single-pole/double-throw; and double-pole/double-throw.

• Single-pole, single-throw (SPST): An SPST has one pole (Figure 3a). When the contact closes, it allows current to flow across the relay. If this relay is normally-open (NO), current only flows when the contact trips (energised). If the contact is configured normally-closed (NC), current will flow until the alarm trips (energises). The choice of normally-open (NO) or normally-closed (NC) is typically selectable.

 
Figure 3: a) Single-pole, single-throw (SPST) relay shown with NO contact de-energised; b) Single-pole, double-throw (SPDT) relay contacts de-energised.

Configuring an alarm trip as either failsafe or non-failsafe is a primary safety consideration. In a safety application, the foremost concern should be the alarm trip’s action in the case of failure. An alarm trip with a relay that de-energises if the input signal exceeds the trip point is called failsafe (Figure 4). This unit’s relay is energised in the normal operating condition. As a result, should the power fail, this unit’s relay operates as if it were in the alarm condition. Failsafe relay action is chosen for the vast majority of alarming applications.

The other relay action is non-failsafe. This unit’s relay is de-energised when the input signal is in the normal condition (Figure 5) and energised when an alarm occurs. In this configuration, the alarm trip will not provide a warning if there is a power failure. Should a loss of power and alarm condition coincide, the alarm would go undetected.

Figure 4: A failsafe alarm trip is energised when it is not in the alarm state and de-energised when in alarm state.

Figure 5: A non-failsafe alarm is de-energised when in normal state and energised when in alarm state.

If the SPDT relay is failsafe, then when the trip point is exceeded, the relay de-energises and sends the contact from NO to NC (Figure 4), turning on the light by completing the circuit between the NC and C terminals. In this configuration, the light needs to be wired to the NC side of the contact.

Setting the deadband

The alarm trip fires its relay at the trip point and the relay resets when the process variable reaches the deadband point. Without deadband, if the process variable was hovering and cycling above or below the trip point, the relay would be chattering on and off, leading to premature failure. By setting the deadband just one or two percent away from the trip point, you can avoid excessive relay wear (Figure 6).

 
Figure 6: A deadband reduces relay chatter.

Contact ratings and precautions

The contact rating of relays used in alarm trips ranges from 1-10 A. A typical annunciator requires only a 1 A relay, while an electrical motor commonly requires a 5 A relay. For an alarm trip to control a higher amperage device, such as a pump, an interposing relay can be used.

To avoid needlessly damaging relays, two precautions must be taken. First, never operate a contact higher than its rating, even if it is momentarily. The rating of the alarm trip’s relay should meet or exceed the device it controls to ensure reliable operations. Second, consider the implication of the load’s behaviour. Capacitive loads create inrush current at start-up that can damage a relay contact, while the arcing created by an inductive load can vapourise a relay contact. Motor loads can have inrush currents five to six times normal run current.

Time delay

In many applications, a momentary over-range signal may not warrant an alarm trip. Some alarm trips can be set with an alarm response time delay that stops the alarm from going into an alarm condition unless the trip point has been exceeded for a specific time period (Figure 7). This can be used to stop false or premature alarms.

 
Figure 7: Alarm time delay stops false or premature alarms.

 
Figure 8: Alarm trip providing 24 VDC power to the loop.

Transmitter excitation

Some limit alarm trips offer the advantage of being able to provide 24 VDC power to a 2-wire (loop-powered) transmitter (Figure 8). This saves the cost of specifying and installing an additional instrument power supply.

Avoid nuisance trips: 2-out-of-3 voting

Some processes are simply too important to rely on a single alarm trip to make a decision. For these, limit alarm trips can be used in a voting strategy.

For example, one plant engineer was using three temperature sensors to monitor the burn-off flame of an emissions flare stack. However, when the wind blew, the flame leaning away from the stack gave a false output signal. The solution was to change the strategy to rely on low readings from two sensors to indicate no flame in a 2-out-of-3 voting scheme. This ladder rung approach creates a ‘flame out circuit’ only in the event that two of the three alarms are tripped. Using an alarm time delay with this strategy will also help prevent false trips (Figure 9).

 
Figure 9: Alarm trips in a 2-out-of-3 voting scheme.