Support-focused enterprise controls: trigger-first design strategy

A trigger-first design strategy enables manufacturers to obtain structured and organized control applications from many independent machine and conveyor suppliers. This method creates a common path for designs to trigger applications while enabling a standard way to develop rule-, template-, and table-based control applications.

By Daniel B. Cardinal March 18, 2016

A trigger-first design strategy means control system designers must develop movement detection and ancillary trigger circuits as the foundation for all controller applications. This means designers group, assign, and isolate trigger circuits for each control point or station. Trigger circuit isolation makes it easy for support personnel, controls integrators, and future programmers to access, understand, and change logic circuits. Isolation and subsequent recognition will provide the foundational basis for developing most latch/unlatch circuit signals for control applications. This trigger-first strategy eliminates the need for different designers to develop redundant and possibly less reliable instruction-based triggers for the same sets of movement and ancillary events.

Redundant triggers are a result of control and information system designers unnecessarily adding sensors and/or triggers to a control system design. The elimination of redundant circuits happens when manufacturers enforce specifications that dictate all trigger circuits shall reside in a common trigger routine. Eliminating randomly programmed instruction-based triggers means prohibiting their use. Exceptions are possible with advanced approval. Through specifications, manufacturers can prohibit the everyday use of instruction-based and coil-blocked triggers. These less reliable triggers cannot exist in a control application without approval documentation describing a need for the exception. Specifications should also dictate that programmers include approval information in a comment area above all exception circuits.

A trigger-first design strategy enables manufacturers to obtain structured and organized control applications from many independent machine and conveyor suppliers. The design process enables a manufacturer to realize that most latch/unlatch circuit signals are set or reset via of a movement detection or subsequent ancillary trigger. This method creates a common method for designs to trigger applications while enabling a standard way to develop rule-, template-, and table-based control applications.

Mechanical engineers were the first to recognize that physically locating and positioning objects after they stop at a station is the key to most designs. Many control system designers have gone along with this premise because presence detection is fundamental to most machines that work on stationary parts. However, if designers re-evaluate the use and effect of movement detection on all aspects of control and information system applications, they will conclude that trigger designs must come first. Any objective examination process makes it clear that movement detection has a significantly greater impact on control and system applications. For example, movement detection enables the development of the following:

  1. Cycle-start and cycle-complete anti-repeat circuits
  2. Anti-repeat pushbutton circuits
  3. Object-occupied memory circuits
  4. Blocking circuits that prevent object collisions
  5. Blocking circuits that prevent machine-to-object collisions or crashes
  6. Shift-register index signals keep data synchronized with moving objects
  7. Station part or object counting circuits 
  8. Station cycle-time collection circuits
  9. Station blocked waiting for downstream process timing circuits
  10. Station starved waiting for upstream process timing circuits
  11. Elemental movement timing circuits for mechanisms and objects
  12. Arming ancillary trigger circuits
  13. Reader application trigger circuits
  14. System-destined messaging circuits
  15. Critical process information used by buffering circuits
  16. Application enabled or reset circuits.

It becomes obvious that movement detection is the cornerstone or catalyst for supporting and spawning a multitude of control and information system applications. Enabling an index signal allows designers to develop shift-register applications. The development of shift-registers enables other applications to synchronize, compare, collect, and send process information. Multiple shift-registers for merging process lines enable tracking and monitoring applications to minimize production losses caused by an out-of-sequence or out-of-balance flow of parts. The development of shift-registers also enables the development of system-level tracking applications.

With all the known movement detection trigger uses, why does it remain as a hidden feature of control system designs? Is it possible that it remains hidden because there are many machine-specific variations? Perhaps it is because management emphasizes the need for designers to be focused on controller features not process similarities. Regardless, this hidden feature forces controls integrators to develop redundant trigger circuits. A signal-free, one-shot circuit is another contributing factor. Why does anyone have to add redundant sensors? Chaos-caused confusion is normally a result of something moving to adversely affect each independently triggered application.

The bigger question is how or whether manufacturers will correct this situation. The answer to all these questions is quite simple. Recognize the problem and put controls in place to make triggers readily available to all designers, programmers, integrators, and support personnel. This starts by controlling how mechanical engineers actuate and locate sensors through machine design specifications. Equally important is bringing all trigger logic circuits to the forefront of control system designs. This way everyone will recognize and understand how they work, the interdependencies that exist between applications, and the expected common outcomes when something moves.

Recognizing the importance of movement detection should be obvious. Developing and enforcing circuit design specifications will be more difficult because each machine supplier will likely object. Objections will revolve around increasing costs associated with changing their standard hardware designs to include additional sensors and input hardware. Regardless, the idea is to understand these costs and provide justification for change. For example, the simple two-step form of movement detection is easier, but it requires two sensors. Some low-cost conveyor companies use release-based movement detection. This cheaper method allows them to escape the design, purchase, install, and wiring costs of one extra sensor per station. This means manufacturers that use release-based designs must knowingly accept the burden of built-in costs of foreseen anomalies.

Before, anomaly costs were unexpected because they did not understand the cause. Armed with this information, manufacturers must recognize and associate anomaly costs with the poor electro-mechanical characteristics of movement detection. When specification analyses are complete, it is more important that everyone understands what they have or do not have and at what cost.

Another obstacle to recognizing the importance of movement detection is breaking down the organizational barriers that manufacturers have set up. These barriers usually separate controls design teams into groups based on machine types. Specifically, one group is responsible for conveyor processes while another group is responsible for station processes. This situation makes overall designs look more chaotic because support personnel do not understand the fundamental reasons for separate groups. The following definitions provide insight into the types of control applications each group designs and implements:

  • Conveyor controls monitors the movements of objects as they travel from station to station.
  • Machine controls are for repetitive movements of mechanisms and processes after objects stop at stations.

The transition between conveyor and machine control is difficult for many designers. Many manufacturers see some obvious differences and use them to justify the need to divide personnel into two groups. A conveyor control group is responsible for conveyor designs, whereas a machine control group oversees static station processes. Design differences revolve around the differing control domains. A conveyor group works in a domain where travel speeds, times, distances, and machine clearances are critical to controller applications, while the machine group works in a domain where cycle times and mechanical sequences are important to applications.

To recognize design similarities between conveyor and machine control systems strategists must separate the qualities that make them different. First, manufacturers must specify that all trigger circuits must come to the forefront of each design. This means specifying the placement of trigger circuits into special routines that are easily viewable and accessible by all design teams. Second, strategists must recognize that conveyor control applications need object-moving signals that enable and disable output devices when objects enter or exit stations. Third, strategists must recognize that machine designs need to contend with the sequential movements of mechanisms and objects at process stations. This means designs need cycle-complete or cycle-start anti-repeat circuits that ensure station processes cycle once per object.

These special anti-repeat signals also provide machine control applications with a way to determine whether to move mechanisms forward or backward in a sequence when machines are physically in a duplicated state defined only by sensor input signals. Fourth, strategists need to recognize the differing ways designers model their individual processes. Conveyor designers use sensor-activation charts that depict a dimensional domain. These charts show the locations of sensors that are critical to moving objects. Machine designers use sequence-of-operation charts that depict a time domain. A time-domain chart shows the individual mechanism movement times that are critical to processing an object that has stopped at a station. In contrast, conveyor designers model the physical locations of events, whereas machine designers model event times.

Mechanical designers create sensor-activation charts using bars to represent activated sensors in a dimensional domain. Specifically, the length of each bar represents the physical length of the actuator. Deactivation bars represent the distance an object travels between two deactivated sensors. Conveyor designers use sensor-activation charts to decide the number and location of possible station triggers. The rising and falling edge of sensor transitions represent potential physical trigger points. Many designs artificially delay enabling triggers, thus changing where triggers activate when objects move. The following definitions describe how conveyor control applications produce movement detection triggers and how machine control applications produce ancillary triggers.

  • Object sensor is a device that activates and deactivates a controller input signal to indicate an object is moving through a station.
  • Process sensor is a device that activates and deactivates a controller input signal to indicate the position of a mechanism.
  • Position trigger is a signal enabled using the rising edge activation of an object sensor.
  • Moving-off trigger is a signal armed by the activation of a position trigger and is directly or indirectly enabled by the falling edge deactivation of an object sensor.
  • Process trigger is a signal armed by the direct or indirect activation of a position or moving trigger and enabled by the activation or deactivation of a process sensor.
  • Release-now trigger is a signal armed by an in-position trigger and enabled when the process cycle completes, and the process machine is physically clear to release the object.

Position, moving-off, and process triggers are the descriptive names assigned to otherwise generic movement detection and ancillary triggers. These triggers fire when object or process sensors activate or deactivate. The release-now trigger fires the moment a station’s control application is ready to release a part from a station and before the object starts to leave the station.

Figure 1 shows the conversion to the step domain for an object entering and exiting the second conveyor station. The individual overlaps represent the simultaneous activation of object sensors. Recognizing the overlap is important because the next step in a trigger-first design strategy involves converting activation bars from the dimensional domain to a step domain. In the dimensional domain, the length of a sensor-activation bar represents the length of the actuator. In the step domain, all object sensor-activation bars have incremental unit lengths based on activation bar overlaps. A bar one unit in length represents the activation and deactivation of one object sensor with no overlapped activations with another sensor.

A bar two unit in length represents the activation of one object sensor whose activation overlaps with another sensor and that other sensor does not have an activation overlap with any other sensors. A bar three unit in length represents a sensor that overlaps at different points with two other object sensors. Some sensor activation bars have extra extensions that represent an object stopped waiting for mechanism-based machine control application to run in a time domain. Furthermore, all descriptive properties stay with each bar converted to the step domain.

Specifying control programs

Figure 1 shows a trigger-firing chart that uses upward pointing arrows. The chart includes an underlying set of step-domain activation bars to show the locations of station triggers. The arrows pointing up from the ends of sensor activation bars show the locations of triggers relative to sensors. The time box next to the in-position trigger represents a process activity that occurs over time after an object stops in the second station. The arrow pointing up from the right side of the time-domain box indicates a process trigger firing after the machine control process completes and mechanisms are in a clear position ready to release an object.

Manufacturers can specify machine control programs to use cycle-start or cycle-complete anti-repeat circuits. A cycle-start design relies on the arrival of a new object at the entrance of a station to set an anti-repeat signal. These designs rely on the departure of an object from a station to reset the signal. If manufacturers specify cycle-complete circuits, designers can begin to assemble trigger logic for the upward pointing arrows shown in Figure 1. The arrows pointing up from the left side or rising edge of bars represent position triggers, whereas the arrows pointing up from the right side or falling edge bars represent moving-off triggers. Lastly, the one arrow pointing up from the right side of the time-domain box represents a process trigger that also occurs at the stop position.

Figure 2 shows the trigger circuits that arm and fire based solely on the sensed movement of objects. A manufacturer’s design specification can control other examinable conditions as needed to provide added trigger flexibility or rigidity. Maintenance mode information creates property field declared parameters for arrowed objects. If designers select maintenance mode, designers can add a normally closed maintenance mode contact to the trigger circuit. If support personnel need to access equipment and reposition station objects, no triggers will fire when in maintenance mode. This provides support personnel with the enhanced flexibility to reliably reposition objects, perform maintenance, and return objects to their original positions without affecting control applications.

When trigger circuit designs cannot reliably support a maintenance mode, manufacturers must absorb extra operational costs. These costs occur when designs force manufacturers to start production shifts early to ensure machine readiness after maintenance personnel move and reposition objects. In many instances, the extra start-up time allows maintenance personnel to debug those control applications affected by the unexpected firing of redundant triggers.

Most control system designers know to create a minimum of two base movement-detection triggers for each station. For a cycle-complete design, this means using an exit-position sensor and another sensor that activates before a moving object activates the exit sensor. This means uniquely identifying the preconditions needed to arm each trigger. They also need to assign a common arm signal that is off before the in-position trigger fires and turns on after it fires. Designers then use the enabled state of the same signal to fire the exit trigger that turns it off. Lastly, they need to select and assign a tag or name to each trigger’s firing signal.

A trigger-first strategy allows designers to use a systematic process of identifying examinable circuit conditions. For an exiting trigger, designers start by selecting a normally opened contact instruction for the exit sensor. If the exit sensor’s activation bar overlaps with one or more downstream sensor activation bars, the designer adds an equal number of opened contacts for each simultaneously actuated sensor. Next, the designer adds a normally closed contact for the first deactivated upstream sensors. For the in-station trigger, designers follow a similar process for identifying conditions. A designer selects a normally opened contact for the sensor that identifies the object in position at the station. If there are any other upstream-enabled sensors when the first sensor activates, the designer adds a normally opened contact for each simultaneously activated sensor. Next, the designer adds one normally closed contact for each deactivated downstream sensor ending with the exit-position sensor.

Figure 2 also shows the two gap-dependent, movement-detection trigger circuits extracted from the firing chart pictured in Figure 1. Notice how the base arming signal is set on when the first trigger fires and is reset after the second trigger fires.

Most manufacturers accept control applications from conveyor suppliers who do not follow a documented, systematic process for designing trigger circuits. This means they can expect to have a large percentage of control applications with reliable triggers and a small percentage with unreliable triggers. The small percentage of applications with less reliable triggers is likely to create the most confusion for support personnel and production loss downtime for the manufacturer.

To reduce the small percentage of unreliable trigger designs, manufacturers must produce a trigger compliance test. To do this, they must force suppliers to submit station-specific trigger firing diagrams for approval. The approval process means each submitted step domain diagram must match a standard activation pattern. An acceptable pattern means it must match an already approved, working, and reliable pattern. This does not ensure the suppliers will design trigger circuits correctly, but it does mean the needed electro-mechanical components will be correct.

To augment this approach, manufacturers need to specify that designers provide a spec sheet for each trigger. The information provided in the spec sheet will enable designers and/or support personnel to understand each type of trigger, the assigned one-shot signal name, its primary purpose, how it is armed, what is used to re-arm it, the need to have it set other arm signals, what timers are possibly needed, and what other special conditions will be programmed to enable it to fire. For example, if the spec sheet shows a requirement to create other arm signals, it specifies the number and names needed.

Figure 3 shows a circuit that creates 160 discrete trigger-arming signals. To do this, the circuit uses an exit-position trigger to load a fixed constant into 10- and 16-bit registers. This sets all 16 bits of each element to an enabled "on" state. For any ancillary, station-specific trigger circuits, designers must select one of the armed signals. When an ancillary trigger fires, the selected armed signal is programmed to go off to create the one-shot. In most cases, the configuration of sensors and the accompanying manufacturer’s design specification affects signal selection process. A cycle-complete, machine design strategy means this station’s exit trigger will set all arming signals for a given station. Similarly, a cycle-start strategy means an entering trigger or the previous station’s exiting trigger sets all arming signals.

Figure 4 shows an example of a position-trigger circuit that fires when an object activates a slow position sensor. Designers select an arming signal created by the circuit shown in Figure 3 as a condition to fire the trigger. Since the next trigger is a position trigger, the precondition is merely a normally opened contact for that sensor. Optionally, designers include a command signal that enables an external circuit to move an object forward towards the Slow Position sensor. The normally closed maintenance mode contact ensures the trigger will not fire while the application is in that mode. When the trigger fires, the circuit resets the slow position-armed signal, which prevents the repeat firing of the trigger.

If the next downstream trigger is a moving-off trigger, the trigger circuit includes supporting setup logic. The setup logic ensures a deactivating sensor reliably fires a moving-off trigger once. The exact configuration of the logic depends on the combination of spec sheet-defined, property fields associated with a specific trigger.

Figure 5 shows a moving-off exiting trigger with supporting setup logic. The logic includes a normally closed contact for the sensor expected to go off. The circuit also includes a normally opened contact that represents sensor power on. This condition ensures the trigger circuit will not inadvertently fire in the event of a power failure. The integrated verify sensor on signal confirms the exiting sensor toggled on while the movement command signal was on. A stabilized, de-bounced, sensor-off signal ensures the exiting sensor has been off for a predetermined amount of time. This filters out momentary sensor deactivations caused by a moving object’s actuator momentarily bouncing away from the sensor. A motor running feedback contact (not shown) provides extra security related to confirming that power is on and that the movement command is electrically enabling the motor to cause movement.

If a conveyor’s control application moves an object into a cycle-complete process station, the design needs a special release trigger. This release trigger identifies the instantaneous readiness of the station’s machine application to release an object from the station. The term "readiness," means the station’s tooling, equipment, and processes are complete and physically positioned to allow the object to have an unobstructed departure from the station.

Figure 6 shows an example of a release-now trigger circuit. The subsequent circuit examples in the next section of this article show how this type of trigger is critical to enabling a conveyor application to release an object from a station.

Daniel B. Cardinal works as an engineering consultant for Insyte Inc., implementing integrated scheduling and part identification applications in the automotive industry. Edited by Chris Vavra, production editor, CFE Media, Control Engineering, cvavra@cfemedia.com.

MORE ADVICE

Key Concepts

Trigger-first design strategy means control system designers must develop movement detection and ancillary trigger circuits as the foundation for all controller applications.

Manufacturers can specify machine control programs to use cycle-start or cycle-complete anti-repeat circuits.

Most manufacturers accept control applications from conveyor suppliers who do not follow a documented, systematic process for designing trigger circuits.

Consider this

What specific applications are trigger-first designs most useful for and what benefits can they provide?

ONLINE extra

See prior stories in this series by Daniel Cardinal linked below.