Secondary HMIs capture data, help integrate, manage

Defining a secondary human-machine interface for automation and process controls can decrease the need for manual controls data capture, increase automation integration, and improve management of related processes. Smart-enabled devices connect sensors to signals to master controls, while dashboards parse and elevate meaningful data. See 7 control loop tests and 4 ways a measurement has value.

By Leah Friberg October 30, 2014

Outside the smart-device layer of automated reporting exists a rich secondary layer of process information that is now at our fingertips, if we can just integrate it into established controls data-management. Seen as a progression, vs. a revolution, the industrial Internet of things currently improves the most important aspects of production. Smart-enabled devices connect sensors to signals to master controls, while dashboards parse and elevate meaningful data pivots to human decision makers, who, in turn, look for indications of process irregularity.

Outside of that critical layer lies a second tier of manually collected data, supporting:

1) Devices that aren’t smart-enabled, and yet still affect overall productivity

2) Smart infrastructure reliability; that is, manual field inspection and correction of sensors, networks, and so forth. 

Inaccessible device health data

The percentage of devices within a plant that fall outside the "wired" environment varies by process criticality, but is thought to be very high on average, around 70%. Seen in reverse, the high-priority devices that are wired represent a small percentage of the total number of things that "go wrong" in a plant. Bluntly stated, there’s an enormous amount of device health data that control teams care about but is not readily accessible.

Parameters that instrumentation technicians collect by hand on nonwired devices are similar to the automated feed. The difference is that the once they’re used for the task at hand, data points are often discarded.

That’s changing. Handheld test tools now communicate wirelessly to smart device apps, which in turn save information to cloud databases for management and access. All of a sudden, we’re looking at that 70% of unaddressed equipment in a different light. The data on this equipment is still collected by humans, but now it has a meaningful interface. The process is a unique mix of portable and managed, allowing us to leverage the more ad-hoc nature of handheld measurement applications, skirting the restrictions of a fixed-mount, networked system while still layering up to the control board. 

Secondary data: 7 control loop tests

The next most critical layer of measurement data is on the control loop. If the whole is only as healthy as the sum of its parts, the control board depends on humans to manually commission, inspect, verify, troubleshoot, and calibrate the dc voltage and mA control signals upon which it relies.

Among control loop test procedures (as suggested in Figure 1) are these seven:

1) Testing remote indicators or HMIs to validate the quality of the mA signal against an input to the control system, on both an ad-hoc or annual basis

2) Testing an entire loop, with test tools connected at the transmitter, power supply, remote indicator, and control element/valve, validating the signal is correct (otherwise known as "shooting the loop")

3) Conducting regulation-required inspections (pharmaceutical, petroleum) that must be documented

Additional common troubleshooting applications include:

4) Tracing control loop failures due to intermittent terminations, blown fuses, wiring mistakes, over-pressurized transmitters, faulty power supplies, bad temperature sensors, and so forth

5) Verifying aberrations in control board readings-a tank that should only be at 70% but reads 90% needs to be checked, and the control operator needs to see the signal originating from the local transmitted while watching the pressure valve data on the board.

6) Simultaneously testing input/output dc voltage

7) Logging multiple control test points simultaneously during a line operation error.

In this kind of human-machine interface, the technician attaches a wireless V dc or mA meter to the device, enables the radio, and transmits the reading to a smartphone app. The technician can take an instant measurement and save it, or monitor the measurement while conducting a test and record the entire sequence. In some instances, the technician may also choose to log over a longer period of time. Multiple measurement tools can be connected at once, as with the comprehensive control-loop test example above, with all measurements transferred to the app.

4 ways a measurement has value

Simply saving data doesn’t inherently add value unless the measurement is:

1) A key indicator of system performance, such as % span

2) Tagged according to the piece of equipment and date

3) Correlated against other data points, such as previous measurements or parallel data streams (the HMI reading compared to the source)

4) Stored where it can be accessed by the extended controls team.

It’s a shift in logic that the regulation-driven plants have already made. In those environments, it’s not enough for an individual technician to take a measurement, finish the job, and walk away; the measurement information can’t be a one-person conversation. Recording those measurements does more than document the accuracy of the system; having that data "on file" also increases the speed, accuracy, and confidence of the team. 

Secondary HMI advantages

All technicians on the team (not just the control board) have access to the universal equipment logs, with all baseline data for every control loop and device. Advantages include:

  • One universal record-keeping method for the plant floor substantially improves communication and consistency within the team (between shifts, skill levels, sites). 
  • No labor-intensive handwritten or data-entered record keeping-and fewer errors.
  • Communication between floor and control board is substantially faster when the board can share the technician’s view of the measurement, as it is happening.
  • Approval cycles are shortened when the supervisor has all the data on hand, instantly.
  • When troubleshooting a failure, starting with a baseline data point for comparison to current state often shortens the time to get to root cause.
  • Overall improved device reliability. 

Management typically determines device priority, parameters to gather, and the role of shared data in the instrumentation maintenance process. The equipment log organization methodology usually maps to the existing naming protocol within the plant, where the holy grail is a 1:1 match, aligning the manually collected data with the automated. (See Figure 2.)

– Leah Friberg is education and public affairs manager, Fluke; edited by Mark T. Hoske, content manager, CFE Media, Control Engineering,


Key concepts

  • Secondary human machine interfaces can capture data, help integrate, manage.
  • Handheld HMIs can increase automation integration, and improve management of related processes.
  • Smart-enabled devices connect sensors to signals to master controls, while dashboards parse and elevate meaningful data. 

Consider this

How much more effective could plant personnel be with a real-time handheld interface into processes, that can transfer key data into other systems?

ONLINE extra November has more information and links to related coverage.