In a petrochemical plant operating near full capacity, the balance of process variables determines not only product quality, but the integrity of the entire facility.
During a routine shift, the flow control loop of a feed line carrying highly corrosive reagents begins to display minor oscillations. In the control room, operators observe only the behavior of the process variable transmitted by the current loop. To the control system, the deviations appear to be ordinary variations, easily compensated for by the flow-regulating valve.
What the team cannot see is what is happening in the field: the valve’s smart positioner is undergoing progressive wear on its internal components, alongside an air line leak. The device features self-diagnostic technology and is actively generating alarms for an impending failure, but this information remains inaccessible to higher-level systems. The analog signal transmits a single variable, keeping the health diagnostics of the asset isolated on the shop floor.
The consequence strikes in the middle of the night: an abrupt valve lockup halts the reagent flow. The automatic safety interlock triggers a total emergency shutdown. The forced shutdown causes a ripple effect, thermal decompression, safety gas flaring, and the mobilization of maintenance teams to replace the actuator under high-risk operational conditions.
In this article, you will discover how to prevent situations like this from ever happening in your operation.
The challenge of integrating field instruments in Industry 4.0
The reason why the plant’s maintenance team failed to predict the valve failure lies in a technological problem that affects more than 80% of legacy industrial facilities worldwide: isolated data.
Although over 42% of unplanned downtime occurs due to direct equipment failures, and approximately 60% of these failures could be prevented through predictive maintenance strategies, most of the necessary diagnostic information usually remains inaccessible at the physical layer of the instrumentation network. In other words, the obstacle is not a lack of data, but the inability to access it.
The HART protocol (Highway Addressable Remote Transducer) was developed precisely to overcome this barrier. It uses Frequency Shift Keying (FSK) modulation to superimpose a digital signal onto traditional analog 4-20 mA current loops, using 1,200 Hz for the binary value “1” and 2,200 Hz for “0”. This bidirectional architecture allows TAG information, calibration data, and self-diagnostics to travel simultaneously with the analog control signal without degrading or introducing noise to the primary measurement.
The bottleneck arises because conventional PLC input and output (I/O) modules were designed for dedicated analog signals. They only read the continuous average 4-20 mA current, using a low-pass filter that discards the high-frequency FSK signal carrying the instruments’ digital diagnostics.
For a long time, extracting HART protocol information in traditional topologies required a heavy, costly infrastructure: large additional racks of dedicated multiplexer hardware, RS485-to-Ethernet TCP/IP signal converters, and extensive secondary analog wiring runs within control panels. In addition to high acquisition and installation costs, this required substantial engineering effort for configuration and commissioning, turning diagnostic data extraction into a project of its own, often unfeasible for medium-sized plants.
Below, see illustrated the difference in data capacity and diagnostic accessibility between traditional field connection solutions and digital networks:
| Communication type | Variables per physical channel | Data rate | Diagnostic level | Risk of downtime due to failure |
| 4-20mA current loop | Only one primary process variable. | Without digital transmission (continuous analog signal). | Low: limited to open-loop current information, requiring manual field testing for advanced diagnostics. | Very high; without wear detection, sensor failures or lockups occur. |
| HART | Analog PV + up to 4 secondary digital variables via periodic commands. | 1,200 bps using analog FSK modulation. | Medium: restricted to proprietary software or local tools. | Medium-High; although diagnostic data exists, extraction is slow and delays decision-making. |
| Industrial Ethernet network (HART-IP / APL) | Unlimited process variables, advanced predictive diagnostics, NAMUR status, and calibration. | Communication with the same instrument remains unchanged. | Comprehensive (if the data is buffered): delivered via open IP networks. | Very low; real-time detection and automated predictive alerts that reduce failures by up to 75%. |
The transition to monitoring via HART-IP protocol
With the advance of connectivity driven by Industry 4.0, the barrier between shop-floor automation and supervisory systems had to be broken down. In this context, the open HART-IP specification has become the benchmark technology for connecting field instruments to modern asset management systems.
By encapsulating the structured commands of the HART protocol directly into IP packets, HART-IP eliminates the need for proprietary physical converters between the asset management layer and the automation system. From the controller, diagnostic information travels over the plant’s existing Ethernet infrastructure straight to supervisory systems, without requiring dedicated secondary cabling. This unlocks new opportunities for optimization and reliability:
Automated commissioning: The initial setup of smart transmitters in the field, which previously took days of manual labor, is reduced to minutes through remote batch parameterization over the IP network.
Valve and actuator health monitoring: Remote analytical routines, such as Partial Stroke Testing (PST), can be performed without interfering with feed lines or exposing field teams to high-risk areas.
Integrated asset management: Systems directly access secondary variables and event logs generated by field sensors, allowing maintenance to be scheduled based on the operational history and wear rate of each individual asset.
Built-in cybersecurity: Aligned with IEC 62443 standard guidelines, the HART-IP protocol incorporates TLS/DTLS encryption, integrated audit logs, and event monitoring via syslog, ensuring that expanding network connectivity does not compromise operational security.
Practices to eliminate bottlenecks in industrial networks
Ensuring that predictive diagnostics and secondary variables reach supervisory systems with integrity and zero delays requires strategy. Simply digitizing analog information does not eliminate the challenges of data packet collisions, electromagnetic noise, or mechanical failures in communication lines. To achieve this, it is necessary to implement key practices, such as:
Implementing proper access management and master contention control: a conventional 4-20 mA current loop natively supports only two HART masters in parallel: the primary master, typically the PLC responsible for real-time control, and the secondary master, such as asset management software or calibration handheld tools. When multiple hosts or scanning buses attempt to access the same instrument without order, digital data collisions occur. Secondary diagnostic variables, such as travel percentage or internal device temperature, are the first to be corrupted or lost, as I/O card priority drivers prioritize primary process measurements via short commands over longer diagnostic requests.
Intelligent selection of the data transport protocol: the HART-IP protocol operates natively over structured IP networks, using TCP or UDP transport. In critical applications with zero tolerance for failure or missing packets, TCP is the right choice. Its built-in handshake mechanisms, packet sequencing, and automatic retransmission of corrupted packets ensure that any dropped data is resent transparently to the receiving system, preserving the logical integrity of the predictive database.
Investing in protection against electromagnetic interference: industrial environments are harsh by nature: variable frequency drives (VFDs), reactors, and large-scale motors generate intense electromagnetic interference (EMI). Analog or Ethernet cables with weak shielding or improper balancing suffer from noise that corrupts isolated communication bits, invalidating entire data packets at the receiver. Robust mitigation requires twisted-pair cables with individual shielding, single-ended shield grounding to prevent ground loops, and proper physical separation from high-power cabling.
Designing redundant networks for physical resilience: continuous communication availability is ensured by designing high-speed ring networks, such as Device Level Ring (DLR) or convergent Ethernet ring topologies. In these architectures, an accidental cable break in the field does not disrupt HART-IP traffic: switching to secondary logical routes occurs within milliseconds, without dropping a single diagnostic record.
How the Altus HART-IP gateway solution works
It is worth noting that, in practice, the vast majority of field instrumentation installed today remains wired in conventional 4-20 mA. The transition to HART-IP typically does not replace this wiring; instead, it takes place at the controller or dedicated gateway level, which consolidates HART data coming from the field bus and exposes it over an IP network to asset management systems.
Faced with the losses caused by unexpected downtime of control valves and the growing need to extract diagnostic telemetry from field instrumentation, such as in the scenario mentioned at the beginning of this article, Altus developed a targeted answer to this challenge: the AX9000 HART-IP Gateway, integrated with ArchiteX software.
The AX9000 is a software-based solution designed to establish transparent connectivity between HART devices and asset management/remote diagnostic systems operating over the HART-IP protocol. Unlike traditional market approaches, it requires no additional hardware inside the electrical automation panel, running directly on a dedicated physical machine or virtualized server connected to the CPU’s industrial Ethernet network. This eliminates costs associated with extra cables, protective fuses, and panel assembly labor.
In the Altus architecture, HART instruments remain connected via input and output modules (such as the NX6014 and NX6134) to PROFIBUS remote I/O units (NX5110, NX5210, PO5064, PO5065), which in turn communicate with the CPU. From the CPU, via the existing Ethernet network, the AX9000 makes HART data available in HART-IP format to asset management and SCADA systems.
The gateway is natively bundled with the installer for ArchiteX (AX8500), offering access to a centralized suite of tools for configuration, remote diagnostics, and monitoring of multi-vendor HART instruments, all from a single web platform.
Software architecture and intuitive interface
The AX9000 features an integrated web-based graphical interface, accessible from any browser securely connected to the plant’s internal network. From this interface, users can configure and monitor the entire data ecosystem across four functional tabs:
Information: Displays the active gateway version, real-time system operating status, and the exact count of HART devices identified on field networks.
Configuration: Brings together server logical parameters, including the definition of the Long Tag transmitted to client software, enablement of logging and alarm-sending services to remote Syslog servers, and license key management.
HART: Environment where the topology of PROFIBUS remote units and analog I/O modules hosting smart field instrumentation is configured. The gateway allows for manual addition of these remote units and control cards or the activation of the Topology Scan feature, which automatically scans all HART channel cards and remote units configured in the plant’s control project, eliminating manual mapping errors.
Status: A comprehensive dashboard showing communication status, manufacturer, model, and HART protocol version for each instrument detected on the bus. From this screen, users can trigger manual scans or configure automatic periodic scanning routines at programmable intervals, ensuring continuous updates of predictive variables in supervisory asset management systems.
In addition, AX9000 licensing is digitally activated and scaled on demand: additional licenses are acquired exclusively for the number of instruments connected and mapped in the gateway’s active topology, avoiding Capex waste.
By eliminating the need for secondary analog racks, the solution reduces plant expansion complexity and unifies diagnostic communication over high-reliability networks. This type of architecture represents the ideal framework for industries seeking to eradicate unplanned downtime and maximize the availability of their critical assets in the era of smart automation.
