What Is the HART Protocol?

Back to Blog

How HART Rides on a 4-20 mA Loop

HART uses Frequency Shift Keying (FSK) to encode digital bits as small ±0.5 mA sine-wave signals at 1200 Hz and 2200 Hz superimposed on the 4-20 mA loop. Because the average of the digital signal is zero, it does not disturb the analog measurement. This clever design meant plants could adopt HART without rewiring: the analog value still drives the control loop while the digital layer adds smart-instrument features on top.

In its standard master/slave point-to-point mode, HART carries one 4-20 mA analog variable plus digital access to several more. A less common multidrop mode fixes the current at 4 mA and puts multiple devices on one pair, communicating purely digitally. WirelessHART (IEC 62591) extends the same command set over a self-organizing 2.4 GHz mesh for monitoring points where running new cable is impractical.

Getting HART Data into SCADA

The great weakness of traditional HART is that the rich digital data often goes unused. A control system reads only the 4-20 mA value into an analog input card and ignores the HART digital layer, so device diagnostics, secondary variables, and configuration sit stranded in the instrument. This is a real issue on oil and gas sites where multivariable transmitters (flow, pressure, temperature in one device) have far more to offer.

To capture it, plants use HART-enabled I/O cards, HART multiplexers, or WirelessHART gateways that read the digital variables and diagnostics and republish them over a routable protocol such as Modbus TCP or OPC UA. From there a SCADA or cloud platform can consume the full instrument picture. A cloud SCADA like Merobix reads those aggregated HART variables and diagnostics over Modbus TCP or OPC UA, so operators can trend secondary variables and see instrument health remotely rather than losing them at the analog input.

The Command Structure Behind the Signal

Everything a HART master does is a numbered command, and the commands come in three tiers. Universal commands are mandatory in every HART device - read the primary variable, read the loop current, read the device identity - so any master can interrogate any transmitter to a baseline level regardless of who made it. Common practice commands are optional but standardized, covering frequent tasks such as range changes and self-test. Device-specific commands are where a manufacturer exposes its own features, which is why full configuration requires the device's descriptor files even though basic reads never do.

The tiering matters for integration planning. If the monitoring goal is variables and health status across a mixed fleet, universal and common practice commands get you there without vendor-specific work. If the goal includes deep configuration or specialized diagnostics, the host or field tool needs device-specific support for each model on site. The full split, with examples of what lives in each tier, is covered in universal, common practice, and device-specific commands.

The Four Dynamic Variables and What to Trend

HART defines four dynamic variable slots - PV, SV, TV, and QV - that a device maps onto its measurements. A pressure transmitter might carry process pressure as PV and sensor temperature as SV; a multivariable flow device can serve flow, static pressure, and temperature from one loop. Which physical measurement lands in which slot is configurable in the device, so record the mapping in the instrument index or the trends will quietly lie to the next engineer. The difference between these slots and the larger set of underlying device variables is explained in device variables vs dynamic variables.

For SCADA, a short priority list covers most of the value: trend the dynamic variables, alarm on the status bit that flags a field-device malfunction, and watch for a loop current that is fixed or saturated, which means the analog value has stopped following the process. Those three catch the majority of silent instrument failures - a plugged impulse line, a degrading sensor, or a device left in a fixed-current test mode that someone forgot to clear after maintenance.

Field Tools and Loop Realities

A HART communicator or modem clips across the loop anywhere - at the transmitter terminals, in the marshalling cabinet, or at a dedicated test point - because the FSK signal rides the entire pair. The one wiring condition is enough resistance in the loop for the signal to develop; the HART specification sets the minimum, and a bench setup with a bare power supply and no resistor is the classic reason a perfectly healthy transmitter refuses to communicate. Routine calibration checks are done exactly this way, as described in trimming a transmitter with a HART communicator.

Two field cautions apply. First, on loops feeding a running control function, connect and disconnect tools deliberately and under the site's permit and management-of-change procedures - a slip that opens the loop drives the input to its failure state, and the process consequences depend entirely on what that input feeds. Second, HART allows a primary and a secondary master on the same loop, and both secondary masters and burst mode change loop timing, so coordinate with whoever owns the host system before enabling either on a live loop. Anything protective or safety-related stays with qualified instrument personnel following site procedure.

Frequently Asked Questions