NAMUR NE43 4–20 mA Signal Levels for Saturation and Faults

NAMUR Alarm and Saturation Ranges in 4–20 mA Loops

A 4–20 mA current loop is one of the most common analog signaling methods used between industrial field transmitters and control systems. A pressure, temperature, flow, level, position, or analytical transmitter converts the measured process variable into a loop current. The receiving device—such as a PLC analog input card, DCS input module, indicator, recorder, or signal conditioner—measures that current and converts it back into an engineering value.

In a conventional transmitter setup, 4 mA represents the configured lower range value, and 20 mA represents the configured upper range value. For example, a level transmitter calibrated from 0 to 10 m may output 4 mA at 0 m and 20 mA at 10 m. Values between those endpoints represent measured values within the configured span. In a linear application, 12 mA would correspond to the midpoint of the calibrated range.

The difficulty is that real process signals do not always stay neatly inside the calibrated measurement span, and transmitters can also fail. A process may go below the lower range value, exceed the upper range value, lose its sensor input, detect an internal electronics problem, or suffer a wiring fault. Without a consistent convention, the receiving control system may not know whether a current slightly outside the normal span is still a process measurement or a fault indication.

That is the purpose of NAMUR NE43 4–20 mA signal levels. The recommendation defines a practical separation between:

• normal measurement within the 4–20 mA span;
• underrange or overrange saturation, where the process is outside the configured range but the transmitter is still measuring;
• defined fault-current regions, where the transmitter indicates a detected device or signal failure;
• intermediate gap regions that should not be treated as normal measurements.

The result is a clearer analog signal language. A current just above 20 mA can indicate process overrange, while a current at or above the high fault threshold indicates a transmitter fault. Similarly, a current just below 4 mA can indicate process underrange, while a current at or below the low fault threshold indicates a fault condition.

What the NAMUR NE43 Recommendation Defines

NAMUR NE43 is commonly used in process instrumentation to standardize how a transmitter represents fault information on a 4–20 mA analog output. It is a recommendation rather than a law of physics or a universal guarantee across every instrument. The transmitter, the analog input hardware, and the control-system configuration must all support the interpretation for it to work correctly.

The key idea is that not every current outside 4–20 mA means the same thing. NE43 separates abnormal but still meaningful process conditions from device faults. A transmitter may be healthy while reporting that the measured process variable has moved below or above its calibrated range. That condition is saturation. By contrast, a transmitter may use internal diagnostics to report that it cannot provide a reliable measurement because of a sensor failure, electronics problem, converter fault, memory issue, or another detected device problem. That condition is a fault signal.

A practical interpretation of the commonly cited NE43 regions is shown below.

This table is useful because it shows why “below 4 mA” or “above 20 mA” is not specific enough. A value of 3.9 mA and a value of 3.5 mA should not be interpreted in the same way. Likewise, a value of 20.3 mA and a value of 21.5 mA have different meanings in a NE43-style configuration.

The normal measuring range remains 4–20 mA. The saturation regions extend just outside that normal span to preserve process information. The fault regions are farther out so that a control system can distinguish a severe diagnostic condition from a simple overrange or underrange event.

Out-of-Range Saturation Signals

Saturation occurs when the transmitter output moves slightly outside the 4–20 mA span while the transmitter is still representing a measured process condition. In this case, the instrument is not necessarily failed. It is telling the control system that the process variable has moved beyond the configured measuring range.

Consider a level transmitter calibrated so that 4 mA equals 0% level and 20 mA equals 100% level. If the vessel is overfilled and the actual level rises above the calibrated upper range value, the transmitter may increase its output above 20 mA. Under the NE43 convention, high-side saturation can extend up to 20.5 mA before the high fault region is reached. The receiving system can then recognize that the process is above the calibrated span, not simply assume that the transmitter has failed.

The same principle applies at the low end. If the measured variable falls below the lower range value, the transmitter may reduce its current below 4 mA. In the commonly cited NE43 interpretation, the low-side saturation region is from 3.8 mA up to just below 4.0 mA. This tells the control system that the process is below the calibrated range, while still keeping the signal separate from the downscale fault threshold at 3.6 mA and below.

This distinction matters in troubleshooting. A loop current of 3.9 mA may indicate an actual process underrange, a calibration issue, a zero shift, or another condition that still leaves the instrument producing a coherent analog output. It should not automatically be treated the same as a low fault current. A current at or below the configured downscale fault threshold, however, should be treated as a device or signal fault according to the system design.

Saturation is also important for process safety and operations. If a flow transmitter is calibrated for a normal operating range but the process briefly exceeds that range, the operator still benefits from knowing that the value is high. A saturated high output tells the system that the variable is above span, even if it can no longer provide a precise scaled measurement beyond that point.

The limitation is that saturation does not provide unlimited measurement capability. Once a transmitter reaches its saturation limit, such as 20.5 mA on the high side, it can no longer express how far beyond the configured range the process has moved using the analog current alone. The signal says “above range,” not “accurately measured at a specific value beyond range.” For applications where accurate measurement outside the normal operating range is required, the transmitter range or measurement technology should be reviewed rather than relying on saturation behavior.

Hardware Fault Current Levels

Modern smart transmitters often include diagnostics that can detect internal and external problems affecting measurement reliability. Examples include sensor element failure, open or shorted sensor leads, analog-to-digital converter problems, memory faults, electronics failures, configuration errors, or other diagnostic conditions depending on the device type.

When such a fault is detected, the transmitter may drive its analog output to a fail-safe current. NE43 is commonly associated with two fault thresholds:

• downscale fault: at or below 3.6 mA;
• upscale fault: at or above 21.0 mA.

These values are best understood as interpretation thresholds used to separate fault signaling from saturation. In practice, a transmitter may be configured for a specific fail-low or fail-high output value that lies beyond the corresponding threshold. The exact behavior depends on the instrument design and configuration.

Fail-safe output selection is often configurable. A transmitter may be set to fail upscale, fail downscale, or follow a manufacturer-defined alarm setting. The appropriate choice depends on the control strategy and the process consequence of a bad signal. For example, in one loop a high current may drive a controller output in a safer direction; in another loop, a low current may be more appropriate. The fail action should therefore be coordinated with the control loop, alarm logic, interlocks, and operating procedures.

Two-wire loop-powered transmitters require special attention. In a two-wire device, the same loop current both powers the transmitter and carries the measurement signal. If the fault current is driven too low, the instrument must still have enough electrical power to maintain the output state or communicate the diagnostic condition as designed. This is one reason low-current fault behavior is more constrained in loop-powered devices than in separately powered instruments.

Four-wire instruments are different. They have a separate power supply and use the analog output mainly as a signal. Because the signal output is not the only source of operating power, a four-wire transmitter or analyzer may have more flexibility in producing a low-current fault output. Even so, the receiving analog input and the control logic still need to be configured to interpret the selected fault current correctly.

It is also important to distinguish transmitter diagnostic faults from wiring faults. A transmitter may intentionally drive the loop to a NE43 fault level when it detects an internal problem. But an open circuit, blown fuse, broken wire, failed power supply, or disconnected terminal can also produce a very low or zero current at the input. The control system may classify both as bad signal conditions, but field troubleshooting must determine whether the source is the transmitter, loop wiring, input module, or power supply.

How Control Systems Interpret NE43 Alarm Thresholds

NE43-style signaling only works if the receiving system can measure and classify currents outside the normal 4–20 mA range. The analog input should not simply clip all values below 4 mA to the lower range value or all values above 20 mA to the upper range value. If it does, the distinction between valid measurement, saturation, and fault is lost.

In a properly configured PLC, DCS, SCADA system, or safety system input, the raw current is evaluated against defined thresholds. A typical configuration may classify the signal as follows:

• 4–20 mA: valid process measurement;
• 3.8–4.0 mA: valid signal quality but below configured range, or low saturation;
• 20.0–20.5 mA: valid signal quality but above configured range, or high saturation;
• ≤ 3.6 mA: downscale fault;
• ≥ 21.0 mA: upscale fault;
• gap regions between saturation and fault thresholds: abnormal or indeterminate signal quality.

The gap regions are important. A current between 3.6 mA and 3.8 mA, or between 20.5 mA and 21.0 mA, should not normally be treated as a valid process value. These regions provide separation between saturation and fault interpretation. If such values appear, the system may flag bad quality, generate a diagnostic alarm, or require further validation depending on the application.

A short delay, filter, or repeated-scan confirmation is often useful before declaring a persistent fault. Analog inputs can momentarily fluctuate because of noise, module update timing, maintenance activity, connection disturbance, or transmitter startup behavior. Immediate fault detection is useful for protection, but excessive sensitivity can create nuisance alarms. The correct delay depends on the process risk, loop dynamics, and whether the signal is used for indication, control, alarm, or shutdown.

Control-system responses should be designed around the role of the measurement. Common responses include:

• alarming the operator that the loop signal is bad;
• marking the process value as invalid or uncertain;
• switching a control loop to manual;
• holding the last good value for a limited time where appropriate;
• substituting a backup measurement if available;
• notifying maintenance or asset management software;
• inhibiting nonessential calculations that depend on the bad input;
• initiating protective action where required by the safety or control design.

These responses are control-system design choices, not automatic results of NE43 itself. NE43 provides a signal-level convention. The plant’s control strategy determines what happens when those signal levels occur.

For critical loops, the alarm thresholds in the control system should be reviewed during commissioning and proof testing. A transmitter configured for upscale failure is not useful if the input card treats 21 mA as a normal high process value. Similarly, a downscale fault is not useful if the signal is clamped, rescaled, or filtered before diagnostic logic sees it.

Signal scaling can also cause confusion. Many systems convert current directly into engineering units using the normal 4–20 mA range. If the system calculates an engineering value for 21 mA, it may display an unrealistic high number unless signal quality logic is applied first. A better approach is to classify signal quality before relying on the scaled process value.

Manufacturer-Specific Fail-Safe Current Settings

Although the NE43 convention is widely recognized, installed instruments do not all behave identically. Some devices described as NAMUR-compliant use manufacturer-specific alarm currents, selectable fail-safe modes, or product-family-specific defaults. The difference may be small in current terms, but it can be significant for alarm configuration.

For example, one transmitter may use a configured upscale alarm value above the 21.0 mA threshold, while another may use a different high alarm value that still falls in the fault region. A device may also offer choices such as low alarm, high alarm, last value, or custom alarm output depending on the model and communication protocol. Even transmitters from the same manufacturer can differ by product family, firmware option, or regional configuration.

This is why the transmitter manual and datasheet matter more than assumptions. Before configuring control-system thresholds, verify:

• whether the device explicitly supports NE43-style output behavior;
• the configured fail-safe direction, upscale or downscale;
• the actual alarm current value or range used by the transmitter;
• the saturation limits used for underrange and overrange;
• whether the analog output is active during startup, warm-up, or diagnostic states;
• how HART, fieldbus, or asset-management diagnostics relate to the analog current;
• whether the input module can accurately detect the selected current levels.

The safest engineering practice is to make the transmitter configuration and the control-system interpretation match each other. If a transmitter is set for upscale fault indication, the receiving system should alarm on the corresponding high fault region. If it is set for downscale fault indication, the receiving system should treat currents at or below the low fault threshold as bad signal conditions, not as valid negative process values.

Maintenance documentation should also record the chosen fail-safe direction and alarm thresholds. During troubleshooting, technicians need to know whether a low current is expected for failure, whether a high current indicates failure, and where saturation ends. Without that information, a valid underrange condition may be mistaken for a failed instrument, or a failed instrument may be mistaken for an extreme process reading.

NAMUR NE43 is therefore best viewed as a shared signaling framework. It provides recognizable current regions for normal measurement, saturation, and faults, but it does not remove the need to check the installed device settings. Accurate interpretation depends on the transmitter, loop power arrangement, input hardware, and control logic all being configured as a complete measurement system.