By Wu Peng, Senior Instrumentation Engineer · 20+ years in process instrumentation · Last reviewed July 8, 2026
A 4-20 mA current loop transmits one process measurement, such as pressure, level, temperature, or flow, as a DC current between 4 and 20 milliamps. The transmitter regulates the current in proportion to the measurement: 4 mA means the bottom of the calibrated range, 20 mA means the top, and 12 mA means exactly halfway. Because current is the same at every point in a series circuit, the reading at the control system is the reading at the transmitter, whether the wire run is 5 meters or 500.
The range starts at 4 mA rather than 0 for a practical reason: a healthy loop never reads zero, so a broken wire or a dead power supply announces itself immediately. This guide covers how the loop works, how to convert between milliamps and engineering units, how to check a loop power budget, what the fault currents mean, and how far the signal can actually travel. There is also a 4-20 mA calculator that runs the scaling math with your own range.
Contents
- What a 4-20 mA current loop is
- Why current instead of voltage
- Why the range starts at 4 mA
- The four parts of every loop
- 2-wire, 3-wire, and 4-wire transmitters
- Converting mA to process value
- Loop budget: sizing the power supply
- Fault currents and NAMUR NE43
- How far a 4-20 mA signal can travel
- HART on top of 4-20 mA
- Instruments with 4-20 mA output
- FAQ
- Request a quote
What a 4-20 mA current loop is
The loop is a series DC circuit with three jobs in it: a power supply provides the voltage, a transmitter regulates the current, and a receiver reads it. The transmitter is the only element allowed to set the current. It measures the process variable and throttles the loop current to the matching value between 4 and 20 mA, no matter what the supply voltage or wire length happens to be, as long as there is enough voltage to go around.
One loop carries one variable. A plant with forty measurement points runs forty loops, which is why the signal is wired in twisted pairs back to marshalling cabinets. The standard descends from the 3-15 psi pneumatic signal used before electronics; the ratio is the same, and so is the idea of a live zero at the bottom of the range.
Why current instead of voltage
Current does not drop along a wire; voltage does. If you send a 0-5 V signal down 300 meters of cable, the receiver sees 5 V minus whatever the wire resistance and connector corrosion take, and the error grows with distance. Send 12.00 mA down the same cable and the receiver sees 12.00 mA, because in a series circuit every electron that leaves the transmitter passes through the receiver. The wire resistance costs voltage, not current, and the transmitter simply works harder to hold the current constant.
Current signals also survive electrical noise better than low-level voltage signals. A twisted pair carrying 4-20 mA runs past motor starters and variable frequency drives with far less trouble than a millivolt thermocouple extension ever could. Noise still matters, which is why shielded twisted pair with the shield grounded at one end is standard practice, but the signal itself has a low source impedance and shrugs off most induced voltage.
Why the range starts at 4 mA
A range of 0-20 mA looks tidier, but a reading of 0 mA would be ambiguous: is the process at the bottom of the range, or is the wire cut? With a live zero of 4 mA, the two cases separate cleanly. A healthy loop sits somewhere between 4 and 20 mA. A loop reading near 0 mA has a broken conductor, a blown fuse, or a dead supply, and the control system can alarm on it instantly.
The live zero has a second job: it powers the transmitter. A 2-wire transmitter runs its own electronics on the loop current, and the guaranteed minimum of 4 mA is what keeps the instrument alive at the bottom of the range. That is why loop-powered transmitters exist at all, and why two wires can both feed the device and carry its signal.
The four parts of every loop
Every 4-20 mA loop contains the same four elements in series: a DC power supply, most commonly 24 V; a transmitter that measures the process and regulates the current; a receiver input at the controller, PLC, or indicator, typically a 250 Ω resistance; and the wire pair connecting them. Each element drops voltage, and the sum of the drops must stay below the supply voltage at the highest current the loop can carry.
The receiver is usually just a precision resistor inside the input card. At 250 Ω, the 4-20 mA current develops 1-5 V, which is what the analog-to-digital converter actually reads. That conversion is also why a multimeter set to volts across the input resistor is a quick way to check a loop without breaking it.
2-wire, 3-wire, and 4-wire transmitters
A 2-wire transmitter is loop powered: the same pair carries both the supply and the signal, and the device budgets its own electronics inside the 4 mA floor. Most pressure, level, and temperature transmitters are built this way because two wires are cheap to pull and intrinsically safe designs are easier to certify. A 4-wire transmitter takes separate power, usually 220 VAC or 24 VDC, and drives the 4-20 mA output as an independent circuit; flow meters with displays, coils, or heated sensors need this because their electronics draw far more than 4 mA. The 3-wire arrangement shares a common negative between supply and signal and appears mostly on compact OEM sensors with 0-5 V or 4-20 mA selectable outputs.
Wiring note from the field: when a 4-wire instrument feeds a loop that is also powered by the receiver side, one of the two sources has to go. Two active sources on one loop fight each other and the reading wanders. Every loop has exactly one active device; everything else is passive.
Converting mA to process value
The signal is linear between the lower range value (LRV, at 4 mA) and the upper range value (URV, at 20 mA). Two formulas cover all of it. Current from process value:
I = 4 + 16 × (PV − LRV) / (URV − LRV) mA
Process value from current:
PV = LRV + (URV − LRV) × (I − 4) / 16
Three worked examples with real configurations. A pressure transmitter ranged 0-10 bar reading 6.3 bar transmits 4 + 16 × 0.63 = 14.08 mA. A temperature transmitter ranged 0-250 °C at 150 °C transmits 4 + 16 × 0.6 = 13.6 mA. Working backward, a level loop ranged 0-2 m reading 12.0 mA puts the level at 0 + 2 × (12 − 4)/16 = 1.0 m, exactly half the span. The 4-20 mA calculator does both directions for pressure, level, temperature, and flow ranges.
| Percent of span | Loop current | Voltage across 250 Ω |
|---|---|---|
| 0% | 4.0 mA | 1.0 V |
| 25% | 8.0 mA | 2.0 V |
| 50% | 12.0 mA | 3.0 V |
| 75% | 16.0 mA | 4.0 V |
| 100% | 20.0 mA | 5.0 V |
One trap: differential pressure flow transmitters. If the transmitter sends raw DP without square root extraction, flow is proportional to the square root of the signal, so 12 mA is 70.7% flow, not 50%. Whether the root is extracted in the transmitter or in the DCS must be decided once, in one place. Extracting it twice is a classic commissioning error. The background is in our guide to the flow rate and pressure relationship.
Loop budget: sizing the power supply
Every element in the loop drops voltage, and the transmitter needs a minimum voltage at its own terminals to run, typically 8 to 12 V depending on the model; the datasheet states it. The loop works when the supply voltage minus every other drop still leaves the transmitter its minimum, at the highest current the loop will ever carry. Size at 22 mA, which covers the 21 mA fault signal with margin.
A worked budget. Supply 24 V; transmitter minimum 12 V; receiver 250 Ω; 500 m of 18 AWG pair, which is 1000 m of conductor at about 21 Ω/km, so 21 Ω of wire. At 22 mA the receiver drops 250 × 0.022 = 5.5 V and the wire drops 21 × 0.022 = 0.46 V. Available for the transmitter: 24 − 5.5 − 0.46 = 18.0 V, comfortably above the 12 V minimum, with about 6 V of spare budget for a fuse, surge barrier, or a second indicator in series.
Datasheets often state the same limit as a maximum load resistance, Rmax = (Vsupply − Vmin) / 0.022 A. For a 24 V supply and a 12 V transmitter that is roughly 545 Ω total for everything that is not the transmitter. Add an intrinsic safety barrier and its end-to-end resistance comes straight out of that number, which is why barriers and long runs sometimes cannot share the same 24 V loop.
Fault currents and NAMUR NE43
NAMUR recommendation NE43 standardizes what the current means outside the normal range, so a control system can tell a saturated reading from a broken instrument. Measurement information lives between 3.8 and 20.5 mA; deliberate failure signals sit at or below 3.6 mA or at or above 21 mA.
| Loop current | Meaning per NE43 |
|---|---|
| ≤ 3.6 mA | Transmitter signals a fault, fail-low setting |
| 3.8 to 20.5 mA | Valid measurement, including saturation just past the range ends |
| ≥ 21 mA | Transmitter signals a fault, fail-high setting |
| 0 mA | Not a transmitter signal: broken wire, blown fuse, or dead supply |
Fast diagnosis by reading. Stuck at exactly 4.0 mA with a process that should be moving: transmitter alive but sensor or configuration problem. Sitting at 3.6 mA: the transmitter itself is reporting an internal fault. Pegged at 20 mA for hours: process above URV or a failed sensing element; check whether it is 20.5 (saturated) or 21+ (fault). Zero: the circuit is open somewhere, so start at terminals and fuses, not at the instrument.
How far a 4-20 mA signal can travel
Far. The limit is the voltage budget, not the signal. Take the 24 V loop above: after the receiver, (24 − 12)/0.022 − 250 leaves about 295 Ω for wire. With 18 AWG pair at 42 Ω/km for both conductors, that is roughly 7 km on paper. Practical installs run shorter because datasheets rate conservatively, cable capacitance slows HART, and long runs pick up more noise; many instrument datasheets state 500 to 1000 m as the routine figure. If a run works out marginal, the fixes are a thicker conductor, a higher supply voltage within the transmitter rating, or a lower receiver resistance.
Where cable itself becomes the problem, tank farms across a road, or rotating equipment, a wireless pressure transmitter replaces the pair entirely and reports on a battery-friendly cycle instead of a continuous loop.
HART on top of 4-20 mA
HART superimposes a small frequency-shift-keyed digital signal on the same two wires, so the loop carries the analog 4-20 mA measurement and a digital channel for configuration, diagnostics, and secondary variables at the same time. The loop needs at least about 230 Ω of resistance for the FSK signal to develop; the conventional 250 Ω receiver satisfies it, which is why a bench setup with just a supply and a transmitter refuses to communicate until you add a 250 Ω resistor in series. Our HART pressure transmitter page covers rerangeing and calibration over HART in detail.
Instruments with 4-20 mA output
Nearly everything we build ships with a 4-20 mA output as standard. For pressure, the pressure transmitter series covers gauge, absolute, and differential pressure service as 2-wire loop-powered devices. For temperature, the temperature transmitter accepts RTD and thermocouple inputs and retransmits 4-20 mA with HART. For level, submersible level transducers run the same 2-wire loop down the borehole or sump. Send the range, supply voltage, and receiver details with your inquiry and the loop budget gets checked before anything ships.
Application example
Building services, chilled-water plant (Southeast Asia). The contractor needed continuous level on a 2 m chilled-water sump at 35 °C, feeding a building management system with a standard analog input. A 2-wire submersible level transmitter was configured 0-2 m as 4-20 mA on 24 VDC, 0.5% FS, so mid-tank reads exactly 12 mA and the BMS scales it without a converter. The client later added a variant with a local display for walk-by checks; both builds run on the same two-wire loop, so the panel wiring did not change.
When you move from the loop theory to physically landing wires, the pressure transducer wiring diagram guide covers the 2-wire, 3-wire, and 4-wire hookups terminal by terminal, with a fault table for loops that read wrong.
The loop is only one of the output options. For how transmitters compare with voltage transducers, millivolt sensors, and switches, read the device comparison guide.
FAQ
How does a 4 to 20 mA signal work?
A transmitter measures the process variable and regulates the loop current in proportion: 4 mA at the bottom of the calibrated range, 20 mA at the top, linear in between. Because the loop is a series circuit, the receiver sees the same current the transmitter set, regardless of wire length within the voltage budget.
Why are we using 4-20 mA instead of 0-20 mA?
The 4 mA live zero separates “process at bottom of range” from “circuit broken”, which both read 0 mA on a 0-20 mA loop. It also powers 2-wire transmitters: the electronics run on the guaranteed minimum 4 mA, so two wires carry both power and signal.
What are the disadvantages of a 4-20 mA signal?
One loop carries only one variable, so large plants pull a lot of copper, and each loop needs DC power. Resolution is limited by the analog signal and the receiver’s converter, and diagnostics beyond fail-low and fail-high need HART or a fieldbus. Multiple loops also have to be isolated from each other properly, or ground loops corrupt the readings.
Is a 4-20 mA signal AC or DC?
DC. The loop is powered by a DC supply, most commonly 24 V, and the current flows in one direction. The only AC-like component that belongs on the wires is the small HART FSK tone, and that rides on top of the DC signal without changing its average value.
Wu Peng, Senior Instrumentation Engineer. 20+ years commissioning transmitters and control loops in refining, water, and OEM equipment. He has chased enough 3.6 mA readings across plant yards to insist on checking the loop budget before the wire is pulled.
Request a quote
Send the measurement, the range, the supply voltage, and what the loop feeds into, PLC input, indicator, or barrier. We confirm the loop budget, configure the range so 4 and 20 mA land where your control system expects them, and return a documented configuration. Tell us the application and we configure one unit, not a shelf part.