Most people know the two obvious transmitter topologies - 2-wire loop powered and 4-wire separately powered - but there is a third option that sits between them and is easy to mis-wire. A 3-wire transmitter takes a separate supply like a 4-wire device but saves a conductor by sharing one wire between the power return and the signal return. That shared common is both the reason it exists and the source of its most subtle failures. This guide explains what a 3-wire transmitter is, why the topology is used, and how its shared return can quietly corrupt a reading.
3-Wire Transmitter in one line: A 3-wire transmitter uses three conductors: a positive DC supply, a signal output, and a single common wire that serves as the return for both the power and the 4-20 mA signal. It is a separately powered device like a 4-wire transmitter but uses one shared return instead of two independent pairs. The saving of one conductor comes at a cost: because supply current and signal current flow through the same common, voltage drop and ground offset on that shared wire can distort the reading.
A 3-wire transmitter is separately powered, so unlike a 2-wire device it does not have to run on loop current and can drive a more capable set of electronics. But instead of the four conductors a fully independent 4-wire device uses - a dedicated power pair and a dedicated signal pair - a 3-wire device folds the two returns into one. The three terminals are typically labelled supply positive, signal (or output), and common. The supply and the signal both reference that single common, which is what lets the manufacturer save a wire.
Because the output is measured between the signal terminal and the shared common, a 3-wire transmitter almost always presents a sourcing, or active, output: the device pushes 4-20 mA out of its signal terminal, and the input card reads that current returning through the common. This is a natural fit for a sinking analog input that expects to receive a sourced signal. It is worth confirming on the datasheet, because the whole point of matching topologies is knowing which device supplies the loop current and which one receives it.
The 3-wire configuration shows up most often on devices that need more than loop power but where a manufacturer or installer wants to trim conductor count - some position sensors, certain temperature and level transmitters, and instruments carried over from older wiring conventions. It is less common today than clean 2-wire or 4-wire designs precisely because of the shared-return issue described next, but you will still encounter it, and recognizing it prevents a wiring mistake.
The weakness of the 3-wire topology lives in the common wire. That single conductor carries the transmitter's full supply return current plus the signal return current, all at once. Every wire has some resistance, and current through resistance creates a voltage drop, so the common is not at exactly the same potential along its whole length. The input card reads the signal as the voltage or current relative to its end of the common, but the transmitter references its end - and if those two ends sit at slightly different potentials, that difference adds directly to the measured signal as an error.
This gets worse the more supply current the device draws and the longer or thinner the common conductor is. A device pulling significant supply current through a long, undersized common can shift the apparent reading by a meaningful amount, and the error is not constant - it changes as the supply current or the signal current changes, so it can look like a drifting or nonlinear fault rather than a wiring problem. A related trap is ground offset: if the common is bonded to ground at more than one point, or the supply and the input card reference different grounds, a circulating current on the shared return injects noise and offset that a properly isolated pair would never see.
The defenses are straightforward once you understand the mechanism. Keep the common conductor as short and as heavy as practical to minimize its resistance. Reference the transmitter's supply return and the input card's signal common to a single, common point rather than to two separate grounds. And where accuracy matters or the run is long, prefer a 4-wire device with independent returns, or add a signal isolator so the input sees a clean, galvanically separated copy of the signal that is immune to whatever is happening on the field-side common.
To wire a 3-wire transmitter, bring the supply positive from your 24 VDC panel to the transmitter's supply terminal, run the signal terminal to the positive of the analog input channel, and land the shared common on both the supply negative and the input channel's return so all three references meet at the card and panel. The critical discipline is that the common must be a single, well-defined return - not one path to the power supply and a separate, unrelated path to the input, which would split the return current and create exactly the offset you are trying to avoid.
On the I/O side, the analog input must be set up to read a sourced (active) signal, since the 3-wire device is supplying its own loop current. A card channel configured to also supply loop power would fight the transmitter's output and give a wrong or zero reading. This is the same source-versus-sink discipline that governs any analog loop, and the 3-wire topology makes it easy to get wrong because two of its three wires connect to power rather than to the signal path. Labeling all three terminals on the loop sheet - supply, signal, common - saves the next person a debugging session.
Once the reading is clean at the RTU or PLC, feeding it into a SCADA system is no different from any other analog point. In a cloud SCADA setup such as Merobix, the local edge device or RTU reads the 3-wire transmitter's output, scales it, and publishes the value to the cloud dashboard alongside every other tag. Where the platform helps is in diagnosis: because the historized value is trended continuously, a shared-common offset that drifts with load shows up as a telltale pattern on the chart, making a subtle field-wiring fault far easier to spot than it would be from a panel meter glanced at once a shift.
A 2-wire transmitter runs entirely on the 4-20 mA loop, carrying power and signal on a single pair, so it must live on under 20 mA total. A 3-wire transmitter has a separate power supply and is not limited by loop current, but it saves a conductor by sharing one common wire between the power return and the signal return. The 3-wire device can drive more capable electronics, at the cost of the shared-return offset issue.
Because both the supply return current and the signal return current flow through the same conductor, and that conductor has resistance. Current through resistance produces a voltage drop, so the two ends of the common sit at slightly different potentials, and that difference adds to the measured signal as an error. The effect grows with supply current draw and with the length and thinness of the common wire, and it changes with load, so it can look like drift.
If accuracy is critical or the cable run is long, prefer a 4-wire device with fully independent returns, which avoids the shared-common offset entirely. A 3-wire transmitter is acceptable for shorter runs with a heavy, single-point common and modest accuracy needs, where saving a conductor is worthwhile. When you are stuck with a 3-wire device on a difficult run, adding a signal isolator gives the input a clean, isolated copy immune to the field-side common.
Merobix reads your field devices into a cloud SCADA - the real thing behind these terms, live in days from any browser.