On a fieldbus segment the wiring splits into two named parts with very different jobs. The trunk is the single home-run backbone that leaves the power supply and carries both power and communication out to the field, and the spurs are the short drop cables that branch off it to reach individual devices. The two are governed by separate length and device-count rules, and mixing them up is a common cause of a segment that passes on paper but misbehaves once wired. This page breaks down what the trunk and spur each are, the limits that apply to each, and how those choices spend a fixed segment budget.
Spur vs Trunk in one line: On a fieldbus segment, the trunk is the main home-run cable that runs from the power conditioner out to the field and is shared by every device, while spurs are the short branch cables that tap off the trunk to connect individual devices. The trunk carries the whole segment's power and signal and can run long distances, but each spur is a short unterminated stub kept far shorter, and the number of devices per spur is limited.
The trunk is the home-run cable that leaves the power supply and terminator at the host end and runs out to the field where the devices are. It is the shared spine of the segment: every device draws its power and exchanges its digital signal through this one pair of conductors, so the trunk carries the summed current of the whole segment and the combined communication traffic. Because it is shared, anything that degrades the trunk degrades every device at once, which is why the trunk is treated as the backbone you engineer first.
The trunk is also the part allowed to run long. It can stretch a substantial distance across a skid or a process unit, and both ends of a segment are terminated, one at the supply and one at the far end of the trunk, so the transmission line behaves properly. Trunk length is bounded because the digital signal attenuates over distance and because the cable's resistance drops the voltage available to the devices at the far end, but within those limits the trunk is designed to reach far.
Practically, the trunk is what you route in conduit or tray from the marshalling area out to a field junction box or a set of couplers. From there the segment fans out. Keeping the trunk continuous, correctly terminated at both ends, and within its voltage-drop budget is the foundation the rest of the topology sits on, because the spurs and devices all depend on a healthy trunk to work at all.
A spur is the short branch cable that taps off the trunk to connect a single device, or occasionally a small number of devices, at a coupler or junction box. Electrically a spur is an unterminated stub hanging off the terminated trunk, and an unterminated stub reflects signal energy back onto the bus. That reflection is tolerable when the stub is short but distorts the waveform once it gets long, so spur length is kept far shorter than trunk length. The longer and the more numerous the spurs, the more the segment's signal quality suffers.
The number of devices allowed on a single spur is also limited, and the more devices you hang on one spur the shorter that spur must be. Many installations favor one device per spur precisely so a fault or short on one drop can be isolated at its coupler without pulling down the rest of the segment. Spur couplers often include short-circuit protection for exactly this reason, so that a shorted spur is disconnected rather than collapsing the shared trunk voltage.
The trade to understand is that spur length and trunk length are not independent. The relevant limit is usually the total cable, trunk plus the sum of all spurs, so every meter spent on a long spur is a meter no longer available on the trunk. A topology with many long spurs eats into the length budget and can also erode signal quality through reflections, which is why designers keep spurs short and push the reach into the trunk instead.
Because the segment has a single interacting budget of length, device count, and current, the way you split wiring between trunk and spurs is a design decision, not an afterthought. A tree or chicken-foot layout groups devices at couplers near the far end of a long trunk with very short spurs, maximizing reach. A daisy-chain or spur-heavy layout spreads couplers along the trunk, which is convenient for scattered devices but spends more of the length budget on spurs. Either way, a segment calculation adds up device current, cable resistance, trunk and spur lengths, and worst-case voltage drop to confirm the last device still has enough voltage to operate.
This matters to remote operations because a fieldbus segment concentrates many measurements onto one shared trunk before they ever reach the control layer. A cloud SCADA platform such as Merobix ultimately reads the process values from those devices through the fieldbus host, but they arrive as digital data off a shared segment rather than as independent 4-20 mA loops. The blast radius of a wiring problem is therefore wider: a trunk fault can take a whole cluster of points offline together, while a shorted spur, if the coupler isolates it, may take only its one device.
For a remote-monitoring team, understanding trunk-versus-spur structure changes how you read correlated outages. Several tags on one skid dropping at once points at the shared trunk, its terminator, or the supply, whereas a single tag vanishing points at that device's spur or coupler. Bringing segment and device health into a central dashboard lets an operator distinguish a backbone problem from a single-drop problem quickly, and lets them see a segment nearing its budget before someone casually adds one more spur that pushes it over the edge.
A spur is kept far shorter than the trunk because it is an unterminated stub that reflects signal, and the allowable spur length shrinks as you add more devices to that spur or more spurs to the segment. The trunk, terminated at both ends, is the part designed to run long. The controlling limit is usually the total of trunk plus all spurs, so spending length on spurs reduces how far the trunk can reach.
The trunk is the transmission line the signal travels along, so it is terminated at each end to prevent reflections and keep the waveform clean. Spurs are short branches off that line and are left unterminated because they are meant to be short enough that their reflections are negligible. Making spurs long defeats that assumption and injects distortion onto the shared bus.
You can, but many installers put one device per spur so a fault on one drop can be isolated at its coupler without affecting the rest of the segment. Adding devices to a spur also forces that spur to be shorter and increases the current the branch carries. Where devices are clustered, a coupler with several short single-device spurs is usually preferred over one long shared spur.
Merobix reads your field devices into a cloud SCADA - the real thing behind these terms, live in days from any browser.