In a dense PCB, a jumper is a small part that still matters. It bridges nets that cannot be routed directly and can rescue a layout. Used badly, it can also take the circuit down. A jumper is not a substitute for a complete route; it is a deliberate short connection inserted where the copper pattern cannot finish the net on its own. The sections below cover jumper types, what they are for, and the details that are easy to miss.
What a Jumper Is
In PCB design, a jumper is a short circuit path made with a heavier conductor or a special connection. It is used when two parts cannot be connected by a normal trace, when the parts are too far apart for a practical route, or when a pin numbering error has to be corrected. In each case the jumper closes a net that the etched copper left open or misconnected.
A normal trace is copper that belongs to the board. A jumper is an added conductor—wire, resistor, solder bridge, or a dedicated PCB feature—that completes the same electrical function. Because it is added rather than etched, its length, its location, and the quality of its joints become part of the net. Those variables are what turn a useful bridge into a source of attenuation, interference, or an intermittent open.

Four Jumper Types
Four constructions are common, and each trades flexibility against process control.
A wire jumper is simple and direct. A short length of insulated or bare wire is soldered between two points. Placement must be checked for interference because the wire stands above the board and can couple to nearby nets or snag during handling. The wire also has its own inductance and resistance; keeping it short limits both.
A resistor jumper can stabilize a signal. The resistance value has to be chosen with care. A low-value resistor used as a jumper adds series damping that can reduce ringing on a short net, but the same resistance drops voltage and dissipates power if the net carries current. The value is therefore part of the circuit, not an arbitrary placeholder.
A solder-blob jumper is flexible. Two adjacent pads or a pair of exposed copper features are bridged with solder. The method depends on soldering skill. Too little solder leaves an unreliable joint; too much can bridge to a neighboring net. The result is hard to control in volume assembly and is more common in rework or prototypes.
A PCB jumper is made through connecting holes. Routing flexibility is high because the connection can be placed where a conventional trace cannot cross, and the holes can be designed into the board rather than added by hand. The plated or non-plated nature of those holes still has to be defined so the jumper conducts only where intended.

Three Functions
A jumper is added for one of three reasons.
It completes a difficult net. Parts that cannot meet on the same layer or in the same area are connected by the jumper instead of by a long detour or an extra layer transition. The jumper finishes the schematic connection that the layout could not route cleanly.
It separates signals. In a high-power circuit, a jumper can keep paths from coupling into each other by giving one net a short, direct bridge while the other remains on its own route. The separation is only as good as the placement; a jumper that runs parallel to a noisy net reintroduces the coupling it was meant to avoid.
It raises reliability. By cutting the number of open or poorly routed connections, the jumper keeps the design stable. An unfinished net is an open. A jumper that closes that net, with sound joints and a short path, removes the open without forcing a board respin. The reliability gain disappears if the jumper itself becomes the weak joint.
Failure Modes
A jumper helps only if these mistakes are avoided.
A jumper that is too long or coiled produces signal attenuation. Extra length adds series resistance and inductance. On a high-speed or high-frequency net the added inductance slows the edge and the coil radiates or picks up energy. The attenuation is not a property of the jumper type; it is a property of the path length and the geometry.
A poor location interferes with other circuits. A jumper routed across a sensitive analog node, a clock, or a high-current switch loop couples into those circuits. The interference appears as noise on the victim net even though the jumper's own net measures correctly in isolation.
A weak solder joint becomes an intermittent open. Cold solder, insufficient wetting, or a joint that cracks under flex opens the net under vibration or temperature change. The board may pass a static continuity check and still fail in use.
Missing or incorrect grounding adds noise. If the jumper is part of a return path, or if it should have been referenced to ground and was not, the return current finds a longer path. The larger loop radiates and picks up interference, so the noise is a grounding error rather than a signal-path error.
Placement and Assembly Rules
Four practices keep the jumper from becoming the failure it was meant to prevent.
Keep it short and straight. Cut length and avoid loops. A short, straight jumper minimizes inductance, resistance, and the chance of coupling to neighboring nets. Any loop that is left in the wire is an antenna and an inductive element.
Stay off sensitive nets. Do not route the jumper across high-frequency or analog circuits. The physical separation that protects those nets is lost if the jumper crosses them, even if the jumper's own net is quiet.
Inspect the joint. Check solder quality under magnification. A joint that looks complete to the eye can still be cold or cracked. Magnification reveals the fillet shape and the wetting that a continuity beep does not.
Verify continuity. Check the net with a multimeter before applying power. The check confirms that the jumper closed the intended net and did not short to an adjacent one. Powering the board first turns an open or a short into a damaged part.
A jumper is small, but it is part of whether the board nets correctly. Used with the rules above—short and straight, clear of sensitive circuits, inspected, and verified—it solves layout problems without adding a new failure. Used as an afterthought, it becomes the attenuation, the interference, the intermittent open, or the noise source that the rest of the layout avoided.