The current rule, formally known as Kirchhoff’s Current Law (KCL), states that the total electric current entering a circuit junction must exactly equal the total current leaving that same junction. In a real installation, this rule changes everything about how you size feeder wires and select shunt resistors, because it dictates that the main supply conductor must handle the algebraic sum of all parallel branch loads, not just the largest single load. Beginners commonly confuse the current rule with Kirchhoff’s Voltage Law (the loop rule) or mistakenly assume it only applies to DC circuits, ignoring how it governs complex phasor addition in AC systems.
The Physics of the Node
At its core, the current rule is a statement of charge conservation. Electrons do not pile up at a wire nut, and they do not vanish inside a solder joint. Every coulomb of charge that flows into a node per second must flow out of it per second. If this were not true, the node would rapidly accumulate a massive static charge, creating an electric field strong enough to arc across your workbench.
When you sit down to analyze a circuit, the first step is identifying your nodes—any point where two or more conductive paths meet. A trace on a PCB, a terminal block in a control panel, or the neutral bar in a subpanel are all nodes. The current rule forces you to treat these junctions as strict accounting checkpoints for electron flow.
Worked Numeric Example: 12V DC Distribution
Let’s look at a common maker scenario: building a 12V DC distribution block for an off-grid telemetry station using a Blue Sea Systems ST Blade fuse block. You have a single main feeder wire bringing 12V into the block, and three separate branch circuits leaving it.
- Branch 1: An ESP32-WROOM-32 dev board running a WiFi sensor node. Peak transmit current draw is 160mA (0.16A).
- Branch 2: A 5V Songle SRD-05VDC-SL-C relay module used to trigger a high-voltage contactor. Coil draw is 70mA (0.07A).
- Branch 3: A 1-meter segment of 12V WS2815 LED strip running at full white. Current draw is 1.20A.
Applying the current rule at the main input terminal of the fuse block:
Imain = IESP32 + IRelay + ILED
Imain = 0.16A + 0.07A + 1.20A = 1.43A
Where You Meet This In Practice
You will run into the current rule constantly across different electrical domains, though it wears different hats depending on the trade:
- PCB Layout (Ground Vias): When routing high-current return paths on a 4-layer board, you cannot rely on a single via to carry 2A back to the ground plane. The current rule dictates that if you need 2A, and your via barrel is rated for 0.5A, you must stitch at least four (preferably six for thermal margin) vias in parallel at that node.
- Home Wiring (Multi-Wire Branch Circuits): In a 120/240V split-phase MWBC, two hot wires share a single neutral. The current rule governs the neutral bar: the neutral only carries the imbalance between the two hots, not the sum, provided the hots are on opposite phases.
- Solar Combiner Boxes: When paralleling four 10A solar strings into a single output, the current rule demands that the output busbar and feeder wire be rated for 40A (plus a 1.25x NEC continuous load derating factor, pushing the requirement to 50A).
The AC Catch: Phasors vs. Scalars
The most dangerous misconception about the current rule is that it only applies to DC, or that AC currents can just be added together like simple numbers (scalars). In AC circuits, KCL applies to instantaneous values and phasors (complex numbers representing magnitude and phase angle).
Consider a 3-phase Wye-connected commercial panel supplying three identical 10A resistive heater loads, one on Phase A, one on Phase B, and one on Phase C. A novice might look at the neutral wire returning to the transformer and assume it carries 30A (10 + 10 + 10).
However, because the three phases are separated by 120 electrical degrees, their phasor sum is exactly zero:
10∠0° + 10∠120° + 10∠240° = 0A
The current rule holds perfectly, but it requires vector math. If you size that neutral wire for 30A, you are wasting copper. If you have non-linear loads (like LED drivers or VFDs) generating triplen harmonics (3rd, 9th, 15th), those harmonic currents do add up in the neutral because they are in-phase. In that specific modern edge case, the neutral can actually carry more current than the phase conductors, a phenomenon thoroughly documented in AC power theory literature.
Conductor Sizing Decision Tree at the Junction
Once you have used the current rule to find the total current entering or leaving a node, you must size the conductor. Use this decision path to select your wire gauge based on the 75°C column of NEC Table 310.16 (assuming copper, 30°C ambient, and up to 3 current-carrying conductors in a raceway).
| Calculated Node Current (Itotal) | Load Type | Required Ampacity (w/ Margin) | Concrete Pick (Copper AWG) |
|---|---|---|---|
| ≤ 1.43A (e.g., our DC example) | Continuous (Telemetry) | 1.43A × 1.25 = 1.78A | 18 AWG (Internal PCB/Chassis) or 14 AWG THHN (Min NEC branch circuit) |
| 15.0A | Continuous (Heater) | 15.0A × 1.25 = 18.75A | 12 AWG THHN (Rated 25A at 75°C) |
| 32.0A | Non-Continuous (Motor start) | 32.0A × 1.0 = 32.0A | 8 AWG THHN (Rated 50A at 75°C) |
| 65.0A | Continuous (EV Charger) | 65.0A × 1.25 = 81.25A | 2 AWG THHN (Rated 115A at 75°C) |
Troubleshooting Apparent KCL Violations
My clamp meter reads 10A on the hot wire and 9.8A on the neutral. Is the current rule broken?
No. The current rule applies to the entire node, which includes the equipment grounding conductor (EGC). The missing 0.2A is leaking to ground, likely through capacitive coupling, degraded insulation, or a slight fault. This exact imbalance is what a 5mA GFCI breaker detects to trip the circuit. If you clamp both the hot and neutral together and read anything other than 0A, you have a ground fault.
Does the current rule apply to capacitors? Current doesn't flow through the dielectric.
Yes, but it requires James Clerk Maxwell’s addition to KCL: displacement current. While conduction electrons do not cross the dielectric gap, the changing electric field inside the capacitor acts mathematically as a current. For practical bench work, you treat the capacitor leads as a standard node where AC current flows in and out, but DC current stops once the capacitor is charged.
I measured the currents at a 3-way junction and they don't add up to zero. What did I do wrong?
You are likely measuring with a standard AC clamp meter on a circuit with heavy harmonic distortion or a poor power factor, and your meter is only reading fundamental RMS magnitudes. To verify KCL in complex AC environments, you must use a true-RMS meter capable of logging instantaneous waveforms, or an oscilloscope with a current probe, and sum the instantaneous values point-by-point across the AC cycle.
