Kirchhoff's Loop vs Junction Rule: Difference

Two conservation laws, so two rules

There are two rules rather than one because there are two separate things a circuit conserves, and the CED files them as separate topics for exactly that reason.

[The loop rule](/glossary/kirchhoffs-loop-rule) is conservation of energy. Essential knowledge 11.6.A.2 says so in one line: Kirchhoff's loop rule is a consequence of the conservation of energy. The statement itself is 11.6.A.3, that the sum of potential differences across all circuit elements in a single closed loop must equal zero,

and 11.6.A.1 supplies the link to energy, $\Delta U_E = q\Delta V$. Carry a charge once around a loop and it returns to where it started, so the electric potential energy it gained has to equal the energy it gave up. A rule about volts is a rule about joules per coulomb.

[The junction rule](/glossary/kirchhoffs-junction-rule) is conservation of charge. Essential knowledge 11.7.A.1 is the parallel sentence: Kirchhoff's junction rule is a consequence of the conservation of electric charge. The statement, 11.7.A.2, is that the total amount of charge entering a junction per unit time must equal the total amount of charge exiting that junction per unit time,

Charge is not created at a node and does not pile up there in a steady circuit, so the rates in and out match.

Everything that follows is a consequence of that split. The loop rule constrains potential differences and needs a journey. The junction rule constrains currents and needs a point. Neither can do the other's job, and a problem that has both unknown currents and unknown potential differences needs both.

Side by side

Neither equation appears on either equation sheet. Render the AP Physics 2 Table of Information and the Electricity box carries $I = \Delta q/\Delta t$, $R = \rho \ell / A$, $P = I\Delta V$, $I = \Delta V/R$, the series and parallel combination rules for resistors and capacitors and the time constant, with no $\sum \Delta V = 0$ and no $\sum I$. The AP Physics C: Electricity and Magnetism sheet is the same in this respect. Both rules are known rather than looked up, which is the practical reason the sign convention below is worth memorising rather than reconstructing under pressure.

The sign convention, stated once and kept to the end

The junction rule barely has a sign problem. Label each branch as carrying current into or out of the node, add the ins, add the outs, set them equal. That is the whole procedure.

The loop rule is where marks are lost, and the fix is to fix a convention before writing anything and then not to think about it again.

Step one: choose a direction to walk the loop. Clockwise or anticlockwise, it makes no difference, but choose before you start.

Step two: guess a direction for each unknown current and draw the arrow. A guess that turns out wrong produces a negative number and nothing else. Nothing needs redrawing.

Step three: apply these four cases as you walk.

Step four: set the total to zero and solve.

The resistor rows are the ones to internalise. Walking with the current means walking downhill in potential, because charge loses energy in a resistor, so the term is negative. Walking against the current means climbing, so the term is positive. The source rows are the opposite way round from what people expect on a first read: crossing from minus to plus is a rise, which is why it is $+\mathcal{E}$ regardless of which way the current happens to be going through that source.

A negative current in your answer means the real current runs opposite to your arrow. That is information, not an error, and rewriting the whole solution to remove the minus sign wastes time you do not have.

Where each one applies: a point against a path

The two rules attach to different features of a schematic, and identifying the feature is most of the work.

A junction is a point where three or more conductors meet. That is where charge has a choice, and where a split or a merge can occur. Two things that look like junctions and are not:

• A bend in a wire. Nothing branches, so no equation is available.
• Two components joined end to end. Charge that enters the first must leave through the second, which is exactly the definition of a series connection at 11.5.A.1.i, and it is why series elements carry identical currents.

A loop is any closed path you can trace through the circuit and back to your starting point. It does not have to be a physically obvious rectangle, and a single element may belong to several loops: 11.2.A.3 states that a single circuit element may be part of multiple electrical loops. What you need is a set of independent loops, meaning each new loop includes at least one element the earlier ones did not.

Counting equations before you start is the habit that prevents thrashing:

• A circuit with $n$ junctions yields $n - 1$ independent junction equations. The last one is arithmetically implied by the others and adds nothing.
• Each independent loop yields one loop equation.
• Together those must number as many as the unknown currents, or the system will not close.

A two-junction, two-loop network therefore gives one junction equation and two loop equations for three branch currents, and it solves. A single-loop circuit has no junctions at all, so the junction rule contributes nothing and the loop rule does all the work by itself.

The case that separates them: one circuit, one rule at a time

Take a $12 \ \mathrm{V}$ ideal battery in series with a $2.0 \ \Omega$ resistor, feeding a parallel pair of $6.0 \ \Omega$ and $3.0 \ \Omega$. Call the three branch currents $I_1$ through the series resistor, $I_2$ through the $6.0 \ \Omega$ and $I_3$ through the $3.0 \ \Omega$.

Now try to solve it with only one rule and watch it fail.

Loop rule alone. Walk the loop through the battery, the $2.0 \ \Omega$ and the $6.0 \ \Omega$: $12 - 2.0 I_1 - 6.0 I_2 = 0$. One equation, two unknowns. Add the second loop, the one containing only the two parallel resistors: $6.0 I_2 - 3.0 I_3 = 0$. Two equations, three unknowns. The loop rule has run out of loops and the circuit is not solved, because nothing has yet said how the current splits.

Junction rule alone. At the node where the branch divides, $I_1 = I_2 + I_3$. One equation, three unknowns, and no way to bring the $12 \ \mathrm{V}$ into the problem at all. Charge conservation knows nothing about batteries.

Both together. Three equations, three unknowns, and it closes: $I_2 = 1.0 \ \mathrm{A}$, $I_3 = 2.0 \ \mathrm{A}$, $I_1 = 3.0 \ \mathrm{A}$. The full working is the second worked example below.

That is the argument for the pairing in one paragraph. The loop rule fixes how much potential each branch has to spend; the junction rule fixes how the charge divides between them. They constrain different variables, so neither is redundant, and this is the smallest circuit in which you can watch each one fail on its own. The series and parallel guide reduces the same network faster using $R_{eq}$, and the reason that shortcut works is that the combination rules are themselves derived from these two rules.

When it costs a mark

Flipping the sign of a resistor term halfway round. The single most common loss. Once you have chosen a walking direction, every resistor crossed with the current is $-IR$ and every one crossed against it is $+IR$, for the whole loop and every loop after it.

Reversing a battery term to match the current. A source's contribution depends on which terminal you enter, not on the current direction. Cross from minus to plus and it is $+\mathcal{E}$ even if the current is being pushed backwards through that source by a stronger one elsewhere.

Redrawing after a negative current. A negative answer already means the current runs the other way. Restarting costs minutes and introduces new sign errors.

Using the junction rule around a loop, or the loop rule at a node. The clearest symptom is a written equation with a mix of amperes and volts in it. Check the units of every term before solving: a loop equation is entirely volts, a junction equation is entirely amperes.

Counting the same loop twice. Two loops that share every element are one loop. If two of your equations are multiples of each other, you have not gained an equation, and the system will refuse to solve.

Writing all the junction equations. With $n$ junctions only $n - 1$ are independent. The extra equation is not wrong, it is just empty, and it can disguise the fact that you are short of a loop equation.

"The current is used up." The junction rule forbids it. Whatever enters a series element leaves it. What a bulb consumes is energy, which is the loop rule's business, not charge.

When one rule is enough, and why that hides the pairing

For most of the circuits you meet, one rule sits idle, and that is why students often finish the unit thinking of them as one idea with two names.

A single-loop series circuit has no junctions. No branch points, no choice for the charge, and therefore no junction equation to write. The loop rule alone determines everything, and the statement "the current is the same everywhere" arrives from the geometry rather than from a rule you applied.

A simple parallel pair across an ideal battery needs almost no loop work. Every branch spans the same two nodes, so every branch has the same potential difference by 11.5.A.1.ii, and each branch current comes straight from $I = \Delta V/R$. The junction rule then just totals them. The loop rule is doing something here, but the something is so immediate that it looks like a definition rather than a law.

Reduction by $R_{eq}$ hides both. Collapsing a network to one equivalent resistance and then unwinding it is faster and is the right method in an exam. It is also why the two rules can feel optional: the combination rules already have them baked in.

The pairing becomes unavoidable in exactly one situation, and it is worth recognising it on sight: a circuit with more than one loop and unknown currents in more than one branch. There, neither rule alone produces enough equations, and the CED's own sample activity for Topic 11.6 in AP Physics C asks students to solve a multi-loop problem and then construct a representation for each loop showing the loop rule and one for each junction showing the junction rule. The two are being taught as a pair because they are used as a pair.

Where this sits on the AP exam

The two rules are consecutive topics in Unit 11, Electric Circuits, which carries 15 to 18 percent of the multiple-choice section over a suggested 12 to 20 class periods.

Topic 11.6, Kirchhoff's Loop Rule, carries suggested skills 1.C, 2.A, 2.C and 3.B. Skill 1.C is creating qualitative sketches of graphs, which is a direct pointer at 11.6.A.4: the values of electric potential at points in a circuit can be represented by a graph of electric potential as a function of position within a loop. That graph is the loop rule made visible, and reading or drawing one is a live exam task.

Topic 11.7, Kirchhoff's Junction Rule, carries 1.A, 2.B, 2.C and 3.C. It is the shorter topic, with two essential knowledge statements against the loop rule's four, and the skills lean toward calculating and justifying rather than sketching.

AP Physics C: Electricity and Magnetism states both rules in the same words at its Topic 11.6 and Topic 11.7, with the loop-rule statement and the potential graph filed as 11.6.A.2.i and 11.6.A.2.ii rather than as 11.6.A.3 and 11.6.A.4. The physics does not change; only the numbering does, so cite the numbers for the course you are sitting.

One boundary statement shapes what these rules will be asked about. Under Topic 11.5 both CEDs state that circuits with batteries of different potential differences connected in parallel will not be assessed, which rules out the hardest classical multi-loop exercises. Expect multi-loop networks with a single source, or with sources in series, rather than two competing batteries in separate branches.

For the quantities the two rules constrain, see voltage vs current, and for what the source itself supplies before any of this begins, emf vs voltage.