01 · Foundational

Node-voltage analysis: foundations

Choose one reference node, make the remaining node voltages the unknowns, express every branch current from those voltages, and let Kirchhoff's current law generate the equations.

115–145 minFoundational
In this lesson
  • Explain why node voltages are a sufficient set of unknowns for many linear resistive networks.
  • Choose and interpret a reference node without treating ground as a physical current sink.
  • Write resistor branch currents directly from node-voltage differences.
  • Construct a nodal equation from Kirchhoff's current law with fixed current references.
  • Solve and independently verify a one-unknown-node network.
  • Recognize when the simple nodal recipe needs extension because of a voltage source between unknown nodes.
Starting question

When a circuit no longer collapses into one obvious series or parallel equivalent, how do we choose a small set of unknowns instead of inventing a current for every branch?

Node-voltage analysis answers that question by choosing electrical potentials as coordinates. Pick one node as the reference, describe every other node by its voltage relative to that reference, express branch currents from voltage differences, then apply Kirchhoff's current law (KCL). The method is not a new physical law; it is a disciplined way to organize laws you already know.

01.1Why node voltages can reduce the number of unknowns

Figure 1.1 — One unknown node is enough to start

The ideal source fixes the input node at 12 V, ground fixes the reference potential at 0 V, and VOUT is the only unknown node voltage. R1 connects 12 V to VOUT; R2 and RL connect VOUT to ground.

verified topology
Loaded voltage divider+−VS12 VR13 kΩR26 kΩRL6 kΩVINVOUTGND
The one question in this figure is: which node voltages are already known, and which single voltage remains unknown?

A branch-current approach could assign a separate unknown current to every element and then add enough Kirchhoff equations to relate them. Nodal analysis instead asks whether those currents can be written directly from a smaller set of node voltages.

For the circuit in Figure 1.1, the source fixes the input node at 12 V and the chosen reference node is 0 V. Only the output-node voltage remains unknown. Once that one number is known, Ohm's law determines all three resistor currents.

This is the central economy of the method: do not make a branch current an independent unknown if an element law already expresses it from node voltages.

01.2The reference node is a coordinate origin, not a physical current sink

Figure 1.2 — A node voltage is defined relative to the reference node

The unknown voltage V is defined from negative-reference GND toward positive VOUT. The arrow follows ADR-0018: negative reference to positive reference.

verified topology
Loaded voltage divider+−VS12 VR13 kΩR26 kΩRL6 kΩVINVOUTGND
No current arrows are shown here: this figure only fixes the definition of the node voltage.

Electric potential is meaningful through differences. We therefore choose one node and define its potential as zero. This is the reference node. Circuit diagrams often mark it with a ground symbol, but that symbol does not automatically mean the node is physically connected to Earth.

If V denotes the output-node potential relative to the reference node, then

V = VOUT − VREF
VREF = 0

Choosing a different reference changes the numerical labels assigned to individual node potentials, but it does not change any physical voltage difference, branch current or power. The reference is a coordinate choice.

01.3Convert a resistor branch into a node-voltage expression

Consider a resistor R connecting node a at voltage Va to node b at voltage Vb. If current is defined positive from a to b, the branch voltage in that direction is Va−Vb. Ohm's law gives

Ia→b = (Va−Vb)/R

This single expression is the workhorse of resistive nodal analysis. It contains both magnitude and sign. If Va<Vb, the calculated current is negative, correctly indicating that the physical current flows from b to a.

For a branch from the unknown node V to ground through R2,

I2 = (V−0)/R2 = V/R2

For a branch from the unknown node V toward a known 12 V node through R1, when we deliberately define current as leaving V,

I1 = (V−12 V)/R1

Nothing special happens because the known node is “a source node.” It is simply a node whose voltage is already known.

01.4Kirchhoff's current law turns those branch expressions into one equation

Figure 1.3 — One simple convention: count every branch current as leaving the unknown node

The three arrows define the positive branch-current references used in the nodal equation. A solved negative value simply means the real current enters the node through that branch.

verified topology
Loaded voltage divider+−VS12 VR13 kΩR26 kΩRL6 kΩVINVOUTGND
No voltage arrow is shown: this figure answers only the current-sign question in Kirchhoff's current law.

Use one sign rule and keep it. In Figure 1.3 all three branch currents are defined as leaving the unknown node. Kirchhoff's current law states that the algebraic sum of signed currents at the node is zero:

I1 + I2 + IL = 0

Substitute the resistor relations directly:

(V−12 V)/R1 + V/R2 + V/RL = 0

This is the nodal equation. There is no separate current-division formula and no need to guess which current is “incoming.” The chosen references and algebra handle the signs.

01.5Complete worked example: solve the one unknown node

Use the values in the figures: R1=3 kΩ, R2=6 kΩ, RL=6 kΩ and the known input node at 12 V. The nodal equation is

(V−12 V)/(3 kΩ) + V/(6 kΩ) + V/(6 kΩ) = 0

Multiply the entire equation by 6 kΩ:

2(V−12 V) + V + V = 0

Expand and collect the terms containing V:

2V−24 V+V+V=0
4V=24 V
V=6 V

Now recover the branch currents from the definitions used in Figure 1.3:

I1 = (6−12)V/(3 kΩ) = −2 mA
I2 = 6 V/(6 kΩ) = 1 mA
IL = 1 mA

The negative sign on I1 means the real current flows opposite its chosen reference: 2 mA actually enters the unknown node from the 12 V side. KCL is then physically transparent: 2 mA enters, while 1 mA leaves through each lower branch.

01.6Independent check 1: reduce the same network by series/parallel reasoning

DC-03 gives an independent route. The two 6 kΩ lower resistors are in parallel:

6 kΩ ∥ 6 kΩ = 3 kΩ

The circuit therefore becomes a 3 kΩ upper resistor in series with a 3 kΩ lower equivalent across 12 V. A symmetric voltage divider gives

V = 12 V · 3/(3+3) = 6 V

The nodal result and the reduction result agree exactly. This cross-check matters: systematic analysis should reproduce the simpler shortcut whenever the shortcut is actually valid.

01.7Independent check 2: power balance

The real current through R1 is 2 mA from the 12 V node toward the 6 V node. The resistor drops 6 V, so

PR1 = (6 V)(2 mA) = 12 mW

Each 6 kΩ lower resistor has 6 V across it and carries 1 mA:

PR2 = 6 mW
PRL = 6 mW

Total resistor absorption is therefore 24 mW. The ideal 12 V source supplies 2 mA:

Psource = −(12 V)(2 mA) = −24 mW

The total signed power is zero. A nodal voltage that satisfies KCL but produces an inconsistent power balance would be a strong signal to recheck signs or arithmetic.

01.8A useful one-node formula: conductance-weighted average

Suppose one unknown node V is connected only through positive resistors to several neighboring nodes whose voltages V1,…,VN are already known. Writing all resistor currents as leaving V gives

Σk=1N(V−Vk)/Rk = 0

Define conductance Gk=1/Rk. Rearranging gives

V ΣGk = ΣGkVk
V = [ΣGkVk]/[ΣGk]

So this purely resistive node voltage is a conductance-weighted average of the known neighboring voltages. With all conductances positive, V must lie between the smallest and largest neighboring voltages.

For the worked circuit, the neighboring voltages are 12 V, 0 V and 0 V, with conductances 1/3 kΩ, 1/6 kΩ and 1/6 kΩ. The weighted average is 6 V.

01.9How to choose the reference node—and when the simple recipe needs extension

Any node can be the mathematical reference, but some choices make the equations simpler. A useful reference often touches many branches, turns several branch voltages directly into node voltages, or coincides with the return node already emphasized by the circuit.

For an ordinary resistive network, after one reference is chosen, each remaining unknown node voltage usually contributes one KCL equation. A node whose voltage is fixed directly by an ideal voltage source to the reference does not need its own unknown.

A special case appears when an ideal voltage source lies directly between two unknown nonreference nodes. The source fixes their voltage difference but does not provide a resistor-like current expression. We do not invent one. Chapter 03 will introduce the supernode, which preserves KCL while adding the voltage-source constraint correctly.

01.10Common errors and mastery practice

ErrorCorrection
“Ground absorbs current.”Ground is the chosen zero-potential node unless the hardware says otherwise. Current follows actual branches.
Using V/R for every resistor attached to the nodeUse (V−Vother)/R. The other terminal is not always ground.
Changing between “entering positive” and “leaving positive” halfway through KCLChoose one sign convention for the equation and keep it.
Treating a negative current as an errorA negative value means the actual current is opposite the chosen reference.
Writing a resistor current through an ideal voltage sourceAn ideal voltage source has no R-like constitutive current law. Use the correct nodal extension.
  1. A node V connects to 10 V through 2 kΩ and to ground through 3 kΩ. Write the nodal equation and find V.
  2. A node V connects to 8 V through 4 kΩ, to 2 V through 2 kΩ and to ground through 4 kΩ. Find V and verify that it lies between the neighboring voltages.
  3. Repeat the worked 12 V example but replace RL by 3 kΩ. Find V and all branch currents.
  4. A student writes (12−V)/3k + V/6k + V/6k = 0 while claiming that all currents are defined as leaving V. Diagnose the sign inconsistency.
  5. Why is choosing a different reference node allowed even though all numerical node-voltage labels change?
  6. An ideal 5 V source is connected directly between two unknown nonreference nodes. Why can you not write its branch current as a voltage difference divided by resistance?
Solutions

1. (V−10)/2k + V/3k=0. Multiplying by 6k gives 3(V−10)+2V=0, so V=6 V.

2. (V−8)/4k + (V−2)/2k + V/4k=0. Multiplying by 4k gives V−8+2V−4+V=0, hence V=3 V. It lies between 0 V and 8 V, as the positive-conductance weighted-average check predicts.

3. (V−12)/3k + V/6k + V/3k=0. Multiplying by 6k gives 2(V−12)+V+2V=0, so V=4.8 V. Currents defined as leaving V are −2.4 mA through R1, 0.8 mA through R2 and 1.6 mA through RL.

4. If every current is defined as leaving V, the R1 term must be (V−12)/3k, not (12−V)/3k. The latter corresponds to a current defined toward V.

5. Physical observables depend on voltage differences, not on the arbitrary zero of potential. Changing reference adds a common offset to node-potential labels but preserves branch voltage differences and therefore currents and powers.

6. An ideal voltage source constrains voltage, not current. Its current is whatever the surrounding network requires, so there is no finite resistance R that supplies a branch-current formula. A supernode or augmented equation is needed.

Chapter 02 will keep exactly the same logic but scale it to several unknown nodes, where the equations naturally organize themselves into a linear system and eventually a conductance matrix.