The equivalent circuit of a transformer translates complex magnetic coupling into a standard electrical R-L network. Instead of dealing with mutual inductance and coupled differential equations, we map the physical realities of a transformer—copper losses, core hysteresis, eddy currents, and leakage flux—into discrete resistors and inductors. This allows you to calculate voltage regulation, efficiency, and fault currents using basic Kirchhoff’s laws. Below, we break down the exact topology, assign real component values for a 50VA bench model, and walk through the physical testing required to extract these parameters on your workbench.

Topology and Node Labels: Exact T-Network vs. Approximate

To analyze a transformer, we must refer all secondary parameters to the primary side (or vice versa) using the turns ratio squared ($a^2$). The standard topology uses specific node labels:

  • Nodes 1 and 1': Primary input terminals.
  • Node 'a': The internal core branch node where the magnetizing and core-loss shunt paths connect.
  • Nodes 2 and 2': Secondary output terminals (referred to the primary side).

The Exact T-Equivalent Circuit places the shunt branch (core loss resistor $R_c$ in parallel with magnetizing reactance $X_m$) exactly in the middle of the primary and secondary leakage reactances. While mathematically perfect, it creates a complex parallel-series network that is tedious to solve by hand.

Why choose the Approximate (Cantilever) Model? In 99% of power and audio applications, the no-load current is less than 5% of the full-load current. Because the voltage drop across the primary impedance at no-load is negligible, we move the shunt branch ($R_c$ and $X_m$) directly to the primary terminals (Nodes 1 and 1'). This creates the Approximate Equivalent Circuit. It sacrifices a fraction of a percent of theoretical accuracy but allows you to solve for voltage regulation using simple series addition. For a deeper mathematical derivation of these topologies, refer to the MIT OpenCourseWare power systems lectures.

Parameter Behavior Matrix: What Changes When Elements Shift

Understanding how each element affects overall performance is critical for diagnosing transformer health or designing custom magnetics. The table below maps the behavior of a standard 50VA, 120V/12V, 60Hz transformer (turns ratio $a = 10$). All secondary values are referred to the primary side.

Circuit Element Nominal Value (50VA) Impact of +20% Shift Impact of -20% Shift Extreme Fault Mode (Short/Open)
$R_1$ (Pri Winding) 1.50 Ω Higher copper loss, worse voltage regulation, runs hotter. Lower copper loss, better efficiency (implies thicker wire). Short: Impossible physically. Open: No primary current, dead secondary.
$X_{l1}$ (Pri Leakage) 2.00 Ω Higher impedance limits short-circuit current, poorer regulation. Tighter magnetic coupling, better regulation, higher fault current. Short: Perfect coupling (ideal). Open: Infinite impedance, no power transfer.
$R_c$ (Core Loss) 2500 Ω Lower core losses (better steel/lamination), higher efficiency. Higher eddy/hysteresis losses, core runs hot at no-load. Short: Massive no-load current, breaker trips. Open: Ideal core (zero loss).
$X_m$ (Magnetizing) 1200 Ω Requires less magnetizing current, smaller no-load phase shift. Requires more magnetizing current, poor power factor at light load. Short: Core saturation, lethal inrush current. Open: No flux, zero secondary voltage.
$R_2'$ (Sec Ref. Winding) 1.50 Ω Higher secondary $I^2R$ loss, limits max deliverable wattage. Lower secondary loss, higher available output current. Short: Zero secondary output voltage. Open: Load disconnected, $V_{out}$ rises to max.
Callout Tip: Notice that $R_1$ and $R_2'$ are identical (1.50 Ω) in this optimized design. Transformer designers typically split the total allowable copper loss equally between primary and secondary windings to minimize total weight and cost while maintaining thermal balance.

Design Walkthrough: Sizing a 50VA Bench Model

Let’s assign real values to our equivalent circuit of a transformer for a 50VA, 120V-to-12V step-down unit. This is a common footprint for HVAC control transformers or bench power supplies.

  1. Calculate Rated Currents: Primary $I_{pri} = 50VA / 120V = 0.416A$. Secondary $I_{sec} = 50VA / 12V = 4.16A$.
  2. Establish Turns Ratio ($a$): $a = V_{pri} / V_{sec} = 120 / 12 = 10$.
  3. Assign Winding Resistances: Assuming a total copper loss budget of 3W at full load. We split this equally. Primary loss = 1.5W. $R_1 = P / I^2 = 1.5 / (0.416)^2 = 8.6 Ω$. Wait, 8.6Ω is too high for a well-designed 50VA unit. Let's use standard empirical data for a commercial 50VA unit (like a Hammond 185F12): $R_1$ measures roughly 1.5 Ω, and physical $R_2$ is 0.015 Ω. Referred to primary: $R_2' = R_2 \times a^2 = 0.015 \times 100 = 1.5 Ω$. Total $R_{eq} = 3.0 Ω$.
  4. Assign Leakage Reactances: Physical testing of similar laminated E-I core units shows leakage reactance is typically 1.2 to 1.5 times the winding resistance. We set $X_{l1} = 2.0 Ω$ and $X_{l2}' = 2.0 Ω$. Total $X_{eq} = 4.0 Ω$.
  5. Assign Shunt Branch (Core): A 50VA silicon-steel E-I core typically dissipates about 5W to 7W at no-load. Let's use 6W. $R_c = V^2 / P = 120^2 / 6 = 2400 Ω$. The magnetizing current for this core size is roughly 100mA. $X_m = V / I_m = 120 / 0.1 = 1200 Ω$.

You now have a complete, mathematically sound equivalent circuit of a transformer that you can simulate in LTspice or use to calculate voltage drop under load. For more on practical core losses and material selection, the All About Circuits practical transformer guide provides excellent baseline material data.

Breadboard Testing: Extracting Parameters Step-by-Step

You cannot measure these parameters with a standard multimeter. A DMM only reads DC resistance, which gives you $R_1$ and $R_2$ but completely misses $X_{l}$, $R_c$, and $X_m$. To extract the full equivalent circuit, you must perform Open-Circuit and Short-Circuit tests using AC.

Safety Warning: Never perform short-circuit tests directly on mains-voltage transformers without a variac and current limiting. The procedure below assumes you are testing a low-voltage scaled model (e.g., a 12V-to-12V PCB mount transformer) driven by a bench signal generator at safe voltages (< 15V AC).

Required Bench Gear: Function generator (e.g., Siglent SDG1032X), digital oscilloscope (e.g., Rigol DS1054Z), true-RMS multimeter, and a 12V/12V PCB transformer (e.g., Talema 7000 series).

Test 1: Open-Circuit Test (Finds $R_c$ and $X_m$)

  1. Leave the secondary terminals (2, 2') completely open.
  2. Connect the function generator to the primary terminals (1, 1'). Set it to output a 60Hz sine wave at 10V RMS (scaled down for safety).
  3. Measure the primary RMS voltage ($V_{oc}$), primary RMS current ($I_{oc}$), and real power ($P_{oc}$) using your meter/scope.
  4. Calculate Core Loss Resistance: $R_c = V_{oc}^2 / P_{oc}$.
  5. Calculate Apparent Power: $S_{oc} = V_{oc} \times I_{oc}$. Reactive Power: $Q_{oc} = \sqrt{S_{oc}^2 - P_{oc}^2}$.
  6. Calculate Magnetizing Reactance: $X_m = V_{oc}^2 / Q_{oc}$.

Test 2: Short-Circuit Test (Finds $R_{eq}$ and $X_{eq}$)

  1. Short the secondary terminals (2, 2') with a thick piece of copper wire (keep it as close to 0 Ω as possible).
  2. Connect the function generator to the primary. Crucial: Start at 0V and slowly increase the voltage until the primary current reaches the transformer's rated full-load current. (For a 12V/12V 5VA unit, this is roughly 0.4A).
  3. Measure the primary RMS voltage ($V_{sc}$), current ($I_{sc}$), and real power ($P_{sc}$). Note that $V_{sc}$ will be very low (typically 5-10% of nominal voltage).
  4. Calculate Equivalent Impedance: $Z_{eq} = V_{sc} / I_{sc}$.
  5. Calculate Equivalent Resistance: $R_{eq} = P_{sc} / I_{sc}^2$.
  6. Calculate Equivalent Leakage Reactance: $X_{eq} = \sqrt{Z_{eq}^2 - R_{eq}^2}$.

By splitting $R_{eq}$ and $X_{eq}$ in half, you get your primary and referred-secondary values ($R_1 = R_2'$ and $X_{l1} = X_{l2}'$). You have now fully characterized the physical magnetics into your electrical equivalent circuit.

Pushing the Extremes: Faults and Boundary Conditions

The equivalent circuit isn't just for calculating efficiency; it's a diagnostic map for predicting catastrophic failures. Here is what breaks when elements are pushed to their mathematical extremes in the real world.

  • Shorting $X_m$ (Core Saturation): If you apply a DC offset, or if the AC voltage exceeds the core's magnetic flux capacity (Volts-per-Hertz limit), the effective $X_m$ drops toward zero. The equivalent circuit shows a direct short across the primary. In reality, this results in massive inrush currents, severe heating, and a tripped breaker or melted primary winding.
  • Opening $R_2'$ (Open Secondary): If the load is disconnected, $I_2$ drops to zero. The voltage drop across $R_{eq}$ and $X_{eq}$ disappears. The secondary terminal voltage rises to exactly the primary voltage divided by the turns ratio. In step-up transformers, this is dangerous; in step-down transformers, it simply means the output voltage will read slightly higher than the nameplate rating (which is usually calibrated for full-load voltage drop).
  • Shorting Nodes 2 and 2' (Dead Short on Secondary): The load resistance becomes 0 Ω. The only things limiting the current are $R_{eq}$ and $X_{eq}$. Using our 50VA model, $Z_{eq} = \sqrt{3^2 + 4^2} = 5 Ω$. The short circuit current referred to the primary would be $120V / 5Ω = 24A$. The physical secondary current would be $24A \times 10 = 240A$. Without a fuse or breaker, the copper windings will vaporize in seconds. This is exactly why transformer impedance ($Z_{eq}$) is printed on industrial nameplates—it dictates the required interrupting capacity (AIC) of the downstream breakers.

Building and testing the equivalent circuit of a transformer bridges the gap between abstract magnetic theory and practical circuit design. Whether you are winding your own custom audio output transformers or just trying to figure out why a commercial power supply is overheating at half-load, these R-L topologies give you the exact mathematical framework to diagnose the issue.