At 60 Hz, a solid 10 AWG copper wire carries current across its entire cross-section. At 100 kHz, that same wire is effectively a hollow tube; the current is pushed to the outer 0.2 millimeters of the conductor. This is the skin effect, and calculating the exact depth of current penetration requires handling numbers in the range of $10^{-8}$ and $10^{-7}$. A standard calculator will round these into useless zeros. To get accurate results on the bench, you must use a conversion scientific calculator—such as the Casio fx-991EX ClassWiz or the Texas Instruments TI-36X Pro—that natively supports engineering notation, metric prefix conversions, and embedded physical constants.
The direct answer for AC skin depth ($\delta$) is calculated using the formula $\delta = \sqrt{\frac{\rho}{\pi f \mu}}$. Below, we break down the derivation, map every symbol, and walk through exact keystrokes and unit tracking to ensure your high-frequency wiring doesn't melt under load.
The Skin Depth Formula and Symbol Definitions
The skin depth formula applies to homogeneous, linear conductors carrying sinusoidal alternating current. It assumes the material's magnetic permeability is constant and that the conductor's physical radius is significantly larger than the calculated skin depth. If you are dealing with non-sinusoidal waveforms (like square waves from a PWM inverter), you must calculate the skin depth for the fundamental frequency and its dominant harmonics separately.
Here is the governing equation:
$\delta = \sqrt{\frac{2\rho}{\omega \mu}} = \sqrt{\frac{\rho}{\pi f \mu}}$
| Symbol | Parameter Name | Standard SI Unit | Typical Value / Notes |
|---|---|---|---|
| $\delta$ | Skin Depth | Meters (m) | Depth where current density falls to $1/e$ ($\approx 37\%$) of surface value. |
| $\rho$ | Electrical Resistivity | Ohm-meters ($\Omega\cdot m$) | Copper: $1.68 \times 10^{-8}$ $\Omega\cdot m$ at 20°C. |
| $f$ | Frequency | Hertz (Hz) | Mains: 50/60 Hz. Switchmode supplies: 20 kHz - 500 kHz. |
| $\omega$ | Angular Frequency | Radians per second (rad/s) | $\omega = 2\pi f$ |
| $\mu$ | Absolute Permeability | Henries per meter (H/m) | $\mu = \mu_0 \times \mu_r$ |
| $\mu_0$ | Permeability of Free Space | H/m | $4\pi \times 10^{-7}$ (or $1.2566 \times 10^{-6}$) H/m. |
| $\mu_r$ | Relative Permeability | Dimensionless | Copper/Aluminum $\approx 1$. Steel $\approx 100$ to $4000$. |
Rearranged Forms and Calculator Setup
On the bench, you rarely solve for skin depth in isolation. You might need to find the maximum allowable frequency for a given busbar thickness, or determine the required resistivity for a custom alloy. A good conversion scientific calculator allows you to store these variables in memory registers (like [STO] A, B, C, X, Y) to avoid re-typing $10^{-8}$ exponents.
Here are the rearranged forms solving for each primary variable:
- Solve for Frequency ($f$): $f = \frac{\rho}{\pi \mu \delta^2}$
Use when: Sizing a transformer core and needing to know the maximum switching frequency before a solid wire becomes ineffective. - Solve for Resistivity ($\rho$): $\rho = \pi f \mu \delta^2$
Use when: Selecting a dummy load resistor material that needs a specific skin depth at RF frequencies. - Solve for Relative Permeability ($\mu_r$): $\mu_r = \frac{\rho}{\pi f \mu_0 \delta^2}$
Use when: Characterizing an unknown steel alloy by measuring its AC resistance vs DC resistance.
Solved Problems with Strict Unit Tracking
Abstract math fails on the workbench. Let's track the units through two distinct scenarios to prove the math resolves to meters.
Problem 1: Standard Mains Copper Wire (60 Hz)
Setup: Calculate the skin depth of a standard copper busbar carrying 60 Hz AC mains power at 20°C.
- Identify Variables:
$\rho = 1.68 \times 10^{-8} \text{ } \Omega\cdot m$
$f = 60 \text{ Hz}$
$\mu_r = 1$ (non-magnetic)
$\mu_0 = 4\pi \times 10^{-7} \text{ H/m}$ - Calculate Absolute Permeability ($\mu$):
$\mu = 1 \times (4\pi \times 10^{-7}) = 1.2566 \times 10^{-6} \text{ H/m}$ - Substitute into Formula:
$\delta = \sqrt{\frac{1.68 \times 10^{-8}}{\pi \times 60 \times 1.2566 \times 10^{-6}}}$ - Unit Tracking (Inside the Square Root):
$\frac{\Omega\cdot m}{\text{Hz} \cdot \text{H/m}} = \frac{\Omega\cdot m}{(1/s) \cdot (\Omega\cdot s / m)} = \frac{\Omega\cdot m}{\Omega / m} = m^2$ - Final Calculation:
$\delta = \sqrt{0.0000707 \text{ } m^2} = 0.00841 \text{ m (or 8.41 mm)}$
Problem 2: High-Frequency Aluminum Inverter Busbar (20 kHz)
Setup: Calculate the skin depth for an aluminum conductor in a 20 kHz switchmode power supply.
- Identify Variables:
$\rho = 2.82 \times 10^{-8} \text{ } \Omega\cdot m$ (Aluminum)
$f = 20,000 \text{ Hz}$
$\mu = 1.2566 \times 10^{-6} \text{ H/m}$ - Substitute and Track Units:
$\delta = \sqrt{\frac{2.82 \times 10^{-8}}{\pi \times 20000 \times 1.2566 \times 10^{-6}}}$ - Final Calculation:
$\delta = \sqrt{\frac{2.82 \times 10^{-8}}{0.07895}} = \sqrt{3.57 \times 10^{-7} \text{ } m^2} = 0.000597 \text{ m (or 0.597 mm)}$
Bench Takeaway: At 20 kHz, current only penetrates aluminum about 0.6 mm deep. A solid 1/2-inch (12.7mm) aluminum bar is a waste of metal; the center 11.5mm carries virtually zero current. You must use flat copper ribbon or Litz wire instead.
Real-World Scenario: The 50kHz Induction Heater Busbar
Formulas are only as good as the data you punch into your conversion scientific calculator. Here is a failure analysis from a recent bench build that highlights what happens when unit conversions are ignored.
The Setup: A maker was building a 50 kHz, 2 kW induction heater. They needed to route high-frequency current from the inverter H-bridge to the work coil. They had a spool of solid 3/8-inch (9.5mm) copper rod and wanted to verify if it was thick enough to handle the 40A RMS current without melting. They decided to calculate the skin depth to see how much of the rod's cross-section would be utilized.
The Numbers Entered:
The maker looked up copper resistivity online and found the value 1.68 (which is the resistivity in $\mu\Omega\cdot cm$). They entered this directly into their calculator as $1.68$, alongside $f = 50,000$ and $\mu = 1.2566 \times 10^{-6}$.
The Outcome:
The calculator spit out a skin depth of 0.291 meters (291 mm). Thrilled that the 9.5mm rod was well within the 291mm skin depth, they soldered the solid rod into the circuit and powered up the induction heater. Within 45 seconds, the copper rod glowed cherry red, the solder joints melted, and the H-bridge MOSFETs shorted out from the massive impedance mismatch and thermal runaway.
What Went Wrong:
The maker failed to convert the resistivity units. By entering $1.68$ instead of $1.68 \times 10^{-8} \text{ } \Omega\cdot m$, they inflated the numerator by a factor of 100 million. Because the resistivity is under a square root, the final skin depth result was off by a factor of 10,000.
The actual skin depth at 50 kHz is 0.29 mm. The 9.5mm solid rod was effectively acting as a 0.58mm thick hollow tube with a massive DC resistance bottleneck at the joints. The correct fix was to use 1/4-inch copper tubing or a flat copper strap, which provides high surface area for the 0.29mm skin depth while shedding heat efficiently.
Unit Mistakes and Realistic Magnitude Checks
When using a scientific calculator for electromagnetic formulas, a single missed prefix will destroy your design. Memorize these common unit traps and magnitude benchmarks to sanity-check your results before you cut any wire.
The Three Unit Mistakes That Break the Math
- Forgetting $2\pi$ in Angular Frequency: The formula requires $\omega$ (radians/sec), not $f$ (cycles/sec). If your calculator doesn't have a dedicated $\omega$ register, you must manually multiply your frequency by $2\pi$. Forgetting this makes your skin depth 2.5 times larger than reality.
- Mixing cm and m in Resistivity: Datasheets often list resistivity in $\mu\Omega\cdot cm$ (e.g., Copper = 1.68). The SI formula demands $\Omega\cdot m$. To convert $\mu\Omega\cdot cm$ to $\Omega\cdot m$, multiply by $10^{-8}$. Use your calculator's metric prefix conversion function (often accessed via [SHIFT] [PREFIX] on Casio models) to automate this.
- Assuming $\mu_r = 1$ for All Metals: Copper and aluminum are non-magnetic ($\mu_r \approx 1$). But if you are calculating skin depth for a steel chassis or an iron-core transformer, $\mu_r$ can range from 100 to 4,000. Because $\mu$ is in the denominator, a high $\mu_r$ crushes the skin depth down to microscopic levels, which is why steel is a terrible conductor for high-frequency AC.
Realistic Answer Magnitudes (The Sanity Check)
If your calculator outputs a number outside these ranges, you have a unit error. Stop and check your exponents.
- 60 Hz (Mains): Skin depth in copper should be roughly 8.4 mm to 8.5 mm. If you get 0.8 meters, you missed a $10^{-3}$ conversion.
- 20 kHz (Audio/Inverters): Skin depth should be around 0.46 mm.
- 1 MHz (RF/Induction): Skin depth drops to roughly 0.066 mm (66 $\mu m$). At this frequency, even 24 AWG wire is mostly dead space in the center.
Mastering the conversion scientific calculator isn't just about passing an exam; it's about understanding the physical reality of the electrons moving through your workbench projects. By strictly tracking units, leveraging engineering notation, and sanity-checking your magnitudes, you ensure that the copper you pay for is actually the copper that carries the load.
References:
1. Georgia State University, HyperPhysics: Electrical Resistivity and Conductivity.
2. Texas Instruments: TI-36X Pro Scientific Calculator Guidebook (Engineering Notation and Constants).






