The Surface Resistance Formula: Symbols, Units, and Assumptions
When designing PCB power traces, evaluating anti-static coatings, or testing insulating materials, you cannot rely on standard 3D bulk resistance equations. The surface resistance formula calculates the DC resistance of a thin, uniform layer by treating it as a 2D plane. The total resistance ($R$) of a rectangular surface layer is the product of its surface resistance ($R_s$, often called sheet resistance) and the number of "squares" in the geometry ($L/W$):
$$R = R_s \times \left( \frac{L}{W} \right)$$
Where: $$R_s = \frac{\rho}{t}$$
This formula applies strictly to uniform, homogeneous thin films where the thickness ($t$) is significantly smaller than the width ($W$) and length ($L$). It assumes a constant temperature (usually 20°C) and DC or low-frequency AC conditions.
| Symbol | Definition | Standard SI Unit | Common Industry Unit |
|---|---|---|---|
| $R$ | Total DC Resistance | Ohms ($\Omega$) | m$\Omega$, k$\Omega$, M$\Omega$ |
| $R_s$ | Surface (Sheet) Resistance | Ohms per square ($\Omega/\square$) | m$\Omega/\square$, $\Omega/\square$ |
| $\rho$ | Bulk Resistivity of the material | Ohm-meters ($\Omega \cdot m$) | $\mu\Omega \cdot cm$ |
| $t$ | Thickness of the conductive/resistive layer | Meters ($m$) | $\mu m$, mils, oz/ft² |
| $L$ | Length of the current path | Meters ($m$) | inches, mm, mils |
| $W$ | Width of the current path | Meters ($m$) | inches, mm, mils |
A "square" is a unitless geometric ratio. A 10mm x 10mm patch of 1 oz copper has the exact same surface resistance as a 10-inch x 10-inch patch. The $L/W$ ratio simply counts how many of these squares are connected in series.
Rearranged Forms for Bench and Design Work
On the bench or in CAD, you rarely solve for $R$ directly. You usually have a target resistance and need to find the required geometry or material. Here are the algebraic rearrangements:
- Solve for Surface Resistance ($R_s$): $R_s = \frac{R}{(L / W)}$ (Use when characterizing an unknown coating with a known geometry).
- Solve for Required Length ($L$): $L = W \times \left( \frac{R}{R_s} \right)$ (Use when designing a thin-film resistor).
- Solve for Required Width ($W$): $W = \frac{L}{(R / R_s)}$ (Use when sizing a PCB trace for a specific voltage drop).
- Solve for Thickness ($t$): $t = \frac{\rho}{R_s}$ (Use when verifying PCB copper plating thickness).
- Solve for Bulk Resistivity ($\rho$): $\rho = R_s \times t$ (Use when identifying an unknown alloy).
Worked Examples with Strict Unit Tracking
The most common point of failure in these calculations is unit mismatch. Below are two solved problems tracking every conversion.
Problem 1: PCB Power Trace (Conductive)
Scenario: You are routing a 5V power trace on a standard FR4 board using 1 oz copper. The trace is 2.5 inches long and 20 mils wide. What is the total DC resistance?
Knowns:
- Material: Copper ($\rho \approx 1.68 \times 10^{-8} \Omega \cdot m$ at 20°C, per GSU HyperPhysics)
- Thickness ($t$): 1 oz/ft² copper = $34.8 \mu m$ ($34.8 \times 10^{-6} m$)
- $L = 2.5 \text{ inches}$
- $W = 20 \text{ mils} = 0.020 \text{ inches}$
Step 1: Calculate Surface Resistance ($R_s$)
$$R_s = \frac{\rho}{t} = \frac{1.68 \times 10^{-8} \Omega \cdot m}{34.8 \times 10^{-6} m} = 4.82 \times 10^{-4} \Omega/\square = 0.482 \text{ m}\Omega/\square$$
Step 2: Calculate the Square Ratio ($L/W$)
$$\frac{L}{W} = \frac{2.5 \text{ inches}}{0.020 \text{ inches}} = 125 \text{ squares}$$
Step 3: Calculate Total Resistance ($R$)
$$R = 0.482 \text{ m}\Omega/\square \times 125 = \mathbf{60.25 \text{ m}\Omega}$$
Problem 2: ESD Bench Mat (Resistive)
Scenario: You are testing an anti-static mat for your soldering station. The mat's surface resistivity is rated at $5.0 \times 10^8 \Omega/\square$. The mat measures 1.2 meters by 0.8 meters. You place your multimeter probes on the opposite ends of the 1.2-meter length. What resistance should you read?
Knowns:
- $R_s = 5.0 \times 10^8 \Omega/\square$ (Note: For insulators, manufacturers often provide $R_s$ directly per ASTM D257 testing standards).
- $L = 1.2 \text{ m}$ (distance between probes)
- $W = 0.8 \text{ m}$ (width of the current path)
Step 1: Calculate the Square Ratio ($L/W$)
$$\frac{L}{W} = \frac{1.2 \text{ m}}{0.8 \text{ m}} = 1.5 \text{ squares}$$
Step 2: Calculate Total Resistance ($R$)
$$R = (5.0 \times 10^8 \Omega/\square) \times 1.5 = 7.5 \times 10^8 \Omega = \mathbf{750 \text{ M}\Omega}$$
Your standard digital multimeter leads have a contact and wire resistance of roughly $100 \text{ m}\Omega$ to $200 \text{ m}\Omega$. If you try to measure the $60.25 \text{ m}\Omega$ PCB trace from Problem 1 with a standard 2-wire DMM, the leads will swamp the reading. You must use a 4-wire Kelvin measurement or a dedicated micro-ohmmeter to accurately verify low surface resistance traces.
Unit Mistakes That Will Break Your Calculation
If your simulation or CAD tool is throwing impossible voltage drops, check for these three specific errors:
- Mixing Mils and Inches in the Ratio: The $L/W$ ratio must be unitless. If $L$ is in inches (e.g., 2.0) and $W$ is in mils (e.g., 10), dividing them directly ($2.0 / 10 = 0.2$) yields a massive error. You must convert 10 mils to 0.010 inches first ($2.0 / 0.010 = 200$ squares).
- Confusing Bulk and Surface Resistivity: Bulk resistivity ($\rho$) is measured in $\Omega \cdot m$. Surface resistivity ($R_s$) is measured in $\Omega/\square$. Plugging a bulk resistivity value directly into the $R = R_s \times (L/W)$ equation without dividing by thickness ($t$) will result in an answer off by a factor of $10^6$ or more.
- Ignoring Temperature Coefficients: Copper's resistivity increases by about 0.39% per °C. A 1 oz copper trace carrying 5A will self-heat. If the trace reaches 70°C, its surface resistance increases by roughly 20% compared to the 20°C datasheet value.
Decision Path: Selecting Copper Weight or ESD Coating
Use this decision matrix to terminate your design choices with a concrete material pick based on your target surface resistance parameters.
| Application Scenario | Target Parameter | Concrete Pick / Action |
|---|---|---|
| Microcontroller logic traces (< 50mA) | Minimize fab cost; standard 4/4 mil routing | 1 oz (35 µm) copper. Keep widths at 8-10 mils. |
| High-current motor driver (> 5A) | Keep $R < 10 \text{ m}\Omega$ to limit $I^2R$ heating | 2 oz (70 µm) copper. Use 40+ mil widths or polygon pours. |
| Sensitive analog/RF shielding enclosures | Prevent static buildup without creating short-circuit risks | Anti-static coating ($10^9 \Omega/\square$). Look for carbon-loaded polyurethane. |
| ESD Bench Mat for component handling | Safely bleed off human body model (HBM) charges | Dissipative rubber ($10^6 \text{ to } 10^8 \Omega/\square$). Must meet ANSI/ESD S4.1. |
Realistic Magnitudes and When the Formula Fails
To build intuition, you need to know what a "normal" answer looks like. If your calculator outputs a number outside these bands, re-check your inputs:
- PCB Copper Traces: $0.1 \text{ m}\Omega$ to $500 \text{ m}\Omega$. (If you calculate $50 \Omega$ for a copper trace, your thickness unit is wrong).
- Anti-Static Coatings / Bags: $10^6$ to $10^9 \Omega/\square$.
- True Insulators (FR4, Kapton, Glass): $10^{12}$ to $10^{16} \Omega/\square$.
When the formula fails:
The surface resistance formula is strictly a DC and low-frequency model. Once your signal frequency exceeds roughly 10 MHz, the skin effect forces current to flow only in the outer few micrometers of the conductor. At 100 MHz, the effective thickness ($t$) of a 1 oz copper trace shrinks dramatically, causing the AC resistance to be significantly higher than the DC surface resistance calculated here. For RF and high-speed digital design, you must abandon this formula and use a 2D field solver to extract AC impedance.
Final Recommendation: For general-purpose DIY and prototype PCBs, default to 1 oz copper for logic signals and 2 oz copper for any power rail carrying over 3A. For bench safety, never rely on bare wood or standard plastics; purchase a verified dissipative mat rated between $10^6$ and $10^9 \Omega/\square$ to protect your silicon.
