The Core Capacitance Formula Epsilon and Symbol Definitions
When designing custom capacitive sensors, calculating PCB parasitic capacitance, or selecting dielectric materials for high-voltage snubbers, you cannot rely solely off off-the-shelf component datasheets. You need the fundamental physics. The parallel-plate capacitance formula epsilon defines the exact relationship between a capacitor's physical geometry, its insulating material, and its ability to store charge.
The foundational equation is:
C = (ε · A) / d
Because absolute permittivity (ε) is rarely listed on material spec sheets, engineers almost always expand this into its practical form:
C = (ε₀ · εᵣ · A) / d
Below is the complete spec-sheet definition for every symbol in the equation. Do not attempt calculations without verifying your units against this table first.
| Symbol | Parameter | Standard SI Unit | Value / Description |
|---|---|---|---|
| C | Capacitance | Farads (F) | The total charge storage capacity. Realistic discrete values range from 1 pF (10⁻¹² F) to 10,000 µF (10⁻² F). |
| ε | Absolute Permittivity | Farads per meter (F/m) | The combined permittivity of the dielectric material (ε = ε₀ · εᵣ). |
| ε₀ | Vacuum Permittivity | Farads per meter (F/m) | A physical constant: ≈ 8.854 × 10⁻¹² F/m. (Source: NIST CODATA) |
| εᵣ | Relative Permittivity | Dimensionless | Also called the dielectric constant (κ). A multiplier representing how much better the material stores energy compared to a vacuum. |
| A | Plate Area | Square meters (m²) | The overlapping surface area of the two conductive plates. |
| d | Separation Distance | Meters (m) | The thickness of the dielectric gap between the plates. |
Dielectric Permittivity Data Table (Real-World Materials)
The relative permittivity (εᵣ) is the variable that dictates whether you are building a low-loss RF resonator or a high-density decoupling capacitor. Below is a data-dense reference table of common dielectric materials you will encounter on the bench or in PCB fab houses.
| Dielectric Material | Relative Permittivity (εᵣ) | Dielectric Strength (kV/mm) | Primary Engineering Application |
|---|---|---|---|
| Vacuum | 1.0000 | ∞ (No breakdown) | High-voltage reference standards, space-rated hardware. |
| Air (1 atm, 20°C) | 1.0006 | ~3.0 | Variable tuning capacitors, spark gaps. |
| PTFE (Teflon) | 2.1 | 60 - 100 | Ultra-low loss RF/microwave PCB substrates, precision audio. |
| Polypropylene (PP) | 2.2 - 2.4 | 40 - 65 | Snubber circuits, motor run capacitors, high-current pulse. |
| FR-4 (Standard PCB) | 4.2 - 4.8 | 15 - 30 | General purpose PCBs, parasitic ground-plane capacitance. |
| Barium Titanate (Class II) | 1,200 - 10,000 | 10 - 20 | High-density MLCCs (Multi-Layer Ceramic Capacitors) for decoupling. |
Reference data compiled from standard material datasheets and Georgia State University HyperPhysics tables.
When the Formula Applies (and Its Assumptions)
The capacitance formula epsilon is an idealized model. Before you plug numbers into your calculator, you must verify that your physical setup meets the underlying assumptions of the equation.
The Three Core Assumptions
- Uniform Electric Field: The formula assumes the electric field lines are perfectly straight and parallel between the plates. This is only true if the plate dimensions are vastly larger than the separation distance (A >> d²).
- Ignoring Fringing Fields: In reality, electric field lines bow outward at the edges of the plates. This 'fringing effect' adds stray capacitance. For macroscopic bench components, fringing adds 2% to 5% error. For micro-scale MEMS sensors or high-frequency PCB traces, fringing can account for 15%+ of the total capacitance, requiring 3D electromagnetic solvers rather than this simple formula.
- Linear Dielectric: The formula assumes εᵣ is constant regardless of the applied voltage. This is true for Class I ceramics (C0G/NP0) and films. However, Class II ceramics (X7R, Y5V) exhibit severe voltage coefficients; a 10µF X7R capacitor might drop to 2µF at its rated voltage, breaking the linear assumption.
Rearranged Forms for Circuit Design
On the bench, you rarely solve for C directly. Usually, you have a target capacitance and a fixed PCB area, and you need to find the required dielectric thickness, or you are measuring a parasitic capacitance and need to deduce the effective area. Here are the algebraically rearranged forms:
- Solving for Plate Area (A):
A = (C · d) / (ε₀ · εᵣ)
Use case: Determining the minimum copper pour area needed on a PCB to achieve a specific decoupling capacitance. - Solving for Separation Distance (d):
d = (ε₀ · εᵣ · A) / C
Use case: Calculating the maximum allowable dielectric thickness for a high-voltage film capacitor to maintain a target capacitance without exceeding the physical winding volume. - Solving for Relative Permittivity (εᵣ):
εᵣ = (C · d) / (ε₀ · A)
Use case: Characterizing an unknown potting compound or conformal coating by measuring the capacitance of a known test fixture before and after potting.
Worked Examples with Strict Unit Tracking
The most common point of failure in these calculations is unit conversion. Below are two complete, step-by-step bench scenarios with explicit unit tracking.
Example 1: Calculating PCB Ground Plane Parasitic Capacitance
Scenario: You are designing a high-impedance analog front-end. You have a 50 mm × 50 mm copper ground plane on an FR-4 PCB. The board thickness (distance to the underlying power plane) is 1.6 mm. The FR-4 has a relative permittivity of 4.5. What is the parasitic capacitance?
Step 1: Identify and convert all variables to base SI units (meters, Farads).
- ε₀ = 8.854 × 10⁻¹² F/m
- εᵣ = 4.5 (dimensionless)
- A = 50 mm × 50 mm = 2500 mm². Convert to m²: 2500 × (10⁻³ m)² = 0.0025 m²
- d = 1.6 mm. Convert to m: 0.0016 m
Step 2: Substitute into the formula.
C = (8.854 × 10⁻¹² F/m · 4.5 · 0.0025 m²) / 0.0016 m
Step 3: Calculate the numerator.
Numerator = 8.854 × 10⁻¹² · 4.5 · 0.0025 = 9.96075 × 10⁻¹⁴ F·m
Step 4: Divide by the denominator.
C = (9.96075 × 10⁻¹⁴ F·m) / 0.0016 m = 6.225 × 10⁻¹¹ F
Step 5: Convert to engineering notation.
6.225 × 10⁻¹¹ F = 62.25 × 10⁻¹² F = 62.25 pF.
Result: This 62 pF parasitic load will severely roll off high-frequency signals if the source impedance is above a few kilo-ohms.
Example 2: Sizing Dielectric Thickness for a Snubber Capacitor
Scenario: You are winding a custom polypropylene film snubber capacitor for an IGBT inverter. You need exactly 100 nF of capacitance. You have a metallized film strip that provides an overlapping area of 0.01 m². Polypropylene has an εᵣ of 2.2. How thick can the dielectric film be?
Step 1: Identify and convert variables.
- C = 100 nF = 100 × 10⁻⁹ F
- A = 0.01 m²
- εᵣ = 2.2
- ε₀ = 8.854 × 10⁻¹² F/m
Step 2: Use the rearranged formula for distance (d).
d = (ε₀ · εᵣ · A) / C
Step 3: Substitute and calculate the numerator.
Numerator = 8.854 × 10⁻¹² · 2.2 · 0.01 = 1.94788 × 10⁻¹³ F·m²
Step 4: Divide by capacitance.
d = (1.94788 × 10⁻¹³) / (100 × 10⁻⁹) = 1.94788 × 10⁻⁶ m
Step 5: Convert to practical manufacturing units.
1.94788 × 10⁻⁶ m = 1.95 µm.
Result: You must source a 1.95 µm thick polypropylene film. Checking the table above, PP breaks down at ~40 kV/mm. A 1.95 µm (0.00195 mm) thickness will fail at roughly 78V, meaning this geometry is only suitable for low-voltage circuits, not an IGBT inverter. You must increase Area (A) to allow for a thicker, higher-voltage dielectric.
Common Unit Mistakes That Break the Math
If your calculated capacitance is off by a factor of a million, you have almost certainly committed one of the following unit errors. For a deeper review of foundational component math, refer to the All About Circuits capacitor primer.
The most frequent error occurs when converting Area from mm² to m². Engineers will correctly convert length (1 mm = 10⁻³ m) but forget to square the conversion factor for area.
Wrong: 100 mm² = 100 × 10⁻³ m² = 0.1 m²
Right: 100 mm² = 100 × (10⁻³ m)² = 100 × 10⁻⁶ m² = 0.0001 m²
This single mistake shifts your final capacitance answer by a factor of 1,000.
Other Fatal Unit Errors
- Confusing ε and εᵣ: Plugging the dimensionless relative permittivity (e.g., 4.5 for FR-4) directly into the 'ε' slot without multiplying by ε₀ (8.854 × 10⁻¹²). This results in a mathematically massive, physically impossible capacitance value.
- Mixing Centimeters and Meters: Using centimeters for distance (d) but meters for Area (A). The formula demands strict adherence to a single base unit system (SI meters). Convert everything to meters before touching the calculator.
- Ignoring Dielectric Thickness vs. Board Thickness: In PCB calculations, 'd' is the distance between the specific copper layers (the dielectric core thickness), not the total thickness of the finished 4-layer or 6-layer board. Using total board thickness will artificially lower your calculated parasitic capacitance.
