Direct Answer
Impedance control matches PCB trace impedance to the system's characteristic impedance to prevent signal reflections, ringing, and crosstalk in high-speed circuits. Without it, signals degrade as they travel along transmission lines, causing timing errors and intermittent failures. Engineers must specify target impedance, tolerance, and reference planes in fabrication documentation, then verify results with test coupons and TDR measurements during manufacturing.
---
The Physics Behind Controlled Impedance
A PCB trace is not just a copper wire; it is a transmission line with distributed capacitance, inductance, resistance, and conductance. When a signal transitions from a driver with a specific output impedance into a trace with a different characteristic impedance, a portion of the signal energy reflects back toward the source. This reflection adds to the forward-traveling signal, creating overshoot, undershoot, and ringing at the receiver.
The characteristic impedance (Z₀) of a microstrip or stripline trace is determined by four primary variables:
- Dielectric constant (Dk) of the laminate and prepreg materials
- Dielectric thickness between the signal trace and the adjacent reference plane
- Trace width and, for differential pairs, trace spacing
- Copper thickness and surface roughness
For a surface microstrip, the impedance decreases as trace width increases or as dielectric thickness decreases. For a stripline embedded between two reference planes, the same relationships apply, but the signal is shielded on both sides, reducing crosstalk and improving immunity to external noise.
The loss tangent (Df) of the dielectric material also matters at higher frequencies. Materials with lower loss tangent, such as Rogers or Megtron laminates, preserve signal amplitude better than standard FR4 at frequencies above 1 GHz. However, for most digital interfaces operating below 10 Gbps, standard FR4 with a well-controlled Dk is sufficient if the stackup is designed correctly.
How Impedance Control Works in Manufacturing
Stackup Design and Material Selection
The manufacturing process begins with the stackup. The fabricator selects laminate and prepreg materials that meet the dielectric constant requirements specified in your design. IPC-4101 defines the specification for base materials, including the electrical and mechanical properties that laminates must meet.
Standard FR4 has a dielectric constant that typically ranges from 3.8 to 4.5 depending on the resin content and glass weave style. This variation is a major source of impedance error. For tighter impedance tolerance, manufacturers may use materials with a more tightly controlled Dk, such as low-loss laminates or materials with a specific glass weave like 106, 1080, or 2116.
The stackup also defines the dielectric thickness between each signal layer and its reference plane. This thickness is controlled by the number of prepreg layers and the lamination pressure. A thicker dielectric increases impedance; a thinner dielectric decreases it.
Trace Geometry and Etching
Once the stackup is fixed, the trace width becomes the primary variable the manufacturer controls to hit the target impedance. The fabricator calculates the required trace width using the stackup parameters and then adjusts the artwork accordingly.
Copper etching introduces a practical constraint: the etch process removes copper from the sides of the trace as well as the top, creating a trapezoidal cross-section rather than a perfect rectangle. The difference between the top width and bottom width is called etch factor. For impedance control, the manufacturer must account for this and adjust the artwork width so the average width produces the correct impedance.
Copper weight also matters. A 1 oz copper trace has a nominal thickness of 35 microns, but the actual thickness can vary by ±10%. Thicker copper increases the cross-sectional area, which lowers resistance but also changes the impedance slightly. For controlled impedance, the manufacturer must know the exact copper weight you intend to use.
Reference Planes and Return Current Paths
Every controlled impedance trace must have a continuous reference plane adjacent to it. The reference plane provides the return current path for the signal. If the plane is split or has a gap beneath the trace, the return current must detour around the gap, increasing the loop area and causing impedance discontinuity.
For differential pairs, the two traces must be coupled to each other and to the reference plane. The spacing between the pair determines the coupling coefficient. Tighter spacing increases coupling, which lowers the differential impedance for a given trace width.
Specifying Impedance Control in Your Design Files
What to Include in the Fabrication Drawing
The fabrication drawing must contain a clear impedance control note. This note should list each net or bus with its target impedance, tolerance, and the layers involved. A typical note might read:
> "Controlled impedance required. 50-ohm single-ended ±10% on layers 1 and 4. 100-ohm differential ±10% on layer 3. Reference planes: layer 2 for layers 1 and 3; layer 3 for layer 4."
The note should also specify the test method, typically TDR (time-domain reflectometry), and whether test coupons are required on the panel.
The Impedance Control Table
A table is the clearest way to communicate impedance requirements across multiple nets. Include the following columns:
| Net/Bus | Target Impedance | Tolerance | Layer | Reference Plane | Type |
|---|---|---|---|---|---|
| USB 3.0 TX/RX | 90 Ω differential | ±10% | L1 | L2 | Differential |
| DDR4 DQ | 40 Ω single-ended | ±10% | L3 | L2/L4 | Single-ended |
| PCIe Gen3 | 85 Ω differential | ±10% | L1 | L2 | Differential |
| Ethernet | 100 Ω differential | ±10% | L4 | L3 | Differential |
This table gives the manufacturer everything needed to calculate trace widths and set up test coupons. Without it, the manufacturer must guess, and the result will likely not meet your signal integrity requirements.
Common Specification Mistakes
Engineers frequently make these errors when specifying impedance control:
- Using a generic impedance calculator without the actual stackup. Online calculators assume a specific Dk and dielectric thickness. Your manufacturer's stackup will differ.
- Forgetting to specify the reference plane. If the reference plane is not stated, the manufacturer may assume the wrong layer, producing incorrect impedance.
- Ignoring solder mask effects. Solder mask has a dielectric constant of approximately 3.5 to 4.0 and can lower impedance by 2 to 5 ohms on surface traces. The manufacturer should account for this in the calculation.
- Not specifying tolerance. A default tolerance of ±10% is common, but some interfaces require tighter control. If you do not specify, the manufacturer will use their standard tolerance, which may not be sufficient.
Verification Methods: Test Coupons and TDR
Test Coupon Design
Impedance verification requires test coupons on the same panel as the production boards. A test coupon is a small pattern that replicates the trace geometry and stackup of the controlled impedance nets. It is placed in the panel margin and removed after fabrication.
The coupon must be designed to match the production trace width, spacing, and reference plane configuration. IPC-TM-650 provides test methods for measuring impedance, including the TDR method described in test method 2.5.5.7.
TDR Measurement Process
TDR sends a fast-rising step pulse down the trace and measures the reflected energy. The reflection coefficient at each point along the trace indicates the local impedance. A 50-ohm system with a 50-ohm trace produces no reflection; a mismatch produces a positive or negative reflection.
The manufacturer measures the coupon impedance and records the values. If the measured impedance falls within the specified tolerance, the panel passes. If it does not, the manufacturer must adjust the process—typically by changing trace width or dielectric thickness—and re-run the test.
When to Verify in Simulation
Simulation tools like HyperLynx, SIwave, or Polar Si9000 can predict impedance before fabrication. These tools use the stackup parameters and trace geometry to calculate expected impedance. However, simulation is only as accurate as the input data. The dielectric constant of FR4 varies with frequency and resin content, and the actual etched trace width differs from the design width.
Use simulation to validate your design before sending it to fabrication, but do not rely on it as the final verification. The TDR measurement on the actual panel is the authoritative check.
Practical Example: A DDR4 Memory Interface
Consider a DDR4 interface running at 2400 MT/s. The data signals (DQ) require 40-ohm single-ended impedance, and the differential strobe signals (DQS) require 80-ohm differential impedance. The stackup uses a 6-layer board with the following configuration:
- Layer 1: Signal (top), 1 oz copper
- Layer 2: Ground plane
- Layer 3: Signal, 1 oz copper
- Layer 4: Power plane
- Layer 5: Signal, 1 oz copper
- Layer 6: Ground plane
The dielectric between layer 1 and layer 2 is a single prepreg layer of 2116 glass weave with a Dk of 4.2 and a thickness of 4.2 mils. The manufacturer calculates that a 5.5-mil trace width on layer 1 produces 40-ohm impedance. For the differential pair on layer 1, a 4.5-mil trace width with 6-mil spacing produces 80-ohm differential impedance.
The fabrication drawing specifies these values, and the manufacturer places test coupons on the panel. After lamination and etching, the TDR measurement shows 39.2 ohms for the single-ended traces and 78.5 ohms for the differential pairs—both within the ±10% tolerance.
If the measurement had failed, the manufacturer would adjust the trace width in the artwork and re-run the panel. This iteration adds time and cost, which is why early collaboration with the manufacturer is essential.
How Impedance Control Affects Cost and Lead Time
Impedance control increases manufacturing cost for several reasons:
- Material selection. Laminates with tighter Dk tolerance cost more than standard FR4.
- Process control. The manufacturer must monitor etching and lamination more carefully to maintain consistent trace geometry and dielectric thickness.
- Test coupons and TDR measurement. Each panel requires coupon fabrication and measurement time.
- Potential rework. If impedance fails, the panel may be scrapped or reworked, adding cost.
Lead time also increases because the manufacturer may need to order specific materials and run additional test cycles. A poorly specified impedance requirement can lead to multiple iterations, further increasing both cost and lead time.
To minimize these impacts, provide complete impedance specifications in your initial RFQ and work with the manufacturer to select materials that balance performance and cost. For boards that do not require impedance control, standard FR4 processing is faster and less expensive.
Common Mistakes and How to Avoid Them
Mistake 1: Assuming All FR4 Is the Same
FR4 is a generic term covering many laminate formulations with different Dk values. A 50-ohm trace width calculated for one FR4 material may produce 55 ohms on another. Always specify the material type or ask the manufacturer for their recommended stackup.
Mistake 2: Ignoring the Effect of Adjacent Copper
Ground pours or power planes on the same layer as a controlled impedance trace can change the impedance by altering the electromagnetic field around the trace. Keep a clearance of at least three times the dielectric thickness between the trace and any adjacent copper.
Mistake 3: Forgetting Via Transitions
A via transition from one layer to another introduces an impedance discontinuity. The via stub, pad, and anti-pad all affect the impedance. For high-speed signals, use back-drilling to remove via stubs or design the via with a controlled impedance structure.
Mistake 4: Not Involving the Manufacturer Early
The manufacturer's stackup and material availability determine what is achievable. If you design a stackup that the manufacturer cannot produce, you will face delays and cost overruns. Share your impedance requirements before finalizing the layout.
When to Involve the Manufacturer
Involve the manufacturer as early as possible—ideally during stackup selection. Most manufacturers provide a recommended stackup for a given layer count and impedance requirement. This stackup specifies the material type, dielectric thickness, and copper weight that will achieve your target impedance.
For complex designs with multiple impedance requirements, request a pre-layout stackup recommendation. The manufacturer can calculate the required trace widths and spacing for each impedance value, saving you time and reducing the risk of errors.
Omini's engineering team can review your stackup and impedance requirements before fabrication, ensuring that your design is manufacturable and that the impedance targets are achievable with the selected materials. This early collaboration reduces the risk of failed panels and costly rework.
Impedance Control in Different PCB Technologies
Flex and Rigid-Flex PCBs
Flex PCBs present additional impedance control challenges because the flexible polyimide material has a different dielectric constant than FR4, and the material thickness can vary during the lamination process. The flex PCB manufacturing process requires careful control of the coverlay thickness and adhesive layers, which affect impedance. For a detailed overview of how flex boards are built, see our article on the Flex PCB Manufacturing Process: An Overview for Beginners.
HDI PCBs
High-density interconnect (HDI) boards use microvias and finer trace geometries, which make impedance control more challenging. The smaller dielectric thicknesses and tighter tolerances require precise process control. HDI technology is often used in conjunction with controlled impedance for high-speed applications. Learn more about the tradeoffs in Decoding High-Density Interconnect (HDI) PCB Technology: An Overview.
Heavy Copper PCBs
Heavy copper boards use copper weights above 3 oz, which significantly changes the trace cross-section and impedance. The thicker copper requires wider traces to achieve the same impedance, and the etching process is more difficult to control. For more on this topic, see the Different Production Process Between Heavy Copper PCB and FR4 PCB.
Double-Sided PCBs
Double-sided boards can support controlled impedance, but the lack of internal reference planes means the designer must use the top and bottom layers carefully. A microstrip configuration on a double-sided board is possible, but the impedance is more sensitive to solder mask and surface variations. See Double Sided PCB: Advantages, Applications, and Manufacturing Process for more context.
Impedance Control and Assembly
Impedance control does not end at fabrication. The assembly process can also affect signal integrity. Solder mask openings, component pads, and via transitions all introduce impedance discontinuities. During SMT assembly, the placement of components and the routing of traces between pads must maintain the controlled impedance as much as possible.
For high-speed connectors and packages, the transition from the PCB trace to the component lead is critical. The pad size and the solder joint geometry affect the impedance. Some high-speed connectors are designed with controlled impedance footprints, and the PCB layout must match the connector's recommended footprint.
The SMT assembly process itself does not change the impedance of the traces, but it can affect the performance of the overall system. Poor solder joints, misaligned components, or excessive solder paste can create impedance mismatches at the component interface. For a review of the assembly steps, see our article on Explained: What are the Steps in SMT Assembly Process?.
> Practical note: When reviewing your assembly drawings, check that the solder mask opening around high-speed pads matches the manufacturer's recommendation. An oversized opening exposes more copper, which lowers the impedance at that point. An undersized opening can cause solder mask encroachment, which increases impedance.
Standards and Test Methods
IPC-2221 and IPC-2222 provide the general design framework for PCB layout, including guidelines for trace spacing and dielectric thickness. IPC-6012 covers the qualification and performance requirements for rigid PCBs, including impedance testing requirements for boards that specify controlled impedance. IPC-4101 defines the requirements for base materials, including the electrical properties that affect impedance. IPC-TM-650 provides the test methods used to measure impedance, including the TDR method.
These standards do not specify exact impedance values—those are determined by your system design. They do, however, define the test methods and acceptance criteria that manufacturers use to verify impedance control.
Final Recommendations
Specify impedance control early in the design process. Choose your stackup and materials based on the impedance requirements, not the other way around. Provide a complete impedance control table in your fabrication drawing, and verify the results with TDR measurements on test coupons.
Work with your manufacturer to select materials that meet your electrical requirements without exceeding your budget. For most designs, standard FR4 with a well-controlled Dk is sufficient. For higher frequencies or tighter tolerances, consider low-loss laminates.
Omini can review your stackup, calculate trace widths, and provide test coupons for impedance verification. Send your design files early and ask for a pre-layout stackup recommendation if you are unsure about the material selection. This collaboration ensures that your impedance requirements are met on the first pass, saving time and cost.
> Engineering handoff note: The Role of Multilayer PCB Fabrication in Modern Electronics Manufacturing before the release package is frozen.
> Engineering handoff note: Interpreting Gerber Files: A Manufacturer's Perspective before the release package is frozen.
> Engineering handoff note: How to Evaluate PCB Manufacturing Risk from PCB Design and KiCad Trends before the release package is frozen.