In the layout of printed circuit boards, width plays an important part, especially in the metal connections fabricated into the board called “traces.” Traces conduct signals to and from the pins of the components soldered to the board and carry varying amounts of current depending on the net. Additionally, the signal’s traces may have special electrical needs affecting their size, and PCB trace widths must follow standard guidelines for error-free manufacturing.
PCB Trace Width Parameters |
||
|---|---|---|
| Narrower trace | Wider trace | |
| Impedance | Increases | Decreases |
| Current capacity | Decreases | Increases |
| HDI routability/fine-pitch breakout | Increases | Decreases |
| Manufacturability | Decreases | Increases |
The Importance of PCB Trace Widths for Fabrication and Assembly
Traces are metal connections of various widths between solder pads, usually fabricated from copper during the circuit board etching process. Although the etching will change based on the board layer fabricated, the desired copper weight (thickness), and the type of materials used to build the board, the basic process is as follows:
- In preparation for etching, the entire board material layer has photoresist applied.
- The traces, pads, and other PCB circuitry elements exposed to the photoresist harden into a protective coating.
- Application of an etchant to remove the photoresist; hardened photoresist remains over those areas that will become the copper circuitry of the board layer
- The copper layer is then etched to remove the unprotected copper, leaving only the traces and other copper circuitry on the layer.
- After removing the protective photoresist chemically, the board layer is composited with other layers to build up the circuit board.
Etching, by nature, is an aggressive chemical process: fabricators must exercise great caution to ensure the traces aren’t over-etched, resulting in traces produced to less than their desired widths. Traces isolated on the board may etch down more than traces clustered together in groups due to the concentration of etching on that one area. Sometimes, it pays to slightly widen isolated traces to guard against this, etching smaller than desired to avoid traces more liable to manufacturing and operational failure. Another essential aspect of PCB fabrication is the thickness of the etched copper, or “weight”: traces with a greater copper weight can not be etched down as much as other traces, as their thickness exposes and attacks more of the copper layer laterally during the etchant’s perpendicular path to the substrate.
Trace widths can also impact the soldering processes used during circuit board assembly. Wide traces in power and ground routing can act like heat sinks, leading to uneven soldering temperatures and poor solder joints. When this happens to large parts, such as high pin-count BGAs or surface mount connectors, the imperfections can be difficult to find and correct, requiring expensive inspection and rework techniques. Designers can avoid this condition by ensuring copper features (traces, polygons, etc.) are distributed evenly about components. As complex as the manufacturing problems associated with incorrect trace widths are, they can have an even larger effect on the electrical performance of the board.
How Trace Widths Impact Signal Transmission
When it comes to the electrical performance of the board, trace widths play an essential part in signal and power integrity.
Signal integrity
Different trace widths can improve signal integrity to control crosstalk, electromagnetic interference (EMI), and other problems associated with interference. Here are some examples:
- Controlled impedance routing – While most digital traces use the minimum trace width, certain high-speed signals must be routed at specific widths to control the line’s characteristic impedance. If the length of these lines has different impedance values scattered through them, the impedance mismatches can create reflections of the signal and lead to distortion and power loss. To prevent this, the trace widths of these lines must be precisely determined based on calculations of the dielectric material of the board, spacing to other signals, and the copper weight.
- Microstrip and stripline – Sensitive high-speed transmission lines must be closely coupled with a reference ground plane to shield their traces. An internal routing layer sandwiched between two ground planes is known as a stripline configuration; traces on the board’s exterior with only the adjacent plane beneath it is a microstrip configuration. The stripline traces will be narrower due to the use of double-ground planes to match the impedance of the lines between the two configurations.
- Analog routing – Analog signals should be short, direct, and wider than other routed traces. This extra width helps keep trace impedance low; along the same lines, minimizing via usage and keeping traces on the same layer is better.
Power integrity
Differing trace widths are also needed to ensure clean power delivery to the components on the circuit board. Here are some areas to keep an eye on during layout:
- Short and direct routing – Just as analog routing, keep traces as short as possible to reduce the chance of these traces behaving like antennas and creating additional noise. Using 45° or rounded corners instead of right angles can reduce length relative to 90° corners.
- Use wide traces – Using wider traces with power routing will help reduce inductance in the line and crosstalk.
- Current and thermal considerations – Power traces conduct different current levels depending on the net – use different trace widths and copper weights as necessary. The amount of heat these lines generate with their current is also a factor; the more current, the more heat. Additionally, the power traces on the external layers of the board can benefit from air-cooling, but internal power routing can not. Therefore, internal power traces need to be wider than their counterparts on the surface layers.
Your Contract Manufacturer Optimizes Trace Width for DFM
Fortunately, most CAD tools today can control for multiple PCB trace widths. Starting with the default nets, layout engineers can set up a trace width value that will apply to each net that isn’t otherwise set up. However, to control individual nets, most systems allow the designer to specify a width and desired spacing for each one in the design.
CAD tools also provide the ability to set up groups or classes of nets. These classes allow designers to specify a group rule without setting up design constraints for every net. For instance, by setting up routing rules for all clock nets, all get routed with the same length. Design constraints like this are beneficial when routing through dense areas such as fine-pitch BGAs, where narrowing the trace width is necessary.
There are many ways to set up your PCB CAD system for trace routing and various trace width requirements necessary to design a circuit board successfully. At VSE, our engineering team has partnered with clients for over 40 years to answer these questions, and we are ready to help.
