.png)
Action list:
In this article, we are going to explore what happens when, instead of a return reference plane, we have another signal trace in the layer stackup. Unfortunately, this topology is becoming increasingly common.
The reason for this is often a lack of understanding of the concepts of fields and signals among designers. Sometimes, stakeholders who do not grasp how electromagnetic fields or printed circuit boards behave, make decisions about the stackup design.
When a PCB design uses two conductive layers, both dedicated to signal traces, we encounter a scenario like the one illustrated in the image here:
In this scenario, the fields from both signal traces are spreading out and interfering with each other. This situation exemplifies a classic mistake in printed circuit board (PCB) design—something that should be avoided at all costs.
I cannot emphasize this enough:
DO NOT CHOOSE THIS TYPE OF STACKUP!
Why is this so critical?
As explained earlier, the fields are no longer confined to a specific area. Instead, they are expanding outward from both signal traces. This means that not only do the fields from one signal trace interfere with the fields from the other, but they also contribute to further expansion of the electromagnetic fields.
The result is a significant increase in electromagnetic interference (EMI), which is precisely what we aim to prevent in low-EMI PCB design. The primary goal of PCB design for low EMI is to contain the fields, preventing them from spreading outwards.
However, with this type of stackup, the fields are doing the opposite—they're expanding and interacting, leading to higher levels of EMI and also Signal Integrity problems. This design approach is counterproductive to what we are trying to achieve.
To address this problem, designers often attempt to compensate by adding copper to both layers and connecting it to “ground” pins.
These copper areas, often referred to as copper pours or copper fills, are placed adjacent to the signal traces rather than directly above or below them. While this strategy might seem like a potential solution, it doesn't effectively resolve the underlying issue of field containment.