Where should the coil be placed?
Reply:
Switching regulators for voltage conversion use inductors to temporarily store energy. These inductors are usually very large in size and must be placed in the printed circuit board (PCB) layout of the switching regulator. This task is not difficult, because the current through the inductor may change, but not instantaneously. Changes can only be continuous, usually relatively slow.
The switching regulator switches the current back and forth between two different paths. This switching is very fast, and the specific switching speed depends on the duration of the switching edge. The trace through which current flows is called a thermal loop or AC current path, which conducts current in one switching state and does not conduct current in another switching state. In the PCB layout, the thermal loop area should be small and the path should be short in order to minimize the parasitic inductance in these traces. Parasitic trace inductance can produce useless voltage offset and cause electromagnetic interference (EMI).
Figure 1. Switching regulator for step-down conversion (with a critical thermal loop as shown by the dashed line).
Figure 1 shows a buck regulator, where the critical thermal loop is shown as a dashed line. It can be seen that the coil L1 is not part of the thermal circuit. Therefore, it can be assumed that the placement of the inductor is not important. It is correct to place the inductor outside the thermal loop-so in the first example, the placement location is secondary. However, some rules should be followed.
Do not route sensitive control traces under the inductor (neither on the surface or under the PCB), in the inner layer or on the back of the PCB. Affected by the flow of current, the coil generates a magnetic field, which will affect weak signals in the signal path. In a switching regulator, a key signal path is the feedback path, which connects the output voltage to the switching regulator IC or resistor divider.
Figure 2. Example circuit of the ADP2360 buck converter with coil placement.
It should also be noted that the actual coil has both capacitive and inductive effects. The first coil winding is directly connected to the switching node of the step-down switching regulator, as shown in Figure 1. As a result, the voltage change in the coil is as strong and rapid as the voltage at the switch node. Since the switching time in the circuit is very short and the input voltage is high, considerable coupling effects will occur on other paths on the PCB. Therefore, sensitive traces should be kept away from the coil.
Figure 2 shows an example layout of the ADP2360. In this figure, the important thermal circuit in Figure 1 is marked in green. It can be seen from the figure that the yellow feedback path is a certain distance away from the coil L1. It is located on the inner layer of the PCB.
Some circuit designers don't even want any copper layers in the PCB under the coil. For example, they provide cuts under the inductor, even in the ground plane layer. The goal is to prevent eddy currents from forming on the ground plane under the coil due to the coil's magnetic field. There is nothing wrong with this method, but there are also arguments that the ground plane should be consistent and should not be interrupted:
u The ground plane used for shielding works best when it is not interrupted.
u The more copper the PCB has, the better the heat dissipation.
u Even if eddy currents are generated, these currents can only flow locally, which will only cause small losses and hardly affect the function of the ground plane.
Therefore, I agree that the ground plane layer, even below the coil, should be kept intact.
In short, we can conclude that although the coil of the switching regulator is not part of the critical thermal loop, it is wise not to route sensitive control traces under or near the coil. The various planes on the PCB—for example, the ground plane or the VDD plane (supply voltage)—can be constructed continuously without cutting.