Issues in embedding components within PCB substrates
It is a trend that has a significant impact on the entire electronics supply chain, a challenge that suppliers at every stage are now striving to meet. Making best use of these developments falls to the design engineering team, who now need access to design automation tools that can offer greater flexibility in the way PCBs are conceived and created.
The design rules of this new paradigm present their own challenges and it is here where EDA tools vendors are now focusing their development efforts, in order to enable more OEMs to gain competitive access to this enabling and evolutionary capability.
There are essentially two methods for embedding components into a substrate: formed or inserted. The former effectively uses patterns of copper plating and resistive thin film to create passive (resistive, capacitive or inductive) components on an embedded (or surface) layer. The latter is the more evolutionary, as it allows discrete components, bare die or even modules to be placed below the surface of the substrate.
There are many benefits to this and perhaps most prevalent is the greater component density it offers. An important aspect of this is the increased need for passive components, particularly capacitors which are needed in direct response to higher operating and signal frequencies. This has given rise to a trend to stack components vertically in order to minimise track lengths. Texas Instruments recently brought a 500mA step-down DC-DC converter to market using this method, to create a module measuring just 2.3mm by 2.0mm and just 1.0mm high.
Component manufacturers must constantly meet demand for new packaging options when bringing products to market, and the widespread use of surface-mount technology (SMT) particularly in passive components lends itself well to embedding components into PCBs. As SMT profiles continue to shrink, these same parts can now be mounted within or directly alongside a die embedded within a PCB; the 01005 (0402) package, for example, measures just 0.4mm by 0.2mm and can be as little as 0.15mm high.
However the method used to provide connectivity introduces further requirements. There are essentially two options here; connections formed with traditional soldering, or using copper vias. When solder is used, general-purpose tin plated multi-layer ceramic capacitors can be used, but this comes with a risk when embedding; secondary heating (when mounting SMTs on the surface, for example) can cause the solder paste to reflow around the embedded component and introduce possible failure.
To overcome this, the industry is beginning to displace soldering embedded components with connectivity through copper vias, but while this scenario avoids the issue of solder reflow, the components electrodes also need to be copper (as opposed to tin) in order to guarantee good connectivity. As a result, the industry is now producing SMT devices with copper electrodes, such as the GRU series introduced by Murata, which are intended specifically for embedding.
In a traditional workflow, the manufacturing stages are often discrete; the bare board is fabricated before being passed to assembly, where component placement machines are used to populate the PCB.
In the embedded component paradigm this changes; these stages are no longer discrete, as components now need to be placed within the PCB while its being fabricated. This presents challenges for both the PCB industry and the manufacturers of production equipment.
Components that are embedded within the substrate are placed within a cavity, either during or after the PCB substrate is fully formed; if the component can be placed after the PCB is complete the cavity is typically open on the surface. If the component is encapsulated within a multi-layer board, the component is completely embedded and must, therefore, be placed by the PCB manufacturer, which is creating a new market opportunity for SMT placement machine manufacturers.
Correspondingly, SMT machine manufacturers must also now consider the demands of embedded component placement. Often the cavity will offer very tight tolerances, perhaps as little as 20m, which creates a need for greater accuracy with SMT placement. For example, the self-alignment effect of solder paste can overcome a level of inaccuracy, but this is not the case with embedded components.
In addition, the force with which components are placed needs to be more closely controlled; damage to surface-mounted SMT components caused during placement can be found through visual inspection, but embedded components are typically not visible and so any fractures incurred could render an entire board faulty. Additional thermal events, such as reflow soldering on surface-mount components, can also compromise the integrity of embedded components.
|Related Articles||Editor's Choice|
|Related Articles||Editor's Choice|