Designing a PCB with Embedded Components in Altium NEXUS
The traditional approach to circuit board design has been to place the components on the top and bottom surfaces of the board. This process is well supported by circuit board assembly houses, which typically use automated pick and place machines to position each component, ready to be soldered onto the surface. The ever increasing demand for smaller and more integrated electronic products, combined with the higher frequencies of the signals within these devices, drives the ongoing research into better ways to fabricate and assemble a circuit.
One technique that delivers both higher density and improved support for higher signal frequencies, is to embed components within the layers of the circuit structure. For example, embedding discrete components directly under an integrated circuit can result in shorter signal lengths, reduced resistance and parasitic inductance, leading to lower noise and EMI, and improved integrity of the circuit signals. These improvements deliver smaller and more reliable products, supporting faster signal speeds and higher bandwidths. Combined with the ongoing improvements being achieved in fabrication processes and technologies, they can also lead to a reduction in the product size, and lower fabrication and board-level assembly costs.
Embedding components places a number of unusual demands on each stage of the process from design to fabrication, to assembly, to testing and maintenance of the finished product.
How can a Component be Embedded?
Embedding components introduces a significant difference in manufacturing a circuit board - no longer is there a simple delineation between fabricating the bare-board; then loading components onto that bare-board during assembly. These two jobs were often handled by different companies, due to their distinctly different process and technology requirements. If embedded components are used, they must be fitted to the board during the fabrication process. While this was once the domain of highly specialized fab houses, the processes are well understood now and there are many fabricators that can manufacture circuits with embedded components.
There are two ways a component can be embedded: an open cavity is created such that the embedded component sits within that cavity and remains visible on the completed board, or they can be placed on an internal layer during fabrication then covered as the upper layers are added to the board during the fabrication process so that they are not visible on the completed board.
There are numerous approaches to fabricating a board with embedded components, the description and image below show one approach.
- The board starts as a double-sided copper-clad rigid core, these copper layers are etched and drilled as required.
- A prepreg+copper layer is applied to each side, which are etched and laser-drilled as required.
- Embedded components are mounted on this layer (one or both sides), using, for example, deposited solder paste and reflow processing.
- A cutout prepreg layer is added, with a cutout to create a cavity for each embedded component.
- A prepreg+copper outer layer is applied to each side, which are etched, laser drilled and thru-hole drilled as required.
The board is fabricated using build-up technology, embedded components are placed and encased as part of the process.
Note the laser-drilled micro vias used to access an embedded component on the lower side of the board.
Designing with Embedded Components
In the PCB editor, components can be placed on any signal layer, not just the traditional top or bottom surface signal layers. If they are placed on an inner copper layer that is covered, the components are called embedded components. There are two approaches to embedding components:
- a user-defined cavity creates the clearance required around the component,
- or for small packages such as 0201, there is no cavity, the component is simply encased as subsequent layers are added, resulting in a bump at each component site in the finished board.
When a component requires a cavity, this cavity can be completely enclosed within the board, or it can extend to one side of the board to create an opening. The image below shows 3 embedded components, the outer 2 have a cavity defined that results in them being open on the top side of the board. The middle component is on a lower layer, resulting in it being completely enclosed. From the designer's perspective, the component placement process is the same for both open cavity and closed cavity components.
Three embedded components, the cavities for the outer two components are open to the surface of the board, the middle one is completely embedded.
This component is completely embedded. To make it easier to interpret the image, the component has been highlighted with blue outlines, the cavity with orange outlines.
Defining the Cavity in the PCB Library Component
If a component is to be embedded and it requires a cavity, the cavity is defined as part of the component footprint in the PCB Library editor. Note that adding a cavity does not stop that component from being used on a surface layer, in this situation the software will ignore the cavity.
To define a cavity:
- Place a Region object on a mechanical layer. The object is placed so that it encloses the 3D body of the component, with sufficient clearance on each side. Check with the fabricator to find out how much clearance is required.
- Edit the Region object and set the Kind attribute to Cavity.
- Confirm that the Layer attribute is a suitable mechanical layer.
- Set the Cavity Height attribute to a suitable height, typically this will be the height of the 3D body plus the fabricator's recommended clearance.
The image below shows the PCB library editor, with:
- the green, selected cavity definition on the layer Mechanical 15,
- the red outlines of the component pads,
- the shaded purple of the 3D body objects that define the capacitor's two pads and body.
The cavity is defined by placing a region object on a mechanical layer, setting its Kind to Cavity and the Cavity Height to the required cavity depth.
Note the shaded purple region, this is the 3D body that sits in the cavity.
Placing and Orienting an Embedded Component
To embed a component, edit the component properties and set the Layer to the required internal copper layer. The direction that the embedded component orients (up or down) is defined by the Orientation specified for that copper layer in the Layer Stack Manager.
Set the Layer for the component in the Properties panel.
Set the Orientation for components on each signal layer in the Layer Stack Manager.
The Interaction Between the Cavity and the Layer Stack
Related article: PCB Layer Stack Management
The Cavity that you define in the PCB Library editor has a Height attribute. This height defines the distance that the software will remove all layers above the surface of the copper layer that the component is placed on.
To simplify the interaction between the cavity and the layers that it cuts through, the software ensures that a layer is not partially cut. If a cavity enters into a layer, such as a dielectric layer, but does not extend all the way through that dielectric layer, then the software automatically extends the cavity all the way through that layer.
The image below shows this, the darker drawing lines have been added to show the corners of the cavity and how the component sits below the surface of the last-cut layer, but the layer cutout continues all the way through that layer. This behavior applies to both internal and open cavities.
If a cavity does not completely pass through a layer the software automatically completes the creation of the cavity through that layer.
Embedded Components, SubStacks and Managed Stacks
The hover image below shows the Layer Stack Manager for a rigid-flex design that includes embedded components. Each separate zone or region of a rigid-flex design can be made up of a different number of layers. To achieve that you need to be able to define multiple stacks, referred to as substacks.
When the Rigid/Flex option has been enabled the Substack Selector button appears, click to select and configure each substack. Hover the cursor over the image to see the Flex substack.
When you embedded a component, the PCB editor has to manage how that embedded component affects the layer stack, not only in terms of how it is displayed but also in terms of calculated data such as solder mask openings and design rule checking. It does this by creating a stack for each unique combination of placed + cut layers needed by the various embedded components included in the design. These stacks are referred to as Managed Stacks.
The Managed Stack is created automatically when a component is embedded within the layers of the board. As managed stacks are created automatically there is no user input needed in their creation and management. The PCB editor checks for embedded components, tests if any of the currently managed stacks are suitable, and if not, creates a new one. The same applies when embedded components are removed, if a managed stack is no longer needed, it is automatically removed. To force the PCB editor to check if new managed stacks are needed, switch between 2D and 3D Layout Modes.
Like User Stacks, Managed Stacks are listed in the PCB panel when it is set to Layer Stack Regions. The image below shows the managed stacks for two embedded components, R1 and C15. Use this feature to examine the extent of each Managed Stack in the X, Y plane.
Examining the Layer Stacks, the two Managed Stacks are visible on the right.