RUSH PCB UK Ltd wants all its clients to be assured that we ship all products from our production facility only after testing them to be in perfect working condition. For this we have a variety of strict quality control procedures at each stage of our PCB assembly process. This includes not only standard services such as visual inspection and automated optical inspection, but also advanced test procedures such as X-ray inspection, and functional circuit testing or FCT. Although each method has its own advantages, for the most meticulous testing method, RUSH PCB UK Ltd recommends In-Circuit Testing or ICT.
In Circuit Testing (ICT)
Working at component levels, ICT allows localizing issues that may be present on the board under test. For instance, ICT can point to a specific device as the cause of the problem. With ICT, it is possible to test the individual voltage and current levels on the PCB, while including a step-by-step program execution.
The above helps in troubleshooting complex boards where the board is still a prototype and the design is not totally verified. Boards not passing the test may need reworking at component level to potentially save the batch. When this happens, our test engineers generate a Design for Assembly (DFA) recommendation to the client.
At RUSH PCB, we work closely with our clients and provide them flexible services tailored to their individual requirements. We have an engineering team to review the specific test requirements for a project, recommend the necessary equipment, and develop the test workflow. We even design test jigs if necessary. We have equipment to handle any type of project.
One of the advantages of using ICT is its speed of test. For instance, a few seconds is all it takes to test a complicated board. Therefore, projects involving large volumes of PCBs benefit exclusively from ICT. Apart from the speed of testing, detection of faults at component levels makes the diagnosis process faster and at the same time, does not involve a skilled operator.
However, for the ICT to be effective and accurate, the process requires a dedicated test fixture and a program. In addition, the PCB design must also allow the test machine and fixture to interface properly with the assembly. To maximize the test coverage and find the maximum number of potential faults, our clients must consider some points when designing the layout for their PCB assemblies. This not only reduces the redesign steps necessary at the prototype stage, it also helps in producing boards that perform right at the first attempt.
Making PCB Assembly Compatible with ICT
Test Pads—have a test pad on each electrical network on the PCB, including on unused IC pins. It should be possible to connect to the test pad via a spring-actuated test pin in the test fixture. For through-hole technology, the test pin can engage the component leg on the solder side.
It is usual to place 0.05-inch (1.27 mm) diameter test pads on a 0.1-inch (2.54 mm) grid, with the test pads spaced 0.1 inch (2.54 mm) from any component, and 0.125 inches (3.18 mm) away from the edge of the PCB. The above dimensions allow using long-lasting standard test pins. RUSH PCB UK Ltd does not recommend using test pads of reduced diameters, as thinner test pins are generally more expensive, requiring more frequent replacements.
Probing—place all test pads ideally on the solder side of the PCB to allow the test pins on the jig to access them from the bottom side. While it is possible to place test pads on the top, the construction of the test jig will become more complicated and expensive as it will require additional transfer probes and wiring.
Solder-Side Components—it is preferable to have no components on the solder side, other than small SMDs. Test fixtures usually have a vacuum plate to hold the PCB assembly from the bottom. It may be necessary to mill the vacuum plate for accommodating components on the bottom side. As milling is an expensive process, it is necessary to restrict the milling for bottom components to only a few millimeters.
Locating Holes—add locating or tooling holes to the PCB (not in the panel), to allow the test jig to locate the PCB in the fixture. Preferably use non-plated holes of 3 to 4 mm diameter. Locating two tooling holes in diagonally opposite corners will allow the test jig to accommodate the PCB unambiguously. Keeping a free space of 5 mm around each hole will ensure the tooling pins of the fixture will not cause shorting of components or tracks during the test.
Pull-Up Resistors—use pull-up or pull-down resistors on all floating pins, rather than connecting them directly to the power rails. For pins that hold other devices to a reset state or high impedance state, the presence of these resistors allows the test jig to control the pins. Tying the pins through pull-up or pull-down resistors also helps in product functioning, as the circuit can reject spurious signals. These resistors also help the test jig in isolating individual components when locating a fault.
Space for Pusher Rods—these are necessary to push down on the PCB when testing. ICT jigs usually have fixtures with 2 mm diameter pusher rods and necessary space should be available between components on the top-side of the PCB under test. Spacing them evenly around the PCB helps the jig manufacturer locate individual positions for the pusher rods.
Programming Devices—although capable of programming devices such as EEPROMs during testing by ICTs, the cycle time per board may go up. RUSH PCB UK Ltd recommends pre-programming such devices before assembly, and allowing the ICT to control them during testing.
Batteries—preferably, fit batteries only after the testing is over. As an alternative, use a removable link to connect/disconnect them during the testing.
Review—reviewing the design to ensure proper functioning is important before committing to a fixture. Moving test pads or components on a PCB can mean a new fixture, leading to time and cost overruns, as an ICT jig can be expensive and take some time to prepare.
We at the Electronic Manufacturing Services (EMS) from RUSH PCB UK Ltd offer advice to our clients on the above points and help them to design their PCB properly for compatibility with ICT. However, once the design meets DFA requirements, ICT provides fast and accurate testing, making the returns worth the investment.
Contrary to the construction of standard PCBs with a metal or fiberglass base, flex PCBs consist of a flexible polymer core and a Polyimide film as a substrate. The advantage of Polyimide is it does not soften when heated, and stays flexible after the initial thermosetting. Unlike several types of thermosetting resins that become rigid after being heated, Polyimide remains flexible, and that makes Polyimide a superior material in flex and rigid-flex PCB construction. RUSH PCB UK Ltd uses upgraded Polyimide film that has good resistance to humidity and tearing.
Rigid-flex PCBs basically connect flex PCB materials to rigid PCB materials. This allows the PCB to bend in certain areas, and to stay rigid in others. Therefore, the board remains strong, but flexible at the same time. When designers want to transfer signals between the rigid and flex parts, they need to design the rigid-flex PCB. While the flexible part of the board resembles a regular flex circuit, the rigid sections may use materials that standard rigid PCBs use, such as fiberglass.
According to RUSH PCB UK Ltd, both the manufacturer and the OEM benefit when they involve in the conceptual design of Printed Circuit Boards (PCBs). Specifically, for flex and rigid-flex PCBs, it is necessary for both to understand the full project requirements and implications up front.
The Concept Design Phase
The project usually starts with discussions between the two engineering teams to develop a broad understanding on several factors such as:
Functionality of the rigid-flex PCB
Objectives of the project related to the rigid-flex PCB
Will the PCB have active components?
Number of interconnects on the PCB
Requirement of special signal capabilities such as impedance control, high current carrying traces, and other factors such as RF design and or shielding and protection requirements, operating temperature requirements, and similar
Shape and size of the PCB
Answers to above questions help to clarify the requirements, based on which, the manufacturer can then give their inputs to the OEM regarding rigid-flex technology best suited to their needs. This mostly relates to the material and methods for high quality rigid-flex PCBs. The first consideration is to decide whether the application really needs a flex-based solution.
While identifying the additional levels of functionality, it is also necessary to look into higher levels of integration between parts. By simplifying complex levels of integrations, it may be possible to reduce the cost of the PCB project.
Completing the design concept phase and determining that the application does require a flex and rigid-flex PCB, the designers must delve into deeper specific details of the design.
Specific Detail Requirements
Although there may be several detail requirements specific to a particular project, those most common to all designs are:
Minimum and maximum bend requirements
Will the bend be permanent or will it flex periodically?
Length of flex between bends
Requirement of stiffeners and type of stiffeners necessary
Based on the requirements in the concept design and specific details the manufacturer can suggest various suitable materials and methods for the high-quality flex PCB capable of meeting the quality and life requirements of the OEM.
In the last decade or so, RUSH PCB UK Ltd has taken the rigid-flex circuit design and fabrication to significant levels of evolvement. For instance, the rigid areas of our rigid-flex designs are capable of the same complexity and density as that of our HDI boards. For instance, just as for our HDI boards, the rigid areas of our rigid-flex designs can have the same fine lines/spacing, high operating temperatures, high layer counts, and compliance to RoHS standards.
In earlier methods of fabrication, manufacturers used several layers of adhesives while fabricating the rigid areas. However, the high coefficient of thermal expansion of adhesives led to a significant amount of stress on vias during thermal cycles that the boards undergo during the assembly cycles, and during operation. As a result, vias placed within the rigid areas would develop cracks in the copper plating.
The adhesives in the rigid-flex system may come from the copper clad flex laminate, the coverlay, and the material that bonds the rigid and flex layers. For solving the issue of reliability of vias, RUSH PCB UK Ltd has made necessary changes in the materials and methods of construction to eliminate or minimize the use of adhesives.
Where earlier manufacturers used layers of copper bonded to the polyimide core with some acrylic type of adhesive, RUSH PCB UK Ltd uses adhesive-less laminates where the copper bonds directly to the polyimide core, eliminating the adhesive bond layers. Not only does this technique allow thinner PCB construction, it also allows for higher flexibility and vastly superior reliability.
Adhesive-less copper clad laminates have further advantages. They can operate at higher temperatures, their copper peel strength is higher, and they exert much reduced stress on vias due to lower Z-axis thermal expansion coefficients.
Similarly, earlier coverlay construction in rigid-flex design used full coverage types that covered the entire rigid area of the PCB. As the coverlay adhesive expanded, it would put vias and other PTH to severe expansion stress. RUSH PCB UK Ltd uses selective coverlay constructions that remain restricted to the exposed flex areas only, while extending to a maximum of 0.05 inches (1.27 mm) into the rigid areas. No via or PTH is placed in this interface area.
RUSH PCB UK Ltd also uses high temperature no-flow FR4 prepregs to laminate the rigid and flex layers together rather than layers of flex adhesive. The structure this technology achieves is dimensionally highly stable. In fact, this stability matches that of standard rigid PCBs.
High quality flex and rigid-flex circuits from RUSH PCB UK Ltd conform to IPC 2223C, the Sectional Design Standard for Flexible Printed Boards. This standard defines the minimization/elimination of the use of adhesives within rigid areas, use of adhesive-less substrates, and the use of partial/selective coverlay construction.
As the demand for portable electronic equipment grows, engineers are increasingly taxed with improving the capabilities of combining functionality with flexibility. This flexibility also takes the form of flexible circuits that fit where no other design solution can help. Along with a significant reduction of interconnects, and a substantially greater freedom of packaging geometry, the integrated hybrid of the rigid PCB and flex circuits allows designers to retain the precision, density, repeatability, and reliability of regular PCBs.
As predicted by RUSH PCB UK Ltd, the market for flexible circuits is going to continue to expand steadily in the future, just as it has been doing for the past three decades. The reasons for this are not hard to find, as, on one hand, flexible circuits continue to support the existing technology so important to different industries, while on the other, advanced flexible circuits are able to comfortably meet the futuristic demands being made by up-coming industries, including the military, avionics, aerospace, telecommunication, consumer electronics, medical, and automotive. To interpret the future of flexible circuits appropriately, it is necessary to consider the subject from three angles:
Newer configurations of flexible circuits 
Newer applications where flexible circuits are useful 
Newer technology being used for manufacturing flexible circuits 
Newer Configurations of Flexible Circuits
Depending on the market demand, manufacturers are always willing to add higher flex layer counts along with prevalent blind and buried via structures, embedded components, integrated connectors, sculptured flex, and more.
Again, based on application, manufacturers can offer flex designs requiring shielding for EMI/EMC in specific areas, asymmetrical constructions, and varying thicknesses between different rigid areas.
Apart from the regular rigid-flex PCB methods of constructions, there have been recent advances in newer configurations of flexible circuits available from different manufacturers. The standard rigid-flex PCB is rather a symmetrical construction, acts as a baseline for building upon, and offers good control over impedance.
Fabrication of standard rigid-flex PCB usually has flex layers at the center of the construction and even layer counts in both, the flex and the rigid areas. Although four to sixteen layers are common in designs, there can be more. Placing the flex layers at the center offers the maximum level of flexibility. However, manufacturers such as RUSH PCB UK Ltd also offer variations in configurations such as:
Odd Layer Counts
Varying Layer Count
Integrated ZIF Tail
Multiple Rigid Area Thickness
Although most designs prefer an even layer construction, manufacturers do offer odd layer counts, and this has its own advantages. For instance, a rigid-flex PCB may have seven layers in its rigid portion, and three flexible circuit layers. Requirements of stripline impedance control mainly drive designs of this nature, where the flex area requires two-sided shielding. The construction in the flex area usually has ground layers on the two outer layers sandwiching a signal layer between them—offering large numbers of interconnects between the rigid sections.
The major factor in the odd layer count design is both rigid and flex areas may have odd layers, and the layers counts in the rigid and flex areas may be mutually independent. Manufacturers offer other variations too—even layer count on one side of the core, and odd layer count on the other. The advantage being higher flexibility and higher reliability of bending both in short-term and long-term. Leaving out unwanted layers also reduces the cost of the design.
Complex design requirements such as blind via construction and or widely different dielectric thicknesses within the same PCB calls for an asymmetrical construction. Manufacturers prefer to shift the flex layers towards the bottom of the stack rather than place them in the middle. Although this does raise some concerns of warp and twist during manufacturing, using hold-down fixtures handles them easily.
Manufacturers offer another variety of construction with varying flex layer counts between rigid sections. For instance, the first and second rigid sections may have four to six layers of flex between them, but have only one or two flex layers between rigid sections two and three. Leaving out unnecessary flex layers helps to improve the bend capabilities significantly for the portion with lower number of layers.
By integrating a ZIF tail into the rigid end of a rigid-flex design, the manufacturer eliminates the necessity of mounting a ZIF connector. This is a boon in high-density designs, as it saves both real estate and cost, while producing a thin design.
High-density designs often require blind and buried vias, whereas close-pitch BGA may require via-in-pad design along with via fill and capping. Although dimensional tolerances of materials and manufacturing methods limit the number of lamination cycles in multi-layer rigid-flex PCBs, manufacturing them with via-in-pad design is possible by placing them within the rigid parts of the board.
Manufacturers also offer flex layers in separately configured independent pairs with an air gap in between. This has the advantage of improving the flexibility of the flex part substantially. Of course, this design is only applicable where there are more than two layers of flex. The absence of adhesives within the rigid areas offers greater reliability of the vias therein, resulting in long-term operation of the board.
Although an expensive and complex stackup design, construction of flex circuits with different thicknesses in multiple rigid areas is possible, but presently limited to two rigid areas with different thicknesses.
Special flex circuits requiring shielded layers for reducing the effects of EMI and RF interference use specialized films rather than copper layers. Using copper layers as shield is an expensive proposition. Instead, the special films act as effective shielding material while keeping the thickness of the flex down, thereby improving the flexibility.
Newer Applications Where Flexible Circuits Are Useful
One of the latest applications that flexible circuits have independently triggered as an explosion is the wearable electronics market. Wearing electronics on the body essentially calls for comfort, and flexible circuits guarantee this. Some examples of wearable electronic applications prevalent on the market are wrist-worn activity and body function monitors, foot-worn sensors, wearable baby monitors, medical sensors, pet monitors, and electronics on worn clothing. By bending and forming flexible circuits to suit the curve of the human body, the applications provide comfort for long wear and use.
Newer Technology Being Used for Manufacturing Flexible Circuits
Manufacturing flexible PCBs still follows the traditional methods of photo-lithography and etching to get rid of the excess copper. To make even thinner flex circuits, researchers at the McCormick School of Engineering at the Northwestern University are dispensing with the copper layer altogether. Rather, they are using a graphene-based ink sprayed onto the substrate in the required pattern to provide the electrical connections.
The advantages of using graphene are twofold. One, graphene can exist as only one atom thick, and its two-dimensional characteristics makes it both flexible as well as transparent. However, the researchers are spraying it on as 14-nanometer thick layers for creating the tracks and patterns. The second advantage is graphene being 250 times more conductive than copper is, only very thin layers are necessary. This improves the flexibility significantly, reduces the weight, and allows for even thinner flex circuits. The next attempt is to allow doping of graphene so that apart from its use as a conductor, it can be used as a semiconductor to make embedded transistors.
The combination of configuration, technology, and application is making rigid-flex circuits a formidable force in the electronics field. It has already overtaken the applications of traditional rigid printed circuit boards, and is threatening the more sophisticated uses of special PCBs in military, aerospace, medical, and consumer electronics industry.
Traditional Printed Circuit Boards (PCBs) are rigid, meaning they are not meant to be bent during use. A different type of circuit board is available for use in applications that need the board to flex or bend repeatedly—flexible circuit boards. Both these are not very useful if the application demands the circuit board be stretched. For this, RUSH PCB UK LTD recommends using stretchable PCB technology. 
Stretchable PCB Construction
Although stretchable PCB technology uses classical processes for production and assembly of such PCBs, the laminate is either Polyurethane or Polyimide. This has the advantage of realizing stretchable PCBs with relatively low investments. For ease of assembly of components on the substrate, manufacturers use one of two methods as follows.
Manufacturers reinforce the laminate locally using an interposer or a special coating. The alternate method is to use Stretch-Rigid technology. Rather than connect two rigid boards with a flexible PCB as in Rigid-Flex construction, Stretch-Rigid technology connects multiple rigid boards using stretchable substrates with embedded copper interconnection traces. The electronic components are soldered on the rigid parts. 
Properties of Stretchable PCBs
PCBs with stretchable substrates are useful for applications that require the PCB to stretch, twist, bend, or any combination thereof. The stretchable substrate is ductile enough to decouple mechanical resonances, which reduces the effort necessary for compensating mechanical tolerances.
Stretchable PCBs come in single or double layers, with Polyurethane being the usual stretchable substrate. Typical base material thickness varies between 90 and 100 µm or 3.5 and 3.9 mil, while the copper weight is usually 0.5 Oz or 17.5 µm.
As the substrate must stretch, manufacturers take special care to give the copper a high peel strength of about 5 N/mm or 456 Oz/in, and a tensile strength of 6 MPa or 870 psi at 50% strain.
The above features of the substrate allow the stretchable PCB a maximum stretchability of 30% of its original length and 10% stretchability for repeated elongations. This however, depends on the structure of the copper pattern on the stretchable substrate. As the maximum allowed temperature for soldering on the substrate is about 150°C, the assembly process uses SnBi solder and FR4 interposers.
This allows a usable operational temperature range of 0 to 100°C for stretchable PCBs. Where the application requires a stretchable substrate of short length and low volume, manufacturers prefer to use Polyurethane as the substrate material. If the application demands a long and high-volume substrate link between the rigid parts, Polyimide is preferable. 
Advantages of Stretchable PCBs
Stretchable PCBs are very useful in the industry where two parts of a machine move relative to each other and must be interconnected electrically. For instance, a sensor executing complex movements on a stationary machine is best interconnected using a stretchable PCB as it allows the sensor to move in multiple degrees of freedom, including linear and rotational. Apart from being able to twist and bend, the stretchable interconnect can also allow the sensor to move linearly away from the machine (stretch) when needed, with a maximum elongation of 30% of its original length.
Therefore, two or more rigid boards connected by stretchable substrates can change their individual positions very easily, can change their positional angles relative to each other, and move apart or come close to each other, while remaining electrically tethered to each other all the time. However, for repeated stretching and contractions, RUSH PCB recommends limiting elongation of stretchable PCBs to 10% of the original length. 
Mechanism of Stretchable PCBs
Although the thermoplastic Polyurethane that manufacturers use as substrate for stretchable PCBs can stretch inherently, copper traces in straight lines on the substrate prevent it from doing so, as copper is not ductile enough for the purpose. Manufacturers use special press and confidential lay-up techniques for bonding the standard ED or RA copper foil on the Polyurethane substrate. Once this is done, they use regular subtractive wet-etching PCB processing steps such as drilling, metallizing, imaging, plating, and etching for fabricating stretchable circuits.
As adding multiple layers of adhesive and Polyurethane substrates reduces the stretchability of the product, stretchable PCBs are mostly double-sided and four layers at the most. To maintain a homogeneous elastomeric construction, manufacturers apply a Polyurethane solder-mask or coverlay on the finished PCB. 
Assembly of a stretchable PCB uses the standard off-the-shelf surface mounting components soldered on its copper tracks. As these components are rigid, the areas where the components are positioned cannot stretch. Therefore, the concept of the stretchable circuit is basically small islands of a rigid nature holding a few SMD components interconnected with conductive copper foil on stretchable substrates. For a mechanically reliable PCB, the manufacturing technique follows a gradual transition from the rigid area to the flexible area and ultimately to the stretchable region.
To allow the copper traces on the substrate to flex without damage, the designer gives the traces a horseshoe shape rather than allowing them to travel in straight lines. The designer then places the horseshoe shapes alternately facing 180°, allowing them to meander along the path the straight trace would have normally taken. When stretched, the horseshoe tracks will uncurl without much stress. Other shapes such as triangular and sinusoidal interconnect traces can also stretch, but exhibit higher stresses, leading to lower reliability. This has led manufacturers to standardize on the horseshoe shape. 
Designers must note that stretching copper traces leads to a change in their resistance. For instance, tests conducted on copper traces with thickness of 15 µm, width 1 mm, and length of 80 mm showed an original resistance of 7.4 Ω, which increased to 13.5 Ω when the trace was stretched by 10%, to 23.8 Ω when stretched by 20%, and to 37.6 Ω when stretched by 30%. However, lab tests have verified that the trace maintained its conductivity even after a 300% stretching. 
Applications that demand the PCB be placed on a non-flat surface are the major users of stretchable PCBs. A conventional rigid PCB cannot be comfortably integrated on a non-flat surface such as that in wearable and implantable devices. Devices such as used in smart textiles, safety, sports and leisure, and biomedical applications often follow irregular shapes, and the printed circuit must follow the shape for proper integration.
Although it is possible to form a flexible circuit in the shape of a cone or a cylinder, only a stretchable circuit can be deformed onto any type of surface, as it has stretchable interconnects. 
For instance, stretchable PCB placed in the sole of a shoe can measure pressure with embedded sensors, collecting data with free movement of the user. Placed inside bandages, the pressure sensors on a stretchable circuit can measure the tightness of the applied bandage. 
A completely new range of electronic devices can make use of stretchable PCBs providing comfortability as their unique characteristic. Apart from the few uses listed above, stretchable PCBs are already being used in applications involving artificial skins, randomly shaped biomedical implants, and conformable light sources.
The rise of the mobile industry on one hand and the increasing demand for wearables on the other, combined with the increasing use of IoT in the industry, has led to the complexity and density of electronic designs to increase substantially in the last two decades. Simultaneously, these demands have also increased the challenges for designers of printed circuit boards (PCBs) tremendously. One of the ways PCB designers are coping with the issue is by embedding electronic components within the PCB substrates. This is fast becoming a feasible step for eminent board manufacturers such as RUSH PCB UK LTD.
Advantages of Embedding
Before starting the design, it is imperative to understand the advantages that embedding components brings, while at the same time considering the drawbacks of adding the fabrication steps leading to the embedding. In fact, there are potential effects on cost and production yield that the design team must factor in when considering embedding components within PCBs. Some of these advantages are:
Reduction in size and cost
Minimizing electrical path lengths
Decreasing parasitic capacitance and inductance
Reducing EMI effects
Improving thermal management
For RUSH PCB UK LTD, innovation in PCB technology comes basically from reduction in size and cost. Embedding components within the PCB substrates help to reduce the size of the board assembly. For complex products, a PCB embedded with components can potentially reduce the manufacturing costs.
High-frequency circuits are highly susceptible to the parasitic effects of long electric path lengths during PCB design. Embedding components within the PCB helps in minimizing electrical path lengths, thereby reducing the parasitic effects to a large extent.
Such reduction in path lengths when connecting embedded passive components to the pins of an IC can decrease the parasitic capacitance and inductance, thereby reducing load fluctuations and noise within the system. For instance, it is possible to place embedded passive components directly underneath the pins of an IC. This not only reduces the via inductance, but also minimizes potential negative parasitic effects, and improves device performance. In fact, embedding components within the substrates of a board allows reduction of path lengths over surface mounting.
It is possible to integrate an electromagnetic Interference shield around an embedded component. For instance, simply adding PTH all around the component can reduce noise coupling from outside. In certain applications, this may even eliminate the need for any additional surface-mounted shield.
It is also possible to add heat-conducting structures to an embedded component for improving thermal management. For instance, embedding thermal micro-vias to be directly in contact with the embedded component can help it to dissipate the heat to a thermal plane on an external layer. Adding thermal micro-vias also reduces thermal resistance, as the amount of heat traveling through the PCB substrate reduces.
One of the major concerns when embedding components within a PCB is the long-term reliability of the design. Solder joints on embedded components formed and placed within the laminates of a PCB can be affected when the PCB undergoes soldering processes such as reflow during assembly of surface mount devices. Embedded components can be an additional problem after manufacturing, as they cannot be easily tested or replaced once they have failed.
What Components can be Embedded?
RUSH PCB UK LTD considers two main categories of components fit for embedding into PCB laminates—passive and active. They are used in different ways and for different applications. As a large majority of embedded components are of the passive category, embedded resistors and capacitors are the most popular.
However, an embedded passive component does not mean that a discrete resistor or capacitor is placed inside a cavity within the substrate of a board. Rather, it is the selection of a specific layer material to form the resistive or capacitive structure of an embedded passive.
Benefits such as listed above make embedded components an alternative to discrete surface-mount passive components. Applications such as series termination resistors benefit from this technology tremendously, a huge number of transmission lines terminate at dense memory devices and ball-grid array (BGA) ICs.
RUSH PCB UK LTD can embed a chip within a PCB, but steps for other manufacturers may vary. Typically, the fabricator has to create space for the body of the IC, and this takes the form of a cavity. Approaches to chip embedding technology may take the following approaches:
CIP or Chip in Polymer: this involves embedding thin chips when building up dielectric layers of the PCB, rather than integrating them within the core layers. The fabricator can use standard laminated substrate materials.
ECBU or Embedded Chip Buildup: this involves mounting chips on polyimide films and building up interconnect structures thereon.
EWLP or Embedded wafer-level package: this involves performing all technology steps at the wafer level. IO area available is limited to the footprint size of the chip, as this technology essentially requires fan-in.
IMB or Integrated module board: this involves aligning the components and placing them within a cavity and using controlled-depth routing to place the cavity within a core laminate. Filling the cavity with molding polymer ensures chemical, electrical, and mechanical compatibility to the substrate. Impregnation of isotropic solder in the polymer helps to form reliable solder joints while laminating the embedded part into the stack.
Component Design Considerations for Embedding
RUSH PCB UK LTD considers layout of components and their physical orientation as important factors when designing for embedded purposes. It is also necessary to select proper substrate materials and compatible components, as this reduces the chances of failure during PCB fabrication.
Selecting specific materials is the key to determine the electrical properties of embedded passives. For instance, an embedded resistor is simply a sheet of resistive film, its dimensions defining the value of the resistance. The resistance of such material is dependent on the resistivity of the material, its length, and its cross-sectional area. Resistive film materials vary in their resistivity, and this directly influences the final resistance value. Therefore, selection of the material is critical to the design and the manufacturing process.
Manufacturers make embedded capacitors by arranging properly dimensioned copper cladding to act as plates, and placing suitable dielectric material in between. Designers calculate capacitance based on the dielectric constant of the material, the permittivity of free space, distance between the plates, and the area of the plates. The final capacitance value increases with an increase in the dielectric constant of the chosen material, an increase in the area of the plane, and decreases with an increase in the plane-to-plane distance in the board layers. Manufacturers use special material for maintaining dielectric strength, with a thin but dimensionally stable dielectric layer for creating embedded capacitors for power supply decoupling.
For making other active components such as ICs, manufacturers and designers select materials that provide substrate durability and long-term reliability of components within cavities. CTE or coefficient of thermal expansion defines the manner in which the material will change during high-temperature events such as reflow soldering of surface-mount components. It is highly imperative for the designer to select substrate material and polymer with matched CTE for filling cavities to maintain the integrity of the board structure.
RUSH PCB UK LTD has two ways of aligning and placing embedded components in cavities—face-up and face-down, with face-down being the preferred process. For a face-down alignment, the cavity depth needs to match the package height, and therefore, the manufacturer can embed chips of different thicknesses on the same layer. This allows for good thickness control for the dielectric material, and accurate component placement during assembly.
Manufacturing Processes for Embedding Components
Individual manufacturers will vary their fabricating processes for embedding depending on the type of PCB and the available equipment at their disposal. Broadly, manufacturing process at RUSH PCB UK LTD for embedding components follow two methods—one, aligning component and placing them within cavities, and the other, molding components into the substrates, building up additional structures thereon.
Manufacturers use different manufacturing and configuration techniques to make cavities in PCBs. Advancement in technology has led to better and more efficient methods of developing cavities for embedding active components. The new methods offer additional benefits such as higher production yields and improved reliability.
Drilling cavities with lasers offers the highest positional accuracy and precision of all methods, as it is possible to control a laser beam precisely for achieving uniform depth and wear as it removes dielectric material. Using a longer wavelength prevents the laser from penetrating copper layers, thereby forming a distinct stop layer. After forming the cavity, the fabricator adds an anisotropic conductive adhesive before placing a component inside the cavity. Application of heat and certain amount of pressure helps to melt the solder particles in the adhesive material, thereby forming reliable solder bonds.
More conventional methods use milling for creating cavities, as milling is more cost-effective than lasers are. Although improved technology allows making miniature milling tools, there is a practical limit to using milling and routing for cavity creation. Even so, milling is more popular as compared to lasers.
Some manufacturers prefer using thin wafer packages, integrating them directly into dielectric layers during the buildup, rather than drilling or routing cavities into the core material. The fabricator begins by die-bonding the thin chip to the substrate, following it up with a layer of liquid epoxy or an application of a laminated RCC or resin-coated copper film as a dielectric. He/she then applies a heated press lamination process, optimizing it to embed the chip without void formation.
Any design with embedded components will require proper documentation for reducing manufacturing time and cost. As the process of embedding components combines component assembly, packaging, and PCB manufacturing into a single manufacturing process, necessary documentation requires layer stack diagrams, NC drill files, fabrication notes, pick-n-place files, and assembly notes for effective PCB fabrication.
Market demand is pushing for high-density, low-profile electronic devices. Manufacturers are complying to this demand with the technology for embedding passive and active components within the board substrate. RUSH PCB UK LTD has successfully broken through potential barriers of reliability concerns and risks to production yields and cost.
Looking at a complicated Printed Circuit Board (PCB) such as the motherboard of a computer, you are likely to find several tracks going nowhere, and terminated rather abruptly. However, a closer inspection, preferably with a magnifying glass, will reveal more details at the point of termination of the track. Most likely, you will see it ending in a small PCB pad, not much larger than the width of the track itself, with or without a hole in its center.
Fig.1: Tracks on a PCB
Fig.2: Close-Up of a Via
In reality, the track does not terminate, but rather continues to travel, albeit on a different layer, hidden under the outermost layer of the PCB. The pad at its end is actually a small pipe through the insulating material, electrically connecting the two parts of the track. In PCB terminology, such an arrangement that allows tracks to continue, but on a different layer, are known as a PCB vias.
Types of Via in PCB
Multilayer PCBs use different types of vias for various purposes. There might be through-hole vias, blind vias, and buried vias in the same PCB. Although the construction of all vias is same, their nomenclature depends on the layers of origin and termination.
Fig.3: Type of Vias
For instance, a via originating from the outermost layer, traveling through the board, and terminating at the other outermost layer is a through-hole via. In its passage through the layers of the board, it may or may not connect to intermediate layers, depending on the necessity of the electrical circuit.
A blind via originates from one of the outermost layers, but terminates on an intermediate layer, and therefore, is visible only on the originating layer. It may or may not connect to other layers in between.
A buried via is not visible from either of the outermost layers, as it originates in one of the inner layers and terminates in another inner layer, possibly connecting other layers in between.
Construction of a Via
By design, a via consists of two outer pads and a copper tube electrically joining them. The two outer pads reside on the originating and the terminating layers of the PCB, while antipads on all intermediate layers allow electrical isolation of the copper tube from the electrical circuits on these layers as it passes through.
Fig.4: Construction of a Via
While the two outer pads and antipads are part of the layout pattern a fabricator etches onto the PCB, an electrode position process forms the copper tube connecting the two. Although in regular multilayer PCBs, you may find through-hole vias, these are less likely in high density interconnect or HDI boards.
Difference Between Plated Through Hole and Via
The major difference between the two lies in their construction process. A fabricator can electroplate a through-hole only after assembling all the layers of a multilayer PCB, since a through-hole spans all the layers, while he can form a complete via, including electroplating it, when assembling each layer pair.
Another advantage with vias in multi layer PCBs is, the designer can either stack or stagger them to suit the requirements of the circuit layout, while he or she cannot do that with a through-hole. Therefore, vias help in increasing the layout density of a board, allowing the designer to reduce the size of and/or number of layers on a multilayer PCB.
Filling a Via
The designer may decide to fill the vias in the PCB during manufacturing. While blind vias require filling to avoid surface dimpling, some designers may specify an additional epoxy filling after lamination to maintain better surface flatness.
You can ask your PCB fabricator to fill the vias with either epoxy or metal epoxy. The choice between epoxy and metal epoxy is that the latter is conductive. Therefore, if you have designed the via with a thermal application in mind, for instance, to disperse heat from one side to the other, filling the via with metal epoxy will be a better choice as opposed to epoxy filling of the non-conductive type. As its barrel always has a layer of copper, a via always retains electrical continuity, regardless of whether the filling is conductive or not.
In highly dense PCBs, especially those with fine-pitch components such as BGA, fabricators fill PCB vias with epoxy and planarize them to make them flat. Flash plating over them makes them perfectly flat and suitable for mounting BGAs.
Other applications may need a Faraday shield on one side of a chip, which could double as a heat sink as well. Stitching the underside of the chip with vias is a standard practice, while filling them up with a conductive epoxy fill, helps in the heat conduction.
When concerned with EMI, you may use multiple vias in the region of a ground strap, filling them up to provide a conductive wall. An impedance-controlled structure may also benefit from closely spaced and filled vias on either side.
Although you will find warnings about placing vias within pads as these can siphon off solder paste while soldering, leaving the joint devoid of solder, you may not have much choice when designing with very closely pitched BGA packages. The space available around the pads may not be adequate for a dog bone, and the only option may be a via in the SMT pad, or partially in it.
You can get over the solder siphoning by having the via epoxy filled, flattened, and plated to encapsulate it. The other side of the PCB via may not be important enough and you can wall it off with a mask.
Sometimes, to attain very high routing densities, it may be necessary for the designer to use landless vias. The trace directly enters the hole without a PCB pad. As the vias do not have PCB pads, the designer can pack in more traces in between adjacent vias.
Stacked Vias and Laser drilling
Even after using blind and buried vias, you may still not have enough room for proper routing. In such circumstances, you may consider using laser drilled micro vias and or stacked vias. The two major benefits of laser drilling are extremely fine holes (sub 0.004”), and excellent registration. Both are obvious benefits for very dense parts.
Fig.5: Laser-Drilled Micro Via
Fig.6: Staggered and Stacked Micro Vias
Laser drilling does not pass through the layers, unlike that in mechanical drilling. It vaporizes the top copper layer, burns through the substrate dielectric layer beneath it, and stops when it touches the bottom copper layer. This accounts for the v-shaped pit as against a straight hole a mechanical drill-bit creates.
For stacked vias, designers place laser drilled vias directly on top of each other. However, designers typically use stacked vias only when board real estate is at a premium.
Vias and Signal Integrity
Just as people do, electrical signals too find it much easier to take a direct route when traveling from point A to point B. A via in the signal path forces the signal to take a detour, and the signal integrity suffers as a result.
For high-speed signals, there is also the challenge of via stubs. For instance, a via taking a trace from L1 to L3, may leave a stub down to L16. A high-speed signal will typically traverse all the way from L1 to L16 before reflecting to L3. This will attenuate the signal, as the effective electrical stub length will be almost double its mechanical length. Designers of thick boards take care of the problem by removing the unnecessary part of the barrel by back drilling. HDI boards do not face this problem, as they use laser drilled micro vias and stack them up to the desired layer.
The above types of via in PCB can increase density and bring down the cost in volume production. Laser drilled PCB vias can increase multilayer density and reduce layer count, without reducing the trace width or trace spacing.
With rise-times of signals on the printed circuit boards (PCBs) continuing to drop, the age-old concerns related to signal integrity is always at the forefront of (PCB) Printed circuit board design. However, with the increasing quantities of printed circuits in high-density interconnect or HDI technology, there are some interesting new solutions.
Signal integrity analysis in PCBs has five major areas of concern:
Electromagnetic Interference (EMI)
Although HDI does offer improvements and alternatives for all the concerns above, it does not provide all the solutions. Signal integrity depends on the materials the PCB uses, and the materials the HDI technology uses, together with the PCB design rules and dimensional stack-up helps the electrical performance including signal integrity. Likewise, miniaturization of the PCB using the HDI technology is a major improvement for signal integrity.
HDI Benefits Signal Integrity
With new electronic components such as ball grid arrays and chip-scale packaging achieving widespread use, designers are creating PCBs with new fabrication technologies to accommodate parts with very fine pitches and small geometries. At the same time, clock speeds and signal bandwidths are becoming increasingly fast, and this is challenging system designers to reduce the effect of RFI and EMI on the performance of their products. Moreover, the constant demand for denser, smaller, faster, and lighter systems are compounding the problems with restrictions placed on cost targets.
With HDI incorporating microvia circuit interconnections, the products are able to utilize the smallest, newest, and fastest devices. With microvias, PCBs are able to cover decreasing cost targets, while meeting stringent RFI/EMI requirements, and maintaining HDI circuit signal integrity.
Advantages of Using Microvia Technology in HDI Circuits
Microvias are vias of diameter equal to or less than 150 microns or 6 mils. Designers and fabricators use them mostly as blind and buried vias to interconnect through one layer of dielectric within a multi-layer PCB. High-density PCB design benefits from the cost-effective fabrication of microvias.
Microvias offer several benefits from both a physical and an electrical standpoint. In comparison to their mechanically created counterparts, designers can create circuit systems with much better electrical performance and higher circuit densities, resulting in robust products that are lighter and smaller.
Along with reductions in board size, weight, thickness, and volume, come the benefits of lower costs and layer elimination. At the same time, microvias offer increased layout and wiring densities resulting in improved reliability.
However, the major benefits of microvias and higher density go to improving the electrical performance and signal integrity. This is mainly because the HDI technology and microvias offer ten times lower parasitic influence of through-hole PCB design, along with less reflections, fewer stubs, better noise margins, and less ground bounce effects.
Along with higher reliability achieved from the thin and balanced aspect ratio of microvias, the board has ground planes placed closer to the other layers. This results in lowering the surface distribution of capacitance, leading to a significant reduction in RFI/EMI.
HDI PCBs use thin dielectrics of high Tg and this offers improved thermal efficiencies. Not only does this reduce PCB thermal issues, it also helps the designer in streamlining thermal design PCB.
Improved Electrical Performance of HDI Circuits
The designer can place more ground plane around components, as they implement via-in-pad with microvias. The increase in routabilty offers better RFI/EMI performance due to the decrease in ground return loops.
As HDI circuits offer smaller PCB design along with more closely spaced traces, this contributes to signal integrity improvements. This helps in many ways—noise reduction, EMI reduction, signal propagation improvement, and lowers attenuation.
The improved reliability of HDI circuits with the use microvias also helps in PCB thermal issues. Heat travels better through the thin dielectrics. Streamlining thermal design PCB helps remove heat to the thermal layers. Several manufacturers make complex enhanced tape BGAs of thin, laser-drilled polyimide films to take advantage of PCB design with HDI.
The physical design of the microvia helps in reducing switching noise. The reason for this decrease is due to decrease in inductance and capacitance of the via, since it has a smaller diameter and length.
Signal termination may not be necessary in HDI circuits as devices are very close together. Since the thickness of the layers is also small, the designer can utilize the backside of the interconnection effectively as well.
Just as the signal path is important in PCB design, so is the return path. Moreover, the return path also influences the resistance, capacitance, and inductance experienced by the signal. As the signal return current takes the path of minimum energy, or the least impedance, the low frequencies follow the path minimizing the current loop.
Miniaturization from using HDI technology provides interconnections with shorter lengths, meaning signals have to traverse shorter distances from origin to destination. Simply by lowering the dielectric constant of the HDI material system, the designer can allow a size reduction of 28%, and still maintain the specified cross-talk. In fact, with proper design, the reduction in cross-talk may reach even 50%.
HDI PCB design not only helps in improving the integrity of signals, but the presence of thin dielectric helps with the PCB thermal issues as well. In fact, HDI technology helps with all the five major areas of concern related to signal integrity.