Materials and Processes Used in Prototyping and Hardware Development
By Anthony Giordano
Here at Tomorrow Lab, we utilize many materials and processes to design our client’s products. Some of these are off-the-shelf components, while others are specially developed by our manufacturing partners. Our method of learn-make-test-break-repeat is fundamental to our development process. The more we learn about how the materials we work with are handled, the more we understand how to make the best possible products.
The goal of this series is to explain a set of common materials and processes we use in our business as we prototype and fabricate products for our clients. Anyone interested in hardware product development should enjoy the encyclopedia approach of this article. Since there are several topics to cover, this guide will be split into three parts; Part One: Electronics, Part Two: Housing Low-volume Methods, and Part Three: Housing High-Volume Methods, and distributed over the next few weeks.
In this section, we cover circuit boards. Our partners at NexPCB, as well as other reliable Chinese manufacturers, are responsible for making our custom PCBs, no matter how small or specific they may be. We develop custom PCBs that are designed to fit into the product, and each decision made by the Mechanical Engineering and Industrial Design team carefully considers how this may effect the electronics. Below you will learn more about the process of creating these incredible electronics that are responsible for many of the day-to-day products you use.
Part One: Electronics
- Rigid Printed Circuit Boards
Printed Circuit Boards (PCBs) are the solution for connecting many electronic components onto a device that requires little to no wires. This is done by creating a substrate that has copper traces layers sandwiched between insulated layers. The top insulated layer is called the solder mask, which is designed to keep the board safe from shorts between copper traces and other nearby electrical hazards. Above that layer is the silkscreen layer, also known as the outermost layer. The silkscreen is reserved for placing labels for components and their pins, brand logos, and sometimes even instructions for use.
PCBs are created in CAD (Computer Aided Drawing) software like CadSoft EagleCad, which is what we use at Tomorrow Lab. These programs allow people to proportionally design PCBs layer by layer. Depending on the complexity of the board, it can be a single-sided, double-sided, or even a multi-layered conducting board. The multiple layers are utilized to save space in the x and y plane, by stacking copper traces on top of each other.
While designing a PCB, engineers can choose components from a carefully curated library of footprints or relative dimensions; or create their own. When the files are ready to send to the factory for manufacturing, they are converted to Gerber files to instruct the machines and employees how to build the boards.
The first information gathered by the manufacturer from the Gerber files is the structure of the layers. For instance, a two-layered board has an epoxy substrate called FR4, which is sandwiched between two copper foil layers. FR4 is a widely accepted grade of flame resistant, fiberglass-reinforced epoxy laminate. Because of its excellent insulation and robustness, this substrate has become a PCB staple to help dissipate excessive heat from the copper traces or components. FR4 is the happy medium for a flame retardant design that is lightweight and relatively strong.
As multiple-layer PCBs are created, the layers are compressed and heated until they bind together. To create Vias (or inter-layer connections), the machine deposits a thin layer of chemical copper on the walls. During this binding, any required holes start with a laser before finishing with a CNC mill. These holes can be as small as 100 microns, which is one millionth of a meter. To really appreciate how small 100 microns is, consider this: the eye of a needle is 1,230 microns.
The machine then gathers the copper trace designs from the Gerber files to create a transparent film with a negative printout of the PCB design. In other words, the transparent sections delineate the copper traces, and the copper areas to be removed are blacked out. Depending on the PCBs intention, a layer of a type of light-sensitive photoresist is placed on each of the copper foils, with the film placed on top of it. The areas where the copper traces should be will allow the UV light to pass through and polymerize (or harden) the photoresist material covering it.
The PCB is introduced to a chemical solution that removes the un-polymerized photoresist, leaving either a positive or negative photoresist material where the copper traces will be. The PCB is placed in an etchant solution that removes the uncovered copper. Finally, the polymerized photoresist that is protecting the copper traces is also chemically removed. You now have a rigid PCB circuit!
References: SparkFun: PCB Basics.
2. Flex PCBs
While rigid boards are more common than flexible PCBs, when the restraints of a project require the electronics to fit into complex enclosures, or need a higher circuit density in a lightweight design, flexible circuit boards are an indispensable resource. This bendable option, however, comes at a higher material cost than standard rigid PCBs. Other downsides include that it can easily be scratched or dented, and flex boards are difficult to repair. Still, their bendable nature makes them very useful in areas like aerospace and medicine where the dimensions and shape of the design are critical. According to Molex, flex PCBs are also touted to maintain signal integrity by minimizing signal loss.
The main material is a very thin, heat-resistant, bendable polymer film. The film is sandwiched between two copper sheets, and supported with a paper laminate. Heat-activated coating allows each sheet to adhere to each other when placed in a high-temperature pressure chamber. Unlike standard rigid PCBs, this flexible material does not use fiberglass to support it. Its construction generally allows for a wider temperature range, especially when combined with Dupont cover lay. When the base is made of polyimide film, thermal expansion is reduced to a minimum. This is another area where flex PCBs have an advantage over the FR4-constructed rigid PCBs. Often the FR4 expands over time, causing stress on the joints of the components mounted on the board, and the alignment to become mismatched, or worse — the pins and/or components can snap off.
The copper traces can be made using polymer thick-film (PTF), which is a very low cost solution; or conventional copper sheets like electro deposited (ED), or rolled annealed (RA) copper foil. RA has permeated the flex circuit industry because of its unique smooth surface, which makes it less susceptible to high-frequency insertion loss, the total RF loss in the dielectric, conductor, and from leakage and radiation.
References: Different Copper Foils for Different Reasons, Molex Copper Flex Products
3. SMD Assembly
SMD (surface-mount design) has revolutionized circuit assembly by greatly reducing the cost of producing PCBs. By placing circuit components on the surface of PCBs, the cost and production time dropped dramatically when this process became widely used in the late 1980s.
Solder paste, the first material placed on the PCB pads, is added on by a precision solder dispensing machine. The most common solder paste is a tin-lead alloy. While there are many lead-free options, they tend to have a higher melting point. The lead-filled solder paste melts at 183°C, while the lead-free melts at ~220°C. When considering utilizing these other pastes, an engineer must adjust the design to include components that fall in this higher temperature range. In addition, lead-free solder pastes require a narrow temperature range to ensure reliable joints.
Next, a machine places the components on the board on top of the solder paste. The process is virtually automated, using robotic machines that do as their name implies — “pick and place.” These pick and place machines, also known as “chip shooters”, utilize suction to lift components in standardized packages and place them. The components are fed to the machine on reels, and a vacuum tip lifts up the components, checks them using an onboard camera, and places them on top of the appropriate paste-filled area.
A single component can be precisely placed on a board in a fraction of a second, and an entire relatively complex board can be completed within minutes. Once the components are placed on the board, usually by a conveyor belt, the board is passed off to a reflow oven, where the boards are heated at the precise temperature to solder the components without damaging them. Once soldered, the boards are automatically, then visually, inspected.
References: Surface Mount Board Assembly Soldering, Surface Mounted Technology Makes Manufacturing Processes Smooth,Automated Production Systems (APS) Gold-Place L40 Pick and Place Machine
Stay tuned for Part Two: Housing: Low-volume Methods which will cover casting, 3D printing, EDM, and thermoform!
Follow Tomorrow Lab on Twitter and Instagram @TomorrowLabNYC.
