Where there is a circuit board, there will be a connector. They have coexisted for almost 80 years and have come a long way since then. While the importance of having signals and power on the outside may be obvious, many of the characteristics of connectors in today's industry are driven by harsh environments, high-performance applications, high-risk conditions, and ultra-compact devices.
This article will delve into a variety of applications that cover the importance of reliability, speed, power capacity, and density for linker-based solutions, as well as a tool or two that may help designers.
The three main uses of PCB connectors are:
1. PCB interconnection: A rigid (or flexible) connection between two PCBS
2. PCB wiring connection: bundled wired connection for peripheral equipment outside the board
3. PCB programming/debugging connections: Connectors (or test point arrays) for debugging or programming, usually for microcontrollers or field programmable gate arrays (FPgas), etc
In these three PCB connections, the form, assembly and function of all connections will vary according to the application. First, PCB mounting connectors come in two main variants: surface mount and through-hole. Of course, there are a variety of surface mount methods, but for all intents and purposes, connectors can be mounted through pin holes or pin pads. This brings us to some of the first trade-offs that should be considered during the design process.
Tradeoffs to consider when designing
Installing a connector through a through-hole has several benefits, but the main benefit is usually a stronger mechanical connection. This is because the pins/leads run through the PCB and are usually soldered to the top and bottom layers. Depending on the application, this can be very important, especially where reliability and safety are critical. If the user can access and cycle frequently, then the through-hole connection will certainly provide better reliability and may reduce the likelihood of the device breaking down over time. Another major advantage of a through-hole component is that it can be more easily probed and reworked for prototyping or servicing.
Surface-mount connectors, on the other hand, can greatly save board space (because the components are only soldered to one layer) and, depending on the application, can also reduce the cost of manufacturing and assembly. Many connectors have alignment posts, but do not require welded connections. In addition, surface mount connectors can often contain a higher density of pins, which likewise helps pack more signals into a smaller area and saves valuable board space.
Because saving board space and optimizing density/volume are common motivations for product development, many designs include multiple boards that are rigid or flexibly connected to each other. This method can not only save board space, but also save more costs. For example, if a design contains a 256-pin Ball grid array (BGA) processor that requires 10+ layers (which greatly increases PCB manufacturing costs), but is connected to many peripherals/connectors (which may only require two or four layers), then a common way to save space and cost is to split them into two PCBS: A smaller embedded PCB with 10+ layers, a secondary PCB with only four layers and most peripheral components (e.g. as connectors). The "MCU PCB" is usually mounted to the "backplane PCB" through a high-density, high-pin number board-to-board connection.
If many satellite PCBS are required for displays and buttons, for example, flexible connections can be used. What is commonly referred to as flexible wiring is common in LCD and motor connections, often helping to achieve tight bending radii (which helps to maintain compactness and low profile) and physical stresses associated with components. The flexible connection can be a separate cable inserted into the connector or can be manufactured directly using the PCB. Figure 2 shows a camera lens that contains many (unplugged) flexible cable connections, while Figure 3 shows the difference in a "rigid flexible" PCB with an integrated flexible cable. Although it is sometimes difficult to design and specify to the supplier, adopting rigidly flexible techniques can greatly help save production costs and enable more reliable and tight-knit designs. However, due to the inability to physically separate the two PCBS, the assembly of the hard-soft board can be more difficult.
Consider the application to which the connector will be applied
While discussing the topic of wired connectivity, let's quickly address some other important things to consider. One theme is ease of use, and the perfect example is the standard USB port. Over the years, USB connectors have improved in all aspects, including current-carrying capability, signal density, and the reversible, directionless connection of USB-C. Conversely, direction-dependent connectors that are keyed and allow only one way, such as the classic USB 2.0 connector, help guide users and prevent misconnections. Locking connections usually provide better mechanical support and may require turning the plug (BNC connector) or squeezing the latch (RJ-45 network cable). Basically, if the connector needs to be reused, then ease of use should of course be the primary consideration for user requirements.
Some applications require high-speed, sensitive connections over long distances, which brings us to fiber optics. There are three main types of fiber connections: single-mode, multimode, and plastic fiber (POF). Multimode connections allow for higher bandwidths, but often suffer in long-distance applications due to their high dispersion and decay rate, making them ideal for shorter LAN-based connections. Single mode performs well over longer distances and is ideal for applications such as RF broadband (your local cable company). In addition, parallel data buses connected via Peripheral component Interconnect (PCI) are generally much faster than serial connections such as USB (although USB-C also allows parallel connections). The speed and performance of the application will determine how these interconnections are defined.
For some applications, such as aerospace and military, poor working conditions may eventually drive these requirements. Some connectors have special protection against electromagnetic interference (EMI), electrostatic discharge (ESD), vibration and/or moisture. A common decision by designers is whether connectors should be shielded. Shielded connectors (which are covered by some conductive metal and may include EMI gaskets) can provide additional protection against unwanted radiation and local magnetic fields, but are generally bulkier or more expensive than unshielded alternatives. Connectors with housing and corresponding pins can be grounded to help prevent ESD introduced by human touch or other local transient sources. Some connectors even include shock-absorbing contacts to help achieve high impact and high reliability applications. Finally, connections that need protection from external moisture usually contain (or allow) seals.
For high-load applications, choosing a connector with maximum current or rated voltage also drives the design process. Some connectors include mixed-signal pin layouts to support data and power connections. Typically, power pins have a larger current capacity and may be a bit thicker, helping to prevent two separate power and data connectors/cables.
Ac voltage applications can also drive the choice of connectors and require a minimum gap or interval between each pin (depending on the maximum voltage). This helps prevent arcing, which can be harmful to both the system and the operator. Connectors will always list power ratings and it is important to follow these specifications during the design process while maintaining a healthy margin.
Finally, sometimes the best solution is to design the connector out completely and simply use bare brass pads for spring-loaded interfaces. While providing a thinner package, it helps to reduce component costs and simplify probing (flexibility based on PCB location). The most common applications include programming or testing interfaces. Embedded designs often include debugging or programming ports, but why have connectors if they're not normally used? Cable technologies such as Tag-Connect enable flatter, lower cost solutions. Although usually used only as a test point, a series of pads are also considered for Pogo-Pin-based connections.
Today, many CAD programs support 3D visualization and mechanical file import/export. One example is Solidworks Electrical (SWE). While traditionally a mechanical design CAD program, Solidworks' electrical kits exist for easy integration with schematic-defined PCBS and associated connections. Designer benefits include electrical design tools that help define interconnections and generate cable specifications, wiring diagrams, and even files that can be used in other PCB design programs to help build netlists while providing complete visualization of system interconnections. While many PCB design programs include 3D visualization capabilities, some allow file interchangeability to help define PCB shapes (via 2D files) and import component packages (via 3D files), which can greatly help check for interference and optimize connector placement.
conclusion
The design of the PCB connection is completely dependent on the application, usually starting with the user requirements, considering the design requirements, and figuring out if any special features are needed. What is often forgotten is that manufacturing requirements must also be considered to ensure that manufacturing and assembly can be carried out with reasonable convenience and cost.
PCB design kits of all shapes and sizes include mechanical-related features that can significantly help reduce the risk of design iterations and optimize connector layouts.
