Connector Selection Checklist: RF, Board-to-Board & Wire-to-Board
Most of the work in how to choose a connector for PCB design is finished before anyone opens a connector catalogue, which is why a connector selection checklist starts with the design rather than with a part number. Three things are usually already settled by the time the layout is frozen: the impedance and frequency the RF path has to carry, the stack height and board outline the enclosure allows, and whether anyone will need to open the unit after it ships.
Treat those three as inputs rather than options. A 50 Ω path running from DC to 6 GHz points at coaxial interfaces and quietly removes ribbon, header and terminal-block families from the shortlist, whatever their unit price looks like. A 1.2 mm gap between two parallel boards is not a preference; it decides which board-to-board pitches stay realistic. If the assembly has to be serviced in the field, a permanently soldered board-to-board pair drops out long before cost is discussed.
Keeping those answers in one place is what makes the checklist worth reusing. It records what the design has already decided, which leaves a candidate list small enough to compare properly.
Connector Selection Criteria That Eliminate Families, In Order
Once those inputs are fixed, connector selection criteria narrow the field in sequence. Running the three questions below out of order is what creates the awkward cases: a part that meets the electrical requirement, cannot be written into a purchase order cleanly, and arrives with drawings that disagree with the footprint.
Signal, power, or both.
The first question is what actually travels through the contact. Signal-only paths with controlled impedance lead toward coaxial families; power-only paths lead toward terminal blocks and heavier headers; mixed paths usually force a compromise on pin count or plating. This single answer typically removes half the candidates.
Board-to-board, or board-to-cable.
A connection that stays fixed between two boards for the life of the product belongs to the board-to-board and FFC/FPC families. One that has to route around a hinge, pass through a chassis wall, or unplug for service belongs to wire-to-board or cable-assembly territory. Leaving this question until late is expensive, because a design that assumes a pluggable harness but specifies a soldered joint is reworked on the production line rather than on paper.
Current, and how often it is mated.
Current sets the minimum contact size and, indirectly, the pitch. Mating frequency sets the contact geometry. A connector mated once during assembly and one mated every time a technician swaps a module are different products, even when they share a 1.27 mm pitch and the same footprint.
The order matters because each answer narrows the next one. Choosing pitch first and working backwards to current is how a 0.5 mm part ends up specified for a load it cannot carry.
What Survives: Connector Selection By Application
After those three questions, what usually remains is one family, occasionally two. The table below is the comparison sheet worth carrying into connector selection by application. The last column is the item to have in writing before an order is raised.
Pitch, current, withstand voltage, temperature range, and contact and insulation resistance figures above come from published Kinghelm datasheets. Mating life is an industry-typical reference range rather than a guaranteed specification, and is worth confirming per part number..png)


Comments
Post a Comment