Issue link: https://iconnect007.uberflip.com/i/1108006
16 FLEX007 MAGAZINE I APRIL 2019 nector suppliers, they are not possible or prac- tical to modify. Therefore, the end user must take steps in the design of the FPC and the housing to ensure the ZIF fingers are well-pro- tected during assembly and in the field. Figure 7 highlights the key elements of a recent ZIF contact failure on early prototypes of a rigid- flex circuit: • A—Blind insertion: This image shows the rigid-flex installed in position. The assem- bly process was challenging and required the operator to insert the ZIF end of the rigid- flex board (R) blindly into the ZIF receptacle under the PCB (L). After locking the ZIF receptacle, the operator would manually bend the rigid-flex into shape • B—Solid ground plane: For signal integ- rity reasons, electrical designers called for a solid ground plane in the bottom side of the flex layer. But this made the flex region extremely stiff and put great strain on the locked ZIF fingers because of the manual bend • C—Vias in the bend region: It is a big no-no to place plated vias of any kind in bend regions because they concentrate stress, and the vias here added even more stiffness to this flex area • D—Cracked finger: This cross-section clearly shows a full crack through the nickel and copper layers. There were several failures during this build cycle, all with the same failure mode Fortunately, as with the previous case stud- ies, these ZIF failures occurred early in devel- opment, so there was time to react. By the fol- lowing revision, each problem area had been addressed: • The solid ground plane was hatched after the electrical team determined an appro- priate pattern to preserve signal integrity Figure 7: These images highlight the construction and configuration of a rigid-flex circuit with a ZIF termination (A, B, and C), and a cross-section of a ZIF finger that cracked after assembly (D).