CircuiTree - March 2009 - (Page 30) Ask the Flexperts By Mark Verbrugge Stress’n the Little Stuff Is there a Web site, article, or book that has different copper trace fractures in flex circuits? I have a flex circuit with broken Cu trace in the zero insertion force (ZIF) connector section, which is reinforced with polyamide stiffener (Figure 1). What would be the potential root causes for this type of failure? Adjacent trace looks ok. variation: finger thickness. Note that in the lower left corner, the finger appears to be better than half the total thickness of the copper trace beneath it. Figure 2 shows a close-up of this area, and we note that the Sorry to say I am not aware of a book showing different copper trace fractures, although I can tell you every good flex design engineer has his or her own collection of pictures of what can go wrong. Even IPC has little published showing typical fractures. Let’s look at your application and a similar one and see if we can address the question of root cause (and I promise I’ll share a photo or two from my past). One needs to consider the end use when designing a flex. Will the application be dynamic or static? What handling is required when installing or servicing the flex in the end unit? Just because we have no intention of bending or forming a circuit does not mean that our circuit is not subject to bending, forming, or twisting forces. Installation may seem innocuous but can indeed apply significant strain if a discontinuity exists in the design. What is a discontinuity? Well, in the flex world, it is a change in material type, material thickness, or etch pattern that can concentrate force in a specific area. These stresses can be enough to damage a conductor without the operator/installer even being aware the damage occurred. Vibration and repeated duty cycles can also impact stress on a connection, resulting in eventual failure. This is just such a case as we have here. So why did the subject circuit fail? One word: discontinuity. The larger the discontinuity, the greater the likelihood of failure. When we look at Figure 1, we see a typical cross-section of a ZIF pattern on a flex circuit—but with one very nontypical 30 March 2009 • circuitree.com Fig 1 ZIF Cross Section Fig 2 Finger Close-Up Fig 3 ZIF Pattern cover material stops at the exact spot where the conductor thickness variation begins. What the design does is magnify light strain from insertion, vibration, or use and concentrates all of the energy in one small area, or stress concentration point. Even seemingly insignificant force on our flex is going to be focused at this very spot, leading to the fracture of the conductor, exactly what we see occurring. In this case, the discontinuity is so sharply defined by the interaction of the thick finger and the termination of the cover that it acted like a knife. The intimate bond between the cover and the conductor transferred all the energy to the thin conductor below, resulting in a break and failure of the device it was installed in. How could we have prevented that in this particular design? Well, the simple answer is by avoiding the discontinuity. We could have done this by an evenly subtractive etch process keeping our finger identical in height to the conductor. Minor variation in plating thickness, such as ENIG, is not a concern. The preferred method would be to extend the cover slightly beyond the discontinuity, allowing the strain to be spread out beyond the finger/trace transition. Cover placement is critical. We want to keep any change in conductor thickness or width encapsulated under our cover to act as a strain relief, much as an epoxy bead acts as a strain relief between a flex circuit and a stiffener. It is also desirable to have the cover extend somewhat into the ZIF connector. Let’s review a similar situation from my past. Figure 3 shows a top view of another ZIF connector designed many years ago. Note the cover does not extend past the transition from a thin conductor to the finger pattern. Can you guess what happened? Same result, only this time the discontinuity was in the plane of the conductor, not the z-axis. The solution was to extend the cover beyond the trace/finger transition zone. Discontinuities in a flex design are, in my opinion, the number one cause for failures in the field. A thorough design review for potential discontinuities is as important as running a complete DRC before releasing any design to production. ■ The Flexperts are Mark Finstad and Mark Verbrugge of Minco. Email: Mark.Finstad@minco.com Listen to Mark’s latest podcast at www.circuitree.com/podcast http://www.circuitree.com/podcast http://www.circuitree.com
For optimal viewing of this digital publication, please enable JavaScript and then refresh the page. If you would like to try to load the digital publication without using Flash Player detection, please click here.