Werner Engelmaier wrote: > > Hi Phil, > The "feathery migration tracks" you are describing are most likely dendritic > growth, sometimes called 'Conductive Anodic Filaments (CAF)'. Dendritic > growth starts as the migration of metal, and other salts, along paths created > by structural imperfections. This migration requires moisture, water in most > cases, but the presence of an electrical potential is not necessary. When > electrical potential is applied, the migration is more rapid and organizes > itself into CAF. CAF leads to a reduction in insulation resistance, and > depending on the currents present can create localized shorts and burn-out. > CAF and electroCHEMICAL migration are two different phenomena. I suspect what Phil is observing is electrochemical migration, which results in the formation of metal dendrites, which frequently have a feathery or tree-like appearance. Terry gave a pretty good description of electrochemical migration, which is sometimes called simply electromigration, which gets it mixed up with non-electrochemical electromigration, as described by Bill Davis, another phenomenon which requires no humidity. In electrochemical migration on circuit boards, moisture is absorbed by hygroscopic contaminants on the surface of the circuit, and if these contaminants are ionic or other contaminant which are ionic are present, a layer of electrolyte is formed. If this electrolyte is present between two adjacent conductors with a dc voltage difference between them, 1.5 V can be enough, metal, usually tin, copper, or silver, can be electrochemically dissolved by oxidation to metal cations or metal complex cations at the anode or positive conductor. If the metal species formed are soluble, they will which move through the film of elctrolyte under the influence of the electric field between the conductors to the cathode, or negative conductor. There, the metal cations are electodeposited at the cathode, but unlike electroplating, which this failure mechanism resembles, a nice smooth deposit is not obtained. The metal redeposits as needle-like or tree-like structures called dendrites. No dc voltage, no dendrite formation. These dendrites like to grow along the little cracks on the substrate surface. Sometimes these dendrites grow from the cathode all the way to the anode and short the circuit. Not good. If one is lucky, the current flowing through the very fine dendrite is high enough for resistive heating to volatilize the dendrite. Chloride and some other anions tend to exacerbate the metal dissolution at the anode because they are specifially adsorbed, that is they form a covalent bond to the metal at the surface and facillitate corrosion. CAF occurs when an electolytic conductive path is formed from the anode to the cathode, usually in voids along glass fibers in epoxy glass laminates. It usually involves higher electrical fields than electrochemical migration, does not involve metal redeposition a the cathode. The conductive path is the metal salt solution, usually copper, not a metal filiment. ################################################################ TechNet E-Mail Forum provided as a free service by IPC using LISTSERV 1.8c ################################################################ To subscribe/unsubscribe, send a message to [log in to unmask] with following text in the body: To subscribe: SUBSCRIBE TechNet <your full name> To unsubscribe: SIGNOFF TechNet ################################################################ Please visit IPC's web site (http://www.ipc.org) "On-Line Services" section for additional information. For technical support contact Hugo Scaramuzza at [log in to unmask] or 847-509-9700 ext.312 ################################################################ ################################################################ TechNet E-Mail Forum provided as a free service by IPC using LISTSERV 1.8c ################################################################ To subscribe/unsubscribe, send a message to [log in to unmask] with following text in the body: To subscribe: SUBSCRIBE TechNet <your full name> To unsubscribe: SIGNOFF TechNet ################################################################ Please visit IPC's web site (http://www.ipc.org) "On-Line Services" section for additional information. For technical support contact Hugo Scaramuzza at [log in to unmask] or 847-509-9700 ext.312 ################################################################