Andrew, I think you will find a common occurance(typically in thicker boards) that the center of the product does not heat up at the same rate as the outer sections due to the thermal conductivity of the laminate being used. I feel we must take into account the TIME that is required for thermal expansion. If the thermal expansion of the laminate is quicker and greater than that of the copper then we may see stress cracks in the center of the board. The secondary concern is of course the tensile strength, elongation and ductility of the copper being plated. The thickness of the plated copper also plays a very large role in this due to the fact that the thicker the copper the higher percent of elongation. As you have pointed out on most thick product the plating is thinner in the center of the hole. Regards, Alan B. Cochrane Multek Inc. ______________________________ Reply Separator _________________________________ Subject: Fab: Interconnect Failure Author: [log in to unmask] at INTERNET_GATEWAY Date: 11/19/96 10:33 AM ---------------------------- Forwarded with Changes --------------------------- From: Engelmaier::(ENGLMRAA) at ~FABRIK Date: 11/18/96 9:06PM To: Andrew P Magee at Rogers-MCD *To: TECHNT at ~FABRIK Subject: Re: Interconnect Failure ---------------------------------------------------------------------- --------- Mr. Engelmaier, Why do you say that the stress of z-axis expansion is greatest near the center? Thermal expansion is measured in terms of expansion per unit length (strain) per change in temp. ( in/in/C ). If two dissimilar materials like copper and epoxy are bonded to each other then they expand as a composite structure based on: Comp. Strain= dT x Sum( Ei x Ai x CTEi) / Sum( Ei x Ai) Where Ei = Tensile Modulus of layer i Ai = Area of layer i CTEi = Coefficient of Thermal Expansion of layer i dT = Change in Temperature The difference between each layer's free strain and this composite strain generates the stress in the material. From Hooke's Law: Layer Stress = Ei x (Comp. Strain - dT x CTEi) As you can see there are no length terms involved in determining the stress. What do you propose is the mechanism for your observation that fractures typically occur near the center? Perhaps the uniformly generated stress is being concentrated by something like differential etchback or thinner copper plating at the center of the PTH. Andy Magee - Applications Engineer Rogers - Circuit Materials Tel: (602) 917-5237 Fax: (602) 917-5256 E-Mail: [log in to unmask] Website: http//www.rogers-corp.com/cmu //////////////////////////// // SERVICE TO THE LINE, // // ON THE LINE, // // ON TIME. // //////////////////////////// ______________________________ Forward Header __________________________________ Subject: Re: Interconnect Failure Author: Engelmaier::(ENGLMRAA) at ~FABRIK Date: 11/18/96 9:06 PM From: [log in to unmask] Date: Mon, Nov 18, 1996 9:06 PM Subject: Re: Interconnect Failure To: Magee, Andrew P; TechNet The stress of z-axis thermal expansion is greatest near the center of a PWB--that is why PTH barrels fracture there. The reason why inner layer separations occur predominantly on the outermost inner layers is the combination of the radial stresses from the expanding resin and the bending moments from the land rotations on the inner layers because the resin can more freely expand in the z-direction away from the PTH. Werner Engelmaier Engelmaier Associates, Inc. Electronic Packaging, Interconnection and Reliability Consulting 23 Gunther Street Mendham, NJ 07945 USA Phone & Fax: 201-543-2747 E-Mail: [log in to unmask] *************************************************************************** * TechNet mail list is provided as a service by IPC using SmartList v3.05 * *************************************************************************** * To unsubscribe from this list at any time, send a message to: * * [log in to unmask] with <subject: unsubscribe> and no text. * *************************************************************************** *************************************************************************** * TechNet mail list is provided as a service by IPC using SmartList v3.05 * *************************************************************************** * To unsubscribe from this list at any time, send a message to: * * [log in to unmask] with <subject: unsubscribe> and no text. * ***************************************************************************