Get one or two assemblies running perfect at test, then place a socket on those "golden" assemblies that you can plug the QFPs into. Re-test the golden units with the components that were removed. If they still pass, keep the two components as golden samples. Test a lot of the QFPs using the golden assembly(s). If you have failures, make sure it is not due to the socket contacts. If you still have failures, you have something wrong in the switching speed, clock frequency or some other performance issue where the component does not work in the given design. If you cull a number of good components on the golden unit, and you get no failures or very few failures, take the good parts and assemble them to 5 or 10 production assemblies using your standard assembly process. If you get failures, but you have good solder joints, then you have either components that fail due to heat issues from reflow, or you have bad boards with fractured vias. You can carefully unsolder and remove the failed components and retest them in the golden unit. If they all pass, something is wrong either with the production boards or the soldering. The only other cause would be some other components on the assembly causing the parts to fail (the design or performance issues mentioned previously). -----Original Message----- From: TechNet [mailto:[log in to unmask]] On Behalf Of Dale Ritzen Sent: Tuesday, July 22, 2008 7:44 AM To: [log in to unmask] Subject: [TN] They solder well... But they don't work! Importance: High Here's a good one for the collective "mind-meld"... We have a two-sided, non-RoHS compliant, all-SMT assembly (no PTH parts) that is reflowed under a leaded profile (cooler reflow temperatures than Pb-free). There is one 44-pin QFP on the topside of the board that consistently solders well (i.e. passes all IPC-A-610 Class 2 inspection criteria), but fails test until it is reflowed with a hand iron, a little flux and wire solder. We have about a 25% pass rate without reflowing the QFP. That jumps to around 95% once the QFP is reflowed. Several different date codes of the device have been tried with similar results. Thinking the ENIG pads under the device leads might be contaminated, we cleaned them on several test case 4-up panels, with no change in yield. Thinking it might be contaminated leads on the devices, our process guys gently scraped off the tinned coating on the leads of several devices and took it down to the Beryllium Copper finish. Same resultant yield. We've re-run the failing assemblies through a higher reflow temperature with extra "tacky flux" on them, trying to get a better reflow and use the action of the flux to purge any contamination that might be interfering in the solder junction. When done, the solder junctions on the device look like a "Target - Class 1,2,3" picture from IPC-A-610, but the yield is still around 25% unless we reflow the junctions with an iron, some flux and wire solder. There may be some things we're overlooking, but it's coming down to where we believe the wire bonds to the lead frame material could be the source of the problem. The intense, direct heat of the soldering iron might be enough to generate a reflow of the wire bond, resolving any bonding issues. But, on several different date codes? I will be getting the vendor involved shortly to test the devices and see if they come up with something. I'm fairly skeptical about doing this as typically the normal answer I get is "all is working as designed". We are all scratching our heads on this one... Any "words from the wise" would be appreciated! Thanks! Dale Ritzen, CQA Quality Manager/ISO Management Representative Austin Manufacturing Services Email: [log in to unmask] --------------------------------------------------- Technet Mail List provided as a service by IPC using LISTSERV 15.0 To unsubscribe, send a message to [log in to unmask] with following text in the BODY (NOT the subject field): SIGNOFF Technet To temporarily halt or (re-start) delivery of Technet send e-mail to [log in to unmask]: SET Technet NOMAIL or (MAIL) To receive ONE mailing per day of all the posts: send e-mail to [log in to unmask]: SET Technet Digest Search the archives of previous posts at: http://listserv.ipc.org/archives Please visit IPC web site http://www.ipc.org/contentpage.asp?Pageid=4.3.16 for additional information, or contact Keach Sasamori at [log in to unmask] or 847-615-7100 ext.2815 ----------------------------------------------------- --------------------------------------------------- Technet Mail List provided as a service by IPC using LISTSERV 15.0 To unsubscribe, send a message to [log in to unmask] with following text in the BODY (NOT the subject field): SIGNOFF Technet To temporarily halt or (re-start) delivery of Technet send e-mail to [log in to unmask]: SET Technet NOMAIL or (MAIL) To receive ONE mailing per day of all the posts: send e-mail to [log in to unmask]: SET Technet Digest Search the archives of previous posts at: http://listserv.ipc.org/archives Please visit IPC web site http://www.ipc.org/contentpage.asp?Pageid=4.3.16 for additional information, or contact Keach Sasamori at [log in to unmask] or 847-615-7100 ext.2815 -----------------------------------------------------