Graham, I wouldn't be happy hand reforming gull wings. You say that the bend will have to start at the component body. This means that a lot of stress will be put on the plastic to metal bond during the bending. Plastic mould compound doesn't have a particularly strong bond to leadframe metals at the best of times. This hand bending of the leads is likely to lead to delamination, giving an easy path into the package interior, wire bonds and die for corrosive contaminants. Gull wing leads are originally formed by die sets that grip the leads before making the bend. The tooling doesn't grip the package at all. This prevents stressing the lead to plastic package interface. Ideally I wouldn't do this. If I had to I would do some die penetration checks on some re-formed parts to see if there has been any delamination. If this shows no problems and you can be sure of the consistancy of the bends you make then it might be ok. You also need to consider the effect of the change in the lead shape on solder joint fatigue. The compliance of the lead is protecting the joint from the differential strains between the package and the board. Changing the lead profile will change the lead compliance. If it makes the lead too stiff then again there could be long term reliability problems. regards, Jeremy Drake Celestica ---------------------- Forwarded by Jeremy J Drake/HQ/CUK/Celestica on 26/08/98 13:15 --------------------------- "Collins, Graham" <[log in to unmask]> on 26/08/98 12:54:26 Please respond to "TechNet E-Mail Forum." <[log in to unmask]>; Please respond to "Collins, Graham" <[log in to unmask]> To: [log in to unmask] cc: (bcc: Jeremy J Drake/HQ/CUK/Celestica) Subject: [TN] reforming a SOIC Technetter's Let's say a customer designed a board with the wrong pad size, too far apart, such that a gull wing lead placed on the pad doesn't have a heel fillet. Obviously a bad thing. A redesign is in progress, but in the interim it has been proposed that the part be reformed to spread the feet a bit further apart (by about 20 thou). Component is a 20 lead SOIC. My questions and concerns are: - the bend will have to occur at the point where the lead enters the component body, otherwise there will be no clearance between the part and the board. How dangerous is this from a reliability perspective (class 3 product, conformally coated)? Should I take the next few weeks off? Flee the country? Anyone have job openings? - do any technetters have suggestions of a half decent way to reform these? I'm currently faced with doing it to about 100 to 150 parts, and doing it by hand does not appeal to me due to the variability. Any other suggestions? I did consider borrowing some of the Freeze-it that Steve's using on his diodes, and shrinking the boards, but the reflow process might disagree with that... thanks! Graham Collins