Printed Circuit Board Assembly & PCB Design SMT Electronics Assembly Manufacturing Forum

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.


Wire bonding pull strength & substrate contamination

Views: 2018

Rosewood

#37581

Wire bonding pull strength & substrate contamination | 3 November, 2005

We've recently encountered some problems with wire bond strengths. I'm looking for anybody that has had some experience with different types of substrate contamination and methods for cleaning. There is no visible contamination on the gold pads, but we cannot maintain good bonds. Our normal plasma cleaning process does not seem to remove this residue. We have verifed our wire bonders and cleaning process and we can relate the issue to specific lot numbers from our laminate substrate supplier. Niether our supplier or us can put on a finger on what is on these substrates.

Thanks.

reply »

#37587

Wire bonding pull strength & substrate contamination | 3 November, 2005

Questions are: * Is the problem 'wire bonds that don't stick' or 'wire bonds that fall or break off later'? * What materials are you working with? * What is your working time after plasma clean? * What does surface analysis tell you about surface contaminants?

reply »

Rosewood

#37611

Wire bonding pull strength & substrate contamination | 3 November, 2005

We have some no sticks, but that is actually a good thing because we have NSD. During process control audits we found that pull strengths to the board dropped off. We build a military application that requires higher bond strenghts. Low pull stregths wires break or separate during temp cycling.

We are bondig with gold wire on gold pads. Laminate substrate is BN300.

We are bonding immediately after plasma clean

Surface analysis does not show any spikes other than Ni Ag

reply »

#37630

Wire bonding pull strength & substrate contamination | 3 November, 2005

There maybe multiple drivers to your problems. * First, decreasing bond quality with time indicates a plating issue. What's the actual thickness of your gold and nickel on the pads? * Second, the no-sticks indicate contamination. If it's organic, plasma may work, but it will only remove ~200-300 angstroms, not a whole lot being removed.

reply »

Boundary Scan

 Reflow System