Check your shop traveler and see you are seeing this trend on lots 
     that may have an extended hold time between copper and solder plate.
     (ie, over night or over the weekend) A lot of process lines include
     a mild H2SO4 dip/deox step in between there to allow hold or shelf
     times.
     
     Groovy

        MURPHY'S LAWS ON TECHNOLOGY


        #1 You can never tell which way the train went by looking
           at the track.

______________________________ Reply Separator _________________________________
Subject: plated Tin/Lead adhesion
Author:  [log in to unmask] at SMTPLINK-HADCO
Date:    1/8/96 4:34 AM


Concerning I.R. reflowed plated Tin/Lead on single and double sided 
PCB's.
What can cause poor adhesion of the reflowed Tin/Lead to the plated 
copper pad?
Our plating process is : acid copper/h2o rinse /fluorboric 
dip/fluorboric Tin/lead plating/ etch/ I.R. reflow.
The "problem" is only seen on one or two panels in a particular lot, 
and comes and goes, sometimes with months in between occurances! 
Sometimes its detected by a random tape test at pre-mask 
inspection...sometimes, "unfortunatly" it's found after the customer 
has assembled the PCB.  The solder joints appear to flow very well, but 
will seperate from the copper if pressure is applied the component 
lead.  
No copper is attached to the solder joint that has lifted. 
The copper pad below is usually somewhat oxidized.
We have explored many possible causes, including excessive hold time 
between steps, etc.
I would appreciate any thoughts on this, if anyone has some ideas...?