Hi Tecnetters, 
I hope you all had a good X-mass.
 
One of my colleagues asked me to present a question to this forum. 
We are manufacturing PCBs with electroless nickel - immersion gold ( ENIG ) finishing, using Atotech chemistry . Most of the boards are very dense SMD ,  fine pitch , BGA etc . All boards are passing through ECT and visual inspection before ENIG .
According to him, we are facing , from time to time , a phenomena of plated ENIG on the Solder Mask and /or on text, resulting in shorts ( which  are extremely small) .
These shorts cannot be found by visual inspection, only by ECT.

The changes we made so far in our production in order to minimize the problem are:
1 ) Higher exposure energies and longer curing time for Solder Mask
2 ) Lowering activation ( palladium ) before electroless nickel
3 ) Improving rinses (including agitation) , after palladium
 
 
But still we are getting these shorts from time to time

We think that considering the ENIG process , we are close to skip most plating problems .
 

If any of you, wise ones, has a practical advice as to eliminating our problem, we will be grateful.
 
 
Best regards,
 
Yehuda Weisz