David, two suggestions that may help with your problem.
 
(1) If you are using board brushing and not pumice scrub then you could have thin slivers of copper breaking away from the edge of the track and causing embedded shorts in the solder resist. This is almost unavoidable with mechanical brushing so I would always recommend pumice scrubbing.
 
(2) Plated shorts can be caused by the laminate weave giving an uneven copper surface. This can lead to tunnelling of the tin-lead or tin plate which results in very fine shorts corresponding to the weave. On half ounce copper this effect is more pronounced because of the reduced time in etch. Oddly enough, better results can often be achieved with one ounce material because the thicker foil is less prone to print through and the increased etch time often means these fine shorts disappear. Look for spikes along the edges of tracks as a sign that this is happening; they may be localised in one area of the panel.
 
Both of these effects are very difficult to see without at least 10X magnification and are almost impossible to find after solder resist application. One method of finding a short of this nature is to apply a current, say 1 amp, across the two shorted nets. You will see a brief flash and the short will carbonise. This makes it much easier to locate and remove.
 
You won't experience (1) with pumice scrubbing and you can reduce the incidence of (2) to zero by getting better surface topography on the laminate, slowing the resist lamination speed, and increasing board entry temperature to the laminator to help the resist to flow and conform to the copper surface.
 
Hope this helps,
 
Paul Gould
Technical Director 
Teknacron Circuits Ltd, UK 
www.teknacron.com
 
-----Original Message-----
    From: Bay State Circuits, Inc. <[log in to unmask]>
    To: [log in to unmask] <[log in to unmask]>
    Date: Thursday, April 01, 1999 2:41 PM
    Subject: PROCESS PROBLEM
    
    
    Chris 
    I am not sure who I should address this to at IPC but I have you on my address list, so your the lucky one.
     
    I am having a problem with an SMOBC board that has .008 spaces and traces running the length of the board (about 18 inches).  It seems that after etch I am getting some small hair like slivers that can not be seen with the naked eye and are only picked up at bare board test.  But at test we might find two shorts, cut them out and then different ones will pop up on retest.  We are being very careful not to over etch but the problem persists.  Do you or anyone else at IPC have any suggestions?
     
    We start on half oz material 
    After etch we scrub to clean the boards, visually inspect and then tacky role during soldermask.
     
    Thank you for your assistance.
     
    David Ducey