DOUG,
        YES INDEEDY THERE IS A HONKING BIG DIFFERENCE, (ALL THINGS BEING 
     EQUAL), BETWEEN THE COEFFIECENTS OF EXPANSION  OF FR4 AND CERAMIC 
     PACKAGE MATERIAL.  POLYIMIDE MATERIALS MAY ALLEVIATE THIS SLIGHTLY, 
     (DEPENDS ON THE DELTA OF YOUR THERMAL EXCURSIONS AND NUMBER OF 
     REPITIONS).  ANOTHER METHOD IS TO USE A CONSTRAINING CORE MATERIAL 
     SUCH AS COPPER-INVAR-COPPER, (Cu/In/Cu).  AND OF COURSE THE ULTIMATE 
     IS DOING BOTH A HIGH TEMP MATERIAL AND A CONSTRAINING CORE.  
     DISADVANTAGES ARE INCREASED COST AND FEWER QUALIFIED FAB HOUSES.  
     ADVANTAGES ARE THAT YOU DON'T HAVE TO SWEEP UP ALL THOSE BROKEN 
     CERAMIC PIECES.  ANOTHER POSSIBILITY, DEPENDING ON ENVIROMENT, IS TO 
     HAVE COMPLIANT LEADS ATTACHED TO THE COMPONENTS.  AGAIN COST IS AN 
     ISSUE AS IS ELECTRICAL PERFORMANCE.
     
     
     
     REGARDS,
      
     ROB BUTTERWORTH
     ADC VIDEO SYSEMS
     WALLINGFORD, CT
     [log in to unmask]
     


______________________________ Reply Separator _________________________________
Subject: FAB: Heat Expansion
Author:  [log in to unmask] at internet-mail
Date:    6/26/96 9:25 AM


Has anyone had problems with the different rates of expansion among IC 
chips made of ceramic material, copper for traces/pads, and say FR4?
     
The result would be cracks in the traces/pads?
     
Doug
     
[log in to unmask]