Delete comment from: Ken Shirriff's blog
Hi,
I visited the Stuttgart factory as part of a school trip in 1985 or 1986.
They explained the construction of those ceramic multichip carrier modules. I remember that the process explained to us was quite involved.
The base material was a film made from the ceramic powder with some plastic material to hold it together as a flexible strip. Very much like todays ceramic filaments for FDM 3D-printing.
Many layers of this material were prepared and stacked on top of each other.
Each layer had holes punched in a pattern as vias to the layer below. They then printed the tracks of the circuitry on it in a conductive paint/paste that also filled the holes to create the through-connects.
The stack of these layers would then be fired in a kiln to form the final ceramic part.
I remember them talking about the way the material shrank during the firing process. This was hard to control and had some variability in it. So much that they took precise digital pictures of the final modules and used image-processing to measure the final dimensions. Those were stored in a database on an individual part bases. When it came to populating the modules with chips the data would be recalled and the placing process would take the variance into account.
I remember that I was very impressed.
Jan 25, 2021, 4:08:05 AM
Posted to Examining a technology sample kit: IBM components from 1948 to 1986