C
Chuck Simmons
I did not want this to polute the current related thread so I start
anew.
There is an amusing little battle where I contrive to work when I can
stop laughing long enough. I am kind of responsible for the "servo chip"
because I was one of the architects and I know the part inside and out.
A couple of years ago we noticed that hot swapping our read channel chip
would blow one of the analog inputs on the servo chip. Thinking nothing
of this, I declared hot swapping read channels to be a "NO-NO." We
carried on happily and quit blowing servo chips by never doing the
NO-NO.
Just yesterday, our read channel design center complained that they were
blowing servo chips. I was asked about that and I said to tell them to
stop doing stupid things. They did not like this and insisted that we
turn the servo chip to fix their problem. I said no that we did not have
any critical fixes on our list and everything would wait for shrink. The
design center said that it was hampering them because servo chips failed
when they hot swapped read channels and that the trouble would cause
servo chip yield issues. They had found the NO-NO.
The servo chip yields above 90% in blind packaging and is better than
99% packaged after wafer test. It passes ESD testing perfectly. It is
really good. We found the solution. We sent the read channel design
center a tray of servo chips with a note not to call us, we would call
them.
In my entire career, it has always been that expectations should be low
on component hot swaps.
Chuck
anew.
There is an amusing little battle where I contrive to work when I can
stop laughing long enough. I am kind of responsible for the "servo chip"
because I was one of the architects and I know the part inside and out.
A couple of years ago we noticed that hot swapping our read channel chip
would blow one of the analog inputs on the servo chip. Thinking nothing
of this, I declared hot swapping read channels to be a "NO-NO." We
carried on happily and quit blowing servo chips by never doing the
NO-NO.
Just yesterday, our read channel design center complained that they were
blowing servo chips. I was asked about that and I said to tell them to
stop doing stupid things. They did not like this and insisted that we
turn the servo chip to fix their problem. I said no that we did not have
any critical fixes on our list and everything would wait for shrink. The
design center said that it was hampering them because servo chips failed
when they hot swapped read channels and that the trouble would cause
servo chip yield issues. They had found the NO-NO.
The servo chip yields above 90% in blind packaging and is better than
99% packaged after wafer test. It passes ESD testing perfectly. It is
really good. We found the solution. We sent the read channel design
center a tray of servo chips with a note not to call us, we would call
them.
In my entire career, it has always been that expectations should be low
on component hot swaps.
Chuck
