Back to the HMCAD1520 FA.
This is sample 6A1 after initial partial decap. It's still covered in the north and easy because I guessed how big the die would be and apparently underestimated its size.
The good news is that it's gold ball bonded, so more time in the acid shouldn't risk damaging the bond wires at all.
Time to re-mount it with a bigger mask and try again.
Back, my instance had some server issues and was down for a bit last night.
After some additional etch time, we have this. The die is fully exposed and we can see 88 bond pads, of which 84 are bonded. The package only has 48 pins so there's a lot of multiple bonds (we can see many bonds from the die surface to the ground pad as well).
Even without higher magnification, we can see obvious 8-way symmetry in the die layout, consistent with the "8 ADCs in a trench coat" architecture of the HMCAD1520.
Straight out of the acid, the die was filthy so I sonicated for a few seconds. When that didn't do the trick I left it in for a whole minute, which cleaned it up a lot more but still left some surface debris. The remaining residue is pretty stubborn.
Seen here with Mitutoyo 5x/0.14 objective and combination of low-angle and coaxial illumination
@azonenberg
Out of context this sentence is beautiful:
"Straight out of the acid, the die was filthy so I sonicated for a few seconds"
This could be from a Mobius comic or a short story from any NewWave SciFi writer
and the pics are beautiful too
Andrew Zonenberg
Amro has been