Next GDEMU batch will be delayed. Somehow in our attempt to speed up the production we overstressed the TQFP pins and they cracked. This is the first time I see something like this and quite frankly the exact cause is yet unknown. It could be something related to (too fast?) heating and cooling cycle during soldering. Point here is though the cracks aren’t always so bad as to open the circuit right away and so some devices might seem to work now only to fail due to vibrations during shipping. Below some photos of the issue.
Part of the 2nd pin gone missing completly in the ultrasound cleaning step (so there had to be two cracks) plus barely visible cracks in the 1st and 4th pins:
These pins were suspect and sure enough even a little bit of pulling force caused a full separation:
All GDEMUs in this batch that I have ready and tested will undergo another visual inspection, focused on finding these cracks. This is really time consuming though and it’s hard to say how many (if any among the working ones) devices are affected by this. Obviously this is also a lesson learned and hopefully won’t be repeated.
In the meantime I will probably have more Rheas and Phoebes for sale, as usual I’ll let you know when the orders are about to open.
EDIT: GDEMU problems have been mostly resolved but it took a lot of time, as expected. I will open Phoebe orders on Saturday – Rheas and more GDEMUs will hopefully follow soon.
EDIT 2: Phoebe orders
open closed. Hopefully almost 24 hours was enough for everybody to get one – but if not, more will come in future.