4 ms·
I think it's one blob of standard-cell logic on the left with power routing through the middle. I don't think there is any visible partitioning into functionali
by kens 4y ago
I think it's one blob of standard-cell logic on the left with power routing through the middle. I don't think there is any visible partitioning into functionality in this logic. The large amount of analog circuitry in the right half is a bit of a puzzle. It looks like some large capacitors, maybe a charge pump for something?
- addaon 4y agoThis is outside my wheelhouse, but I assumed that the large capacitor area was a (trimmable) capacitor bank that could be put in parallel with the load capacitor on the (external) quartz oscillator, allowing calibration to counter some unit-to-unit variant in crystal frequency.
- kens 4y agoThat's a possibility, but I don't see any laser-trimming marks on the big capacitor. Another option would be selecting capacitors with fuses, but I don't see any fuses or capacitors with powers-of-two sizing. With the complexity of the circuitry in the right hand side, I think there are a bunch of analog things going on.
- makomk 4y agoThe calibration for quartz crystal variance in these watches is done much, much later in the manufacturing process using solder bridges on the PCB - people have reverse engineered it. I think there are watch oscillator chips out there with one-time programmable calibration (particularly for analog watches) though.