Not great, but something. If you count just getting them to start, it's 19, but really only 3 were properly driven.
The NiCd's were charged in about a day, so the current drain on the CR2477 was fairly high (for a coin cell), and not much energy was extracted to the NiCd's.
I have some others charged over longer periods, but I'm saving those for my #Coin Cell Jump Starter .
After this test, I decided to see how many screws the same cells could drive if charged fully in a normal charger.
Answer: at least 27! That's all the holes I had. So, the coin cell charged the NiCd's to no more than 11%. I suspect it was less than that, though, since the screwdriver still had enough juice to remove all those screws after I stopped the video!
If I end up with some extra cells after the jump starter experiments, I might try this again.
I always thought this screwdriver was lousy, but I never tried it with anything but alkaline cells. With NiCd's, it's actually not that bad...