Hacker News new | ask | show | jobs
by stephengillie 3984 days ago
What Coin has managed to do in such a thin package is really impressive. It’s a shame they weren’t able to do it more reliably.

It's a really cool idea. I actually wonder what the technological hurdle was that held them back - mechanical (reliably magnetizing the coils, easy-to-break components due to narrow size), software (programming issues), or QA (reliable SoCs, reliable builds, reliable solders)?

3 comments

I suspect flexing the PCB caused the usual problems of hairline cracks in solder joints, resulting in intermittent faults.
If that's true, I wonder if something like a paper circuit would work? e.g., http://www.instructables.com/id/Paperduino-20-with-Circuit-S...
Flexible substrates are not a problem - the industrial solution is polyimide (Kapton) or just really thin FR4. The problem is inflexible components. As soon as you bend a curved surface attached to a small flat rectangle (IC), it pulls at the joins.

You might get better results with bare die+wirebonding, then encapsulating with slightly flexible epoxy. That would allow the bondwires to flex and take up the bending. Doesn't help with the discrete components or the battery.

It would have been nice if they shared any of that info. Maybe more of us would be more sympathetic.
Very often, the merchant would just assume I was using an Amex e.g. Amex Black....even though the display read VISA. This, of course, caused a decline...not worth the trouble to argue.