VSELECT: Voltage spike at boot (now fixed!)

[Ancyker]

Moving the voltage change up slightly in the code eliminates the 5V spike entirely, as seen here:

As seen in the above image, it takes about 2ms to charge the capacitor/fully turn on. Moving the voltage change earlier on in the code has eliminated the 5V spike as the capacitor is still charging/turning on is still happening and hasn't reached 3.3V yet when the switch occurs. This results in a smooth transition to 3.3V.

Even if the capacitor has been partially discharged, there is still no spike. The image below is the result of me toggling the power switch off and back on again as fast as I could:

Also, for anyone curious, here are some more images of how VSELECT behaves:

When going from 3.3V to 5V it takes about 2ms:

When going from 5V to 3.3V it takes about 15ms:

Your results may vary depending on the bulk capacitor you use, but as long as it's the value specified by the build list it should be fine. So, as of firmware V13.2, it is no longer needed to hot-plug 3.3V carts. You can power off the OSCR when connecting the cart if you wish to as that is now the safer method.