Hi, basically Konrad provided scheme which I had in my mind and well explained :)
Just one more thing about using 3.3V instead of 5V - you might have problems with limited crystal frequency. Then you might need to reduce internal clock settings and also reduce programming bitrate. Roger, I agree that they should "survive" more, but as you correctly mentioned, that current will sum, so it is better to keep it low. Martin S. On Wed, 2 Jun 2021 at 00:04, Rogier Wolff <[email protected]> wrote: > On Tue, Jun 01, 2021 at 09:56:13PM +0200, Martin Stejskal wrote: > > Thing is that every GPIO suppose to have some > > protection diode (in generic - "every" IC - there are exceptions of > > course). They typically stand like ~1 mA without problems. > > That diode is quite beefy! In general ("Typical") it can take about > 10mA! > > That doesn't mean that it is safe to do so. Suppose you have two pins > doing that 10mA, then that 20mA will end up on the 3.3V rails. Now if > there is more than 20mA of power draw on the 3.3V, but if the 3.3V > stuff goes in power-save or something like that, you can easily > over-volt your 3.3V stuff. Really easy to mess up. So a safe margin, > say a factor of ten is reasonable. > > Somehow, on the RPI, I suspect that the IO cells of the broadcom CPU > can be a bit less "beefy" than "typical". Another reason to keep a > safe margin. > > Roger. > > -- > ** [email protected] ** https://www.BitWizard.nl/ ** +31-15-2049110 > ** > ** Delftechpark 11 2628 XJ Delft, The Netherlands. KVK: 27239233 ** > f equals m times a. When your f is steady, and your m is going down > your a is going up. -- Chris Hadfield about flying up the space shuttle. >
