On 12/06/18 11:17, Mark J wrote:
> In message 
>            mm0fmf  wrote:
>
> [snip]
>
>>>>>
>>>>>      Uhm... 0v between "motion pin" and 5v pin means the "motion pin"
>>>>> is at
>>>>> 5v also...
>>>>>
>>>>>      BUT! RPi GPIOs are 3.3V inputs...
>>>>>
http://www.mosaic-industries.com/embedded-systems/microcontroller-projects
>>>>>
/raspberry-pi/gpio-pin-electrical-specifications#rpi-gpio-input-voltage-an
>>>>> d-output-current-limitations
>>>>>
>>>>>
>>>>> A 5V signal on that "motion pin" could easily result in burning out
>>>>> the RPi
>>>>> GPIO (if you are lucky, only the one GPIO will be killed)
>>>>> http://www.notenoughtech.com/raspberry-pi/rpi-gpio/
>>>>>
>
> [snip]
>
> The use of even a high impedance voltmeter between 5v and 3.3v GPIO pins
> worries me - maybe this has already been covered, but...
>
> If that meter is registering 0v between 5v and a GPIO pin, then that
> implies /both/ leads are at 5v, so the 3.3v pin has 5v on it, all relative
> to 0v. Can the use of a voltmeter in this way kill an IO? My feeling is
> that a voltmeter should always be used from 0v to 5v or 3.3v, to avoid
> this issue. Maybe I'm wrong...
>
In general IIRC MOS inputs have a parasitic diode to the local rail that
if it gets forward biassed will shunt snall amounts of current back into
the power supply.

Large amounts of current fuck it (and the device, often)


--
Microsoft : the best reason to go to Linux that ever existed.

--- SoupGate-Win32 v1.05