digital io - simple buttons - how to safely use?

General Discussion
1

hello community,

In the docs there is this warning:

NO PINS ARE TO BE DRIVEN UNTIL AFTER THE SYS_RESET LINE GOES HIGH. DO NOT APPLY VOLTAGE TO ANY I/O PIN WHEN POWER IS NOT SUPPLIED TO THE BOARD. IT WILL DAMAGE THE PROCESSOR AND VOID THE WARRANTY.

So how to wire up a simple switch ?

i did not find the SYS_RESET Line mentioned anywhere else in the header-pinouts.

the only ones that are similar in naming are VOUT_VSYS and nRESET - there is no explanaiton on this site what exactly the pins are doing..
VOUT_VSYS is not searchable as in the table it is then written as VOUT - VSYS that is very confusing.
a text search for RESET does only bring up the warning-boxes and the nRESET pin.

i then have looked up the TechLab schematics

and there the buttons are just wired to the 3.3V rail

image
image463×415 18.6 KB

so they are driven high via the pull-ups in the moment the 3.3V Pins have power…

in that way i have no idea what the warning wants to tell us - or if the TechLab is not following best practice here..

i am open to all sorts of hints and tipps :slight_smile:

Sunny greetings

Stefan

2

No the TechLab is fine here, as 3v3 is powering the both the TechLab and the PB2..

The issue is when the PB2 (or any BeagleBoard.org) device is in an un-powered state (so Vcc = GND), and you supply a voltage signal (or potential) to any of the I/O, it will blow the internal I/O and destroy the part. After 3v3 and sys_reset trigger the PB2’s I/O is in a more protective state.

Regards,

3

If I’m not completely mistaken, that warning is copied wholesale from BBB,
and that did have a SYS_RESET pin on P9.

A guess at why that pin would be to avoid messing with the boot parameters on LCD_*,
but that just speculative on my part.

In general, it is unwise to drive any electronics above the positive rail,
unless explicitly told it’s safe to do so.

4

The copied BBB warning is pretty misleading here, especially with no matching SYS_RESET on the header. The powered-off case is the bit that tends to bite, not the button wiring itself.

5

Like everything we do here, the Docs are open-source too,
so you’re free to post a PR if you have improvements to share. :wink:

As long as you follow the advice of Robert,
you won’t hit the “powered-off” case…