Absolute Maximum pulse time of an output pin to ground o +5V without damage?

I think this forum is as good as any (because of power) to ask this question. If not please be free to move it where it's more relevant.

I need to know the Absolute Maximum pulse time of an output pin to ground o +5V without damage.
Couldn't find it on the Atmega328P datasheet anywhere and I've seen some datasheets that specify that (mostly power application ones) but I'm guessing AVRs should have them anyway.

Please don't recommend using transistors, it's not an issue of design or performance, it's for an experiment/stress test.
Actually, for my purpose it could be in the microsecond range (1-100 ?s) and I'll be more than happy, but want to be on the safe side "just in case".

Thanks for your time

Is there even such as a thing?

Yes there is. The IC is made of physical transistors and they have tolerance ranges respecting their properties.

It's just that I'm not that deep into the physics of electronics to know my way around that answer.

This might be of interest

http://forum.arduino.cc/index.php?topic=176968.0

Totally relevant, thanks Erni

Whew, 12 pages, I better get started :smiley:

2bam:
Totally relevant, thanks Erni

Whew, 12 pages, I better get started :smiley:

I just skimmed those 12 pages and concluded that was waaaaaay beyond me. Would make as much sense if I stuck it into Google translate and turned it into Finnish.

Note that driving the input capacitance of a transistor, which charges up, or discharges down, and then looks like a open circuit, is different than shorting a pin to +5 or Gnd, which will take all the current the pin can support and Not look lke an open circuit at all.

You should take steps to not exceed the datasheet ratings:

29.1 Absolute Maximum Ratings*
DC Current per I/O Pin ................................................ 40.0mA

*NOTICE: Stresses beyond those listed under “Absolute Maximum Ratings” may cause permanent damage
to the device. This is a stress rating only and functional operation of the device at these or
other conditions beyond those indicated in the operational sections of this specification is not
implied. Exposure to absolute maximum rating conditions for extended periods may affect
device reliability.

  1. Although each I/O port can source more than the test conditions (20mA at VCC = 5V, 10mA at VCC = 3V) under steady state
    conditions (non-transient), the following must be observed:
    ...
    If IOH exceeds the test condition, VOH may exceed the related specification. Pins are not guaranteed to source current
    greater than the listed test condition.
  2. Although each I/O port can sink more than the test conditions (20mA at VCC = 5V, 10mA at VCC = 3V) under steady state
    conditions (non-transient), the following must be observed:
    ...
    If IOL exceeds the test condition, VOL may exceed the related specification. Pins are not guaranteed to sink current greater
    than the listed test condition.

Thanks for your answer.

User BillO pointed out in a PM that he shorted pins for periods longer than 200us and the chip was still functioning to date, and suggested me on how to build a stress test rig (short pins until they fail)

When I have some time I'll do it and share the results.

I like having working parts. Smoke yours if you want to. I do know that driving NPNs (to sink current for an LED) with no base resistor will break a pin after a while, having done so.

The problem is hard to characterise without subjecting chips to electron microscopy - one of the
failure modes due to high currents is electromigration - nothing to do with thermal overload, just
the current density and duration of the stress. The atoms in the metalization are knocked out
of place by energetic electrons (current densities are already very large in IC metalization due
to its extremely small cross-sectional area).

en.wikipedia.org/wiki/Electromigration

For want of a nail....

build a stress test rig (short pins until they fail)

When I have some time I'll do it

Must have a bigger budget than I do..... 8)

No big budget, just wasting a 2 dollar microcontroller and some time programming the tests :stuck_out_tongue:

My intention is not to do it all the time, of course, just wanted to know if an unintentional short in the microsecond range would impose a real risk on a pin/the whole uC