0
\$\begingroup\$

I have a micro-controller, for which i have a datasheet specifying the ac/dc characteristics of that mcu. But in this datasheet there are no absolute maximum ratings only the operating conditions are available.

According to datasheet, MCU I/O can have voltage signal in range -0.3V to 3.6V. I have test case in which almost 8V appears at these pins of MCU for 10 milliseconds.

Now i was wondering, if this signal can affect the reliability of device or damage it. Or if this signal persist for longer duration or repeats itself more than one times, can this voltage 8V (higher than rated 3.6V) enough to damage the device??

Thanks in Advance..

\$\endgroup\$
1
  • \$\begingroup\$ Where is the datasheet (link)? What company is the manufacturer? I can't believe a serious manufacturer would release a datasheet without an Absolute Maximum Ratings section or the phrase Stresses beyond those listed under “Absolute Maximum Ratings” may cause permanent damage to the device.. Whether they tell you the limit or not, there is definitely a threshold beyond which they cannot guarantee their product is not permanently damaged. \$\endgroup\$
    – MarkU
    Aug 6, 2018 at 6:28

2 Answers 2

2
\$\begingroup\$

Generally, no.

Inside the chip package are diodes to prevent ESD damage.
A simplification of these diodes:

enter image description here

This limits the maximum and minimum voltage on the pin to Vdd + Vf and Vss - Vf. Where Vf is the forward voltage of the diode. For example, some chips specify a maximum of: Vss - 0.3V and Vdd + 0.3V.

There are exceptions.
SAR or Delta-sigma ADC pins are absolute intolerant of any out of spec condition. When analog mode is enabled, different specifications apply.

Some chips offer FT pins, five volt tolerant pins. These pins can be used as input for a 5V signal. Changing the pin mode to anything else than input with 5V applied, damages the chip.

For example, the STM32F051 states the following for FT pins:

VDDIOx + 4.0V

(3) Valid only if the internal pull-up/pull-down resistors are disabled. If internal pull-up or pull-down resistor is enabled, the maximum limit is 4 V.

If you exceed the limits on other pins, you are injecting current into the chip and the power rails. Which is also allowed, for a tiny amount of +/- 5mA total on the entire chip.

Notice the -5mA, you can also inject negative current into the pin. Eg: pull current by going below Vss-0.3V.

See table 17 and 18. (dm00039193)

The other exception is a switched capacitor ADC, but this is a specialty part. You'd know about the ratings.


In order to pass your test case without damage, you have two options.
1. Reduce the voltage of signal.
2. Limit the current below the injected limit.

\$\endgroup\$
1
  • \$\begingroup\$ Insert a 1Kohm resistor between the high-voltage and the MCU pin. Then measure the MCU pin's voltage during over-voltage. \$\endgroup\$ Aug 6, 2018 at 12:25
3
\$\begingroup\$

If the voltage exceeds the "absolute maximum ratings" specified in the datasheet of your chip, then yes - it will be destroyed (though it still may seem to work... for a while).

\$\endgroup\$
2
  • \$\begingroup\$ I had a chip once, where an overvoltage only destroyed the adc. It was working well, but always showed the maximum value. \$\endgroup\$ Aug 6, 2018 at 6:57
  • 1
    \$\begingroup\$ I put an output of a fire-detector on the GPIO of a XMEGA256 once, did not realize that the output was 9 volts. Worked for a couple of days, then the microcontroller got fried. \$\endgroup\$ Aug 6, 2018 at 10:07

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.