Running a 3.3V STM32F0 on 6.5V

[David Koster]

Posted on March 31, 2018 at 19:35

I'm using an STM32F0 in a high volume low power project running on a LIR2032 that can be charged via USB. The STM is specd at 2.4 - 3.6V, with 4.0V as the absolute maximum. When the battery is fully charged however the voltage will be 4.2V, 0.6V out of spec.

Now, if I power the device with 6.5V for a few days it actually still runs fine, except for a drawing scary high currents (81uA in standby compared to 3.12uA on 3.3V). What exactly happens when you apply a too high voltage to a device like this, and what are the actual risks?

Thanks in advance!

David

[waclawek.jan]

Posted on April 01, 2018 at 10:19

This is the same as saying 'I've hit the wall with my car and while the bumper broke the engine runs just fine, so I think it's OK that my brakes won't work and I'd just stop every time by hitting the wall'.

The maximum VSS is a design limit and while there may be some headroom, it's a safety margin with no telling how much it is and what will happen if you'll operate permanently in it. The mcu is a very complex assembly of parts, even if integrated and in 'solid state'; and the limits apply to them all individually as well as to the particular assembly of them. There are many characterizing steps (e.g. measuring MOS oxide breakdown under a wide manufacturing and operating parameters variations) and the design is adjusted to that design limit at many many steps, so there's no single description of the processes beyond the design limit (and there doesn't need to be any). There may be slight impacts which will accumulate over time leading to a catastrophic failure (a pipe or a hinge bent often beyond it's normal range while the hitting the wall may break one day), and there may be modes of operation where the margin is different than what you'd experienced before (in winter the pipe may be more rigid and hitting the wall with the same speed as was OK in the summer will break it). There may even be complex interactions.

So there's only one, simple and single and ultimate guideline for operating beyond maximum allowable limits: don't.

Use an LDO.

JW

[S.Ma]

Posted on April 01, 2018 at 13:43

In semiconductor, transistor use a certain oxyde thickness which determines the breakdown voltage.

So, after any designer smart transistor construction, using a too high voltage will result in boom.

Below this limit, on ANY semiconductor product datasheet, there are 2 tables. The Absolute Maximum Rating, and the Normal Operating Condition. When you stay within the normal operating condition, your chip is guaranteed to wear and tear and break after typically 10+ years. When reaching absolute maximum rating, your chip may still work... for 1000 hours typically. So here it is like using a normal car tire in a A380 plane landing gear.

[T J]

Posted on April 02, 2018 at 02:10

I use this part for that

AP2114H-3.3TRG1  good for 1A, and very low dropout.

it is a pnp type regulator.

otherwise, to save your pennies,

you could use a diode drop of minimum 0.6 to guarantee 3.6V max, or two diodes, since you can work down to 2.7 without issue.

but please be sure to add some caps,  10uF and 1uF and 3x 0.1uf after the diodes should be ok.