Questions about current injection on STM32G431CBT3 pins.

[ABido.1]

Hello, i have some questions about current injection on

STM32G431CBT3 pins.

On datasheet Table 15 it is reported:

IINJ(PIN) -> Injected current on FT_***, TT_xx, NRST pins ->-5/0

NOTE3:

Positive injection (when VIN > VDD) is not possible on these I/Os and does not occur for input voltages

lower than the specified maximum value.

On Table 52, However, it is reported:

"Positive injection" sometimes NA and sometimes 0mA... which is the difference between NA and 0mA?

If the Voltage on a TT_a pins is higher than supply (but lower than 3.6V)

the current is injected (due to protection diodes on input) or not?

[Simon.T]

Hello @ABido.1​ 

As explained on the AN4899, the FT or TT pins don't have a clamping diode.

In the TT_a there is some physical effect that allow the current injection. Otherwise if you do this you will disturb the analog part. This is why it is written 0mA instead of NA.

Best regards,

Simon

[ABido.1]

Hello STALA1 and thanks,

In the AN4899 it is reported "0 mA means that current injection can damage the GPIO and induce STM32 malfunction."

"Warning: positive current injection is prohibited for a TT or FT GPIO

defined as 0 mA"

While in your explaination it is reported " if you do this you will disturb the analog part.".

Obviously. in normal applications, i don't have current injection but

this could be possible during startup since there is an analog monitor (please see attached) of the 12V that is the main power source (3V3 for STM32 is obtained from a step down from 12V).

Before 3V3 startup the uc is unpowered (Vdd=Vdda=0V) but the 12V is rising so

signal AIN_12VMON_CH1 >Vdd (0V).

This means that the uC can be damaged (not disturbed since uC is unpowered)?

Please let me know.

[Simon.T]

Hello @ABido.1​ ,

When I said that " if you do this you will disturb the analog part.". I mean in the TT_a IOs there is parasite diode which create a link between the IO and VDDA, if the user inject current he will increase the VDDA voltage and disturb the analog peripheral supplied by this voltage. Otherwise it's a parasite diode so it is not really resistant from the outside world. This is why the current injection is defined as 0mA.

Concerning your application. I have some questions:

• Which converter do you use when you said :"3V3 for STM32 is obtained from a step down from 12V" ?
• On which IO AIN_12VMON_CH1 is connected ? Is it FT or TT ?

[ABido.1]

Hi STALA_1

about your first question

the 3V3 for Vdd and Vdda is obtained from a BUCK/Step-down DCDC converter 12V (in) ->3.3V (out)

The P/N is not relevant.

The analog input is on a TT_a pin.

That's why i'm asking for it,

But, how can you imagine, during startup this problem is always present (please see attached)

It is a matter of accuracy or device damaged?

I'm talging about a Iinj of 0.1mA or lower

[Simon.T]

Thanks for you chronogram, I understand more the purpose.

To answer to your last question. It is matter of both. In fact in the datasheet it is written 0mA. It means ST doesn' t recommend using the product with positive injection. If a customer does that, he will be outside of the product spec and may have permanent damage.

Concerning your issue during start-up. I have two way to handle it :

• The first one is to use FT IO instead of TT_a IO. In the datasheet there are some ADC input on FT IO.
• The second one is to put a Schottky diode between the AIN_12VMON_CH1 and the VDD/VDDA. With this, AIN_12VMON_CH1 will clamp at the VDD level and positive injection won't be possible