STM32F446 VREF+ power up condition

[mabe_rs]

Hello everyone,

I am in the process of designing a PCB with an STM32F446ZET6 and I have a question regarding the voltage on the VREF+ pin. In the datasheet on p. 76, footnote 4 states that the condition "VDDA-VREF+ < 1.2 V" needs to be respected. Table 74 on p. 139 also states the same condition.

On my PCB design, I would like to power the microcontroller (VDD, VDDUSB and VDDA) with a 3,3 V linear regulator. To improve the precision of the ADC measurements, I want to add a voltage reference and connect its output to the VREF+ pin of the microcontroller. The issue I am facing now is that the voltage reference has a startup time of typically 2 ms, whereas the linear regulator starts up in typically 500 µs. This means that during power-up there will be a time frame where the condition will probably not be met, since the linear regulator is already running and the voltage reference is still starting up. Is should be noted that after power-up the VREF+ pin will be supplied with 2.5 V, so the condition will be met during normal operation.

Sadly, the datasheet does not mention what happens when this condition is violated during power-up. Will the MCU or the ADC be damaged? Or do I just need to wait until the reference is fully started up before using the ADC? Table 85 on p. 145 of the datasheet lists that "VREF+ <= VDDA" should be the case when using the DAC. This leads me to believe that the MCU will not be damaged, but could you please provide more information regarding the power-up sequence?

Thank you.

Best Regards

[ONadr.1]

I think that every designer of AD converters takes into account the fact that the reference sources ramp up significantly more slowly and takes this into account in the design. Operational Voltage values in RM are considered as operational values, not transient. If you want to be sure, you can connect a small resistor between the MCU pin and the output of the reference source to limit any current.

[mabe_rs]

Hey ONadr.1,

thanks for your answer. I agree with you, ST probably took that into account and made sure that the ADC/µC will not be damaged in case this condition is not met. Adding a series resistor to be on the safe side sound like a good idea, I will think about that.

However, I would really appreciate if someone from ST could give me a definitive answer. Also, it might be possible that I need to reset the ADC or that the first x measurements might be incorrect if this condition is not met during startup. Instead of figuring this out on my own, it would be better if ST could add a short additional description to the datasheet/reference manual.

Best Regards