VDDA/VSSA connection to VDD/VSS

[Pavel Krupets]

RM0038 Reference manual

STM32L100xx, STM32L151xx, STM32L152xx and STM32L162xx advanced ARM®-based 32-bit MCUs

Figure 8. Power supply overview

(LQFTP100)

has the following note:

1. VDDA and VSSA must be connected to VDD and VSS, respectively.

If I am using separate supply for VDDA should I still connect them?

Table 58. ADC pins

has the following:

2.4 V ≤ VDDA ≤ VDD (3.6 V) for full speed

It means VDDA can be less than VDD.

Manual is a bit confusing. Please help! Which one is a correct statement:

1. VDDA doesn't need to be connected to VDD and can be different from it
2. VDDA must be connected to VDD

[MikeDB]

Connect the VSSes together at a single star point near the power supply main capacitor. You can leave the VDDes separate but it's a good idea to have a Schottky diode from VDDA to VDD such that if VDDA rises above VDD it raises VDD as well (or collapses itself).

[Pavel Krupets]

Thank you!

By the need of adding diode do you mean that VDDA and VDD must have approximately the same voltage?

Table 58. ADC pins

has the following:

2.4 V ≤ VDDA ≤ VDD (3.6 V) for full speed

From which I concluded that VDDA can be less than VDD.

To be concrete this is what I am trying to achieve:

I need to run ADC quite fast and to do so I need to have VREF+ = VDDA for 16 MHz ADC. I also need VREF+ to be 2.5 volts (well a bit more than 2.4 which is also a requirement for 16MHz ADC). So I also need VDDA to be 2.5 volts.

For communication lines I do want to have 3.3 volts (to be compliant with other MCUs) so I was thinking to have VDD at 3.3 V and VDDA and VREF+ at 2.5 V.

[MikeDB]

No - VDDA can be any voltage you like lower than VDD. Problems only occur when VDDA goes higher than VDD, for example during turn on and off, and the diode helps make sure this isn't the case.

[MikeDB]

And yes - 2.5V for VREF and VDDA, and 3.3V for VDD sounds a good choice to me. The diode is just a safety measure and costs a few cents

[Pavel Krupets]

Thank you!!!

[Chris1]

Note that the datasheet states, "It is recommended to power VDD and VDDA from the same source. A maximum difference of 300 mV between VDD and VDDA can be tolerated during power-up and operation."

[Pavel Krupets]

Table 58. ADC pins

 has the following:

 2.4 V ≤ VDDA ≤ VDD (3.6 V) for full speed

Why is this in the reference manual then?

If they need to be from the same source why do we need separate pins and parts of manual talking about less noisy sources?

So I assume VDDA and VDD should be the same. Or might need to look for a different MCU.

[MikeDB]

Yes it does seem I'm using one of the few STM32 processors (the H750) that does allow a huge range of different voltages. Sorry for misleading you.

I assume what they think is you will use an RC or LC network feeding the VDDA pin so that you clean up the dirty VDD a bit. But it's certainly not the way to get a really clean supply

[MikeDB]

Ah sorry I hadn't checked his actual processor. The one I use says they are independent except during power up/down when VDD is below 1V.

[jose luu]

VDDA could be derived using a low noise LDO from VDD