need help understanding Vref / Vdda

[deep_rune]

I'm trying to find what voltage reference my ADC is using, as it is not going up to 0xFFF. Along the line, I have become quite confused about the Vref - I used to use other chips where the Vref is just a reference voltage for the ADC, but with STM32 it seems it's more complicated.

I'm confused by the function of the Vref / Vdda pin on the STM32L412 - is this a power pin, or a voltage reference pin? I looked through the reference manual but the information was ambiguous. If it's a power pin, how much current does it need? And how does the Vref work?

[TDK]

VDDA is a power pin. It must be the same as VDD for most (all?) STM32 chips.

VREF+ is a voltage reference pin.

On smaller packages, these two pins are bonded together, which means VREF+ = VDDA. This means it is both a power pin and a voltage reference pin.

[Tesla DeLorean]

There is an internal intVRef (1.25V as I recall) that can attach to the ADC and can be used as a calibration point

[deep_rune]

ok.. so if my chip runs off 3.3V but I want to run the ADC of a 3v reference, I can only do this on a chip with separate VDDA and Vref+ pins?

[TDK]

That is correct.

[waclawek.jan]

My take on the topic.

JW

[DCarr.1]

Hello TDK. I'm currently having some ADC problems on my STM32L496RGT6 and I'm trying just about everything I can think of (If I knew how to link to my question where I describe what's going on I would; the subject is "STM32L496RGT6 ADC")

How sure are that VDDA must be equal to VDD? I can't find anything about that in the data sheet.

[TDK]

VDDA and VDD are independent for the L412 and L496 per the datasheet. Apologies for the misinformation. Gotta be an answer somewhere else.

[DCarr.1]

Thanks a lot; I'm relieved to hear that.

At this point, I've got a trivial application which reads the ADC channel and the VREF channel. The ADC channel is several tens of millivolts down (depending on the applied voltage) and the VREF channel is coming out about several millivolts down from the expected value with a couple of millivolts of noise.

Can you imagine a way to explain this as a software issue or should I start looking more closely at the board?

[Tesla DeLorean]

I'm *not* an analogue guy, but I don't think any of this is marketed as high accuracy/precision stuff, and is apt to have all manner of offsets and biases. And is built on a CMOS MCU process. The VDDA runs the PLL/VCO doesn't it?

I'd consider it suitable for measuring analogue joysticks and generating 50 Hz servo PWM signals. Or getting a reasonable RSSI measurement from an RF downconverter.

For precision/critical stuff I'd perhaps be looking at different means, and getting the analogue leaning guys to design it.