Using ADC1 VBAT pin on STM32U535 to measure the battery supply voltage of my system

[L4sj4s]

Hi,

My battery (++ LTO/Bucks) that runs everything on my PCBA is connected to VBAT (pin 1). Before the connection I have a voltage divider factor of roughly 1.374, but what I read on the pin via ADC1 channel 18 (what VBAT is routed to internally) seems to be non-sensical. With a battery voltage of 3.6V I measure roughly ~1160 mV (after multiplying by 4 and 1.374). When I tried replacing the resistors in my voltage divider by smaller ones I got higher measurement values.

VBATEN is enabled in ADC12_COMMON_CCR. Also tried to enable MONEN in PWR_BDCR1, but that didn't make any difference.

What's could be the issue here?

[TDK]

Sounds like the cap isn't being charged up fully. Set sampling time to max.

If that's not it please show the code used.

What voltage divider are you talking about here? There's an internal one but you certainly can't change the resistors on it.

[L4sj4s]

I have sampling time set to max, and I can correctly measure temperature on channel 19 (so I know the code is correct). I even tried to route my battery voltage to a different ADC channel (PA6) and then I could read it correctly. 

To clear up the confusion about the voltage divider: I have an extra voltage divider (not the internal to the MCU that gives VBAT/4) because my battery voltage can be up to 4.2 (which is above what the VBAT pin is rated for).

Removing the voltage divider entirely I get the correct VBAT measurement, so I have concluded that it is the culprit, and this is therefore a hardware design error where the two voltage dividers is not getting along.