High AD noise on STM32F103 (bluepill)

[pilot]

This is on STM32F103C8 (bluepill)
The A0 is connected to a voltage divider between bluepill's 3v3 and GND,
And the output is very noisy. adding a delay did't do not much good

```
void setup()
{
Serial.begin(9600);

// Enable ADC1 and configure it
RCC->APB2ENR |= RCC_APB2ENR_ADC1EN; // Enable ADC1 clock
ADC1->CR2 |= ADC_CR2_ADON; // Turn on ADC1

// Configure ADC1 to use the internal reference voltage (1.2V)
ADC1->CR2 |= ADC_CR2_TSVREFE; // Enable internal reference voltage
ADC1->SMPR2 |= ADC_SMPR2_SMP0; // Set sample time for channel 0 (PA0)

// Calibrate ADC1
ADC1->CR2 |= ADC_CR2_CAL; // Start calibration
while (ADC1->CR2 & ADC_CR2_CAL); // Wait for calibration to complete
pinMode(PC13, OUTPUT);
pinMode(A0, INPUT_ANALOG);

// configure I2C 2 for Switching controller communication
Wire.setSDA(PB11);
Wire.setSCL(PB10);
Wire.begin();
delay(1000);
}

void loop()
{
static uint32_t last_time = 0;

if (millis() - last_time > 500)
{
// Start ADC conversion
ADC1->SQR3 = 0; // Select channel 0 (PA0)
delay(3); // Wait for ADC to stabilize
ADC1->CR2 |= ADC_CR2_ADON; // Start conversion
while (!(ADC1->SR & ADC_SR_EOC)); // Wait for conversion to complete
int T1 = ADC1->DR; // Read the ADC value (12-bit)
Serial.println(T1);
last_time = millis();
}

delay(10);
}

```

output:
1964
1963
1964
1960
1985
1968
1976
1963
1985
1976
1972

What can be done to make it better?

[gbm]

Increase sampling time, add external 100n capacitor, maybe provide cleaner Vdda.

[pilot]

dumb me, it was the capacitor - thank you.  This was a quick thrown together test.

What I really am trying to understand is the devices 1.2v Vref ,  if the AD range is actually 1.2v or 3.3v , while the internal vref is in use ... and whatever I actually need any code to "make it work" or if it is active all the time.