STM32F401 ADC VrefInt and Vbat read issue

[LoMo-MoMo]

During simple task I need to measure Vrefint and Vbat, so, I open CubeMX and configure ADC to read Vrefint and Vbat

Than I setup a DMA to read values...

I write a really simple program:

/* USER CODE BEGIN PD */
#define	ADC_BUF_SIZE 20
/* USER CODE END PD */

/* USER CODE BEGIN PV */
uint32_t	adc_buffer	[ADC_BUF_SIZE];
/* USER CODE END PV */


int main(void)
{

 /* USER CODE BEGIN 1 */

 /* USER CODE END 1 */

 /* MCU Configuration--------------------------------------------------------*/

 /* Reset of all peripherals, Initializes the Flash interface and the Systick. */
 HAL_Init();

 /* USER CODE BEGIN Init */

 /* USER CODE END Init */

 /* Configure the system clock */
 SystemClock_Config();

 /* USER CODE BEGIN SysInit */

 /* USER CODE END SysInit */

 /* Initialize all configured peripherals */
 MX_GPIO_Init();
 MX_DMA_Init();
 MX_ADC1_Init();
 MX_USART2_UART_Init();
 /* USER CODE BEGIN 2 */

 /* USER CODE END 2 */

 /* Infinite loop */
 /* USER CODE BEGIN WHILE */
 while (1)
 {
 HAL_ADC_Start_DMA (&hadc1, adc_buffer, ADC_BUF_SIZE);

 HAL_GPIO_WritePin(LD2_GPIO_Port, LD2_Pin, GPIO_PIN_RESET);
 HAL_Delay (200);
		
 HAL_GPIO_WritePin(LD2_GPIO_Port, LD2_Pin, GPIO_PIN_SET);
 HAL_Delay (200);
 /* USER CODE END WHILE */

 /* USER CODE BEGIN 3 */
 }
 /* USER CODE END 3 */
}

But I faced with really strange issue

When I configure Rank1 and Rank2 as Vrefint and run my program I get such results (i.e. ref value is 1492)

(value stored at 0x1FFF 7A2A close to this value, so, reading is correct)

 

When I configure Rank1 and Rank2 as Vbat and run my program I get such results (i.e. Vbat value is 1021)

But when I configure Rank1 as Vbat and Rank2 as Vrefint I get such values (also I can configure Rank1 as Vrefint and Rank2 as Vbat):

i.e. Vbat is 1021 (as previously) and Vrefint is 656 (previously 1492)

I try to change sampling time - doesn't help

I try to change resolution of ADC - doesn't help

I try to disable DMA - doesn't help

I try to use Enable/Disable Continuous Conversion Mode - doesn't help

I try to disable DMA and read ADC with polling mode - doesn't help

I try to use manual channel selection Vrefint like this

 sConfig.Channel = ADC_CHANNEL_VREFINT;
 sConfig.Rank = 1;
 sConfig.SamplingTime = ADC_SAMPLETIME_144CYCLES;
 if (HAL_ADC_ConfigChannel(&hadc1, &sConfig) != HAL_OK)
 {
 Error_Handler();
 }

than run ADC, get result in polling mode, than change ADC channel to Vbat

 sConfig.Channel = ADC_CHANNEL_VBAT;
 sConfig.Rank = 1;
 sConfig.SamplingTime = ADC_SAMPLETIME_144CYCLES;
 if (HAL_ADC_ConfigChannel(&hadc1, &sConfig) != HAL_OK)
 {
 Error_Handler();
 }

and Run ADC again in polling mode - doesn't help

I try to add pause for 10ms, 100ms, 1 seconds between conversions - doesn't help

 

I can remove Vbat from measuring and add more ADC channels to sequence (like IN0, IN1, Vrefint, IN4) - in this situation I get Vrefint 1492, but when I add Vbat to such sequence (i.e. IN0, Vbat, IN1, Vrefint, IN4) I get Vrefint as ~600, or some other values, but never 1492

Firstly I try it on my own PCB (STM32F401RCT) and faced with this problem, but now I run NUCLEO-F401RE board and see the same problem...

I think I'm doing something wrong, but I doesn't understanding what I missing in such simple task...

[LoMo-MoMo]

CubeMX project file

[LoMo-MoMo]

I try just the same project / firmware for NUCLEO-L476RG demo board

only Vbat measuring (1306)

only Vref measuring (1434)

Vrefint (1434) and Vbat (1305) measured together in a raw

So, the same project/software works correctly for F476 and doesn't work for F401

How to ask some official guys from STM about such situation?

[waclawek.jan]

This is consequence of a bizarre limitation in some (most?) 'F4, that you need a software intervention - switching on/off ADC_CCR.TSVREFE and ADC_CCR.VBATE - between measuring VBAT and VREFINT (and internal temperature sensor).

JW

 

[LoMo-MoMo]

Previously I try manual switching, but it doesn't help

But now I know where to look at :)

Thanks!

Despite adjustments to settings like sampling time and resolution, the problem persists. Have you encountered similar issues with ADC channels on the STM32F401, and what solutions have worked for you?

[LoMo-MoMo]

I will try it today :)