How to run ADC in F0 at interrpupt mode on multichannel?

[raptorhal2]

DMA interrupt when the count decrements to zero ?

I'm reluctant to provide more help without seeing substantial effort by yourself to solve the problem.

Cheers, Hal

[BBart.16]

My currently conf is:

My code:

/* USER CODE BEGIN PV */
int reading[];
/* USER CODE END PV */
...
/* USER CODE BEGIN PFP */
void HAL_ADC_ConvCpltCallback(ADC_HandleTypeDef* hadc){
 
	int i;
	for(i = 0; i<3; i++){
		reading[i] = HAL_ADC_GetValue(&hadc);
	}
 
}
/* USER CODE END PFP */
...
/* USER CODE BEGIN 2 */
	HAL_ADC_Start_IT(&hadc);
 /* USER CODE END 2 */

Debbugger says, thats not entering into a ConvCplt callback function.

And if i stop manually, of course reading 0, 1, 2 are empty.

ADC interrupt is also Set in NVIC section.

In the Description of HAL and low-layer drivers is written, that:

"

ADC conversion by interruption:

Activate the ADC peripheral and start conversions using function

HAL_ADC_Start_IT()

Wait for ADC conversion completion by call of function

HAL_ADC_ConvCpltCallback() (this function must be implemented in user

program)

Retrieve conversion results using function HAL_ADC_GetValue()

Stop conversion and disable the ADC peripheral using function

HAL_ADC_Stop_IT()

"

I made all of them, except stopping(last point), becouse im running on continous conversion.

[Tesla DeLorean]

Pretty sure HAL_ADC_GetValue() doesn't work like that.

WHY can't you just use DMA to fill the reading[] array?

[vivek yadav]

1: There is no meaning of putting HAL_ADC_GetValue in a loop and if you are putting array to save data register value in continuous mode than logic is not correct.

2: First, check your ADC working properly or not, by checking EOC bit of ADC status register.

3: Second, Confirm your ADC Interrupt is working or not. Have you enabled EOCIE bit and NVIC ADC Interrupt properly? you can also write your handler for ADC interrupt.

[BBart.16]

OK, i cant do it, it's too difficult. Lets try the Polling mode. Now my conf looks like this:

And my code:

/* USER CODE BEGIN PV */
uint16_t reading[3];
/* USER CODE END PV */
 
...
 
 /* USER CODE BEGIN WHILE */
	while (1) {
 
		HAL_ADC_Start(&hadc);
 
		HAL_ADC_PollForConversion(&hadc, 200);
		reading[0] = HAL_ADC_GetValue(&hadc);
 
		HAL_ADC_PollForConversion(&hadc, 200);
		reading[1] = HAL_ADC_GetValue(&hadc);
 
		HAL_ADC_PollForConversion(&hadc, 200);
		reading[2] = HAL_ADC_GetValue(&hadc);
 
		HAL_ADC_Stop(&hadc);
 
    /* USER CODE END WHILE */

Now, readings from 3 channels are completely random. I set three pots, first(CH2) have grond on output, second(CH3) have some from 1 to 3V, and the third(CH4) have Vcc potential(3.3V).

If i stop debugger at HAL_ADC_Stop(&hadc), readings looks like completely random, eg:

0: 1720

1: 1720

2: 4030

always first 2 readings are from CH3, and 3rd reading from CH4.

Wheres CH2? And why CH3 are doubled?

Theres any way to SIMPLE read more channels than one?

[raptorhal2]

Theres any way to SIMPLE read more channels than one?

Yes. As clive pointed out, set up DMA ADC conversion to place the 3 conversions in a buffer. Poll the DMA transfer count, and when it decrements to zero, read the buffer values. Or, you can set up a DMA interrupt to trigger when the DMA count is zero.

There is a DMA ADC example in the library package. Usually the example converts one ADC channel, you will have to modify it to convert 3 channels.

Regards, Hal

[BBart.16 · Accepted solution]

Okay, i have working code. I just did what says the example(before i wasn't about examples in documentatnion).

If somebody will also problem with it, heres solution.

Cube conf:

And code:

/* USER CODE BEGIN PV */
int data[5];
/* USER CODE END PV */
...
 /* USER CODE BEGIN 2 */
 
 
  if(HAL_ADCEx_Calibration_Start(&hadc) != HAL_OK){
	  Error_Handler();
  }
 
 
  if(HAL_ADC_Start_DMA(&hadc, data, 5) != HAL_OK){
	  Error_Handler();
  }
 
 
  /* USER CODE END 2 */
...
 /* USER CODE BEGIN WHILE */
  while (1)
  {
 
	  HAL_Delay(1);//for debugger stopping only
    /* USER CODE END WHILE */

Thanks all for help!