I2C issues with ATH10, HAL_I2C_Master_Transmit giving error but HAL_I2C_Master_Recieve doesn't

[BadEngineer]

I have a NUCLEO-F103RB and I am trying to communicate with my AHT10 sensor. I understand the basics of I2C but i'm a newbie to stm32. 

For some reason when i do HAL_I2C_MASTER_TRANSMIT it seems to return a failure but HAL_I2C_MASTER_RECIEVE doesn't return any error at all. 

The null_a_char function basically just puts '/0' into the array so i can reuse it next time.

My code is below. 

while (1)
  {
	  null_a_char(buf);
	  buf[0]= 172;		//0xAC
	  ret = HAL_I2C_Master_Transmit(&hi2c1, 112, buf, 8, 10000); //Returns HAL_ERROR

	  null_a_char(buf);		// Just puts /0 in the buff 
	  buf[0]= 172;		//0xAC
	  ret = HAL_I2C_Master_Receive(&hi2c1, 112, buf, 2, 1000);	// Returns HAL_OK

	  null_a_char(buf);
	  ret = HAL_I2C_Master_Receive(&hi2c1, 113, buf, 6, 1000);	// Returns HAL_OK


	  // this returns buff[0]=24, buff[1]=220, buff[2]=105, buff[3]=0, buff[4]=0
	  // buff[5]=0, buff[6]=0
}

From my research, it seems the AHT10 sensor is on address 0x38. I have attached from what I think is the datasheet for the AHT10 as well as a couple pictures of my setup as well as my .ioc file. 

I'm a hobbiest and don't have anyone to turn to, please help. 

My AHT10 sensor with the 2 pull up resistors (both connected to positive rail 

The 2 wires connecting back to my nucleo board. 

Image of my .ioc file

I have attached the AHT10 datasheet (i could be wrong but this is what i found). 

[TDK · Accepted solution]

> ret = HAL_I2C_Master_Transmit(&hi2c1, 112, buf, 8, 10000); //Returns HAL_ERROR

Looks like the device is only expecting 3 bytes for this command (0xAC, DATA0, DATA1), but you're sending 8. It's probably NACKing on the unexpected bytes. Why are you sending 8? Or am I misreading the data sheet?

[Karl Yamashita]

It's not very clear, but it looks like the trigger data needs to be 0x33, 0x00.

#define AHT10_SLAVE_ADDRESS (0x38 << 1)

#define TRIGGER_MEASUREMENT 0xAC

uint8_t aht10_data[6] = {0};

// update for Trigger measurement

aht10_data[0] = TRIGGER_MEASUREMENT;

aht10_data[1] = 0x33;

aht10_data[2] = 0x00;

// send trigger measurement byte plus 2 bytes

HAL_I2C_Master_Transmit(&hi2c1, AHT10_SLAVE_ADDRESS, aht10_data, 3, 100);

// Datasheet indicates MCU must wait for the measurement to be completed.

HAL_Delay(100);

// read temperature and humidity data, 6 bytes

HAL_I2C_Master_Receive(&hi2c1, AHT10_SLAVE_ADDRESS, aht10_data, 6, 100);

[BadEngineer]

@Karl Yamashita 

Thank you for the code. 

Yes you are correct i also had to pass it 0x33. 

My follow up question is, in terms of the return data. I am getting the following in my buff[] or for your aht10_data equivalent. 

buff[0] = 24, buff[1]=149, buff[2]=177, buff[3]134, buff[4]=104, buff[5]=181. 

Am i correct to assume that buff[4] and buff[5] and half of buff[3] is tempreture?.
When i plug the numbers in (after converting them into binary) i get the following (420021/2^20)*200-50=30.1C

Am i correct in assuming this?

[Karl Yamashita]

I couldn't find anything that indicates how many bits there are for temperature and humidity, but if is is 20 bits, then your calculation looks correct.