BME680/88 Reading Data Registers

[cweaver]

I'm having some issues reading both the MSB and LSB data registers from the BME688. I'm using two single-byte read I2C transactions to read first the MSB and then LSB registers. When I reset the sensor and read values, there are no issues, but after running the sensor, something is messed up. I currently don't have access to a logic analyzer but wanted to know is there a requirement to read both data registers in the same transaction?

Thanks for any advice!

[BSTRobin · Accepted solution]

Hello cweaver,

You could read one or more sensor data. This is BME68X sensor API code, it demonstrated reading the raw data from the sensor.

/* This internal API is used to read a single data of the sensor */
static int8_t read_field_data(uint8_t index, struct bme68x_data *data, struct bme68x_dev *dev)
{
int8_t rslt = BME68X_OK;
uint8_t buff[BME68X_LEN_FIELD] = { 0 };
uint8_t gas_range_l, gas_range_h;
uint32_t adc_temp;
uint32_t adc_pres;
uint16_t adc_hum;
uint16_t adc_gas_res_low, adc_gas_res_high;
uint8_t tries = 5;

while ((tries) && (rslt == BME68X_OK))
{
rslt = bme68x_get_regs(((uint8_t)(BME68X_REG_FIELD0 + (index * BME68X_LEN_FIELD_OFFSET))),
buff,
(uint16_t)BME68X_LEN_FIELD,
dev);
if (!data)
{
rslt = BME68X_E_NULL_PTR;
break;
}

data->status = buff[0] & BME68X_NEW_DATA_MSK;
data->gas_index = buff[0] & BME68X_GAS_INDEX_MSK;
data->meas_index = buff[1];

/* read the raw data from the sensor */
adc_pres = (uint32_t)(((uint32_t)buff[2] * 4096) | ((uint32_t)buff[3] * 16) | ((uint32_t)buff[4] / 16));
adc_temp = (uint32_t)(((uint32_t)buff[5] * 4096) | ((uint32_t)buff[6] * 16) | ((uint32_t)buff[7] / 16));
adc_hum = (uint16_t)(((uint32_t)buff[8] * 256) | (uint32_t)buff[9]);
adc_gas_res_low = (uint16_t)((uint32_t)buff[13] * 4 | (((uint32_t)buff[14]) / 64));
adc_gas_res_high = (uint16_t)((uint32_t)buff[15] * 4 | (((uint32_t)buff[16]) / 64));
gas_range_l = buff[14] & BME68X_GAS_RANGE_MSK;
gas_range_h = buff[16] & BME68X_GAS_RANGE_MSK;
if (dev->variant_id == BME68X_VARIANT_GAS_HIGH)
{
data->status |= buff[16] & BME68X_GASM_VALID_MSK;
data->status |= buff[16] & BME68X_HEAT_STAB_MSK;
}
else
{
data->status |= buff[14] & BME68X_GASM_VALID_MSK;
data->status |= buff[14] & BME68X_HEAT_STAB_MSK;
}

if ((data->status & BME68X_NEW_DATA_MSK) && (rslt == BME68X_OK))
{
rslt = bme68x_get_regs(BME68X_REG_RES_HEAT0 + data->gas_index, &data->res_heat, 1, dev);
if (rslt == BME68X_OK)
{
rslt = bme68x_get_regs(BME68X_REG_IDAC_HEAT0 + data->gas_index, &data->idac, 1, dev);
}

if (rslt == BME68X_OK)
{
rslt = bme68x_get_regs(BME68X_REG_GAS_WAIT0 + data->gas_index, &data->gas_wait, 1, dev);
}

if (rslt == BME68X_OK)
{
data->temperature = calc_temperature(adc_temp, dev);
data->pressure = calc_pressure(adc_pres, dev);
data->humidity = calc_humidity(adc_hum, dev);
if (dev->variant_id == BME68X_VARIANT_GAS_HIGH)
{
data->gas_resistance = calc_gas_resistance_high(adc_gas_res_high, gas_range_h);
}
else
{
data->gas_resistance = calc_gas_resistance_low(adc_gas_res_low, gas_range_l, dev);
}

break;
}
}

if (rslt == BME68X_OK)
{
dev->delay_us(BME68X_PERIOD_POLL, dev->intf_ptr);
}

tries--;
}

return rslt;
}

https://github.com/BoschSensortec/BME68x-Sensor-API/blob/master/bme68x.c

[cweaver]

Thanks! I ultimately found a bug in my program outputting occasionally incorrect SPI transactions.