Hi teritama, [quote userid="386772" url="~/support/data-converters-group/data-converters/f/data-converters-forum/1454223/ads1118-about-k-type-thermocouple-measurement-accuracy/5582825#5582825"]And about cold junction compensation, the customer don't performed cold junction compensation. [/quote] Unless the cold junction is at 0℃, cold junction compensation must be performed for accurate temperature measurements when using a thermocouple. This is likely the reason the error rate is higher at temperatures close to 0℃. [quote userid="386772" url="~/support/data-converters-group/data-converters/f/data-converters-forum/1454223/ads1118-about-k-type-thermocouple-measurement-accuracy/5582825#5582825"]I guess to resolve this issue, they need to use external temperature sensor for example RTD or sensor device. Because they want to place K-type Thermocouple little away from ADS1118.[/quote] There are many ways to determine the cold junction temperature aside from the ADS1118 internal temperature sensor. RTD measurements are often used to get a more accurate temperature reading for the cold-junction measurement. There are also thermistors and other semiconductor temperature sensors that can be used to get a cold junction measurement. Regardless of how it is done, the cold-junction measurement must be accurate. Any error in the cold-junction measurement directly adds to the error in the thermocouple measurement. The following app note explains the principle in detail: A Basic Guide to Thermocouple Measurements (Rev. A) To get an accurate temperature measurement of the thermocouple voltage, you would add the thermocouple voltage to the equivalent cold-junction voltage, and then convert the sum back to temperature. Best Regards, Angel
↧