03-14-2013 11:58 AM
So i searched far and wide to find an answer to this, and figured somebody on here would have the answer. Id like to know if setting a smaller signal input range(closer to what temps we are testing at) such as 0-200f then just a covering the complete thermcouple range say -100f to 600f is benefiicial. In other words does this narrow range allow the unit to work more accurately because it only has to "search" through a smaller temp range, compared to a larger range.
03-14-2013 01:13 PM - edited 03-14-2013 01:27 PM
I guess it depends on what you are using to measure the thermocouple with. A "commercial" unit designed to read thermocouples should have no problem reading the entire range it is designed for.
If you are "rolling your own" A to D then sure you would get a higher accruacy limiting your range.
For instance an 8 bit A to D has 255 possible "steps".
If you are measuring a 100 degree range that would be .3921 degree/bit resoultion
If you are measuring a 300 degree range that would be 1.176 degree/bit resoultion.
03-14-2013 01:14 PM
What hardware are you using to measure the temperature and how are you measuring in your program?
I always choose a range to measure instead of the full range of the TC, it just makes me feel better.
The multifunction DAQ cards will switch gains depending on the min/max voltage that you are reading. I am guessing that if you set up a TC channel in MAX it would automatically switch to the highest gains because it would know that TC readings are low voltage.
Dedicated TC modules like the ones that the cDAQ uses do not have multiple gain settings. Basically because it is set up to only read TC's and not higher voltages.
So then it would come down to if you have 16-bit, 18-bit or 24-bit resolution on the measurement hardware for actual accuracy.
I am sure someone will correct me if I am wrong.
03-14-2013 03:11 PM
Thanks for the replies, guess i left some details out.The reason i ask is that the prevoius person had a themocouple setup(NIMAX) range of -100 to 750 f and i figured it would be less work to calibrate, and potentially increase the resolution in the "real" range i am testing within. Im using an sxci 1000 chassis with a sxci1600 connected to my pc and the readings are performed using an 1102's connected to TC 2095 with thermocouples. So acording to RTSLVU it does sound like i will have better resolution if im limiting the signal input range on the thermocouple setup(NI MAX) closer to the actuall range i am testing which is in the 0-200f range.