Multifunction DAQ

Showing results for 
Search instead for 
Did you mean: 

thermocouple amplifier/input spec for usb-6259

Hi, I am using the M-series 6259 USB board and I am connecting Thermocouples directly to the
analogue inputs. After calibration everything works find, however for documentation purposes
I would like to know if there are any specs for the thermocouple analog inputs,  ? for example
the expected accuracy / tolerance, limitations.
I have looked in the tech spec and manual but cannot find anything directly relating to a thermocouple
If anyone can point me in the right direction I would be grateful
0 Kudos
Message 1 of 4

Hello Sandra,


The accuracy of your Thermocouple reading will depend on the code width provided by your device, in this case an NI USB-6259. The resolution of your Analogue Input channels is 16 bit, which will mean you have 216 (65536) measurable divisions over your voltage range. The widths of the divisions are what will dictate the accuracy of any voltage measurement.


Different Thermocouple types will return different V per oC change, for example:

K Type: 41µV/ oC

J Type: 56µV/ oC

T Type: 52µV/ oC


This means to detect a 1oC temperature change using a K Type thermocouple you will need a code width under 41µV to definitely detect it. To do this on your device you will need to have your voltage range to –1 V to 1 V, which has a code width of 32 μV.


The different Input Ranges and Nominal Resolutions for the NI USB-625x can be found here: M Series User Manual


This resource may also be helpful - ITS-90 Thermocouple Database


Hope this helps!




Message 2 of 4


I also have the USB-6259 card. 

I have some thermocouple wires (type K - Chromel and Alumel)

I make the thermocouple by spark welding the two materials.

I connect it to ai0 of the DAQ ( in the Differential mode (Floating Signal Sources Not Connected to uilding Ground) as in manual on page 4-14)

So one of the thermocouple ends go to terminals 1 and 2. 


In LabView I use the DAQ Assistant to display the collected data. But there are problems: 


1. how exactly should I go through the calibration process?

I have a thermocouple based electronic termometer and compare this to my thermocouple - how many points and over what range of temperatures do I have to go to have good calibration?  Do I need to do it many many times over the whole range I am interested in ?


2. my signal is very distorted. (it seems that noise changes the reading by +- 15 Celsius) 

Do I need to use signal conditioning? What kind? If yes - will it not be easier just to buy NI 9211A or NI 9211


Thanks in advance

0 Kudos
Message 3 of 4

Hi RafalM,


Good afternoon and I hope your well today.


Thanks for your post.


I would like to recommend that in the future you start a new post/thread for your own issue, espcially if the thread is no longer active. This will just ensure that you do receive the best support possible in the shortest amount of time! 


The calibration process from the USB 6259 can take place in a few modes. But this is referring to 5 Volts on the input being read by DAQmx as 5 volts, with the smallest possible error. If this is what you meant then please refer to B/E/M/S Series Calibration Procedure for NI-DAQ mx. The easiest way to calibration a vitural channel is in the DAQmx assistant for the task your setting up - there is a calibration tab. This is known as self-calibration. 


However, I feel you you mean how to calibrate 25c to 25c in LabVIEW/DAQmx. Well, the reading you read back will be in Celsius (C). and then if the channel is calibrated, then this will be more accurate. One type of signal condition is Cold-Junction Compensation that can be any AI channel - which in your DAQmx configuration you can spare which channel is CJC. For more info please refer to Required Signal Conditioning for Thermocouples.


I hope this finds you well, 


Kind Regards
James Hillman
Applications Engineer 2008 to 2009 National Instruments UK & Ireland
Loughborough University UK - 2006 to 2011
Remember Kudos those who help! 😉
Message 4 of 4