Wrong reading from 9213

[VladFialkov]

Hi all

I have a NI-9213 connected with a cDAQ-9171 and a J type thermocouple.

I am testing the unit and measuring the room temperature. In LabVIEW it shows about 23 degrees celsius, and in other room temp measurment devices it shows about 26 degrees (which I tend to think is the more accurate room temp).

Does anyone knows why?

Does it have to do with calibration?

Is it not meant to measure room temp?

Any kind of help would be greatly appreciated.

Thanks in advance,

Vlad.

[TwinTwo]

Do you know if it has autozero channel on?

If you check the NI USB-9213 user guide and specifications you can see on page 17 the section of Temperature Measurement Accuracy. There could be several reasons for the lack of accuracy in your room temperature measurement. Look at the image attached as reference.

[VladFialkov]

Hi TwinTwo, thanks for the reply.

Excuse my ignorance but what is an autozero channel?

And if I am reading the figure you have attached it should have about 1 or less degree error?

By the way, I don't need super accurate measures (I can afford to have 2-3 degrees off), but I want to make sure that it's just that and it won't fake more than that at higher temps (around 50 or so, which is about the same as room temp error wise by the attached figure).

The system will alert overheating.

Thanks

[TwinTwo]

Hello VladFialkov,

Your device has 18 channels in which 16 are thermocouple channels and then you have 1 internal autozero channel and 1 internal cold-junction compensation channel. Your NI USB-9213 has an autozero channel that is useful to compensate for the offset error. Normally it assumes this channel is on but you can set it off in software. So basically you use autozero channel if you care about accuracy. Your NI USB-9213 also supports high resolution and high speed timing modes, high resolution optimizes accuracy. I would like to encourage you to look at the manual so you can check all this information. 

http://www.ni.com/pdf/manuals/372499b.pdf 

You are reading the figure well, but that is the typical error so it does not mean that is exactly your case. For the best accuracy you should enable autozero channel and also minimize the thermal gradients. Look at the attachment for minimizing thermal gradients, following those recommendations could lead to some difference in your measurements. It won't be a bad idea to try them out. 

Have a great day!