Can CJC be used for more than one thermocouple measure?

[User_1979]

Hi,

I am planning to use a 6221 PCI card with SCB-68 to make 3 temperature measures (2 on a Copper surface that will be controlled at 40°C and 1 for air at 20°C) using T thermocouples and the Cold Junction Compensation of the SCB-68. From what I have understood, once the switches have been correctly set, AI0/AI8 are used by the CJC. So I must use AI1/AI9 in differential mode for my first Thermocouple.

1°) Will Thc n°2 and Thc n°3 (respectively placed in AI2/AI10 and AI3/AI11) be also compensated by CJC?

2°) Does anybody no what kind of accuracy I can expect with T thermocouple + CJC compensation? Ideally I would like to reach a +/- 0.2 °C accuracy.

Thanks a lot,

User

[Henrik_Volkers · Accepted solution]

1°) Will Thc n°2 and Thc n°3 (respectively placed in AI2/AI10 and AI3/AI11) be also compensated by CJC?

Yes

2°) Does anybody no what kind of accuracy I can expect with T thermocouple + CJC compensation? Ideally I would like to reach a +/- 0.2 °C accuracy.

First take a look into the 6221 spec.  than into the SCB-68 CJC spec and finally to the spec of the TC used.  Coarse guess: +-2°C  without individual calibration. 

To meet 0.2°C uncertainties only for the type T TC (without the CJC and EMF measuring device) you would need a stirred liquid bath calibration at the point of interest.

0.2°C is for  RTDs like Pt100 ....

[User_1979]

Dear Henrik,

Thank you for both answers.

Up to now, I have been using Thc with transmitters (that have a drift of 0.7 °C each 10°C) and doing a calibration. Since the transmitters are quite expensive I am looking into the CJC solution. In this case is this what I should be doing ? :

1) make sure the SCB-68 is roughly at 20 °C (which will be the futur temperature room)

2) connect the thermocouples directly to the AIs, use the CJC compensation and make a calibration according to my reference.

I didn't really understand this sentence :

To meet 0.2°C uncertainties only for the type T TC (without the CJC and EMF measuring device) you would need a stirred liquid bath calibration at the point of interest.

Why without CJC and what is "EMF" ?

Thanks a lot,

User 

[Henrik_Volkers]

1) make sure the SCB-68 is roughly at 20 °C (which will be the future temperature room)

more important is an constant temperature evenly distributed @ the CJC connectors (TC-copper wire), the uncertainty of the CJC goes directly into your measurement. The TC integrate (due to the Seebeck-Effect) the differential temperature all the way from its measuring end to the CJC and create an EMF that is measured as a voltage at the CJC. To determine the temperature of interest at the tip junction you need to measure the voltage AND the CJC temperature.  

2) connect the thermocouples directly to the AIs, use the CJC compensation and make a calibration according to my reference.

I didn't really understand this sentence :

To meet 0.2°C uncertainties only for the type T TC (without the CJC and EMF measuring device) you would need a stirred liquid bath calibration at the point of interest.

Why without CJC and what is "EMF" ?

 The uncertainty of the CJC temperature measurement and of the EMF (voltage) measurement is not included in this uncertainty. I did these measurements with a HP3458 (8,5 digit) high precision DMM in the past, using using stirred ice baths and even a triple point cell as CJC... You CAN resolve some mK resolution with a quite low absolute uncertainty, but it involves much more hardware and time than you might want to invest 😉

I really encourage you to do an uncertainty analysis, the first one is the hardest, however you will learn a lot for your future.

A good source for TC knowledge:
Manual on the use of thermocouples in temperature measurement,
ASTM PCN: 28-012093-40,
ISBN 0-8031-1466-4

255 pages dedicated to TC 😉

(Page1): 'Regardless of how many facts are presented herein and regardless of the percentage retained,

                all will be for naught unless one simple important fact is kept firmly in mind.

                The thermocouple reports only what it "feels." This may or may not the temperature of interest'

Message Edited by Henrik Volkers on 04-23-2010 10:35 AM

[User_1979]

Thank you very much for taking time to give a detailed answer and for your advice. I will think about your information and see what solution i'll choose. I didn't realise EFM was FEM in french...

Regards,

User