Thermocouple not reading correctly using SCC-TC01

[RJ75]

We have set-up several thermocouple inputs that are conditioned by SCC-TC01.  These are professional grade thermocouples of type J and K.  It appears that the the readings from the thermocouples is incorrect (-1.15 mV at room temperature).  This means the temperature it is reading is negative.  We tried switching the positive and the negative, but the reading is still negative.  The gain through SCC-TC01 appears to be 100 so we tried to correct the measurement that way, but it does not seem to work (we tried multiplying and dividing by different powers of 10).  We thought that maybe the thermocouples were faulty, but using new did not correct the problem.  We have been successful in using the same data acquisition board in measuring other signals such as for load cells, so this problem with the thermocouples is puzzling.  When we increase the temperature around the thermocouple it also does not appear to respond as significantly as it should.

Thanks,

Rob Jackson

[Neal M]

Hello Rob,

For starters, I have a few questions about your setup.  What SC Connector block do you have? What DAQ card do you have the SC hooked up to?  What are you using to measure the voltage (e.g. a Test Panel in MAX, a Task created in MAX,  or possibly a VI)? 

Using a test panel directly from the SCC module will give you exactly what the module is seeing.  You can get this by right clicking the device:



If this measurement is correct, there my be a setting in the task that is causing the unexpected reading. If this is the case, I would reccomend making sure the settings in your task are correct.  Most of these rely on the thermocouple, but Table 4 in the SCC Quickstart Guide has some reccomendations.

If the Test Panel does not see the correct voltage, again make sure that the thermocouple is connected correctly.  Also try the module in a different slot to rule out the chance the slot is the issue. Also, make sure that your thermocouple is recieving excitation current if it needs it.  I hope this helps, post back if it doesnt.

Message Edited by Neal M on 10-15-2007 01:19 PM

Message Edited by Support on 12-09-2008 10:53 AM

[RJ75]

Thank you for the suggestions.  I've checked the test panel and the reading seems to be the same.  We are using Sc-2345 and PCI-6221.  I don't think that the thermocouples need an excitation.  We have tried Type J and Type K thermocouples and we have a problem with both.  I also tried switching the module to another port, but that did not seem to help.  I noticed that the port that the TC had been on was reading -1.124 V when the TC was removed.  Is there be an offset from the system that I need to account for.  I thought that there might be before, but it did not seem to solve the problem.  Thanks.

[Neal M]

Hello again Rob,

The floating signal (the -1.124V) that you see on the port when you remove the TC is expected and is caused by lack of grounding for the signal.  This shouldn't cause an offset when the TC is attached.  If there is an offset, I would expect the thermocouple to react to changes in temperature like expected, just with incorrect initial values.

The next thing to try is to set up a task so that we are sure that the CJC value is "built in".  The easiest way to do this is to right click on DAQmx Tasks under the Data Neighborhood in Measurement and Automation Explorer and then select Create New DAQmx Task»Acquire Signal»Analog Input»Temperature»Thermocouple then select the correct SCC and module and name your task. Once the task is made, make sure the CJC is set correctly (Built In) and the type is correct.  Then save and run it.  This should take any offset or issue with the CJC being constant out of the equation.



You said that you tried the module in a different port and this didn't work, do you have another module you could try?  Also, has this module been used sucessfully in the past? When you said the thermocouple did not respond like you expected, what was the response?  If you could post a screen shot of the response that may help as well. Hopefully this will bring us closer to a solution.

Message Edited by Neal M on 10-16-2007 05:00 PM

Message Edited by Neal M on 10-16-2007 05:00 PM

[RJ75]

Thanks.  I thought we had figured out the problem, but that is not the case.  We tried using the Measurement and Automation like you said and it is giving us this error:

Error -200077 occurred at DAQ Assistant

Possible Reason(s):

Requested value is not a supported value for this property.

Property: AI.Thrmcpl.CJCSrc
You Have Requested: Built-In
You Can Select: Constant Value, Channel

Channel Name: Temperature

I thought we had tried setting up the channels before and don't remember having this problem, but maybe we just left it as Constant Value.

Is there a way to fix this?

[Neal M]

Hi again RJ75,

It says in the SCC Quickstart Guide for the SCC-TC Series that they have a built in CJC:



 The fact that you are getting this error makes me think that the device may be configured incorrectly.  I would recommend that you go through the quickstart guide and make sure everything is set the way you are expecting. For troubleshooting, you can set the CJC can be a constant, but if the constant is not correct, you may see incorrect readings. The LabVIEW Help has some good documentation about taking thermocouple measurements (CJC and other considerations).  You can also use a voltage task in Measurement and Automation Explorere (MAX) to read the voltage on the thermocoupe directly, but if there are CJC considerations, then this will not be taken into account.  It might be good though to see if the thermocouple responds the way you would expect with the voltage task.

The TC Series User Manual also has some troubleshooting tips, specifically how to remove offset (pg 9).

Lastly, you said that you thought you had solved this problem, what made you think that?  Were you able to get it to work at all?

Message Edited by Neal M on 11-11-2007 12:44 PM

Message Edited by Support on 12-09-2008 10:54 AM

[RJ75]

Thanks, I will look some of these things over, although I have been trying to work from the manual when trying to fix the problem.

When I thought it was fixed, I had set up the labview to read the voltage of the TC and then I had multiplied that by 10 (since it says the module provides a gain of 100).  That seemed to allow the TC to react similar to how I thought it should to changes in temperature.  However, I also had to provide an offset which changed each time I started the VI again (it could be due to changes in ambient temp).  I also noticed that still the voltage always remains the same sign even if I change the polarity of the wires (this is puzzling).

I am using the 'new' labview routines.  I have set up TC before with no problem, but that was using the old 'classic' labview.

Thanks,

ROb

[mallorim]

Hi Rob,

You mentioned that the offset voltage you had to provide in your program changed each time you started the VI, and indicated that this was possibly due to changes in ambient temperature. This would be compensated by the CJC sensor, which is built into your SCC-TC01. Since MAX does not allow you to select the built in CJC Sensor, the SCC may be configured incorrectly, possibly due to the SCC-TC01 being assigned to the incorrect port or some other mistake. You did not mention in your previous post the results of going through the quick start guide and ensuring the configuration is correct.  This would be a helpful trouble shooting step. Since the device is for signal conditioning, if the configuration is off, the resulting data will be off.

You had also mentioned that you are using the “new” LabVIEW routines. If you are referring to programming with DAQmx instead of Traditional DAQ, this will not be a concern. In MAX, where the test panels are being run, it does not matter if you were using DAQmx or DAQ, since the test panel is looking at the actual hardware.

I hope that this helps. Please let me know if you have any further questions I can clarify or if this solution does not take care of your missing VI issue. Thanks-

Mallori M

[RJ75]

Thanks for your help so far.  We are still having problems.  We tried using the CJC thermistors built into the modules and they seem to be working properly.  They are reading close to room temperature.  We then tried to add this to raw voltage from the TC but that does give the correct result (as outlined in the manual).

The CJC's are still not showing up when we try to set-up a NI-DAQmx task.  I should also say that the VI we use is not using NI-DAQmx tasks, it is using the voltage directly from the module through the channel ai0.

The raw voltage from the module is about -1.33 V.  When we short it out, it goes to about -1.3 V.   We tried to use the potentiometer built into the module but it did not seem to be able to adjust it enough.

Any more ideas?

[Neal M]

Hello again RJ75,

I have a couple questions and a couple more things you can try:

Questions:
1) How do you know the internal CJC is working? Are you using a property node to read the values?
2) When you say you are not using DAQmx Tasks, do you mean you are not making them in MAX, or you are using your own driver? Maybe if you post screen shots it would help me understand what your program is doing (and how it is doing it).

Things to try:
1) Attach a known voltage source to the input to simulate a known temperature.  This could help you see how the signal is conditioned.
2) Use a Constant CJC value (room temp) in a task made in MAX to see if the task is reading the temperature correctly.  This would let us see if the issue is with the task/programming setup.
3) Try the load cell module in the slot (and change the configuration in MAX) to make sure that the SCC chassis is still working correctly.
4) If possible, try another TC01 module to eliminate the chance it is an issue with the actual module.

Hopefully at the least we can get more information about what is going on.