I am confused by instructions for calibrating strain gages.

[jimsteinmeyer]

Hi all,

In my quest to learn all things  some things LabView I am attempting to calabrate my strain gages prior to running a test.  Using the Measurement and Automation program I open the NI-DAQmx tasks and run the Channel Calibration Wizzard. I have 5 SCC-SG02s in my signial conditioning block with a total of 10 gages, Does this mean 10 channels? In the Wizzard step one asks for a reference value of the measurement. Where do I get this reference value? Then I am instructed to setup the signial to take a measurement at this value, again I ask, how? If I enter numbers in the reference value slots to procede, I next come to a message that my values did not encoumpass the channel's range and endpoints, how do I determine what they are? And do I need to encoumpass them? Is there a tutorial that covers this with examples and better descriptions or do I need to keep asking questions of the forums?

Thank you for your help.

[Sarah_Coleman_NIUK]

Hi Jim,

Hopefully this will clarify things for you but if not just post back.

I have 5 SCC-SG02s in my signial conditioning block with a total of 10 gages, Does this mean 10 channels?
Yes, you need to calibrate the measurement of each channel. In this case you have 5 modules with 2 channels each.

In the Wizzard step one asks for a reference value of the measurement. Where do I get this reference value?
The reference value for the measurement should be at least the low end and the high end of the signal you're measuring. So lets say you expect to read a value of 50mV you should calibrate for 0mV and 50mV at a minimum. You can enter as many calibration references as you want and it's recommended you do at least 2.

Then I am instructed to setup the signial to take a measurement at this value, again I ask, how?
I believe it's already set up to take the measurement. You can pause and start it with the button on the right of the wizard.

If I enter numbers in the reference value slots to procede, I next come to a message that my values did not encoumpass the channel's range and endpoints, how do I determine what they are?
The range of the SG02 is +/-100mV.

And do I need to encoumpass them?
It suggests you calibrate if for values that encompass the full range of the device but it will let you procede with the calibration if you don't want to do that.

Is there a tutorial that covers this with examples and better descriptions or do I need to keep asking questions of the forums?
The following two tutorials provide more information on the Calibration Wizard and the theory behind calibration:
System Calibration
NI-DAQmx Professional Tools

Hope this helps.

Regards

Message Edited by SarahB on 07-06-2006 09:14 AM

[jimsteinmeyer]

Thank you Sarah,

So if I enter a 0 for my first reference number the wizzard shows an Uncalibrated number of .0019.  Next I enter .025 for 25mV and I still have .0019 for an uncalibrated number. In fact no mater what number I use for referance the uncalibrated number is the same, can this be correct? When I attempt to save this calibration I recieve Error-200939 which suggests "the channel calibration table scale pre-scaled values is not supported because one of the values appears more than once in the array".

Further, in the settings screen of the strain setup, do I also enter 100mV and-100mV in the signal input range locations since that is the range of the SG02? I kept getting errors there until I lowered those values to 84/-72.2 which the error message informed me were the most allowed. The tutorials you indicated were a good start, but they were far to general and left me with the idea the authors intended the users to open a call for the full information.

[jimsteinmeyer]

Going further, I have talked with an applications engineer who suggested that I create a known strain on a gage (preferrably on the ones I will be using for my testing, which is not pratical) and use that for the calibration. Is this correct? As I understand from him, and he seemed a little uncertain, I would enter the known strain value in the referance section and let the calibration wizzard determine the "uncalibrated" value. The reason I am questioning this is that I have gages in locations where I can not hand calculate what the strain from a known weight would be. My understanding is that the calibration is to compensate for any induced strain in applying the gage and any line resistance that could vary with different leads. The applications engineer thought I should have recieved a chart with expected voltages with knowen strains, but I have not seen any of them with the gages we have.

As I understand Sarah, I would enter the first value at zero, and then find one near the maximum of what I expect to find in the test. Do I then enter the calculated strain or what do I do?

Any suggestions?

[Jarrod_B.]

Hello Jim,

The type of calibration you are trying to do is a multi-point device calibration.  Let me first point out that this is not your typical strain gauge calibration.  Normally when we speak of calibrating strain gauges, we are talking about doing null and shunt calibrations (offset and gain adjustments).  The process you are going through is used to adjust the calibration constants stored in the EEPROM of your DAQ device.  This is done to correct for any errors in your measurement device. 

If this is what you really want, what you have described is the correct procedure.  You would apply a known strain value to the channel, enter this value in the reference field, and then take the measurement to obtain the uncalibrated value.  The minumum number of points you would need to do is two, the endpoints of your full range.  The more points you take, the more accurate the device calibration. 

What I am assuming you really want to do is null and shunt calibrations.  For more information about how to do this with the SCC-SG modules, please refer to the SCC-SG Series Strain-Gauge Modules User Guide.  Specifically, you should read the section SCC-SG0X and SCC-SG24 >> Calibration.  This will describe how to perform null and shunt calibrations with the SG02.

I hope this information was helpful.  Let us know if you have any further questions.

Best regards,