How to perform shunt calibration with a load cell and NI 9237

[Ali2001]

Dear All,

I require a bit of help with my load cell application. I am measuring tension/compression force and will be using a full bridge set up . I will be using NI9237 with RJ50 and NI-PXI with built-in FPGA. I'd like to know how I should perform the shunt calibration.

I will appreciate your help in advance,

Ali

[OKors]

Are you using Signal Express only or do you have LabVIEW or any LabVIEW experience? I know there's a couple shunt cal VI's out there but I never had much luck getting them to work with the SCXI platform.

Also, since you didn't mention it, I assume you have everything wired already and this is not a "how do I wire the NI9237 so I can perform a shunt cal" question?

I struggled with this for a long time. You can read this post HERE but basically the short of it is you cannot shunt cal a load cell in Signal Express unless you have a way to convert the microstrain output to force by using a custom scale.

I work with a lot of load cells and my workaround has been to use the mV/V sensitivity or, when provided, the Slope and create a custom Linear Scale each time. Fortunately we have our own Metrology lab that calibrates each one of our load cells after every project; so I've convinced the lab to add the Slope to the cal sheet which I enter directly in the Custom Linear Scale.

If this is an off-the-shelf load cell, like from Omega or something, you should have a datasheet which should have the sensitivity. You can derive the slope using the formula m=y/x

m= slope

y= full-scale output of load cell

x= mV/V sensitivity

If this is going in the right direction let us know and I'd be happy to walk you through in more detail the procedure I use here in our lab.

[Ali2001]

Hi

Thank you for the response.

I think I put my post in the wrong category ( I am not dealin with Signal Express, just when I was looking at the google results on shunt calibrations I led to this category), however, we do have LabVIEW and a little bit of experience with that. Right now we are hooking up our servo-hydraulic system including the loadcells, I think there should not be any issue in wiring as we are following the instructions in the NI 9237 manual. We need to perform the shunt calibration first, browsing the NI website I got this post (http://zone.ni.com/devzone/cda/tut/p/id/3445#toc0) which recommends to do the shunt calibration through MAX. I'd like to know if I am on the correct track or not.

Regards,

Ali

[OKors]

I work primarily with servo-hydraulic controls as well so regardless of if you're using SigEx or not I still may be able to help if you haven't figured it out already.

Besides our vehicle dynamics hydraulic test simulators (MTS Flex Test) I work a lot with the 2-channel MTS 407 controllers.

The 407 is it's own digital closed-loop hydraulic servo controller so it has its own signal conditioning. When I need to record the load and displacement signals from the 407 I connect to the BNC outputs and scale the 10v signal coming from the 407 into corresponding load and stroke values. No need to shunt because as mentioned, the signal conditioning happens in the controller.

Let me know if I can be of any more assistance or you'd like me to elaborate on my process.

[OKors]

BTW, Signal Express uses the MAX environment for channel configuration, so as far as your question goes the channel setup is all the same.

[JimMacD]

OKors,

I am trying to hook up a Load Cell from Omega (  ) to a NI-9237 Input module. The module is in a cRIO chassis, so I have to do my own scaling (no DAQmx). The load cell sensitivity is 2.153 mV/V

However, I'm not getting the expected mV signal when I apply a known load.

The test I am doing is placing the Load Cell between 2 thick bars of metal, then standing/balencing on the top bar so all of the weight is transferred through the load cell. The total weight should be about 200 LBs, however, I am only reading back 0.55mV in the NI 9237, which equates to about 25LBs.

This makes me think something is not wired correctly.

Any ideas?

JimM

[JimMacD]

I forgot to include the Load Cell model #: http://www.omega.com/pptst/LCGD.html

[OKors]

The wiring looks pretty straight forward if you follow the instructions in this guide.

My guess is that it's a scaling issue

Questions:

Why is DAQmx unavailable?

Your doing your own scaling... How are you currently scaling the load cell?

Are you using Signal Express?

If so please post your project file.

Once I understand a little more about your system I can probably help if it's a scaling issue.

[JimMacD]

OKors,

To my knowledge, the DAQmx VIs are to be used on a PC based system, and use the configurations setup in MAX. They are not meant to run on a cRIO target. The same goes for SignalExpress, not compatable with cRIO in the same way a "cDAQ" system would be.

That being said, I am reading an analog voltage from my load cell, via the 9237 in the cRIO chassis (using Scan Mode). I beleive this is returning units of volts. I will attach a .zip file with my project so you can see the scaling I am doing, nothing fancy.

The load cell data sheet is also in the .zip folder.

I have connected the load cell to the 9237 as follows:

Red (EX+) = 6

Black (EX-) = 7

Green (Out-) = 3

White (Out+) = 2

[OKors]

This question may be better answered in the LabVIEW or Signal Conditioning forum.

Sounds to me like you're missing the scale. I'm not sure how you'd to this in your environment but in SE there's a Custom Scale option. From there I use either Map Range or Linear (y=mx+b slope) depending on the lab equipment or transducer type being used.

If you can figure out how to create a custom scale you should have enough info in your load cell's datasheet to plot a slope.