01-14-2009 03:23 AM
I'm triying to make strain & temperature measurements. I use a SCXI 1001 Chassis w/SCXI 1520 and 1314 modules. I set it up on Quarter Bridge I configuration, and get data proportional to the deformations I cause. I use the Acq Wheatstone Bridge Samples (with Calibration).vi I found in the examples, with an small change due to sample&hold errors.
But the examples shows as physical units "lbs", what I can't understand.
Besides, I don't know how can I make sure that the scaling is correct, since I don't clearly understand what the map scaling is even reading the help (is it just a differente way to define a linear scaling? two points instead of slope and zero value?)
Thanks,
usuario.
Solved! Go to Solution.
01-14-2009 11:08 PM
The example VI gives you a couple ways of defining your custom scale. Which one is used is dependent on which tab is selected at the time the program is started. One method (linear) gives slope and intercept. For that example, they assumed the translation was from Volts in the prescaled to psi in the scaled. Perhaps for a pressure gauge.
The other method is 2 points or Map Ranges. A linear correlation between x1,y1 and x2, y2. If you said x1 is 0 and x2 is 1, then y1 would be intercept and y2 would be equal to the slope.
It's all just math. For that example, the conversion was from volts to pounds such as you might use for a load cell. LBS is just a label. Another example might relate 4 to 20 mA as the input to 0 - 100% as a valve output.
The subVI that creates the scale as two more polymorphic instances for a polynomial equation mapping, or for a piecewise table mapping.
01-15-2009 01:18 AM
thanks Ravens Fan,
I'm reading strain measurements to know the actual strain, not for it's use as tranducers, so didn't think of that possibility.
So, the scaling I must set is actually "1" for all values, right? Or asked another way, the readings I get from the example, without any scaling, are E (deformation), right?
And how do I set the scaling to 1?
Usuario.
01-15-2009 10:10 AM
I would set the tab to the linear scaling, set M =1 B=0. Apply a known load and see what the results are. See if they make sense. I don't know whether the output would be volts, millivolts, strain, or microstrain. The unit label could be changed accordingly.
I'm not using this VI for my strain measurements. I set my channels up as a task in MAX. That lets me do my null offset calibration in MAX rather than inside the program.
I'm not sure why this example for strain gages uses the Create Channel instance of Custom Voltage with Excitation. I would think Analog Input - Strain Gage would make more sense. That would handle some of the gage scaling within itself and not necessarily need the custom scale.
01-16-2009 02:32 AM
Hi Ravens Fan,
I've tried to, based on the example, make may first non-express acquisition .vi, and seems to work! And I guess that the units I get are strain, since the subvis ask for enough information to make the transformations themselves (gage nominal value, bridge type, etc).
Thanks for your support. About the null offset calibration, I don't see any differences between making it on the .vi or on MAX. Actually, I make it on the program since I don't know working on MAX very well (even less than in LabView).
Usuario
01-16-2009 09:55 AM
I'm glad it's working now.
It is definitely worth learning to do these kinds of things without using the Express VI's. They are a good way to get something up and running quickly. And if you right click and select Open Front Panel, you can see how they work in the background. But what they give you in ease of use can hurt you in flexibility and performance.
In the end there should be no difference in whether you use MAX to set up the functions or do it in LabVIEW itself. Whichever your most comfortable with and makes sense for your application. With MAX, you could adjust some parameters without having to modify your actual VI. Doing it in your VI can save you from requiring a user to set up the acquisition task separately.
Please mark whichever reply in the thread best answered your question and was the solution to your problem.