ATI force balance calibration proble

[mopen]

Dear folks,

I have a question regarding the ATI Gamma Force transducer. website :

http://www.ati-ia.com/products/ft/ft_models.aspx?id=Gamma

I am gonna use it for airfoil force measurements in low speed wind tunnel. The problem is that each time i try to calibrate it statically using dead weights i get appreciable interference (50%) between the calibrated force and the other forces, i use sting of length 50 cm that protrudes from the center of the balance. Also i get inaccurate values for the force. I use dead weights between 100 g to 600 g for calibration. So the questions is this balance valid for airfoil force measurements and why it gives this large inteference?

I appreciate your help.

mopen

 

[Sparweb]

You may have to provide a photo or a diagram of your apparatus. I think you mean that you have two or more of these scales, installed so that they support a balance-arm mechanism reaching into the wind tunnel, and capable of measuring loads with small deflections. Good so far. But unless I can see the mechanism I can't determine if there is a flaw in the arrangement of the arms and supports.


STF

 

[btrueblood]

"i get appreciable interference (50%) between the calibrated force and the other forces"

Let me try to rephrase your question and make it more technically correct: You apply a known, calibrated force in a certain direction (let's say a -x axis force), and the transducer gives you a signal on multiple axes, i.e. x, y, and z axis forces, and probably some moment values as well? This is normal, and expected: the transducer is offset by a moment arm from your calibration force application, so moments must exist and be accounted for. Further, no transducer can be perfectly aligned, nor can your calibration weight/pulley system be, so some crosstalk between axes is going to occur. Do realize that the output matrices may change appreciably when your actual test model is mounted and the weight of the model adds to other forces acting on the transducer.

You need to construct a set of matrices, in which the calibrated forces are "solved" as a set of linear equations operating on the transducer outputs, you will end up (if you do it right) with more sets of calibration values than you have transducer outputs. You prepare a best fit straight line, or least-squares fit, for all of the calibration data. You should then have a set of equations that (when measured outputs for each channel of the transducer are entered) can solve for your desired x/y/z forces and moments.