Measuring strain with cRIO 9237 module. Range problem.

[igleo]

Hi everyone,

I have a problem related to a temperature drift of 9237. I was trying to heat up the module itself with a blow dryer and saw an output increase in approx. 30,000 - 35,000 points (binary units - before the conversion to nominal values). This is about 0.2% of the whole range of 2^24=16,777,216 and this is within the specs. The problem I have is that the strain gauges I use give us about the same change in measurements (30,000-60,000 points) that is also approx 0.2-0.5% of the full working input range. Obviously I have a huge error in measurements. I use half-bridge configuration (one active and one dummy for temperature compensation). I'm not saying this is the problem of the module, but maybe you can suggest any way to widen the range of the strain gauge output so it would make the drift caused by temperature shift negligible. Does 9237 have any restrictions for strain gauges characteristics (sensitivity, etc)?

I was thinking about increasing the excitation voltage to 10V (max value, we have 5V now), but as I understand, it will increase the module input range. As a full-scale range is 25mV/V:  we have now +-125mV (with 5V excitation), but with 10V excitation we're going to have +-250mV.

Thank you for you suggestions!

[Brian_R]

Typically when you use a half-bridge configuration with a dummy gauge, the affects of uniform temperature change are voided.  Is there a chance that because you are using a blow dryer, you do not have uniform temperature across the material you are measuring.  What material are you measuring and how much strain are you putting on it?

[igleo]

Thank you for your reply!

You are absolutely right. The dummy gauge is for temperature drift compensation only. The gauges are located about 3 meters (9 ft.) apart from cRIO. I was heating up the cRIO, not the gauges. The reason I was doing it is the hot weather we had the other day. During those days I noticed a drift on all my strain channels, that is why I was using a blow dryer to simulate this drift.

The material is a steel rod (some helicopter parts). The strain is about 600-1000 kg.

 The problem is that the unbalanced bridge creates voltages of about 0.5% from the whole 9237's input range.

[Brian_R]

Since you are using a half-brige configuration, make sure you set the "Half-Bridge Enable" property of the analog input set to True.  Otherwise, the other half of the voltage divider will float and that floating signal may vary depending on the temperature.  The hardware will complete the voltage divider in the module, which might explain why you are seeing these effects.

[igleo]

Yes, of course I've done it.

[Brian_R]

I will try to reproduce your problem.  I can't think of anything else that would be causing this.

[Brian_R]

I would suggest checking the this article on temperature affects on strain measurement.  You can also view this document on general strain measurement. 

Since your signal and the temperature drift are on the same orders of magnitude, I would suggest trying to increase your signal-to-noise ratio.  You could possibly try amplifying your signal.