A problem while performing "Shear strain gage" measurement

[ramshree]

Hi guys,
I am performing a test to determine the shear modulus of a particular composite material and also a well known material like aluminum. But from my results, I see that for some reason I am having to multiply with a factor of 2, in order to achieve the desired results.
Here is a brief description of the procedure:
1. Test fixture used: Iosipescu test fixture. ASTM D5379
2. Test specimen used: Iosipescu test specimen 3.0 in X 0.75 in.

The fixture is subjected to a load by means of a 10kip MTI testing machine. The strain gages are read by means of the data acquisition machine. A filter is used for noise reduction. I guess... that is it... I think I am not missing anything..I have performed quite a few experiments to know that I am constantly missing only by a factor of 2. For eg, If I were to get say 0.80 Msi for a particular object as the Shear Modulus, I would end up actually get 0.40 Msi. The way the shear modulus is measured is by taking the slope of the obtained shear vs strain curve. (only in the linear zone) For this particular material its almost like 0.1 - 0.3 % strain.

Any ideas as to where the factor of 2 is coming from? I really appreciate your time. Thank you very much.

 

[Biggadike]

Its not somthing silly like the machine is set to give change about a neutral axis is it? I'm just stabbing in the dark here but I would suspect that a close examination of exactly what you measuring equipment is physically doing to the test piece geometry might provide the answer. Have you also tried checking the deflection during test with a clock dial indicator to give you a second opinion?

 

[Kiranpatel]

Maybe I am thinking along the wrong lines competely, but from my school days, I remember that we used to put in strain gauges in pairs at 90deg relatively to each other, primarily to eliminate the error due to mis-alingment. As strain gauges measure only linear strains, the shear strain was obtained as (S1-S2)/2. In your case, S2 might be 0 and S1 is whay you measure. Could this be the source of the factor of 2?
Maybe, I am talking rubbish but my curiosity is aroused and I will research this for my own interest.
Kiran

 

[kafeel]

This is just a check. Have you connected the strain guages in a bridge with 2 active arms or 4 acitve arms or with just the one active arm? depending on how you have connected it you will need to make suitable adjustments to the aquisition system(if it outputs strain directly), if you read just the voltage, then you may want to check your calculations. i may be completely wrong....

 

[Kenneth]

Okay, two more "obvious" ones:

1) Is your actual strain gage excitation voltage half of what you're using in your calculations/conversions?

2) Are you using the correct "gage factor" for your strain gage?


(I apologize if these are too obvious...)

 

[ramshree]

Firstly answering: Biggadike, I am very sure that the source is not from the way the testing machine is setup. I am positive on that.
Answering Kiran Patel:
Kiran, presently I am using 2 "shear strain gages". One on each side of the specimen. Each Shear gage itself is a combination of +45 and -45 orientation.
Kafeel and Kenneth:
I am pretty sure that this is not the possible source as I am sure of the connections and stuff like that. For example, to verify what is happening, we performed that same test with the same strain gages, (2 half bridges) and used a "strain Indicator" boxes. These are supposed to give us a ball park of the material property we are looking for and not surprisingly they were pretty close to what we would expect from the materials. And these readings were not off by any factor of 2. This problem only is coming when we are collecting data off the file generated from the computer after the testing is completed using the actual test procedure.

So at this point I have successfully eliminated most of the precarious possibilities in determining the source of error. Now we think its either to do with the software that we are running, or something like that.. something really silly like that. I will inform of you guys as to what exactly is the reason. I am sure I will be able to go on from here now.. as I have eliminated a lot of sources of errors, I think.

Thank you very much for your time guys and I really appreciate it.