Composites and Strain Gauges 2

[gwolf]

I have a small carbon/epoxy structure upon which multiple strain gauges have been placed. The locations and orientations of the gauges were designed (not by me) to detect and identify a large number of natural frequencies up to a few KHz. There is a single gauge rather than a rosette at each location and the gauges do not line up with the fibres on the surface. It will be necessary to accurately assess strain amplitudes.

How should I relate the strain gauge results to fibre axial strains?

 

[DrADK]

Dear gwolf,

Don't you have the information sheet for the strain gauges? In such as sheet you can find info how to relate the frequency readings ton strains. All manufacturers use certain formulae to pass from frequency or voltage readings to strains.

In case you don't have that sheet, could you please tell me who the manufacturer is, since it is more than certain that such information can be obtained on the manufacturers web site!

Pardon me for not saying "hello", since I am a new user here!!! Hello Everybody!

It has become appallingly obvious that our technology has exceeded our humanity
- Albert Einstein

 

[gwolf]

Hi DrADK,

The point is not the calibration of the strain gauge but how you make real sense of the strains from a single gauge attached to a composite ply weave when the gauge is not aligned with one of the weave primary directions.

 

[DrADK]

My fault then......I thought that the main problem was that you couldn't relate the frequency readings to strain values.

Is there any possibility to attach a simple layout of your experiment, in order to make clearer what the situation is?

It has become appallingly obvious that our technology has exceeded our humanity
- Albert Einstein

 

[gwolf]

No, sorry, security....

 

[SWComposites]

gwolf -
since you only have axial gages then you do not have sufficient information about the entire strain state at the gage location to allow you to calculate the strain in any fiber orientation. Therefore your only option is to analyze the structre (probably via FEA), predict the strain at the gage locations in the direction of the gage, and once the analysis/model is sufficiently correlated to the strain gage data, you can use the model to predict the fiber diection strains at the gage locations. After you have gone thru all this agony you need to go bludgeon the fool who was too cheap to properly instrument the test article (why is it that people insist on saving $1.98 on strain gages and yet will spend $$$$ on fancy analysis and fabricating expensive test articles??)

 

[gwolf]

SWComposites - yes, I have my bludgeon ready.

I already do have an FE model so one gauge is enough at each location. What bothers me is how a uniaxial gauge will respond on a composite surface when the gauge is not aligned with one of the ply directions. It will pick up some e11 and some e22. Added to this there is resin on the outside which will smear any reult - not sure how big a factor this will be.

My ABAQUS model is set up to output results in the ply local directions not global directions (I am using orthotropic solids because the structure is too complex for shells and I am using one solid/ply or ply property group)

The root of this is that I have no experience of strain transformations with composites and if given a choice would have aligned the gauges with the fibres....

And yes, this component would be much better suited to a metallic solution but these designers just see "black metal" with very low density and high strength. It will probably fall to pieces via through-ply stresses.........

Thankful for any advice.

 

[SWComposites]

Generally strain gages will give acceptable readings on composite materials. However, if you have a fabric ply on the surface, the length of the gage should not be smaller than the tow width; I have seem some anomalous readings where a gage was small compared to the size of the weave. You should be able to output the strains at the surface of your model and then rotate them to whatever direction you need using standard strain transformation equations (similar to stress transformations except there are factors of 2 in the shear terms).

 

[gwolf]

SWComposites,

OK, thanks, we have a big weave but an even bigger gauge, I'll try the usual equations and see where it gets me.

 

[JimNDT]

Dear gwolf:
I am with a company that manufactures measuring systems for optical full field vibration analysis. They are into non-contact measurements of 3D deformations and strains. This technique might also make the use of strain gauges obsolete in your application. And the contactless measurement avoids to falsify the measured results since no additional mass is applied to the test specimen that could shift natural frequencies and amplitutes.


You might have a look at

http://www.dantec-ettemeyer.com

and follow the link to the vibration product lines.

I am sure that they can provide help with your callenging task.

Regards
Jim

 

[gwolf]

Thanks JimNDT, Very interesting. Looks like it would replace strain gauges and an FE model with a single measuring device.

Unfortunately it would be 100% impossible to use your device to measure the component in-situ, in use, which is where we need to do it. I'm sorry but I can't elaborate.

I do, however, take careful note of your suggestion because this measuring technology was previously unknown to me and looks great.

Regards,

Gwolf.