Preferred Shear Testing Method for Lamina Data

[EnginerdNate]

I'm working on obtaining lamina level data for a material system we are using in one of our products. Looking at past programs at my workplace, we have used both the D3518 and D5379 tests for lamina (0/90 balanced woven composite) level shear property data.

Is one of these tests considered superior to the other/Do you have a personal preference?

Does the answer above change if the intended use is for developing a shear strain allowable vs obtaining accurate shear modulus data or both?

I've talked with the older employees around my office and no one had a straight answer for me on why we'd choose one over the other.

Thanks,
Nathan

 

[SWComposites]

Both will give accurate moduli.
But the lamina shear strength from either is basically useless for laminate strength prediction.

 

[EnginerdNate]

I am using max principle strain predicted vs open hole tension/compresson allowables (tested in the minimum fiber direction) for the majority of the layers in the laminate but I have a significant portion of the laminate that's biaxial (0/90 plies only) and this approach did not seem to make sense for that layer. In the 0/90 layer I had planned to check Tensile/Compressive in the primary fiber directions as well as shear to get around this. Do you have a better suggestion? At past jobs we used interactive strength criteria but it's my understanding those are basically considered obsolete/incorrect at this point in time.

Modulus data is for informing the FEM property cards and for input into CLT for derivation of the expected laminate properties for the non 0/90 layers. Good to hear either test is fine for that part.

 

[SWComposites]

Ok, you are correct you need a shear strength for the 0/90 layup. I would use the v-notch test method, since it has way less edge effects. And would use an onset of nonlinearity stress rsther thsn ultimate stress, to be conservative (if that gives negative margins, then add 45 plies).

 

[EnginerdNate]

Yes, I had planned to use .2% offset.

I may be stuck with the D3518 on this particular program (legacy compliance...) but will keep that in mind for the future. Hopefully the margins will be high enough to not merit a ton of additional investigation.

Do you have a strong preference between the two V-notch tests? The D7078 looks a lot more robust to me due to it's clamped shear load introduction vs loading in bearing on the sample edges. It would seem to me that that would avoid issues with out-of-plane deflections/bending in thinner samples. Some of our materials are painful to get much more than ~1/8" panels out of due to low ply thickness.

 

[SWComposites]

Yes, D7078 is generally better for in-plane shear tests.

 

[EnginerdNate]

That was my impression. The other looks optimal for replacing short beam shear interlaminar tests because of the smaller sample size. Laying up a D7078 sample for interlaminar testing sounds... onerous.

 

[SWComposites]

There is no real benefit for using a v-notch test for interlaminar shear strength over SBS test unless the material has thru thickness reinforcements.
However, if you need to measure interlaminar shear modulus, then the smaller v-notch specimen can be used.

 

[EnginerdNate]

Interesting. I had read a number of opinion pieces that seemed to think that the SBS tests imparted localized compressive stresses in the laminate that skewed results.

 

[Compositepro]

My understanding is that SBS is a relatively cheap and convenient test that is useful to compare the shear strength of different materials but the data is never used to design or analyze structures.

 

[SWComposites]

Oh yes, we often grit our teeth and use SBS derived values for analysis. It’s usually conservative, so not really an issue. However, if you have primary structure with significant interlaminar stresses you should be conducting design specific tests to validate the analysis and properties.

 

[EnginerdNate]

At my last job we specifically chose our material system because it had well established A-basis material properties data you could purchase. It came back to bite us when the big customers who paid to develop those properties came in and bought up all the stock every couple of years when they had to produce stuff for their government programs.

We are using SBS on my current project but it's not considered primary structure and so far my interlaminar margins are +HIGH so I'm not overly concerned.