Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations cowski on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Fiberglass 7781 Dry Fiber Material Properties

Status
Not open for further replies.

zukis

Aerospace
Jan 19, 2015
1
Hi All,

I am new to composite materials my background is in metals. I am running stress analysis on fiberglass/silicone composite material. The fiberglass dry fiber is emended into the silicone and cured together. For stress analysis purposes I have calculated the modulus of the rubber using the instron machine. I was wondering if there is existing data available for fiberglass 7781 since its so widely used in the industry.

I am looking for the strength of 7781 (0/90 direction) in the E1 E2 E3 Nu12 Nu13 Nu23 G12 G13 G23 for dry fiber.

Can you please help or provide me direction to look up this information without running a instron test because of lack of testing time.

Thank you in advance for all your help.

Zuki
 
Replies continue below

Recommended for you

Berkshire's data is sound as far as it goes. It will give you undamaged tensile strengths for room temperature. Just ignore the silicone and the tensile strengths won't be a mile out. You probably want to use a fiber volume fraction of 0.4 to 0.5 but you can use the thickness in the datasheet as an approximation. So, tensile dry RT strength is about 450/0.009 = 50 ksi or 350 MPa. Depending on the amount of silicone the thickness may be a bit more so the 0.009" might well be 0.011" with a corresponding 20% reduction in strength...Also bear in mind that strength data is average for E-glass (which is part of the 7781 fiber weave definition). If you have S-glass or some other glass in an eight-harness satin weave you will need to apply some approximate factors. Also the fabric strength given is not an allowable strength.

That data also doesn't give you any of the moduli that you want. You can look at other Hexcel datasheets for an idea of the stiffnesses and Poisson's (their resin datasheets often have woven glass as a reinforcement). You can also do simple checks ignoring the silicone. Glass fiber's modulus is about 72 GPa and as a first approximation the E will be about 0.5*0.5*72 = 18 GPa in the 1- and 2-directions. This is a bit low; the 50% (one of the 0.5s I've multiplied by) of fibers at right angles will help a bit more than I've allowed for, for glass. 7781 will typically be about 20 to 23 GPa in the 1- and 2-directions in tension. With a matrix as floppy as silicone the compression modulus will almost certainly be significantly less and the compression strengths will be much less. A test or manufacturer's data for a similar material (if silicone is unavailable then another elastomer would be a good alternative, if available) are needed to quantify this.

Other Hexcel data of interest include (Hexcel are one of the great heroes of composites, freely publishing a great deal of useful data. Their arch-rivals Cytec are also a useful source.)

The silicone matrix is a bit of a problem. I'd argue that ignoring it is the best bet, but that does leave you with a problem for some of your desired properties (if you ignore the matrix then G12 is zero). Finding the properties of the silicone as you have started to do helps. Tensile testing material at 45° to the fibers will give you some useful shear data but do not trust it. It's a first approximation only.

The silicone properties of use area a basic tensile modulus and strength and some shear measurements, along with Poisson's ratio (it's yet another approximation but you can assume the matrix is isotropic as a first approximation). Where silicone is an elastomer, nonlinear behavior must be allowed for. A silicone supplier website should help a lot.

NASA CR 198229 Micromechanical Models for Textile Structural Composites micromechanics ( is a useful reference with guidance on what acceptable approximations can be made. It is not for elastomeric matrixes but is better than nothing.

We often use E3 = E2_for_unidirectional_material, E2_woven = E1_woven, S2_woven = S1_woven, &c. (in that datasheet S1 is 570 lb/in =~ S2 of 450 lb/in, which is about the accuracy of such approximations; also note that a very rough approximation for strength from scratch (a glass fiber is roughly 2200 MPa strong) gives 2200 * 0.5 * 0.5 * 0.8 (strength knock-down for woven) = 440 MPa or 64 ksi, Cf. 50...that's about as good as it gets).

See also NASA Technical Memorandum 107165 "Simplified Micromechanics of Plain Weave Composites" (
If there is any heat or wetness in the environment then matrix-dominated properties will be reduced. This also applies to compression properties. Testing or data for a similar material are needed.

For all these sorts of simple approximations their fairly gross inaccuracies need to be borne in mind.

Even though the NASA data may have micromechanics solutions for the properties you'd like, things like the G12 and G13/G23 will still need some similar data or measurements. You might be able to approximate them with the NASA info but silicone is not typical as a matrix resin (for structures) and you need to treat any micromechanics-derived numbers as essentially wrong. If you can find any manufacturer's data with a similar matrix you'll be much better off.

You haven't said what you need these properties for. Any numbers derived using the above cannot be used for anything like allowables. For examples of allowables, see for instance the AGATE database at NIAR NCAMP, but bear in mind none of their data is for a horrible elastomer-like matrix such as silicone.

The above may save you test time but is not anywhere near as reliable as testing. If you can, find some manufacturer's data for a similar material.

Well, good luck...
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor