are layered Uni carbon and carbon weave similar? 2

[magnasanti]

Lets say you have a plain weave carbon fabric and wanted the same anisotropic properties lets say at 0 and 90 degrees but without the downfalls of a weave such as crimp. So you used unidirectional fabric at these specific angles. Would the result be similar but surperior to the weave since there is no crimp and fibers are as flat as possible?

 

[berkshire]

Yes.

You are judged not by what you know, but by what you can do.

 

[SWComposites]

Yes, no, maybe.

It depends. What do you mean by"result"? What is your definition of "superior"?

 

[magnasanti]

The question arrived because I told a friend that weave is just two or more layered uni fabrics at specific angles But the uni has the advantage of superior physical properties since there is no crimp and the fibers are as flat as possible.The superior properties can be lets say thinner,less weight, and compression/tension.

His response was A layered fabric of two uni-directional sets of fibers would be rather unusual. They are usually single layers with a few fibers to hold the fabric in position. A typical woven fabric is very different as you should understand by now because of the crimp and the fact that none of the fibers are in pure tension at any time. You would have to pull the fibers flat to have that happen and the weave doesn't allow that unless you are displacing the other fiber into an even longer path length.

 

[berkshire]

The superior properties can be lets say thinner, less weight, and compression/tension.
If you are using two layers of unidirectional, this is mostly true. However for the most part the woven fabric will have better peel strength.
Depending on your application this may or may not be a factor.
B.E.

You are judged not by what you know, but by what you can do.

 

[blakmax]

If you are using ONLY two layers (one 0 one 90) then it won't work because the laminate will not be symmetric and will distort when cured. You would need at least four plies 0/90/90/0 to prevent distortion due to coupling effects.

Regards

Blakmax

 

[magnasanti]

Might be a dumb question but is the same true for all angles with maybe the exception of 0/0?

 

[tbuelna]

As others noted, your laminate should have a ply stack that is balanced with regards to fiber orientation. Otherwise it will likely distort during cure. Besides balancing the orientation of your uni plies, it is probably a good idea to add woven plies at the outer layers and occasionally between the inner uni plies to help stabilize them.

If you are going to put mechanical fasteners or holes thru the laminate, it is a good idea to transition the laminate stack to all woven material in the location where the holes/fasteners will be. Uni materials do not like to have holes put thru them.

 

[blakmax]

There are three coupling effects which may be a bit of a trap for those familiar with isotropic materials.

The first is shear coupling. In isotropic materials direct stresses do not produce shear strains. In any composite with other than 0 or 90 plies the application of a direct stress will produce shear strains.

The second is coupling between direct loads and bending or twisting. One usually does not expect to see an in-plane load produce out of plane twisting, but for non-symetric laminates this will occur. For the mathematically inclined essentially if any element of the matix is non-zero then the laminate will twist and/or bend under direct loads. To prevent this for every ply at a given angle and of a given material must be matched by a corresponding ply at the same distance below the mid-plane at the same angle and the same material. This is termed a symetric laminate.

The third coupling is between bending and twisting, where a pure bending moment produces laminate twisting. Again for the mathematicians this is because the D16 and D26 terms are non-zero. To prevent coupling between bending and twisting you need a balanced laminate where for every ply of a given material at a given distance from the mid-plane there must be an equivalent ply of the same material and at the same distance below the mid-plane but at the opposite angle (e.g. +30 is matched by -30).

Now clearly there is a conflict between the requirements for balanced and symmetric laminates. In practice the coupling between bending and twisting decreases with the third power of laminate thickness, whereas the coupling between direct loads and bending or twisting decreases with the square of laminate thickness, so in practice it is more common to use a symmetric laminate.

Hope this helps.

Blakmax

 

[RPstress]

Non-crimp fabrics (NCFs) consisting of UD stitched together even on a speciallized Liba machine or similar can still have a surprising amont of crimp due the stitching shoving tows sideways, in-plane crimping if you will. I have seen NCF material with lower stiffnesses than five harness satin woven.

Also, a five or eight harness satin weave ply is quite similar to two plies of UD; you should be careful to balance warp-side down on one side of the ply centerline with weft-side down in the other half.

As noted above by berkshire, woven material can exhibit greater toughness. The CAI value for woven laminates is often higher than for UD ones. I have no CAI data for NCF to oompare that. Also, fiber bridging usually gives a higher value of ILSS for all-0° laminates. This course cannot happen with NCF or woven. With multidirectonal UD a lower than anticipated ILSS can be a result.

I have also seen some counter-intuitive behavour from a laminate of mixed UD and woven. Heater mats for a high-temp repair without autoclave pressure caused disbonding between the UD and the woven. Even harness-satin weaves have no 1-direction and 2-direction thermal mismatches. UD certainly does.

A woven laminate (even harness satin) very rarely has any measurable bend-twist coupling, whereas UD (even if NCF) always does if any angleplies are present.

To summarize my and others' above points, the answer I would give is 'Yes, quite often, but usually at the expense of toughness and some other more subtle laminate characteristics.'

 

[magnasanti]

BlakMax by the following does this mean that the coupling of either system decreases as the laminate gets thinner?

Now clearly there is a conflict between the requirements for balanced and symmetric laminates. In practice the coupling between bending and twisting decreases with the third power of laminate thickness, whereas the coupling between direct loads and bending or twisting decreases with the square of laminate thickness, so in practice it is more common to use a symmetric laminate.

 

[blakmax]

It is common practice to use a symmetric laminate, and yes for thin laminates there is an increase in coupling between bending and curvature, but for thin laminates the plate stiffness is usually low so they can bend.

I would always opt for a symmetric laminate.

Regards

Blakmax

 

[ESPcomposites]

Well, if you are only using 2 uni plies, then you will have a theoretical "problem" with coupling in/of the [A], , and [D] matrices (discussed by blakmax). But that is probably just an academic case. From a practical standpoint, the stresses (say after curing) needed to get the part flat would be minimal (since the bending stiffness is minimal). So let's consider where such a thin laminate might be considered. Perhaps the skins of a honeycomb structure. But the core would easily stabilize it. Residual stresses to force it flat should be minor. Of course, one might not be using uni-plies for such a case (rather a fabric might be used). The point I am making is that for any practical, unsupported laminate, in will not be that thin. As such, you would have more uni-plies to increase your chance of making a symmetric/balanced/etc. laminate. On the flip side, you can create some real problems in real laminates (a 0/90 is probably not real) if the coupling properties are not designed properly (that should be obvious).

Some of things to consider:
-If there is double contour in the part, the "better" material may be the fabric. This will depend on the manufacturing methods and amount of curvature.
-From a producibility standpoint, fabrics may be easier to drill (or least the outer layer)
-An outer layer of fabric may improve the scratch/peel resistance
-Damage resistance (such as compression after impact) can be increased by using fabrics (mentioned earlier)
-I think blackmax mentioned this too, but for certain loading conditions/scenarios, a fabric placed on the bond surface may be better than a uni-ply
-Guessing here, but there may be benefit of the fabric for in-plane damage resistance due to fatigue loading. This could be caused by the natural interlinking of the fabric which could resist matrix cracking.

Brian

http://www.espcomposites.com

 

[magnasanti]

Thanks for all the impressive and detailed reply's. The original question was more of a thought experiment which you guys have taught me something about. The actual scenario is a bit different and takes into account coupling of bending and twisting which is my main concern. I would like to expand and detail the project which is an odd transducer. Blakmax has taught me that a balanced lamina is in order something like 2x2 twill bias right? and the coupling or the degree which bending motion will effect twisting motion is dependent on its thickness the thicker the better or less twisting correct?

 

[magnasanti]

The transducer as you can see the coil pushes up bending the laminas outward