Interlaminar shear stress in composite laminates 1

[ggburne]

Hi. I'm having trouble to determine interlaminar shear stress in sandwich plate laminates. Does anyone has a procedure or program to calculate interlaminar shear stress in sandwich plate of 3 layer?

 

[SWComposites]

If the facesheets are thin compared to the core, then the core will essentially carry all of the thru-thickness shear load. And the shear stress distribution in the core is essentially constant. So as a first approximation, neglect the facesheets, and calculate the shear stress in the core using s = V/h (V = thru-thickness shear load/width, h = core thickness)

 

[ggburne]

bump ! The thickness of the top and bottom plates is 4 mm and the core is 20 mm. The top and bottom plates are made of steel and the core is made of elastomer. I need to know the correct procedure to determine interlaminar shear stress due to uniformly distributed load which is applied in normal direction to the sandwich plate.

Length = 2400 mm
Width = 1400 mm

I couldn't find any book or article that provides even a basic formula of interlaminar shear stress. I'm not sure by using s = V/h would be sufficient as that is just a normal stress formula.

 

[RPstress]

The interlaminar shear stress is essentially the same as the normal shear stress, due to it being the complementary shear to the normal. They're the same.

SW's method is fine, if a little conservative for your thickish, stiff skins. The shear varies linearly from zero at the outer surface of the skin to the interface value. It is then effectively constant across the core, although there is a tiny bit of the usual rectangular-section peaking with a very slightly higher value at the neutral axis (a really stiff core cf. the face sheets will increase this small peaking; with steel face sheets and an elastomer (=low modulus) core, it's not a worry).

A slightly more accurate value for the interface shear value is to divide the shear force by the distance between face sheet centerlines.

For an actual structure you need to do structual analysis to assess the shear/unit width, and then apply the shear stress formula as we've described. For a simply supported flat rectangular plate the peak will be in the center of the plate, and the formula in Roark will be adequate.

An old copy of Allen, "Analysis and Design of Structural Sandwich Panels" or Zenkert, "An Introduction to Sandwich Contruction" will clear things up. (I'm told that Tom Bitzer's "Honeycomb Technology" is also good.)

In fact, our old friends at Diab and Hexcel may well help:

http://www.diabgroup.com/europe/literature/e_pdf_files/man_pdf/sandwich_hb.pdf

http://www.hexcel.com/NR/rdonlyres/...B3-7EFF685966D2/0/7586_HexWeb_Sand_Design.pdf

With a flexible elastomer core watch out for the large shear component of the deflection due to the normal loading, and the comcommitent decrease in buckling loads due to any in-plane loading.

 

[ESPcomposites]

RPStress, thanks for the info. I would usually just go with the simple calc like SW proposed. But it is worth looking into it further, even if only for personal interest. I suspect in practice I may stick with the simple formulation for convenience though.

Brian

http://www.espcomposites.com

 

[ggburne]

Wow RP thank you so much, same goes to SW. Those links really help a lot !

"For an actual structure you need to do structual analysis to assess the shear/unit width, and then apply the shear stress formula as we've described.

For a simply supported flat rectangular plate the peak will be in the center of the plate, and the formula in Roark will be adequate"

For the statement above, when you say structural analysis to asses the shear/unit width, isn't it just the shear force divided by the width of the plate?

As for the formula in Roark, is it under formulae table in the section "Flat Plates" ?

Lastly to clear things up, for the dimension given above, can I just ignore the top and bottom face and just directly calculate the shear on the core using the Roark's formula?

 

[SWComposites]

"shear force divided by the width of the plate" - if you are referring to the in-plane shear force, then No.

you need the thru-thickness shear force (denoted as Q or V, typically) due to an out of plane load, such as a pressure load. Then use that shear force (lb/in) to get the shear stress as stated above.

 

[ggburne]

Thank you for your replies. I'm still new in composite analysis and trying to learn as much.

Another thing, is there any specific value or range of values of interlaminar shear stress than can cause failure in sandwich laminates?

 

[SWComposites]

Yes, a value that is higher than the core or adhesive shear strength.

 

[RPstress]

http://www.hexcel.com/NR/rdonlyres/...7D8B547CA/0/HexwebAttributesandProperties.pdf

The "min" values of strength are a reasonable guide, although they are not proper B- or A-basis. A 'good' bond to the skins will be better than the core strength, though be very careful with heavier cores, especially with larger cell sizes. (Bond strength more or less goes with cell peripheral distance.)

Note that foams will be significantly weaker than honeycomb of the same density, especially when used hot and wet. Diab, Gurit, ALCAN Airex and Rohacell have many foam material specs with 'representative' strength data, which is usually more than the typical minimum (again, especially when hot and wet). On the plus side with foams, the bond to the skins is stronger than the foam, at least ordinary polymer foam. You might get in trouble with the heavier syntactic cores, but not the usual stuff up to about 3 or 400 kg/m³ (19 to 25 lb/ft³).