Ply based strain allowables

[garyjh]

When creating a laminate and then selecting the analyse button you can see the effective elastic engineering constants and the laminate failure allowables (both stress and strain Fu1, Fu2, Fsu12, eu1, eu2 & esu12). These are LAMINATE properties. You are using CLT for the elastic constants and the generated moduli are considered an accurate estimation. However, how are the strengths (Fu1, Fu2 & Fsu12) predicted?

When I then use this laminate in a composite PLY BASED analysis it uses the LAMINATE failure allowables for each of the analyses. Is this correct? I would have expected the orthotropic ply material stress and strain allowables to be used for a PLY based analysis. With the LAMINATE properties used in the LAMINATE based analysis.

However, when the laminate based analysis are selected in the failure tab it does not read across the pristine laminate properties from the created laminate. Why does it not read across the effective laminate failure allowables (Fu1, Fu2, Fsu12, eu1, eu2 & esu12)?

The following are values that are typical of carbon epoxy UD tape and are follows:

Orthotropic Et1 = 128 GPa, Et2 = 8 GPa, Ec1 = 120 GPa, Ec2 = 9 GPa, G12 = 3 GPa, Ftu1 = 2300 MPa, Ftu2 = 20 MPa, Fcu1 = 1200 MPa, Fcu2 = 150 MPa, Fsu12 = 90 MPa, etu = 0.018000, ecu = 0.011000

Hypersizer generated QI layup allowables E1 = E2 = 48 GPa, G12 = 18 GPa, Ftu1 = Ftu2 = 126 MPa, Fcu1 = Fcu2 = 362 MPa, Fsu12 = 89 GPa, eu1 = eu2 = 0.002633, esu12 = 0.004943

(It should be noted we consider the generated strength & strain values to be extremely low. We would expect tensile values higher than this and the tensile strength to be higher than compression strength. This material with this layup has been tested with Ftu1 = Ftu2 =600 MPa and Fcu1 = Fcu2 = 300 MPa. This would give strain allowables etu1 = etu2 = 0.012500 & ecu1 = ecu2 = 0.006250. Hence the question above about how are the strength allowables generated. We have also used the same material to create a Hypersizer laminate with 30% ±45° plies, 65% 0° plies, 10% 90° plies and the 90° direction has considerably higher strength and strain allowables than the 0° even though there is only 10% plies in the 90° direction!)

The ply based analysis uses the QI layup generated strain allowables of 2633 microstrain which is not the strain allowable of an individual orthotropic ply?

Your reasoning and logic for using the laminate allowable for a ply based analyses would be appreciated, as currently we are ignoring the ply based analyses except for OHT/OHC where we can input the damage tolerance strain allowables.

Thanks.

[Phil]

When creating a laminate and then selecting the analyse button you can see the effective elastic engineering constants and the laminate failure allowables (both stress and strain Fu1, Fu2, Fsu12, eu1, eu2 & esu12). These are LAMINATE properties. You are using CLT for the elastic constants and the generated moduli are considered an accurate estimation. However, how are the strengths (Fu1, Fu2 & Fsu12) predicted?

The process to determine the effective Fu1 is to apply a 1 lb/in load in the x direction and calculate the ply-by-ply margins (based on your selected failure mode on the left hand side of the form).  The minimum margin for any ply is used to determine the strength of the laminate by Fu1 = 1 * (MSmin + 1) lb/in.  (this is then converted to SI units for display).  Note that depending on the laminate, it can be highly dependent on which failure method is chosen on what the effective strength will be.

Also note that these effective engineering constants are listed for reference only.  These are not used in the analysis on the sizing form at all.  The only time these properties are used is if you create an "effective laminate".  If you are using discrete laminates in your analysis, then these effective engineering constants are not used.

When I then use this laminate in a composite PLY BASED analysis it uses the LAMINATE failure allowables for each of the analyses. Is this correct? I would have expected the orthotropic ply material stress and strain allowables to be used for a PLY based analysis. With the LAMINATE properties used in the LAMINATE based analysis.

Actually, no, when doing PLY BASED analyses, HyperSizer uses the orthotropic ply material stress and strain allowables.  It does not use the effective engineering constants.

However, when the laminate based analysis are selected in the failure tab it does not read across the pristine laminate properties from the created laminate. Why does it not read across the effective laminate failure allowables (Fu1, Fu2, Fsu12, eu1, eu2 & esu12)?

For LAMINATE BASED analysis, HyperSizer uses the strain allowables entered on the Laminate Allowables Tab of the orthotropic material.  Again, not the effective engineering constants from the Composite analysis form, rather the laminate based allowables on the orthotropic material itself which can be functions of %0s, %45s, etc.

The following are values that are typical of carbon epoxy UD tape and are follows:

Orthotropic Et1 = 128 GPa, Et2 = 8 GPa, Ec1 = 120 GPa, Ec2 = 9 GPa, G12 = 3 GPa, Ftu1 = 2300 MPa, Ftu2 = 20 MPa, Fcu1 = 1200 MPa, Fcu2 = 150 MPa, Fsu12 = 90 MPa, etu = 0.018000, ecu = 0.011000

Hypersizer generated QI layup allowables E1 = E2 = 48 GPa, G12 = 18 GPa, Ftu1 = Ftu2 = 126 MPa, Fcu1 = Fcu2 = 362 MPa, Fsu12 = 89 GPa, eu1 = eu2 = 0.002633, esu12 = 0.004943

(It should be noted we consider the generated strength & strain values to be extremely low. We would expect tensile values higher than this and the tensile strength to be higher than compression strength. This material with this layup has been tested with Ftu1 = Ftu2 =600 MPa and Fcu1 = Fcu2 = 300 MPa. This would give strain allowables etu1 = etu2 = 0.012500 & ecu1 = ecu2 = 0.006250. Hence the question above about how are the strength allowables generated. We have also used the same material to create a Hypersizer laminate with 30% ±45° plies, 65% 0° plies, 10% 90° plies and the 90° direction has considerably higher strength and strain allowables than the 0° even though there is only 10% plies in the 90° direction!)

The ply based analysis uses the QI layup generated strain allowables of 2633 microstrain which is not the strain allowable of an individual orthotropic ply?

Your reasoning and logic for using the laminate allowable for a ply based analyses would be appreciated, as currently we are ignoring the ply based analyses except for OHT/OHC where we can input the damage tolerance strain allowables.

Thanks.

The effective strengths are really a function of which failure mode is chosen.

However, again, when HyperSizer does PLY BASED analysis on the sizing form, it is actually going ply by ply and using the orthotropic ply stress and strain allowables to calculate margins and is not using these effective engineering constants.

One other thing to point out is that you said you use ply based allowables for OHT/OHC and I just wanted to make sure you are aware that you can do these on a LAMINATE basis as well as there are laminate based OHC and OHT curves that can be entered.

Hopefully this clears things up.  Let me know if you still have questions.
Thanks,
Phil

[garyjh]

Phil,

I am using effective laminates to give a rough order of magnitude of thickness and therefore mass.

With regard to the strain allowables used for the ply based analysis is it because I am using effective laminates, because the report states the strain allowable used for the MOS calculations (max strain, tsai, etc. ) and it is equal to the stress allowable of the created laminate divided by the elastic constant. As the strength allowables are so low it is giving a low strain allowable which is sizing the structure I am looking at.

I have repeated the analysis with a laminate with the same layup and which gives approximately the same thickness as the effective laminate component results and the strain allowables used are, as you state, the orthotropic strain.

This, I surmise, means that when using effective laminates you can not accurately perform ply based composite strength analysis, except for the OHT/OHC damage tolerance allowables which are user defined, and is in part due to the fact that ply strains can not be obtained as the effective laminate has in effect no plies.

Or the user has to be fully aware of where the allowables are derived from and there possible inaccuracies (like I do now) enabling them to choose to use them or not because the MOS & critical failure analysis vary considerably between effective laminates or an actual laminate of the same % ply orientations & thickness.

Can you please provide an example of the calculation of the effective Fu1 with, for example, max strain theory and the Tsai-Wu interaction theory.