Fatigue assesment of CFRP parts

[CarlesFEA]

Hi,

Could someone recomend me some references were I can get fatigue data for UD T300 and Twill 2/2 T300? So far I have been able to find several databases (mainly for windmill applications) but none of them showed T300 characteristics:

http://prod.sandia.gov/techlib/access-control.cgi/1997/973002.pdf

http://www.wmc.eu/optimatblades.php

Does the Composite Materials Handbook CMH-17 (

http://www.cmh17.org/)

contain some "rules of thumb" to asses fatigue (i.e: sigma_max @ 1e6cycles = 40%UTS...)? I have acces to MIL-HDBK-17-3F but chapter related to fatigue is empty...

Thank you in advance,

 

[ESPcomposites]

Two situations are common: At fastened joints and delamination growth (due to impact or at edges).

There is a decent amount of information with regard to fatigue if you look in the mechanically fastened joints section.

There is also some information with regard to delamination growth speckled through the handbook (but CMH has recently reorganized things so I can't point you to a section).

The general "rule of the thumb" is that if the structure is sized to notched allowables you don't usually have a fatigue problem unless either (a) there are many cycles such as in a rotorcraft or (b) the fatigue level loading is very high.

Brian

http://www.espcomposites.com

 

[berkshire]

This may help, Sailplane wings are similar to wind turbine blades.
B.E.

http://papers.sae.org/872459/

The good engineer does not need to memorize every formula; he just needs to know where he can find them when he needs them. Old professor

 

[ESPcomposites]

If you can be more specific, it would help. Are you talking about unnotched, notched, or a fastened joint? What is the layup and matrix? What is the application? Data for just UD tape probably isn't very useful.

Bakis, C.E., Simonds, R.A., Vick, L.W., Stinchcomb, W.W., "Matrix toughness, Long-term behavior, and Damage Tolerance of Notched Graphite Fiber-Reinforced Composite Materials", ASTM STP 1059, 1990, pp 349-370

From the above reference : T300/5208 (0/45/90/-45)4S notched specimen has a fatigue life of about 130 MPa at 1e6 cycles, R=-1, 10Hz. Of course you need to consider environmental degradation, layup, actual loading, actual configuration...but that is a point of reference. Some rules of thumb can be found in MIL-17 and CMH-17 as mentioned above.

Brian

http://www.espcomposites.com

 

[ESPcomposites]

Crews, Bolt-Bearing Fatigue of a Gr/Ep Laminate. Use of T300/5208:

http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19800020242_1980020242.pdf

Brian

http://www.espcomposites.com

 

[CarlesFEA]

berkshire and ESPcomposites,

Thank you for your interest.

We are using composites for almost first time to create a gripper: imagine like a 2.5x22 beam:

....%%%%%%&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&BBBB&&&&&&&&&&&&&&&&&&&
..%
%
..%
....%%%%&&&&&&AAAAA

with a fixed support at A and a vertical force at B

It is not for any critical application (automotive, space, aero,...) but for some paper machinery
Not many details have been provided by our local supplier concerning plys characeristics: only that a stack of 4 (6 in the "%" area) are being used. UD Carbon T300 500g/m2 epoxy (Vf=51%) @ 0 are covered on the outside by a layer of woven Carbone Twill 2/2, T300 200g/m2 epoxy (Vf=47%).
We have a FE model that correlates prety well with ForceVsdeflection measures that have been done.
Main stresses are found far from the fixed support (at the transition between % and &) so I think it would be more like an unnotched situation..

 

[ESPcomposites]

Another good paper is:
[R8] Whitehead, R.S., “Certification of Primary Aircraft Structures', 14th Symposium of the International Conference on Aeronautical Fatigue (ICAF), June, 1987.

The unnotched and notched data tend to converge at high cycles (free edge delamination plays a role).

So the structure is such that it can not take an impact? Sizing to unnotched allowables, although not unheard of, is relatively rare. But I am not familiar with your application and you will have to make that call. Just be aware that if it gets damaged, even by a low energy impact, that the residual strength can be dramatically reduced.

Brian

http://www.espcomposites.com

 

[CarlesFEA]

ESPcomposites,

Thank you for your interesting comments.
Our part is not supposed to see any impact, just a cyclic vertical force.

Going through some papers found at the NASA archives, I have found:
*for a [+-45]s, R=0.1 @10Hz sigma_max=52%UTS = 90MPa @1E6 cycles (at "Stifness reductions during tensile fatigue testing of grpahite/epoxy angle-ply laminates")
*for a [0,90,+45,-45]s, (notched) , R=0.1 @10Hz sigma_max=80%UTS @1E6 cycles (at "Fatigue damage in notched composite laminates under Tension-Tension cyclic loads")
*for a [+45,90,-45,0]s, (notched) , R=0.1 @10Hz sigma_max=70%UTS @1E6 cycles (at "Fatigue damage in notched composite laminates under Tension-Tension cyclic loads")

 

[ESPcomposites]

Again, have a look at MIL-17 or CMH-17. For Tension-tension, and compression-compression the fatigue strength is very high - 80-90% of UTS. But for fully reversed, it is much lower (about 35% of UTS). You didn't state the loading condition, but you probably don't want to just consider the best case scenario. Also, is there any chance the part could be dropped, mishandled, banged, etc.? If so, be careful about using the full unnotched capability.

Brian

http://www.espcomposites.com