Ftigue Information 3

[cirokos]

Hello there,

I am trying to find Wohlers curves or S-N curves for aluminum 5052. More specifically I want to calculate the fatigue life of a component made from this material with the Miners law but I cannot find fatigue information about the material.

Any help would be appreciated.

Thank you.

 

[rb1957]

I think these are "OEM" holes, standard to core, to perforate the cells.

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[cirokos]

Tmoose :

rb1957 is right. That's the function of the holes.

 

[SWComposites]

Am I willing to pay for tests? No.
I can refer you to test labs.
If you answer the questions I posted above maybe we all could help a bit more.

 

[cirokos]

SWComposites :
Because I am not managing the project so I cannot share the information publicly. Also, I cannot perform any tests by myself. I was just looking to see if there is any information online I could use.
The material is probably: Aluminum 5052-H19 Foil any other specifications are not very significant I guess.

 

[rb1957]

ok, but someone has asked you to look into the problem.

You're looked (and may look more) but no information readily to hand.

You believe you understand the problem.

I doubt you'll find much information from other products (like panels with heavier cores or 2024T3 core). You just may get information from Hexcel.

You could propose some design changes based on your good sense and understanding your projects limitations.

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[cirokos]

Yes, I was asked to look into the problem. Fatigue tests were performed and will be performed.
I will try to contact Hexcel to see if they can provide any information.

 

[SWComposites]

if you are getting fatigue cracking after a relatively low number of cycles, then you probably need to use a higher density core. (yes, people will complain about the weight impact .....)

 

[cirokos]

SWComposites :
Cracks appear in a relatively low number of cycles from those tiny holes. Do you think that the cause may be the low density of the core?
How can you justify that?

 

[rb1957]

"low density core" = thin walls. Higher density core = thicker walls (= less stress).

I'd look to change material as well.

Is this a space application ? Why vent the cells ?

Yes, I'd talk to Hexcel, see if they've had other people having the same/similar problem.

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[SWComposites]

well, the applied fatigue stress level must be high enough to initiate and grow cracks. A higher density honeycomb core will have thicker foil, therefore the stress will be lower, and fatigue life should be improved.

 

[cirokos]

I could perform the same analysis with a material with different properties (Young's modulus, Poisson ratio ect.) to see how the behavior of the stresses changes ( if it gets lower).
I am working the problem on Ansys APDL code so it is easy to change these parameters.

I don't really know the application of the core/panel. The vents are there to help moisture escape ( or something like that).

Ok yes, I get it now! ( lower thick lesser the stress the walls can edure.)

 

[rb1957]

are you seriously modelling the honeycomb core cells ??

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[cirokos]

For the examination of the stress concentration around the vents, I am modeling only one unit cell.

Most of my analysis is done in one unit cell, but I have also created a model of the whole geometry!

 

[SwinnyGG]

I guess that's one way to do it

You'll have much better results (better in this case meaning more likely to represent something not completely impossible in the real world) modeling honeycomb panels based on bulk panel properties.

Modeling a single cell is going to tell you basically nothing about real behavior... there's way too much localization in a honeycomb panel. Every single cell in the panel is loaded differently.

 

[rb1957]

but I guess you had the failure before you started modelling it ?

so your modelling is to try to make sense of the failure ??

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[cirokos]

SwinnyGG :
My material properties are based on the bulk material. How would you approach the problem?

rb1957 :
No, I got the failure mode after I had created the model. Also, the core fails by buckling of the cell walls when static load is applied. This type of failure (from the vents) is happening only in the fatigue experiements.

 

[SwinnyGG]

Bulk panel properties != bulk material properties

You need to talk to the manufacturer of the panels. They are your only chance at getting the information you're asking us to give you.

No, I got the failure mode after I had created the model.

So... you've created and run a simulation, and you're seeing failure around your vent holes

Or you have actual physical parts you have tested and you're seeing failures

The way you're explaining your scenario is extremely confusing

 

[rb1957]

how close is your fatigue load to your design limit load ?

ok, the panel fails (everything fails at some time). Is it failing near the manufacturer's capacity ? I assume you are using a prefab panel, and not building your own panel from core and face sheets.

If you are making your own panel, is it failing near a load at a similar commercial panel would fail ? If similar, then good. If lower, then there are some tricks to be discovered.

presumably the sequence was ...
you designed the panel,
modelled it, showed it was good,
built the panel, tested it, and it failed.
and now everyone is looking at you ? Sorry, painful place to be.

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[cirokos]

I didn't design anything or had any involvement in the panel's design and composition.
Because I was brought to the project last minute, don't have many information about those questions I will get back with more details.

 

[rb1957]

ok, but how about answering the technical questions ...
how close are your fatigue loads to your design limit load ?

is this a commercial panel ??

if so, is the failure close to the commercial allowable ?
if so, what does the manufacturer say about this ??

but it sounds like it's a custom panel. Then
is it similar to a commercial panel ?

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.