Shear Stress in Dual Core Honeycomb

[Hood91]

Hi All,

I can't seem to quite figure out how this works. I'm analysing a honeycomb panel which has a higher density core above a lower density core (for local crushing issues).

When this panel is subjected to bending, is the core shear distributed out depending on the shear modulus and depth of the different cores or is it a usual shear distribution as with using only one core? I can make arguments for both so I'm wondering what the correct answer is and perhaps a source or background reading into why.

To note, there is a single ply between the two core layers to aid manufacturing.

Looking forward to reading your views on this!

Thanks,
Matt

 

[GregLocock]

I think you need to go back to (a) the stress strain diagram in bending, and (b) equivalent sections as in reinforced concrete and (c) free body diagram



Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[Hood91]

I'm not sure it works the same as in equivalent sections because the honeycomb obviously doesn't take any end load and so all tension / compression must be taken in the outer skins. The tension and compression must be equal and opposite, which is leading me to believe that the shear force through the honeycomb depth would be as if a single density honeycomb was used.

But that thinking contradicts that the stiffer material picks up more load...

Are you saying that you would distribute the shear based upon the shear moduli and thickness of the two honeycomb cores?

 

[GregLocock]

Hmm. I can see the puzzle. I'll need to think about it.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[rb1957]

assume the two core weren't tied together.

if you assume both bend (deflect, curve) the same then the stiffer one will react more of the load. but this has each developing bending stress fields, so you'd get an incompatability at the interface.

so what if some of the bending was reacted by a couple between the two core ... the upper one would be in compression and the lower one tension. one idea for the couple is to completely cancel the bending stress,so both have zero stress on the interface. I think this couple may not balance, but I think you can juggle things (what you could call "polishing the tu?d") ...

or you could test it !?

another day in paradise, or is paradise one day closer ?

 

[ESPcomposites]

How thick are the two cores? Have you checked to see if the lower density core can react the entire shear load? If so, then you probably don't need to determine the actual distribution.

Brian

http://www.espcomposites.com

 

[rb1957]

are there face plies between the two cores ? even more indeterminacy ...

can the stiffer core handle the load alone ? or the more continuous core (it sounds like one core might be for local reinforcement).

have you tried an FEA solution ?

another day in paradise, or is paradise one day closer ?

 

[Hood91]

I guess no one knows for sure!

Yes, ESPcomposites - that is the approach I had taken. I was more asking as a matter of interest as I wasn't sure how to analyse it if, for example, the lower density core was not able to take the entire load.

I may create an FE model to see the result that gives, but I assume that the FE model would split the load according to the stiffnesses.

 

[stressebookllc]

It may be fair to say that the load would split up to the point where the 1z or 2z stiffnesses are the same between the cores, and beyond that point the weaker core would fail and then, the panel is not the same anymore. FE Model could be a good approach, but it would be an idealized bond between the cores, which may be OK as long as the bond strength is not critical for failure.

Aerospace Stress Analysis and FEA Courses

http://www.stressebook.com

Stressing Stresslessly!

 

[aerostress82]

There is a single ply between the 2 cores, but the real question is what are there at the top and bottom of these 2 cores? Laminates? Of same number of plies? What is the expected bending behavior of this structure? Tension and compression at the top and bottom faces? Where is the bending stresd expected to be zero? At the cores somewhere or outside the corr region? Where is/should be your neutral axis?

And last one:
Is this a new design you are working on? Or is it already stress certified and you are trying to do more digging into it by understanding the whole structural behavior?

Spaceship!!
Aerospace Engineer, M.Sc. / Aircraft Stress Engineer

 

[Hood91]

There are aluminium skins top and bottom. The top skin (atop the denser honeycomb) is thicker. It has been designed this way to stop crushing (the panel must withstand vehicle driving loads).

If I imagine this as a beam in 3 point bending, I would expect the behaviour to be the same as if it were a single core? Tension and compression in the top and bottom faces, with the honeycombs transferring the shear. The denser honeycomb is say 1/3 the depth of the low density honeycomb so the N/A will be within the low density honeycomb. Because of this, I expect the ply in the middle to take some end load as it's not on the N/A... however I planned to assume this was negligible as the ply can't carry much load?

This is already stress certified, I am just trying to understand the whole structural behaviour.

 

[aerostress82]

The top skin being thicker is somehow making me think that the N/A could be located significantly at a higher location than assumed. So, the N/A might be right at the ply plane as that regular Inertia formula is Ixx_NA=Ixx+A*y[sup]2[/sup]. I'm sure you know what I'm getting at. So it would be better to calculate the exact N/A location and proceed from there on. This is the first step.

Once having found your exact location of the N/A in the structure, it would be good to check how it is bonded/fastened to the underlying structure and what kind of structural behavior this other structure would have.

You may have noticed by now- with the location/function/service life parameters, the structural assumptions / exceptions differ a lot. I could exactly guide you through why it has been designed like that accounting for all static/vibrational parameters, but it doesn't look possible to do this from this far and without checking the design and location of the structure. But above might give you some ideas on how to proceed further into your investigation.



It somehow sounds like an automotive design rather than aerospace. But as you posted this in aerospace engineering forum, I'm assuming we are already looking at an aerospace design.
So, my last question would be what kind of aircraft this is on, and where is this panel exactly located at? If it is a helicopter, I'm definitely not that familiar as my primary experience is on commercial aircraft.

Hope I could pinpoint exactly what you were asking. Try to find the stress analysis report for it if you can. Maybe someone in the company retrieved it at some point?

Spaceship!!
Aerospace Engineer, M.Sc. / Aircraft Stress Engineer