How to size stiffeners?

[sdar79]

Let's say I have an aluminum skin, carrying a compressive load over distance "a" from frame to frame, and supported by longitudinal stiffeners equally spaced at width "b". I am sizing everything, having the required loading in lb/in. After selecting a skin thickness I can get the buckling stress of the skin based on the dimensions "a" and "b", assuming that all sides are simply supported (Bruhn chapter C5). How do I size the stiffeners? I know to check for crippling, but there must also be some minimum moment of inertia for the stiffener to provide a simple support, below which it is no longer effective at restraining the skin. Bruhn offers such an equation for stiffening a shear web in equation C10.8, but I can find nothing in Bruhn or our OEM manuals on sizing a stiffener for a panel under compression, or combined shear/compression. Can anyone offer help or point me to a reference?

 

[SWComposites]

Stiffeners are sized to a) prevent overall panel buckling (not the skin between stiffeners), b) carry the panel compression loads when the skin is buckled (see effective width section in Bruhn), c) provide arrestment of skin cracks/cuts (damage tolerance analysis).

 

[rb1957]

as above, the effective skin works with the stiffener to resist the compression loads. crippling (and long column) strength of the stringer (and skin) needs to exceed the applied compression.

remember the skin can buckle in shear (diagonal tension) and this adds load into the stringer.

you are correct … there should be some design guidance about stiffener size verses skin thickness … a small stiffener on a heavy skin won't support the skin (this may be in the diagonal tension section (since it relates to the ability of the stiffener to resist the skin shear buckle.

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

 

[LiftDivergence]

These may be of some assistance:

NACA TN 455
Comparison of Three Methods for Calculating the Compressive Strength of Flat and Slightly Curved Sheet and Stiffener Combinations

https://ntrs.nasa.gov/search.jsp?R=...s=N=0&Ntk=All&Ntx=mode+matchallany&Ntt=tn+455

NASA-TM-108399
Preliminary analysis techniques for ring and stringer stiffened cylindrical shells

https://ntrs.nasa.gov/search.jsp?R=...=0&Ntk=All&Ntx=mode+matchallany&Ntt=tm+108399

NACA TN 3781, 3782, 3783, 3784, 3785. & 3786
Handbook of Structural Stability (inc. Compressive Strength of Flat Stiffened Panels)

https://ntrs.nasa.gov/search.jsp?R=...=N=0&Ntk=All&Ntx=mode+matchallany&Ntt=tn+3785

There is also a very nice treatment in Theory and Analysis of Flight Structures Robert Rivello
Pages 479-500

I am assuming since it seems like the driving factor for your design is compressive strength, you are not in an area of appreciable tensile stress and so you are not concerned as much with things like residual strength of a cracked panel ie load shedding to substructure. However, this is simply an assumption - there are areas of a fuselage which see both tension and compression during a GAG cycle and crack buckling can be a real concern. So if you need to look at multi-element damage this would be a different ball game.


Keep em' Flying
//Fight Corrosion!

 

[rb1957]

I'd've thought Niu would've had some guidance (for sizing stiffeners).

Are you looking at wing or fuselage stiffeners ?

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

 

[sdar79]

I'm in the maintenance world where we usually design to original capability, but am working on a small mod now where I can't keep the original geometry. This particular panel is mostly shear/tension loaded, but it got me thinking that I don't know how to size stuff if it were a compression panel! Which is why I'm asking. I don't have Niu's book but will see if I can find a copy.

So the best I can tell, the process would be:

- Determine target design stress and sketch initial geometry to match this
- Calculate section properties to include the stiffener plus the effective skin width at the design stress (or 1/2 distance to next stiffener if skin is fully effective). Use this section for the following two calculations.
- Column buckling analysis, with length being from frame to frame, pin-ended (or whatever restraint I feel best), and considering eccentric loading if the stiffener runs out rather than attaching to the frame
- Crippling analysis
- Skin buckling analysis if no buckling is allowed (for us it is generally not allowed at limit load), typically assuming simply supported at frames and stiffeners
- Repeat process until everything has suitable margin of safety

Is this correct? Am I missing anything?

 

[rb1957]

i'm in the mod world too. I get where you're coming from … "equivalent strength".

Your basic problem is that you don't know any detail stresses. The best we can do (IMHO) is
1) calc the allowable stress for the pre-mod panel,
2) assume the load is the pre-mod allowable, (this isn't as easy as it sounds !?),
3) calc load redistribution caused by your mod,
4) reinforce the structure to handle these loads (lick o'paint).

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

 

[Burner2k]

sdar79,
Niu Chap. 14 "Compression Panels" will be your go to reference for sizing panels (skin as well as stiffeners).

>> considering eccentric loading if the stiffener runs out rather than attaching to the frame.
Could you elaborate on the above?

I'd think it would be wise to go through Flabel just to determine & be informed when effective width contribution from skin would be a valid consideration (or not). IIRC, it depends on the inter-rivet spacing and thus the buckling strength. Usually, the spacing is such that the effective skin area contribution would be effective, but from a knowledge perspective, it would be helpful to learn on how to determine the above.

Lastly, on your current mod issue, how are you sizing your stiffeners & skin if they are loaded in tension+shear? I have go through Bruhn or other reference books again, but I do not recall seeing any kind of MS interaction equation for shear & tension combined loading.

 

[Burner2k]

am answering my own question. Bruhn C9.5 gives MS interaction equation for curved sheet panels under combined action of shear & compression. If axial load is tension instead of compression, Bruhn recommends adding a negative sign but still use compression allowable.

 

[SWComposites]

Sdar - yes, you are missing the damage tolerance analysis.

 

[Sparweb]

Am I missing anything?

And shear buckling. Bruhn C5, IIRC
If your cut-out is aft of the spar, there's always shear from torsion (the rudder).
Fuselage bending gives you shear on the sides (particularly around windows and doors).
If you track down a copy of Niu, you can look at Chapter 6 on "large cutouts", in addition to the chapters referenced already.

No one believes the theory except the one who developed it. Everyone believes the experiment except the one who ran it.
STF

 

[sdar79]

Hey all, I was able to find someone with a copy of Niu. Specifically chapter 14 held what I was looking for. Turns out, our OEM manuals do have a procedure for each of the individual steps, but had nothing which say "do these steps when designing a stiffener", which is where Niu was very helpful.