Small pitch distances in composite joints 2

[Felipe_MRB]

Hi,
I would like to know if there is a simple way of calculating the margin of safety of a composite joint due to a pitch distance considerable loss in a conservative way. As for example 2.5D or 3D pitche distances.

The reason is to calculate the margin of safety loss due to manufacturing discrepancies.
For example bearing by pass or the laminate resistance margins of safety.
I was wondering if a linear correlation could be valid. Imagine I have the bearing by pass margin of safety location in the structure. The design pitch is 5D. Due to a discrepancy the pitch is reduced to 3D. Can one calculate the new margin of safety based on a linear interpolation considering a zero margin of safety for 1D pitch, design margin of safety for 5D pitch and see where the 3D pitch would fall?

Related thread:
thread31-419325

 

[rb1957]

Fatigue may be more critical than static loads.

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

 

[SWComposites]

Assuming you are referring to fastener to fastener spacing as pitch, and not edge distance, I would conservatively assume zero strength (NOT zero margin) for pitch = 2D and linearly interpolate to that.

 

[rb1957]

net area ? on 5D the net area factor is 1.25, for 3D is 1.5, or net area stress is 20% higher.

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

 

[SWComposites]

Net area strength does not work well for composite laminates.

 

[rb1957]

ok, why assume zero strength at 2D (and not 1D) ?

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

 

[WKTaylor]

F-MRB... complex subject... RB and SWC best-experienced/equipped to address/explain.

A few added thoughts...

There are related documents available on the subject of mechanical composite joints. DOD/FAA, AIAA and SAE have numerous text-books and papers written on composites in general... as well as mechanically fastened joints in composite structures. Shall we assume You have reviewed existing documents, for example... ???...

AIAA 84-0917 BOLTED JOINTS IN LAMINATED COMPOSITES

AGARD-CP-427 Behaviour and Analysis of Mechanically Fastened Joints in Composite Structures

AGARD-CP-590 Bolted/Bonded Joints in Polymeric Composites

AFWAL-TR-81-3041 {V1, V2, V3] EFFECT OF VARIANCES AND MANUFACTURING TOLERANCES ON THE DESIGN STRENGTH AND LIFE OF MECHANICALLY FASTENED COMPOSITE JOINTS

AFWAL-TR-81-3154 [V-I, V-II] DESIGN METHODOLOGY FOR BONDED-BOLTED COMPOSITE JOINTS

AFWAL-TR-92-4084 ANALYSIS OF BOLTED AND BONDED COMPOSITE JOINTS

NOTE.
Many of these documents emphasize 'bolted + BONDED' joints. When I worked laminated composite repair [years ago] it was evident that composite surfaces with fasteners were far more reliable when thin-film paste adhesive was added to the joint... fay surfaces and fastener holes... 'typical metal structure wet assembly' practices. The reason for this is NOT the added shear strength [unreliable]… it was (a) the added adhesive reinforcement [coated-onto and in intimate contact-with] of the never-perfectly-smooth 'fuzzy' surfaces of the 'drilled holes'... and (b) rigid [liquid] shimming between never-perfectly-matching composite fay-surfaces. Both (a) and (b) eliminate small voids/gaps and added rigidity to the mated-parts so the fasteners could 'do better their job'... at least that was the theory as explained to me and observed over-years of 'field' use.

Regards, Wil Taylor

o Trust - But Verify!
o We believe to be true what we prefer to be true. [Unknown]
o For those who believe, no proof is required; for those who cannot believe, no proof is possible. [variation,Stuart Chase]
o Unfortunately, in science what You 'believe' is irrelevant. ["Orion", Homebuiltairplanes.com forum]

 

[Felipe_MRB]

First of all, thanks for the answers.
I am relatively new in this subject, composite joints, and trying slowly to gain a deeper knowledge.
WKTaylor, the main documents I read are listed in the end of this text. I will look after the ones you listed. Thanks.

Basically what was annoying is that I saw in document [3] that the best joint strength to laminate strength ratio was about 2.5 to 3D pitch. This was against the recommended pitch which was told me of 5D, to design composite structures. So why 5D?
But yesterday reading document [1], I found the following information again about the best ratio of joint strength to laminate strength:

“Having the peak so defined is usual for highly orthotropic laminates, for which it is found by test that the best are obtained with a bolt pitch of about 2.5 bolt diameters. For near isotropic patterns, however, the more usual behavior is given by the intersection of curves E and I, in which case the maximum strength is obtained at about 3d pitch and is a tension failure. Trying to force a bearing failure in such a laminate to prevent a catastrophic failure mode means increasing the bolt pitch to nearly 5d and accepting a reduced joint strength.”

So, to prevent catastrophic failure, one should design the joint to fail in bearing. That makes sense to me. And this also tells me, but I am not entirelly sure of this, that 3D pitch can even make the joint stronger, increasing the static margin of safety in a non conformity repair.
So if I have a damaged hole, and the solution is to repair it by drilling a larger hole and installing a bigger fastener, if my design pitch is 5D and the repaired pitch falls to 3D, I might have increased the joint margin of safety locally.
Is this argument right?

References:
1. LJ Hart-Smith, “Mechanically fastened joints for advanced composites: Phenomenological considerations and Simple Analysis” Douglas paper 6748 A, Douglas Aircraft Company, USA, 1978.
2. L. J. Hart-Smith; "Bolted Joints in Graphite-Epoxy Composites," Douglas Aircraft Company, NASA Langley Contract Report, NASA CR-144899, January 1977.
3. WD Nelson, BL Bunin and LJ Hart-Smith, “Critical joints in large composite aircraft structures”, NASA CR 3710, Aug 1983.

 

[SWComposites]

rb - because he wanted something conservative. To estimate something less conservative one needs detailed info on the joint configuration which was not provided.

Felipe - no, reducing the pitch does NOT make the net tension strength go up. I have generated lots of test data (proprietary, sorry) backing this up. Does bearing strength go up faster than net section strength drops with increased diameter? Maybe, depends on lots of things. A
lot of Hart-Smith’s stuff was based on theory not test data, and the data in those old reports are for very old 1970s brittle epoxy resin systems. And the data may also be skewed by lousy test methods, including unclamped pin bearing data. Basically, don’t use it. You need specific joint test data for your materials, layups, thicknesses, fasteners, dismeters, pitches, etc to properly sort it out. And the critical failure modes are a function of the specific joint loading, and the failure modes will change with diameter and pitch and end distance. Basically you cannot make generalizations as can be done for metallic joints.

 

[WKTaylor]

F-MRB... comments, only... no wisdom here... SWC is 'the-expert', here.

It appears that Your primary concern is Gr-E composite structure... hard to drill/match-drill to close-tolerance and high quality under almost any circumstance.

Going from 5D to 3D pitch is a LOT of oversize drilling to eliminate damage... without a scarfed composite repair to rebuild the damage... then restore [drill-back-to] the original hole size.

I got spoiled working F-15s with Be-E composite panels that were step-laminated onto the edge of a titanium 'picture-frame structure' that was then conventionally fastened to spars/end-ribs. No holes in the virgin composites.

Regards, Wil Taylor

o Trust - But Verify!
o We believe to be true what we prefer to be true. [Unknown]
o For those who believe, no proof is required; for those who cannot believe, no proof is possible. [variation,Stuart Chase]
o Unfortunately, in science what You 'believe' is irrelevant. ["Orion", Homebuiltairplanes.com forum]

 

[schopenhauer]

A bit more in depth can be gained from a stress concentration analysis around the hole. I usually always plot the stress distribution around the hole, for that, you can use Lekhnitskii's equations. The maximum stress concentration factor depends on the laminate stacking sequence, for highly orthotropic laminates, the factor can go up to 7.

That is only half the analysis, having a high stress does not necessarily mean that the laminate will break. Basically, you need that high stress to act over a portion of the laminate (there is a characteristic distance a0 for each laminate). There are (to my knowledge) two simple criteria, the average stress criterion and the point criterion. These are simple to use, but you need material data to know how "sensitive" your laminate is to defects (the a0 distance). There are some papers with this information for some materials, but it is usually difficult to get.

So, if you want it cheap, stay to the rule of thumb! If not, you need to perform tests to get to know your material/configuration.

Hope this helps, this topic is well explained in the ESA standards for composites.