Stress Concentration - Figure 8 Holes

[graemew]

My problem concerns an upper wing skin afflicted by a series of .03" deep x.09" dia "drill starts" near to/ adjacent to, a series of .128" dia rivet holes in .080" th 2014T6. I believe that it was caused by the misplacement of a production drilling fixture (by the OEM many moons ago during construction). The critical loads for stress are compressive, with tensile load cases being roughly 50% of the comp. I hold out no hope of obtaining the original production concession document from the OEM since the type is out of production.
Many of the "starts" are spanwise from the adjacent hole but some are at other angles to spanwise and are close/overlapping.
I have been investigating the problem using ESDU data sheet 75007 which provides data for adjacent holes at 0, 90 and 45 degrees, but it becomes invalid for the figure 8 cases.
Some years ago I saw a report dealing with this issue which dealt specifically with various figure 8 holes at various angles to the load direction. I believe that it may have been in association with some work done by BAe on a fatigue program on the Buccaneer.
Does anyone know of any source material that would be of use?
I am trying to avoid doing a significant repair to the area because it has major fittings nearby. In any event, because the prevalent load is compressive, I need only to square off the fatigue aspects associated with the lower tensile loads in order to resolve my problem.
If figure 8 data is not available, I will probably spotface the surface and try to justify the loss of strength of the skin in compression. Not a path that i find desirable in the circumstance, since I believe that it will be OK as is.
Any thoughts?

 

[jetmaker]

graemew,

Out of curiosity, how do you propose on repairing this even if you can show the presence to be satisfactory? Do you just plan on leaving the drill starts "as-is"?

Most SRM repairs allow 10% blend of wing skins, and since you are only 0.03" deep in a 0.080" thick skin, this kinda falls within that requirement. Also, is the 0.128" hole size the final size? This seems like a very small rivet for a structural application. If it is not the final hole size, can you pull the hole when opening it up to final size to remove the drill start?

Thanks.

jetmaker

 

[graemew]

thanks jetmaker
I have had the drill starts radiused to remove the sharp corners and to clean-up any corrosion at the base of the holes.
The .128 holes attach secondary nacelle structure. To clean up the "drill starts" I would need 5/16 or 3/8" fasteners. Something of an overkill I feel.
I have contemplated freeze-plugging, and it is still a live possibility, but it will cause significant disruption to the work-flow and I would prefer not to go that way if I can avoid it. The spot-facing option is more preferable and I have stressing data that I can use to support that path. The best path is still the do nothing path. It has been like this for 35 years and is still OK, it was fine until it was discovered (not that I adhere to the "ignorance is bliss" school of thought ) and it seems reasonable that it will probably continue to be OK. I need to wrap it up with a technical review however.
The fact that it is mostly seeing compressive load looms large in my mind and I feel guilty in not being able to answer the question.
Do you have any suggestions?

 

[737eng]

Don't know if this will help but-
Considering Gross stress and not net stress. An open hole in an infinite plate under tensile loading has a stress concentration factor Kt of 3.0. If the figure 8 is perpendicular to the load path, this would considerably increase your stress concentration to about a Kt of 4.0, however, if your figure 8 is parallel to your load path you will actually decrease the stress concentration to about Kt of 2.5. This data comes from an Airbus Structure training manual. i believe I have a Boeing training manual that states basically the same (Fig 8 perpendicular to load increases Kt, Parallel to load decreases Kt. It's the same as watching water in a river flow around rocks!!

 

[rerig]

Jetmaker,Graemew, small problem with the decimal. 0.03 isn't < 10% of 0.08. (almost 40% material depth is more than a drill start)

Graemew, stepping up from 0.125&quot; to 0.375&quot; is quite drastic for any fastener increase. Normally you'll have an EM problem.

Suggestion, calculate as if you were spotfacing the area. This would show conservative numbers for material removal. Then do nothing, having already radiused the &quot;drill start&quot;.
You would have a non-uniform stress around the hole, but this would have the least impact on overall structural stresses.(ie. 737eng) And as you have noted, it has been like this for many years without the damage enlarging. (The basis for damage tolerance repairs) Dont forget to apply sealant to the dimple area.

Rerig

 

[jetmaker]

rerig...

Thanks for the catch....

As for the possible repair, I would suggest taking an insurance cut on the drill start to remove a crack that may exist but is below the limits of NDI detectability. After ensuring a smooth repair finish (63RA or better), shot peen the reworked area and the adjacent hole.

To justify this repair, the smoother finish, in combination with the shot peening, will offset any material removed from the cleanup operation. Therefore, the fatigue detail is no worse than what you had, and therefore should expect an additional number of cycles equivalent to the current number of cycles.

jetmaker

 

[AeroE]

If the holes are not drilled through, don't do anything but a little clean up and polish of any burrs or sharp edges. The reason the part has lasted for so long is because there are no holes and the stress concentrations are nearly non-existent, maybe Kt of 1.1 or 1.2; i.e., the stress field is essentially uninterupted by holes!

 

[graemew]

Thanks for all the input, AeroE, jetmaker, 737Eng and Rerig.
My customer demands a level of rigour that demands substantiation of the final configuration. I have settled upon spotfacing the affected areas and using the ESDU stress concentration data to help satisfy the requirements in a fatigue analysis.
AeroE; I feel about the same as you do about it, but for the life of me I could not find any stress concentration data to deal with hemispherical recesses adjacent to holes and so I end up spot facing so that I end up with a step if a defined shape, far enough from the hole so that I do not need to concern myself with interactions.
Regards