Fastener Load Fraction for Stacked Repair

[northeast1976]

Hi All,

I am trying to determine the nominal stress of the aircraft skin under the tripler given the following example.

Example:
- Doubler has 4 rows of rivets. These 4 rows of rivets goes through doubler and aircraft skin.
- Tripler has 2 rows of rivets. These 2 rows of rivets goes through all layers (aircraft skin, doubler, and tripler)

Based on the Table 17.6 (attached: Book Structural Shear Joints by George Hahn, p224), if the fraction of the nominal stress in the aircraft skin just after the 4th rivet row of the doubler is

0.63 = 1 - (Row 1 + Row 2 + Row 3 + Row 4 ) = 1 - (0.19 + 0.06 + 0.06 + 0.06),

then does this mean that the fraction of the nominal stress in the aircraft skin just after the 6 rivet row of the doubler is

0.4725 = [1 - (Row 1 + Row 2 + Row 3 + Row 4)] * [1 - (Row 5 + Row 6)] = 0.63 * [1 - (0.19 + 0.06)]

I would like to know if this principle can be extrapolated to apply for a tripler.

Thank you for your expertise.

 

[rb1957]

your table is headed "derived from 2 spring model with previously reported values". i'd exercise caution in applying this unless your design is similar. in any case i'd interpret the "4 row" as being 2 rows either side of CL, i think your case has 4 (or 6) rows either side of CL. i don't think you can extrapolate the table the way you have. i really don't think the way you account for the 5th and 6th rows is right ... the stress in the skin drops 20% due to 2 rows of rivets into the tripler ?

it is easy to build a complince model, refer Niu "airframe structural design" pp234, of a joint and get the "right" distribution. although going for 6 rows is going to be a bit of a pain.

why interested in the tripler fasteners ? ... load transfer peaks at the edges (unless you've been very careful with the stack-ups) so the outer rows are critical. oh, ok, to get the stress in the skin. one approximateion is to say "fully effective" so the stress in the skin on the CL is tsk/(tsk+tdblr+ttplr)* applied skin stress. this is probably ok for a static calc but is probably unconservative for F/DT. IMHO i think the tripler does little to reduce skin stress, and is mostly there to make inspection of the cut-out edge difficult; if you Have to have one it's better on the IML. I usually account for load transfer at the 1st two rows and that's enough; even then i'd typically analyze the skin under the dblr with the undoubled skin stress (ie no doubler) which is horribly conservative. i have a two row compliance model which gives me a "real" stress on the CL. Normally the load transfer at the edge creates the crtical inspection program.

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