stressiterior123
Aerospace
- Apr 26, 2010
- 8
This is very general question, there is a guidline which says 4D pitch requirement but I was wondering how does it affects the joint strength?
Thanks
Thanks
Tek-Tips is the largest IT community on the Internet today!
Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!
-
Congratulations cowski on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!
How pitch affects the joint strength? 5
- Thread starter stressiterior123
- Start date
- Status
-
1
- #2
You want to avoid 'toilet paper' pitch which minimizes material between fasteners. If you do a study of strength of fasteners and base material vs pitch, the 4-6D solution will jump out at you. I used it flange joints.
0.190 D was not used because installers are ham handed enough to break them. 0.250 D meets that test.
0.190 D was not used because installers are ham handed enough to break them. 0.250 D meets that test.
-
1
- #3
my 2c ... there is a small Kt effect and a net section effect if the holes are close together, and there's an inter-rivet buckling concern if they're too far apart. Another consideration is crack growth; under a repair dblr i like to have 1 pitch detectable (hard to argue with that not being detectable) and 1 pitch for repeat inspection, smaller pitches might also make the inspection harder.
4D < pitch < 8D are typical bounds.
4D < pitch < 8D are typical bounds.
-
2
- #4
ESPcomposites
Aerospace
I think a few things could be more clear:
- PITCH is the distance between fasteners (in line with the load).
- WIDTH is the distance between fasteners (perpendicular to the load). This is the distance that plasgears and rb1957 are referring to.
For the WIDTH, the basic failure modes are net section failures and bearing/fastener failures. If the width is too far apart, then you will have high bearing/fastener stresses. If the width is too close, then you will have a net section type failure (i.e the "swiss cheese" that plasgears referred to).
You can run some simple checks for "net section" and bearing and fastener type modes to come up with joint optimization. This depends on the material system and the allowables (i.e. metallic vs. composite). The numbers above are reasonable, but are not written in stone. The other thing to consider is that you might want to have a larger width in the event that there is an error during manufacturing (misdrilled holes, etc.). You may also want to bias it towards a more graceful bearing type failure mode by increasing the width.
For the PITCH, it is not nearly as critical (as should be obvious right now). However, if the joint has multiaxial loading or shear, then it becomes the width. The pitch itself would have a small affect to the fastener distribution, which is usually more important in fatigue of metallics and static strength of composites.
Brian
- PITCH is the distance between fasteners (in line with the load).
- WIDTH is the distance between fasteners (perpendicular to the load). This is the distance that plasgears and rb1957 are referring to.
For the WIDTH, the basic failure modes are net section failures and bearing/fastener failures. If the width is too far apart, then you will have high bearing/fastener stresses. If the width is too close, then you will have a net section type failure (i.e the "swiss cheese" that plasgears referred to).
You can run some simple checks for "net section" and bearing and fastener type modes to come up with joint optimization. This depends on the material system and the allowables (i.e. metallic vs. composite). The numbers above are reasonable, but are not written in stone. The other thing to consider is that you might want to have a larger width in the event that there is an error during manufacturing (misdrilled holes, etc.). You may also want to bias it towards a more graceful bearing type failure mode by increasing the width.
For the PITCH, it is not nearly as critical (as should be obvious right now). However, if the joint has multiaxial loading or shear, then it becomes the width. The pitch itself would have a small affect to the fastener distribution, which is usually more important in fatigue of metallics and static strength of composites.
Brian
ESPcomposites
Aerospace
Minor correction: It looks like rb1957 was referring to pitch by indicating inter-rivet buckling, which would be a real concern there. For the pitch, the Kt usually improves due to "shadowing", but is worse for width (though very minor and hardly noticeable for practical joints).
Brian
Brian
The OP doesn't specifically state that he's asking about rivets. When discussing the Kt effects of the hole spacing its important to differentiate between types of fasteners, specifically hole filling and non-hole filling. Hole filling fasteners (solid rivets, interference Hi-Loks) can be placed closer together due to the smaller size and magnitude of the stress concentration.
An aside, I've never heard the term "width" as ESP is defining it. I can see the benefit of making the differentiation, but haven't seen it done. Is this a common thing?
An aside, I've never heard the term "width" as ESP is defining it. I can see the benefit of making the differentiation, but haven't seen it done. Is this a common thing?
ESPcomposites
Aerospace
YoungTurk,
Have a look at Peterson, Chart 4.32. For an open hole, at 4D, the effect is about 1%. At 3D is at about 2%, which is pretty minor.
Direct comparison of the pin loaded hole is more challenging since the width is very important to the Kt (although there is still little "interaction" amongst Kt's). I am in the process of writing a series of technical papers that is closely related. Aside from the standard data, I have built FEM's, some of which go all the way to use contact.
As far as "width" goes, there may be other terms (spacing, etc.). I wanted to better distinguish the two parameters since they were being mixed together a little bit. I have been on a Boeing program for the last several years and that is what they use. I have not checked to see if Lockheed or others use different terms.
Brian
Have a look at Peterson, Chart 4.32. For an open hole, at 4D, the effect is about 1%. At 3D is at about 2%, which is pretty minor.
Direct comparison of the pin loaded hole is more challenging since the width is very important to the Kt (although there is still little "interaction" amongst Kt's). I am in the process of writing a series of technical papers that is closely related. Aside from the standard data, I have built FEM's, some of which go all the way to use contact.
As far as "width" goes, there may be other terms (spacing, etc.). I wanted to better distinguish the two parameters since they were being mixed together a little bit. I have been on a Boeing program for the last several years and that is what they use. I have not checked to see if Lockheed or others use different terms.
Brian
-
1
- #10
ESPcomposites
Aerospace
rb1957,
Yes, I plan on publishing the results. This is related to a closed form bearing-bypass solution that I developed. Part of the problem has been the discrepancy between the bearing Kt reported from test data (i.e. Peterson) versus that is which is commonly used in analysis programs (BJSFM).
Paper 1: Demonstrate the bearing Kt difference between test, cosine distribution (series solution based), cosine distribution (FEM based), contact solution (FEM based).
Paper 2: Difference between a finite "unit" strip versus the as installed configuration with multiple holes (FEM based)
Paper 3: Closed form bearing-bypass solution that uses any of the selected curves from Paper 1 and combined with the effect of Paper 2.
I originally went down this road as a means to allow analysts to simply solve bearing-bypass problems for my book, though I am not sure I will implement the results (I want stay as "standard" as possible)
Hart-Smith also has a closed form approach which is pretty good, but somewhat dated.
Brian
Yes, I plan on publishing the results. This is related to a closed form bearing-bypass solution that I developed. Part of the problem has been the discrepancy between the bearing Kt reported from test data (i.e. Peterson) versus that is which is commonly used in analysis programs (BJSFM).
Paper 1: Demonstrate the bearing Kt difference between test, cosine distribution (series solution based), cosine distribution (FEM based), contact solution (FEM based).
Paper 2: Difference between a finite "unit" strip versus the as installed configuration with multiple holes (FEM based)
Paper 3: Closed form bearing-bypass solution that uses any of the selected curves from Paper 1 and combined with the effect of Paper 2.
I originally went down this road as a means to allow analysts to simply solve bearing-bypass problems for my book, though I am not sure I will implement the results (I want stay as "standard" as possible)
Hart-Smith also has a closed form approach which is pretty good, but somewhat dated.
Brian
ESPcomposites
Aerospace
"as opposed to using a compliance model with the fastener stiffness of your choice ? "
I suspect you are referring to the distribution of fastener loads in the joint? That is what it seems like.
I think I was less than clear (too close to me). What I was referring to was the local stress field AFTER you have already established the bearing and bypass loads.
Brian
I suspect you are referring to the distribution of fastener loads in the joint? That is what it seems like.
I think I was less than clear (too close to me). What I was referring to was the local stress field AFTER you have already established the bearing and bypass loads.
Brian
- Status
Similar threads
- Question
- Locked
- Question