Riveted joint strength...

[dsb]

Whilst researching riveted joint strength I stumbled across this piece...

http://www.rvplane.com/pdf/RivetLength.pdf

Which I found informative. However, I'm still curious as to the relative strength of joints made with small rivets and fine pitch as compared to larger rivets and a coarse pitch...

I also find it curious that no one seems to be using structural adhesive... Am I missing something?

Thx,
Dave

 

[orion]

Good article - thanks. And the nice thing is that the results were pretty much as predicted by theory and engineering practices.

Regarding bonding, you might use the search function herein and look for topics on adhesives, aluminum bonding, etc. We've discussed this numerous times so there a bit here already. And of course if you have any further questions, don't hesitate to ask.

The bottom line though is that so far there doesn't seem to be an adhesive system that works reliably well with aluminum without having the builder go through some involved procedures in preparing the material's surface. The ideal bonding material would work with only minimal prep and degreasing - so far there have been a few identified (epoxy and methalcrylate - and one very knowledgeable gentleman here is a proponent of urethane adhesives) but the tests I've done so far only show significant promise but nothing yet that I'd hang my life on.

 

[dsb]

Yes, I read a couple of threads on here about adhesives...

I don't think I want to set up for PAA in my garage...

I've been looking at Aeropoxy ES6228:
http://www.ptm-w.com/dynamicdata/data/docs/aeropoxy es6228 bulletin.pdf

Given that the .032 failed at 2300psi, it seems that it wouldn't take a perfect bond to be as strong as the bondant...

 

[orion]

The Aeropoxy material does look promising but really, it's not shear you're worried about. It'll be more than strong enough for most airframe applications. But the problems tend to be peel, susceptibility to environmental degradation and sensitivity to vibration (which goes back to peel). These bonds are very notch sensitive and even a small amount of peel onset could potentially result in catastrophic debond.

To see how applicable the material is, bond some test coupons in a configuration that will allow you to peel them apart. Let them cure for a sufficient amount of time (maybe two months or so) or aid the full cure with a few hours in a low temp oven. Then grab the ends and peel the material apart.

Ideally what will fail is the epoxy so that the surface of both coupons will have a coating of the epoxy. Some bare spots are OK. But if the peel occurred in such a way that one of the coupons has little or no epoxy on it, then you'll know that the material is poor in peel and will require a choice of a different product or the installation of numerous "chicken" rivets.

 

[Captain_John]

Whilst researching riveted joint strength I stumbled across this piece...

http://www.rvplane.com/pdf/RivetLength.pdf

Which I found informative. However, I'm still curious as to the relative strength of joints made with small rivets and fine pitch as compared to larger rivets and a coarse pitch...

I also find it curious that no one seems to be using structural adhesive... Am I missing something?

Thx,
Dave

Dave, I aren't a en-ga-near but... a bunch of small rivets in fine pitch makes an incredibly strong joint. I know that we are speaking in relative terms here but the AN426-3-X rivets we use in our RV's are nothing but pure magic!


A row of them and some perpendicular structure is stronger than a virgin sheet!

What exactly are you planning? I think that you are better off with more small rivets than fewer larger ones because the stress will be spread out more evenly across the span.

CJ

 

[dsb]

The Aeropoxy material does look promising but really, it's not shear you're worried about. It'll be more than strong enough for most airframe applications. But the problems tend to be peel, susceptibility to environmental degradation and sensitivity to vibration (which goes back to peel). These bonds are very notch sensitive and even a small amount of peel onset could potentially result in catastrophic debond.

To see how applicable the material is, bond some test coupons in a configuration that will allow you to peel them apart. Let them cure for a sufficient amount of time (maybe two months or so) or aid the full cure with a few hours in a low temp oven. Then grab the ends and peel the material apart.

Ideally what will fail is the epoxy so that the surface of both coupons will have a coating of the epoxy. Some bare spots are OK. But if the peel occurred in such a way that one of the coupons has little or no epoxy on it, then you'll know that the material is poor in peel and will require a choice of a different product or the installation of numerous "chicken" rivets.

Funny, but way back when, building AL tub race cars, we always used adhesive on the rivet joints to keep the vibration from rattling the rivets out. I understand what you are saying about peel, hopefully I can model that appropriately.

Cpt.John, I are a fizz-a-sist... Trying to understand and model the stresses and strains in an aircraft, particularly sheet metal aircraft, especially in aerobatic mode. I have the preview plans for an RV-4, to help develop the model, the rest has been research to this point.

Speaking of research, I was directed to this study:
http://www.agingaircraftconference.org/all_files/22/22b/146_doc.pdf

Not particularly enlightening in and of itself, but some of the references are interesting...

 

[Captain_John]

Cool, I are an ed-yu-cator... currently.

My background is Electrical Construction and Maintenance. I also have a background in technological sales. It is varied, but I consider myself a student of life.

Anyways, is your goal to design your own plane? I don't see many (there are exceptions like the Tiger and the Citation Jet) that have bonded skins.

Personally, if it is not going to be pressurized I maintain that there is no need to use any adhesives. A good row of structural rivets has been the standard for years.

Are you considering the -4? It would make a fine choice, IMHO.

Getting parts more or less prefabbed is hard to beat!

CJ

 

[dsb]

"A good row of structural rivets has been the standard for years."

True, but a riveted joint will always be a percentage of the strength of the material being riveted. If you want/need a stronger joint you have to use stronger material. Using adhesives or a combination of rivets and adhesives you should be able to obtain a larger percentage of the joined materials strength in the joint. It would seem that adhesives offer the opportunity to exceed the joined material's strength in certain aspects (which poses problem's as well...).

" Are you considering the -4? It would make a fine choice, IMHO."

If I won the lottery I'd buy an Extra... I have Pitts plans, but I'm not willing to commit to a single seat plane. I've talked to the Laser people about their 2 place. It's a tube fuselage with wood wings, as are the Staudacher, Panzl and Rhin designs, and all well north of the $100K mark. Given that the Sukuoi's and Zlin's are or have been sheet metal designs, I thought I would see what I could come up with. So, since the RV-4 has been used in IAC competition and is 2 place I thought I would start there. The main problems are the low Va, which is a result of the low Vs and the modest load rating, and the low Vne, which is apparently flutter limited. At this point the RV-4 is just a point of entry, a known quantity to run analysis on. It may turn out that the -4 needs very little, and the modifications relatively minor... Or not...

Please don't think I'm knocking the RV-4, or think that I know more than R.V., it's just that you can't fly it like an Extra in it's present form.

" Getting parts more or less prefabbed is hard to beat!"

Agreed, provided they are the 'right' parts...

 

[Othman]

The main parameters in designing a riveted joint are rivet type (form, material and diameter), sheets/parts being joined (material and thickness), rivet pitch and row spacing (if multiple rows are required).

If we consider a long lap joint, we need to size the rivets and pitch appropriately to carry the loads that have be transferred from one sheet to the other. This would be considered as a running load (lbs/in).

The ultimate (failure) running load carrying capability of a single row of rivets can be calculated by the rivet allowable load divided by the pitch.

W_allow (lbs/in) = rivet allowable (lbs) / pitch (in)

The rivet allowable load depends on the rivet type, material and diameter, and the sheet material and thickness. There are published joint tables that provide rivet allowables for many combinations of joints. The industry standard is the document titled Metallic Materials Properties Development and Standardization (MMPDS), which replaced MIL-HDBK-5. The first release, MMPDS-01 used to be available for free (someone may be able to provide that for you... 75 MB PDF), but now the latest version is for sale only (-03).

Riveted joints may be shear limited or bearing limited. In the case where small rivets are used in thick sheets, the rivets will fail in shear before the sheets will fail in bearing. On the other extreme, for a joint with large rivets and thin sheets, the sheets will fail in bearing or fastener tear-out before the rivet fails.

In a well designed joint, bearing failure should occur first (since it’s not a catastrophic failure), but without using rivets that are too large for the joint (this is better seen when looking at the joint allowable tables that are attached).

The attached joint table is for MS20426AD rivets (flush head) in dimpled 2024.

Some rules of thumb (not always true):

• Rivet diameter = 3 x material thickness of thicker sheet (but should not use less than D= 0.125”, or -4 rivet for primary structure).

• Rivet Pitch Minimum ~ 3D - 4D, Maximum 8 D (- 13D). Depending on the load in structure, the joint may experience inter-rivet buckling if the pitch is too high, so try and keep to the lower end of the Max Pitch.

• Refer to AC43.13-1b for some generic guidance.

Riveted joints are always weaker than the original material. They may be stiffer, but never stronger. If the rivets don’t fail you will end up tearing the thinnest sheet right along the rivet line.