Structural difference between engineering adhesives and steel rivets

[Enginkrb]

Hello all,

I'm coming to eng-tips because from what I've experienced, this forum is a trendsetter and trustworthy voice in the engineering community.

What I want to gather is how professional engineers feel about structural adhesives versus steel riveting in any structural application from past experiences. (ie: ANCA - ANaerobics for machinery adhesives and CA - cyanoacrylates or instant adhesives)
Also, I want to understand is what “in-house” resources do managers provide to designers to remain cutting edge? Or, where do designers and managers go for new design methods or applications?

I'm gathering this information as a learning experience for a potential customer and would like to share professional insight with them. Any information would help out here!

Thanks a lot all and I look forward to hearing from this community,

-Steve

 

[Tmoose]

You did not mention if the components being assembled are metal or something else. The Composite Materials Handbook Series, MIL-HDBK-17 requires the bonding of secondary structures (not the original composite fabrication) to also include some mechanical fastening.

From here in my armchair I view structural bonding as mostly a significant enhancement to mechanical fastening. I don't think I'd want to zoom down the road or thru the woods relying on structural bonding to support my carcass.

http://lh5.ggpht.com/_JVNmHcqXh9k/S...9uKs/s400/IMG_1766.jpg?source=mog&hl=en&gl=us

 

[dhengr]

Enginkrb:
We have some long term history with many mechanical fasteners so we know what to expect out of them, when designed and installed properly. Adhesives, on the other hand, have a fairly short history for real structural bonding. There are thousands of minor variations on the theme with adhesives, minor variations in chemical make-up, etc. and they are susceptible to degradation by UV light and other environmental contaminants, chemicals in the air and rain water, environmental aging, etc. Then, rivets, bolts and the like, transfer the total load at several discrete locations; while adhesives transfer the same total load over a distributed area at a much lower unit stress. Adhesives usually work well when loaded in tension or shear, but you must guard against pealing or ripping (a highly concentrated tensile loading) at bonded edges. Thus, mechanical fasteners are often used for initial clamping, during bond curing and then to protect the edges or load transfer at high stress points. The two systems work differently too. Many adhesives creep over time, some get quite brittle over time, but they start picking up load almost immediately through shear in the bond line/plane. While often times bolts must allow the joint to move a bit to bring them into bearing in the bolt holes. Slip critical bolted joints don’t allow much joint movement to take their loadings and many rivets tend to fill the oversized holes during their driving, so they don’t have much movement to bring the rivets into bearing. The upshot of this is that the mechanical fasteners and the adhesive fastening system do not really (or often) share the total loading, at the same time, or in a well defined proportion. Each different application must be studied and designed.

As an aside.., ‘trendsetter and trustworthy’ are two completely different things in the engineering business. Yes, engineers and their opinions are expected to be ‘trustworthy’ in their dealings and designs, and in their effort to protect the public and their clients, the whole “health, safety and welfare of the public” thing. ‘Trendsetting’ has a slightly funny/bad connotation in my mind, as I read it today. It certainly shouldn’t mean majority opinion rules or democratic vote controls all the best solutions. A good engineering education, good mentoring during much of your career and then experience and good sound engineering judgement are far more important, in doing good engineering work, in my mind. If I were to rewrite your first sentence, as regards Eng-Tips it might go something like this ‘a very good source of world wide engineering info. and knowledge, and generally trustworthy.’ It is really a good place for mentoring amongst experienced and co-equal engineering professionals; and a very good place for young engineers to learn. You should still vet what you read here, there are no guarantees here, and there are some pretty zany opinions and ideas floatin around. You do get to know the members who you can trust for meaningful answers and opinions, based on the preponderance of their other posts.

 

[blakmax]

Enginkrb

Adhesive bonding is in many ways far superior to riveting a joint. For a start, drilling holes in a sheet of metal (or composite) will always result in high stress concentrations which can result in fatigue or failure at the joint. In an infinitely wide joint, there is no stress concentration for an adhesive bond. In the case of a localised joint such as a repair in the centre of a sheet, there is a stress concentration associated with load attraction which may be between 18% to 28% depending on the comparative stiffness and shape of the repair being bonded. In comparison the stress concentration for a mechanical joint is of the order of 300% increase in local stress at the fastener.

Adhesive bonds are always considerably stiffer than mechanical joints. Mechanical joints are more compliant because of fastener bearing in the sheet, fastener shear bearing, fastener bending and fastener rotation. This adds up to produce considerable displacement in the joint. In an adhesive bond all load is transferred very close to the end of the joint. (It is NOT as many people think evenly distributed over the bond.) As a consequence of this difference, for repair of cracks the defect must always be cut out prior to applying a mechanical repair. In a bonded repair, you can leave the crack alone and in fact cutting it out or "stop" drilling result in a shorter fatigue life than leaving the crack alone.

Another factor to consider is that for thin adherends it may be possible to design the joint such that the adhesive is actually stronger than the metal. This is true for aluminium alloys up to about 0.14 inches thick. In other words, the metal breaks, not the adhesive.

The down side with adhesive bonding is that it is process sensitive. The importance of correct execution of processes can not be overstated. There are many misconceptions about preparing to bond. Many people believe that the surface must be roughened to enable the adhesive to key into it. That is not true. Adhesive bonds depend upon chemical bonds formed at the interface. These chemical bonds establish the strength of the bond and also have a direct effect on the long-term survival of the bond. It is well known that bonding surfaces must be clean and this is to remove any contaminants which would inhibit the formation of chemical bonds. The surface must be clean, but must also be chemically active to enable bonds to form. The abrasion process actually removes the surface oxides exposing a fresh chemically active surface; it is not to roughen the surface. Also, the cleaning process must be the first process step to prevent contaminants being ground into the surface. It is important that the surface is not “cleaned” after abrasion because that transfers contaminants over the surface. Also important, once the surface is clean and chemically active it is essential that the surface be bonded as soon as possible. Delaying the process will result in formation of thick oxide layers which will weaken the interface.
Now the longer-term survival of the bond also depends on the durability of the interface. For metals, the most common form of bond failure is caused by degradation of the interfacial bonds and typically this is due to hydration of the oxides formed just prior to cure of the adhesive. Typically Al2O3 will hydrate to Al2O3.2H2O. For this to happen the chemical bonds between the adhesive and the surface dissociate, thus leading to disbonding. If this is to be prevented, then the surface must be treated to prevent hydration AT THE TIME THE SURFACE IS FIRST TREATED.

Regards

Blakmax