|
Rivets are a unique animal with a lot of controversy surrounding them both pro and con. Personally, I have taken apart hot riveted joints that were tight as all get out and some where when the rivet head was cut off, the body fell out on its own. One problem lies with the type and quality of the alloy used for the rivet and the other problem is the installation. I've read a lot of different tech articles on this subject and each has it's own slight bias except for one that went into detail on subject having actual testing done on various types of joint using various types of rivets. (Can't recall where I read this but I will keep my eye out for it as I clean up my office). Destructive and non-destructive testing was done on the joints and compared to the same joints done with SMAW and weld joining and bolting. Overall, the welded joints faired far better than the riveted and bolted ones for shear and tensile strength as well as vibration but the bolted joints also did better than the riveted ones too. The riveted joints failed most often because of loose connections perpendicular to the body of the rivet. Reasons stated for the failure was when hot rivets were used, the rivet contracted as they cooled and the shrink allowed for movement in the joint. Each alloy tested a bit different as the shrink rate changes with the alloy but none of them were anything to brag about. A couple of the previous articles suggested that the shrink rate during cooling is what adds strength to the joint but the latter article that did the testing proved this to be false. When cold rivets were used, the failure rate remained about the same for most joints but some failed faster than hot rivet joints. Most of this testing resulted from the rage over the Titanic which used all riveted construction. From the info given, the testing of these joints seemed to back up the theory that most investigators had prior to this that when the ship hit the iceberg, some of the rivets failed from shear stress and the plates peeled open like a zipper. A few years ago, I took down an old bridge that was installed in the mid to late 1930's. The bridge was being replaced because you could see movement at the joints when a vehicle passed over it. Because some of the pieces were cast iron and not steel, the company decided to just replace it and reduce the liability and loss risk. When I started cutting the old bridge apart, I took note to a lot of rust streaks coming from between the joints. Once some of the upper support members were cut free, the entire bridge became very shaky and you could see a lot of movement between the pieces at the riveted joints. As I cut the bridge up, I took note to a lot of joints that showed serious wear around the rivet heads and on the plates that was caused by movement of the plates rubbing together. After seeing this condition, I related to the findings of the test article as fitting the description of this bridge. Catastrophic failure of this bridge was immanent considering the looseness of the joints and amount of wear I saw on the plates themselves. Once the upper supports were removed, the deck began to sag in the middle. Granted the span was only 60 feet and width 12' but it sagged at least 12" in the middle in a matter of minutes and I could hear the rivets popping as the sag increased. It took about an hour for it to collapse under its own weight and as the rest was cut up for scrap, many of the rivet heads were found to be sheared off and the bodies of them very loose in the joints. Having seen this with my own eyes, I'm not too impressed with riveted joints. Any joint will fail no matter if it's bolted, welded or riveted once the load and stress limits of the material have been exceeded but the way this bridge came down so quickly and in the manner it did, it makes you think. The bridge was replaced with a new modular steel bridge. Each section was welded construction and got bolted to the other sections to make the assembly easier and faster. Once assembled, welds accompanied the bolts on every joint between sections. The new bridge has an expasion joint at each end and one in the middle that the old bridge did not have. It does give a little when the haul trucks go over it but nothing like the old one. The main beamd in my shop have riveted flanges but did not show any signs of movement like I found on the bridge. I decided to weld them anyway just to eliminate the possibility of a failure. Personally, I go with weld as my first choice and bolts as the second.
| 
