The sample showing the "phone book" laminations on the strap material, what caused the laminations? Is it just poor quality wrought iron?

Good question. A little insight as to why I went to the trouble. I was mostly interested in the mode of failure rather than the hard numbers. The joint should fail (in a perfect world) with three rivets in single shear. In the test the failure was two rivets in single shear and tearing of the welt strap. I was some surprised but then if I were sure of the failure mode I wouldn't have gone to the effort.

In the test the first two rivets sheared at about the same time as the welt tore. So dividing the up the resultant forces was problematic. I did however break a sample of welt strap by itself. It broke at 22,000 +/_ psi. It wasn't elastic at all, the curve was mostly straight with out a visible yield point. The curve looked like cast iron. It also came up so quickly that by the time it started to pull it was over. Strain rate set a little high.

Anyway the joint was expected to break at about 70,000 lbs.(f) but failed instead at 58,200 lbs.(f).

Hope this answers your question. It was very instructive for me to do this.

The laminations are a result of poor material, probably dirty when faggoted with less then good chemistry in the furnace used. Among other things this allowed infiltration of moisture into the iron and corrosion. An etching and photomicrograph would yield exciting results were one to undertake such an effort.

A little more history. When riveted joints are calculated, a properly designed joint will fail in the tearing between rivet hole in the outermost row. So when people calculate the strength and efficiency of riveted joints it is common to just do the calculation for that failure as a short cut. When I came across this joint I had never seen it before and it wasn't covered in the usual books. I hunted around and found the joint and the list of failures. In so far as I like math I did all the calculations: tearing between outer row of rivets, tearing between inner row of rivets and shearing outer row of rivets, shearing of three rivets (single shear), crushing in front of three rivets, and tearing at the inner row of rivets and crushing in front of one rivet.

The three rivets in single shear are the low value (at least inn the case I was dealing with). I then reviewed the previous calculations done through out the life of the boiler and found that the convention of using tearing in the outer row had been used exclusively. The assumed factor of safety of 5.2 for the longitudinal seam was in reality about 3.1. This boiler was a bomb. To compound matters finding that the strength of the welt strap was well below the assumed value of 45,000 psi (NBIC C-4010) was chilling.

So I cut out the seam and the rest is history..

Thank you, this has been informative and educational.

What were your test seams from?

Sounds like a use what you can repair.

Its sounds like with proper materials you could do the double seam.

The sample is from the first course of an 1888 Cooke locomotive. The locomotive is Nevada State Railroad Museum's #8. The repair was made in 1991. I have been sitting on this sample for years. The seams were removed from the barrel and repairs made with code material and appropriate design and engineering. It is instructive to note that the original workmanship of the Cooke boiler is poor.




The rivet holes have clearly been punched without follow up reaming prior to driving. Baldwin and Porter made much better boilers. The #8 is currently out of service for repairs. Yet another story.

(it is necessary to disclose that the rivet/sheet interface has been enhanced to more clearly show the fit-up)