Reading through locomotive accident reports on Google books from around 1920, I ran across lots of references to problems with autogenous welds on boilers.

Being a non-welder, I have a few questions: How was autogenous welding done back in that era compared with how it is done today? Is it acceptable to do autogenous welding on boilers today?

Some excerpts from the reports follow.

Example of weld failure on the SP in Tucson:


Example of weld failure on the C&A:


Opinion of report author (1923):


Source:
Annual Report of the Director, Bureau of Locomotive Inspection, Issues 1-13

__________
Footnote: Regarding the Chicago & Alton fatalities noted above, I looked at Shirley, IL, on Google maps and noticed a tiny cemetery along the C&A mainline just outside of town. I suppose there's a chance the remains of those three trainmen killed are buried there.

I don't know the correct answer, but I can tell you I've seen some rough looking stuff from the 1940's. It was basic "stick welding" i.e. electric welding.


It is standard practice. All modern commercial boilers are welded, and the construction and repair of locomotive boilers includes welding.

(comment deleted by the author.)

In the 1920s, weld strength, ductility, and reliability were godawful by modern standards. The Master Boilermakers' Association (a railroad industry group) published annual convention proceedings annually. The proceedings are verbatim transcriptions of the official presentations and subsequent discussions. There was a lot of discussion about locomotive boiler welded repairs over the years. In the 1920s, welds had to be at least 25% as strong as the base metal... on the official welder qualification test. In the field, under typical sub-optimal conditions (i.e. the foreman barking at you, hot and tired, difficult access to the weld area, etc) you can bet the welds were <20% of the base metal strength. That's >5X weaker than a modern weld.

Also the crack-detection methods were crude at best. So often the entire crack wasn't even detected. Modern methods (dye penetrant test, for example) are far more capable of definitively locating cracks than the old slosh-it-with-black-oil-and-kerosene, wipe clean, then dust-it-with-whiting-and-hit-with-a-brass-hammer test.

So the ICC had very good reasons to discourage welding as a safe repair technique, unless it was used only to help seal the edge of a riveted joint. That way you might develop a small leak if the weld failed, but the rivets would hold everything together.

Just before WW2, the U.S. Navy switched from riveted submarine hulls (dangerous when depth-charged - the rivets tend to break under shock and become projectiles inside the hull) to welded sub hulls. Ditto for merchant ship construction. The WW2-emergency Liberty Ships had welded hulls, but they had a lot of trouble at first with steel quality, cracking, and strain relief. At first designers didn't realize how much stress could build up in a huge welded assembly from the hot welds cooling and contracting, and they failed to provide stress relief structures for this. The earlier Liberty ships were nicknamed "Kaiser coffins" for a grim reason. Similarly, locomotive boiler welds only start to become trustworthy starting in the late 1930s under unusually well-controlled conditions. General adoption of recognizably "modern" welding techniques, training, and equipment is generally a post-WW2 phenomenon.

- Doug Debs

Autogenous welding is NOT stick welding, at all. It is defined as a welding process using no filler metal. It's fairly common now in tig welding, 'back then' I am thinking perhaps it is done with a carbon-arc torch.

I believe autogenous welding is more commonly known as forge welding.

Forge welding, carbon arc, oxy-fuel, and TIG are or can all be autogenous weld methods.

I would imagine that the welds in this case would have been performed using the oxy-fuel process.

Greetings:
Brendan is correct.
A major factor to remember is that in the early days of electric welding, the electrodes were bare wire, that is "uncoated". Later, the electrodes were coated with a mineral compound which acts as a flux, allowing the operator to easily strike and hold an arc and protecting the freshly deposited weld metal while cooling. This type of electrode is used to this day.
I have been told that the railroad industry was slow to adopt coated electrodes due to cost.
J.David

Yesterday I found a 1921 article about welding on boilers in a publication called "Welding Engineering". It said this:

" By 'autogenous welding' is meant any form of welding by fusion, that is, where the metal of the parts to be joined or added metal used for the purpose is melted and flowed together to form the weld. Such welding is accomplished by the oxy-acetylene, hydrogen or other flame processes, or by the electric arc; no distinction is made between any of these processes. Autogenous welding may be used in boilers in cases where the strain is carried by other construction... "