Coated Electrode

Write it down: molten steel dissolves oxygen and nitrogen from the air, and a solidified weld cannot spit them back out. The coating's job is to keep air away.

Cut three 40 mm coupons from the steel flat bar. Sand all three bright with 220 grit and weigh each to 0.01 g.

Mix 10 g borax powder with water, a few drops at a time, to a thick cream. It should coat a rod without running off.

Paint coupon A with the paste and let it dry fully. Leave coupon B bare. Coupon C is the untouched control — it is never heated.

Heat coupon B with the propane torch until it glows dull red. Hold for 60 seconds. Let it cool on the firebrick.

Repeat exactly with coupon A: same flame, same colour, same 60 seconds. Only the coating differs.

Look at both. The bare coupon carries loose grey-black scale; the coated one is covered by a glassy layer that has excluded the air.

Wire-brush or tap the loose scale off both coupons and weigh each again. Record the change.

Subtract each final mass from its start mass. The bare coupon has lost measurably more steel to oxidation than the coated one.

Once fully cold, tap the coated coupon's glassy layer. It detaches cleanly — the same self-releasing behaviour a welder relies on when chipping slag.

Record them: shield with gas, cap with slag, stabilise the arc with easily ionised salts, and alloy the weld with powdered additions. Borax does only the first two.

Write down honestly what this experiment does not reproduce: a real electrode's coating burns with the arc at 3000 °C. Yours is a static oxidation test.

Steel stops glowing long before it stops burning you. A dull-red coupon is around 600 °C; at 400 °C it looks like cold steel and will still cause an immediate full-thickness burn. Move hot coupons with tools, never fingers, and check with the infrared thermometer before handling.

Borax is a flux, not a food. Sodium borate is classified in the EU as toxic for reproduction and is restricted in consumer products. Handling small quantities of the powder with clean dry hands and washing afterwards is the standard precaution; do not create dust, and keep it away from food preparation areas and young children.

A real point about real electrodes. Welding electrode coatings absorb moisture from the air, and that moisture becomes hydrogen in the weld, which causes delayed cracking days after the job. Low-hydrogen electrodes are stored in heated ovens for exactly this reason. It is the clearest example in welding of a defect that appears long after the welder has gone home.

Swedish patent SE 27,152, dated 29 June 1907, Oscar Kjellberg. Swedish patents of this era are not carried by the public patent-image databases, so unlike the contemporary American patents there is no scanned sheet to check the claim text against. The number and date are consistent across the documentary record, and that is stated rather than dressed up as a primary source.

Kjellberg was a shipyard man, which is why he cared. He worked at a Gothenburg shipyard where arc-welded repairs kept failing, founded ESAB in 1904, and spent three years on the coating. The first electrodes were dipped by hand and dried on racks.

What the coating actually does, in order. As the arc approaches, the coating decomposes and releases carbon dioxide and other gases that physically displace the air from around the pool. The mineral remainder melts to a slag less dense than steel, so it floats to the top and caps the solidifying weld against oxidation and against cooling too fast. Add potassium or sodium salts and the arc becomes easier to strike and far more stable; add iron powder and the deposition rate rises; add chromium and nickel and the weld metal can be made stainless. One consumable, four jobs.

The consequence was the welded ship. Riveted hulls needed overlapping plates and hundreds of thousands of rivets, each a hole in the plate. Welded hulls are butt joints: lighter, cheaper, faster and watertight without caulking. The first fully welded seagoing vessels appeared in the 1920s, and by the Second World War the Liberty ship programme was building 2,700 welded hulls. Those ships also taught the industry about brittle fracture the hard way — a welded hull is one continuous piece of metal, so a crack has nothing to stop it, which riveted seams had quietly been doing all along.