Resistance Welding

Record the question: when current passes through two touching pieces of metal, where does the heat appear? Predict an answer before measuring.

Measure the resistance along a single 50 mm strip of nickel, then across two strips laid one on the other. The contact reading is higher. That difference is the whole principle.

A handheld meter cannot resolve sub-milliohm contact resistance accurately. Record it as a comparison, not an absolute — the ranking is what matters here.

Set the welder on a non-flammable surface. Put on safety glasses. Molten metal can spit from the joint even at battery-welder power.

Overlap two nickel strips by 5 mm. Weld at the welder's mid setting. Record the pulse duration the machine reports.

Clamp the current meter around one electrode lead and repeat the weld. Record the peak current in amperes.

Make three welds at short, medium and long pulse settings, current unchanged. Label each coupon with the marker.

Repeat three welds pressing lightly, normally and hard. Force changes contact resistance, so it changes the heat even at identical current.

Pull each coupon with the force meter until it fails. Record the peak force and whether the joint tore a plug of metal out or simply peeled.

Plot pull-apart force on the vertical axis, pulse time on the horizontal. Expect a rise, a plateau, then a fall as the joint is burned through.

Immediately after a weld, read the temperature of the joint and of the electrode tips. The joint is far hotter. That is the claim from step 1, confirmed.

Look at a torn joint under the magnifier. A sound weld shows a round recrystallised nugget; a cold one shows only flattened contact marks.

No arc, so no arc-eye — resistance welding is one of the few electric welding processes with no significant ultraviolet output. The real risks are different: expelled molten metal from an overheated joint, which is why safety glasses are worn throughout, and burns from the workpiece, which stays hot long after it stops glowing.

The current is large but the voltage is very low, typically one or two volts across the electrodes, so shock is not the hazard people expect. The hazard is the energy source feeding it: a shorted lithium or lead-acid cell can deliver hundreds of amperes into a dropped tool. Keep the leads tidy and disconnect between sessions.

US 347,140, “Apparatus for Electric Welding”, Elihu Thomson, granted 10 August 1886. The sheet is signed “Elihu Thomson” alone. Figures 3 and 4 show two wire ends before and after upsetting; the claims cover the clamping jaws, the single-turn secondary and the mechanism that maintains pressure as the metal softens.

Thomson found it by accident, and said so. He was demonstrating induction with a transformer whose secondary was a single heavy loop. Shorting it did not blow anything up; it heated the joint white and fused it. The invention is the recognition, not the phenomenon — blacksmiths had been fusing metal by heat and pressure for three thousand years, and forge welding is the same physics with a fire instead of a current.

The three numbers. Everything in resistance welding reduces to current, time and electrode force. Joule heating gives Q = I²Rt, so heat rises with the square of current but only linearly with time — which is why the process is fast and why a small current increase matters far more than a long pulse. Force is the counter-intuitive one: pressing harder lowers contact resistance and therefore lowers heat, so more force is not more weld.

Where it went. Thomson's company merged into General Electric in 1892. Spot welding reached the motor industry in the 1930s and is now the dominant joining process in car manufacture by sheer count — a typical unibody carries between three and five thousand spot welds. Seam welding, projection welding and flash butt welding are all the same patent's physics with the geometry changed.

What it cannot do. Both sides must be reachable by an electrode, the sheets must be thin enough for the current to cross before the surroundings heat up, and highly conductive metals such as aluminium and copper are difficult precisely because they do not resist enough. That last limit is why the aircraft industry went to blind rivets instead — a problem the blind rivet solved in 1916.