What Actually Sets A Spot Welder Output: Transformer, Arm And Secondary Loop

Two spot welders with the same number on the nameplate can behave like different machines. One welds 1.5 mm galvanised sheet at a comfortable setting; the other needs everything it has and still marks the surface. Buyers usually compare the kVA figure, and the figure is real, but it describes the transformer at its own terminals, not the current that arrives under the electrode cap. Between those two points sits an arm of copper, a set of bolted joints and a loop of busbar, and on a long-throat machine that path can take a visible share of the output.

This article is about the three parts we design and build ourselves, and about the questions worth asking before a machine is specified. We have lost a tender on a nameplate comparison and won the follow-up order once the customer put a current meter on both machines, so the subject is not academic to us.

A spot welder is rated in kVA at a stated duty cycle, usually fifty per cent, and that figure is a thermal statement about the transformer. It says how much apparent power the transformer can pass for half of every minute without exceeding its insulation class. It does not say what the machine will deliver at the tip, and it does not survive a comparison between an AC machine and an inverter machine without translation.

A conventional AC welder works at mains frequency, which means a large laminated core to carry the flux. A medium-frequency inverter rectifies the mains, switches it at around a thousand hertz and passes the result through a transformer that is a fraction of the size and weight for the same output. The current is then rectified again at the secondary, so what reaches the electrode is direct current with a fast rise.

The practical consequences are worth knowing even if the electronics are not. Inverter output is far easier to hold constant as the load changes, the three-phase input is balanced, and the same kVA buys more useful current because the waveform is not spending part of every half cycle crossing zero. What it does not buy is immunity to a long secondary path. That part is pure copper engineering, and it is the same on both technologies.

Every millimetre of copper between the transformer and the cap carries the full welding current. Resistance in the secondary loop is small in absolute terms and large in proportion, because the current is measured in tens of thousands of amperes. Doubling the length of the arm roughly doubles its resistance, and the voltage drop along it is current multiplied by that resistance. Part of the kVA the transformer produces is therefore spent heating the arm rather than welding the part.

This is why a long-throat machine and a short-throat machine are not the same machine with a different reach. If a panel needs a metre of throat, the arm has to be built with a larger cross-section than the short version to keep the loss in proportion, and it has to be water cooled, because copper that carries this current heats up whether or not the weld is being made.

There is a mechanical side to the same problem. Electrode force on a standing machine is several kilonewtons, and a long arm deflects under it. The force that reaches the tip is therefore lower than the force set at the regulator, and the deflection changes the geometry of the joint. We measure force with a load cell at the electrode tip on every machine before it ships, not at the air regulator, because on the long arms the difference is not a rounding error.

The photograph below is the loop on one of our extended-throat machines. A solid copper busbar runs from the transformer to the lower electrode holder, and the connection is bolted and shimmed to set the alignment. At the moving end the current has to cross a gap that opens and closes, so flexible copper braid takes the movement while the busbar stays rigid.

The bolted joints are the part that fails quietly. Two copper faces bolted together have a contact resistance that depends on cleanliness, flatness and torque, and that resistance rises as the joint heat cycles. A joint that measures well when the machine is new can be a hot spot a year later, and the symptom is not a fault light: it is a machine that welds well in the morning and poorly after lunch, because the resistance has drifted with temperature.

The habits that prevent it are unglamorous. Faces are cleaned and not touched with bare hands, torque is set and recorded, and the loop is checked for temperature rise after a duty-cycle run rather than by hand after a single weld. On service visits we ask for the temperature of the joints, not just the condition of the caps.

Current density is current divided by contact area, and contact area is set by the cap geometry. A cap that has mushroomed from its original face diameter spreads the same current over more area, so the density at the interface falls and the nugget shrinks. Nothing on the machine reports this, because the machine is still delivering the programmed current.

The joint between cap and holder matters for the same reason. A taper fit that is scratched, greasy or bottomed out on the wrong face has a smaller effective contact area, which shows up as a hot holder and a cap that fails early. Cooling water flow is part of the same circuit: the cap is cooled through the holder, and a restricted flow path turns a normal duty cycle into a short cap life.

Because the losses are invisible from the outside, we measure them. The table is the acceptance sheet we fill in for each machine, and it is the list worth asking a supplier for if you are comparing two offers that look identical on paper.

A spot welder is a short, heavy circuit with a press attached, and the nameplate only describes the first part of it. When two machines that should be equivalent are not, the answer is usually in the arm, the joints or the cap rather than in the transformer. Those three parts are also the ones that can be inspected and measured, which makes them a fair basis for comparison.

Requirements vary by destination market. Always confirm against the current edition of the applicable standard and with your certification body.

PDKJ continues to provide customized laser welding and resistance welding solutions based on different materials, workpiece structures, welding requirements, and production applications.

• Spot Welding Machines
• Medium Frequency Inverter Seamless Spot Welding Machine

Send us your part drawing or a sample, and our engineers will run a welding feasibility trial on your own material and recommend the equipment configured for your application.