Extended Data Fig. 1: Solar loop experiment circuit simulation.

(a) Experiment circuit diagram. The plasma part of the circuit is represented as an inductor and a time-dependent resistor. The plasma inductance is assumed to be 50 nH, which is obtained by simplifying the plasma loop as a half circle loop of wire with 5 cm loop major radius and 1 cm minor (wire) radius. The voltage and current spikes are both peak functions, so the corresponding resistance change is presumed to be also a peak function. We use Gaussian function \({R}_{plasma}={R}_{0}\exp (-a{(t-{t}_{0})}^{2})\) to represent the transient change of the plasma resistance where R₀ is the peak resistance value, and t₀ is the resistance peak time, and a is related to the full width at half maximum (FWHM). They are chosen according to the relative voltage spike amplitude, voltage peak time and the voltage spike FWHM. In the simulation, R₀ = 0.4Ω, t₀ = 3.65 μs and a = 5 μs ⁻² are used. The corresponding plasma resistance is plotted in (b). (c, d) Voltage and current measurement from experiment Shot # 9258. As shown in (a), the voltage measured in (c) is the voltage across the plasma part and an extra inductor. We also measured the voltage across the plasma part by connecting two voltage probes directly to the top electrode and bottom electrode and then subtracting the two voltages. The voltage trace across the plasma is similar but has a several kV larger voltage spike compared with (c). (e, f) Voltage and current curves from the simulation. Voltage and current spikes similar to the experimentally observed spikes are reproduced by the transient resistance increase.