In the automobile industry, safety and comfort equipment require more and more switches which are operated over a wide range of electrical, mechanical and ambient conditions. In the low electrical power range, i.e., near minimum current and voltage, shorts and fluctuating arcing may occur. Using a testing apparatus capable of rapid measurements with high temporal resolution (0.2 ns) we have measured and analyzed arc durations on break in a pure resistive circuit near minimum arc current and voltage O < I < 1 A, and V = 14 VDC. For all contact materials tested (Ag, Cu, Pd, Au, Sn, Ni), average arc duration values versus current are grouped around two exponential curves corresponding to two anodic arc regimes. The analytical expressions for these curves are found to be in good agreement with the multiburst arc operating model [T = t <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">B</sub> exp(I//i <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">B</sub> )]. The lifetime and current intensity of the burst (t <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">B</sub> , i <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">B</sub> ) are deduced for various contact materials. In addition, the histogram of arc duration fluctuations is well fitted by multiGaussian distributions centered around a multiple of the burst time. This confirms the burst mechanism and introduces the concept that arc duration values are discrete. On the other hand, the plot of arc occurrence in the first regime clarifies the significance of the minimum arc current (100%) and yields the current limit for arc appearance (0%). Finally, high voltages (14-28-42 V) and inductances (1 μH-10 mH) can produce long arc durations with the result that we may enter into the cathodic arc.