Following activation, S4-based voltage sensing domains (VSD) undergo a voltage independent transition, known as relaxation, shifting their voltage dependence to more negative potentials. The molecular determinants for relaxation remain unknown. However, studies on Ci-VSP suggest that relaxation likely involves secondary and tertiary structure rearrangements within the VSD. Since relaxation is seemingly an intrinsic property of VSDs, we sought to identify regions of the domain involved in the triggering of relaxation. For this, we focused our study on the loop connecting the S3 and S4 segments (S3-S4 loop). We argue that, like the S4-S5 linker coupling the VSD to the pore domain, the S4 segment can do electromechanical work on the rest of the VSD using the S3-S4 loop as “coupling element”. Several studies have proposed that the S3-S4 is helically structured. Hence, we hypothesized that the movement of the S4 segments can be readily transmitted to the S3 segment and the rest of the VSD causing relaxation through a “rigid” S3-S4 loop. Conversely, a “flexible” S3-S4 would be able to absorb the movement of the S4 segment, dissipating mechanical energy and diminishing relaxation. To test this idea, we determined amplitude and timing of relaxation by performing cut-open voltage-clamp recordings of potassium currents from ShakerIR S3-S4 loop proline-to-alanine mutants. We reason that, since prolines disrupt helical structure, introduction of a “helix-friendly” alanine will increase the rigidity of the loop. Consistently, our data indicate that the proline-to-alanine mutations increase the magnitude of relaxation. Hence, we proposed that S3-S4 loop's prolines act as “hinges” conferring flexibility to this region. Similar results were obtained with glycine-to-cysteine S3-S4 loop mutants of Ci-VSP, suggesting that the S3-S4 segments may have similar function. In summary, we conclude that the S3-S4 helix seemingly acts as a mechanical stress dissipater in VSDs.