Aim By a combined electromyographic (EMG), mechanomyographic (MMG) and force (F) analysis, the electromechanical delay during muscle relaxation (R‐DelayTOT) was partitioned into electrochemical and mechanical components. The study aimed to evaluate the effects of fatigue on R‐DelayTOT components and to assess their intersession and interday reliability Intraclass correlation coefficient (ICC). Methods During tetanic stimulations, EMG, MMG and F were recorded from the human gastrocnemius medialis muscle before and after fatigue. The latency between EMG and MMG ripple cessations (R‐Δt EMG‐MMGR, electrochemical R‐DelayTOT component); between MMG ripple cessation and F decay onset (R‐Δt MMGR‐F, first R‐DelayTOT mechanical component); and between F decay onset and maximum MMG negative peak (R‐Δt F‐MMGp‐p, second R‐DelayTOT mechanical component) was calculated. Results Before fatigue, R‐Δt F‐MMGp‐p was the major contributor (61.9 ± 1.7 ms, 75%) to R‐DelayTOT (82.7 ± 1.0 ms), while R‐Δt EMG‐MMGR and R‐Δt MMGR‐F accounted for 16% (13.3 ± 1.2 ms) and 9% (7.5 ± 1.0 ms) respectively. After fatigue, R‐DelayTOT, R‐Δt EMG‐MMGR and R‐Δt MMGR‐F increased by 11, 41 and 67%, respectively (P < 0.05), whereas R‐Δt F‐MMGp‐p did not change. Consequently, the relative contribution of R‐Δt EMG‐MMGR, R‐Δt MMGR‐F and R‐Δt F‐MMGp‐p, to R‐DelayTOT changed to 20 ± 2, 12 ± 1 and 68 ± 2% respectively. Measurement reliability was always from high to very high (ICC 0.705–0.959). Conclusion Fatigue altered the processes between neuromuscular activation cessation and force decay onset, but not the second mechanical component (cross‐bridges detachment rate and series elastic components release). This combined approach provided reliable measurement of the different R‐DelayTOT components and it may represent a valid tool to get more insights on muscle electromechanical behaviour.