Abstract: During decay heat removal, the sodium flow rate through the core of sodium-cooled fast reactors (SFRs) determines the removal ability. This flow rate is a balance between the buoyancy-driven force generated by the sodium density variation in the primary loop and the flow resistance during the natural circulation. The resistance loss consumption in wire-wrapped bundles occupies more than 90% of the total head loss in the natural circulation loop. Consequently, the frictional resistance in wire-wrapped bundles at low velocities is one of the essential parameters in designing passive residual heat removal systems (PRDRS) for SFRs. However, there are a few experimental data in this regard. In this paper, a liquid column manometer was improved by adopting image processing techniques, and volumetric flow rates were obtained by measuring the mass flow rates. In addition, a constant-head tank replaced the pump to drive water through the assembly, reducing the fluctuations significantly in water flow during frictional resistance measurements. Then, the frictional pressure drops of a 37-pin wire-wrapped bundle assembly made from polymethyl methacrylate were accurately measured using a water medium at Re=200-1 100. Experimental results show that the frictional factors of the measured assembly decrease with increasing Re and are 0.337-0.094 for Re=200-1 100. The transition Re from laminar to transition flow of the measured assembly is 370, and the friction factors of the assembly decrease faster with the increasing Re in the laminar flow region. The rate of decrease decelerates in the critical zone. At low flow rates, the entrance length of the assembly increases as the flow velocity increases and decreases as the flow begins to transform from laminar flow in the assembly. Although the flow channel inside the assembly is very narrow and complex, at low flow rates, the flow inside the assembly still exhibits the same frictional resistance characteristics as inside simple geometries, such as a circular tube. Among several available empirical formulas for the frictional factors of wire-wrapped rod bundles, the UCTD formula is closest to the measured values in this paper. It overestimates the frictional factor by about 7% in the laminar flow region and is generally consistent with the measured values in the critical zone from laminar to turbulent flow. Furthermore, both UCTD and CTS formulas overestimate the critical Re. The UCTD formula overestimates the critical Re less (370 vs. 560) than the CTS formula (370 vs. 826). The experimental results can guide the design and optimization of PRDRS of sodium-cooled fast reactors.