In high-temperature superconducting (HTS) power devices, the presence of iron cores changes the magnetic field profile around the HTS coil windings, potentially affecting their AC loss characteristics. AC loss measurements for HTS coil windings coupled with an iron core using the electrical method can lead to a significant error, owing to the indirect estimation of the iron core loss through using a copper test coil. To investigate the cause of the experimental error and the influence of an iron core on coil AC losses, transport AC losses of REBCO double pancake coil (DPC) assemblies coupled with an iron cylinder were measured. A 40-turn 1DPC and an 80-turn 2DPC assembly wound with 4 mm SuperPower wire were employed in the measurements. To ensure the same iron core loss using the HTS coil assembly and the copper coil, 2D finite element method simulations were conducted iteratively to design the iron core and the copper coil to get the same local magnetic field distributions in the designed iron core for the two cases. The main cause of the error is due to the difference in local magnetic flux densities in the iron core generated by the HTS coil assembly and the copper coil even when the ampere-turns of the coils are identical. We showed that the simulation-guided measurement method can assure accurate AC loss measurement in the HTS coil assemblies coupled with iron cores. Compared with the AC losses in the 1DPC and 2DPC coil assemblies without the iron cylinder, the presence of the iron cylinder significantly increases the coil losses. Frequency dependence is observed in the coil AC losses of the 1DPC and 2DPC assemblies when coupled with the iron cylinder. This is due to the eddy current induced in the iron cylinder generating a magnetic field, which influences the coil AC loss.