The Modular Multilevel converters are gaining momentum in power converter-driven power systems. Their reduction in the Total Harmonic Distortion (THD) and their flexibility of control make them very versatile compared to conventional two-level converters. However, as it happens with any power converter-based system, they are very exposed to ground faults due to their accelerated aging process caused by the semiconductors switching. Ground faults (L-G) are caused due to insulation degradation. Depending on the grounding topology of the electric systems these faults can be more or less harmful and more or less detectable. In case of isolated grounding configurations, a first ground fault does not affect the operation of the system. However, a second fault will incur into a Line-to-Line (L-L) fault, which will harm the system considerably. Also, in case of a rigid grounding, the L-G faults have also high fault currents which will damage also the system. However, its detectability increases due to this fact. In this technical report, the experimental validation of a ground fault location method developed by Jose M. Guerrero et al. (2023) is described. This method uses a DC midpoint high resistance grounding configuration, and some voltage measurements to detect the fault, discern the phase in fault and locate the module in fault. In this case, the grounding configuration proposed is an intermediate case between isolated and rigid groundings. This fact enables the normal operation condition of the system with a first fault as the fault current is limited by the grounding resistor. Afterwards, with the abovementioned voltage measurements, the fault is detected if the grounding voltage is up to a certain Root Mean Square (RMS) value. The faulty phase is discerned comparing the phase angle between the AC voltages and the grounding resistor voltage. Finally, the location inside the arm is located considering the relative value of DC component respecting to the AC main harmonic in the grounding resistor voltage. The DC bus voltage measured will be recorded with severity estimation purposes in further works. The experimental validation was carried out in the SINTEF laboratories, that have been used under the EriGrid 2.0 project. The experimental setup used was a 60 kVA 12-levels three-phase Modular Multilevel Converter (MMC) with intermediate terminals that allow performing ground faults inside the submodules (SM). More than 400 tests were conducted from the June 6th to the June 13th 2024 in order to validate the method. The faults considered involve all the fault positions (24 submodules, DC positive and negative poles and AC side fault) for the three phases and for fault resistances of 0, 2.2, 4.4, 6.6, 8.8, and 11 kΩ. The ground faults have been performed utilizing a galvanically isolated ideal grounding in order to not affect the rest of the laboratory during the tests. The results are promising. They allow to detect the submodule in fault with an accuracy of ±1 submodule in the intermediate zone, which can be admissible in a 24 submodules-per-phase MMC. Also, the experimental validations have allowed the user group to realize about the non-linearities of the problem, due to not all the capacitors involved in the fault circuit are equalized. Experiments have proven that voltage differences among capacitors can be up to 1 V, which can finally lead to a total of 25 V of difference between faults, i.e., an induced error of a submodule: 4.16%. With the experimental data collected, further works should aim to obtain a deeper theoretical understanding about the location method in real MMCs. Furthermore, additional methods and intellectual property can be generated.