Microgrids are a promising solution for providing electricity access to rural populations in the Global South. To ensure such microgrids are affordable and effective, careful planning and dimensioning is required. Numerous tools exist to assist with this planning and dimensioning, however, they generally require microgrid electricity data, such as load profiles, as an input. Unfortunately, such electricity data is scarce for microgrids in the Global South and the little data that is available has a low temporal resolution. Therefore, we introduce high resolution real-world electricity data from three microgrids in the global south. The data is collected from microgrids in the Democratic Republic of Congo, Rwanda, and Haiti, and has a temporal resolution of up to five seconds. Furthermore, we include data from both residential and industrial microgrids, and consider microgrids with renewable generation from hydropower and photovoltaic arrays. We describe and analyse the characteristics of the recorded data and show how it can be used to derive flexible load profiles with varying temporal resolutions, useful for affordable microgrid planning and dimensioning.