Electronic – Obtaining 240V from 480V 3-phase

The circuit will both be at higher voltage (well the parts of it) and will self-adjust. The reasoning behind this is pretty simple: First, the phases are out of phase, or to explain it simply, their voltage peaks happen at different times, so while one is at its maximum voltage, the two other phases will be at lower voltage than the one at maximum is, so basically, the load will be at the voltage between phases. Since we don't have the neutral conductor, the current which was going through it will have to go through one of the phases and that's how the circuit will "self-adjust". You can see that from the simulation here.

When the switch is closed, all 3 loads are at the same voltage of 5 V compared to the ground. Once the switch is open, the voltages at the loads will change. So the lightest load will drop the least voltage and the greatest load will drop the most voltage.

Any help would be appreciated

You can measure phase-neutral voltage and if the supply and load are balanced then you can infer line voltage by: -

line volts = phase volts\$\times\sqrt3\$

This should be one of the easier measurements yet you say you can calculate "real power, reactive power and apparent with ease". This does make me think that you have used one of these measurements, and the RMS measurement of current and back-calculated phase/line voltage and that's when you are seeing a discrepency.

If this is so then I suspect your current measurement may be at fault either through incorrect use of a current transformer or some ratio being incorrect. Or, it could be the \$\sqrt3\$ thing mentioned above.

If you need any further help with this you should consider detailing how you make the other measurements and why you believe voltage to be incorrectly calculated/measured.