For clarification:
You have 4 sets of 500 mcm conductors.
By a 'set' you mean a _single_ conductor for each phase (plus the neutral if applicable). So a set might have a single phase A conductor, a single phase B conductor, a single phase C conductor, and a single neutral conductor.
When you put your clamp meter around a _single_ phase A conductor, you measure 169A. All of the 'phase A' conductors are pretty much equal.
When you put your clamp around _two_ conductors, you still measure 169A. You believe that the _two_ conductors that you are measuring at the same time are both 'phase A' conductors.
Do I have the above correct?
1) The _total_ phase A current is simply the sum of the current flowing on each of the phase A conductors. Thus _if_ the 169A measurement is a true value, then you should see a total of 776A on all four conductors of phase A, and 338A on any pair of conductors.
2) If the system is balanced across the phases, then you would see a similar current flowing on each of the other phase conductors. An interesting fact: if you have 'X' amps flowing on a phase A conductor, and 'X' amps also flowing on a phase B conductor, and you measure both conductors together in your clamp, you will still read 'X' amps. This is because the phase A current is not in phase with the phase B current.
Given the information that you've provided, I would suspect a measurement error. Here are the possibilities that come to mind:
1) Perhaps the meter is 'maxing out' at 169A, and the real values are higher. If you have 180A on one conductor, and 355A on a pair of conductors, but the meter is maxed out, then you will read 169A in both cases.
2) Perhaps your phase identification is wrong, and that while you believe you are measuring two phase A conductors, you are actually metering across different phases.
It is also possible that you have correct measurements of the current flowing in the conductors, but that the process of measurement is changing the current flow.
1) Perhaps one of the conductors has a bad termination or a break. When you move the conductor to make the measurement of two conductors at the same time, you interrupt the current flow in that conductor.
2) Perhaps the meter itself introduces enough impedance that it changes the current flow distribution. While you are not directly introducing the meter into the circuit, it is magnetically coupled, which means a slight additional impedance on that branch of the circuit. (I would consider this very unlikely with any sort of normal clamp meter, but I'm throwing it out there as a reminder that the act of measuring something changes it, if only slightly.)
-Jon