[time-nuts] Can one use a two channel TC with an internal reference to measure two independent oscillators to establish the quality of the internal reference?

[Erik Kaashoek]

During evaluation of a two channel TC that can measure simultaneously the
phase of two external oscillators the need came to understand the quality
of the internal reference while eliminating (some) uncertainty of the
performance of the two external oscillators.
The 3 cornered hat method comes to mind as a way to get insight but I am
sure I do not understand this all good enough.
Some practical info on the setup:
Two channel TC with 4 ns clock resolution and capable to measure phase
against it's internal clock using interpolation over maximum 50000 phase
measurements per second.
The external clocks being measured are a Rb with unknown performance and an
OCXO with unknown performance.
Any advice on how to create a better understanding of the performance of
the internal reference using the above setup?
Erik.