positive
to
both
we
that's the idea yes, amplitude is on the vertical axis and phase is on the horizontal axis we're only comparing the horizontal axis, and (almost) ignoring the vertical unless to "select" a reference point
the phase of a single signal has to do with the 360deg (2*pi rad) between two instances of a repetitive cycle whatever the amplitude, the 360deg is a given constant for every cycle of this signal whatever the time is to go through the full 360deg doesn't matter too, so ignore the aspect "time" too (disconnect "time" from "phase" for a moment in your mind)
now, given the initial 5Hz and 10Hz of this thread, starting from a specific coinciding point of two signals (assume zero crossing) one can start counting the completed cycles as 360deg for every signal the 5Hz sig will have done 360deg after 1 cycle, whilst the 10Hz will be at
720deg (= two cycles) at the finishing point of this first 5Hz cycle thus the phase difference of both signals will always increment, no matter how long you look at them (*unless* you take a new reference point and start counting from scratch but that's a new measurment) an integral part of a pll is a phase comparator (usually phase detector combined with an integrator) giving the "measure" for the phase difference to be able to adjust a vco which is another part of a pll