Re: confused about PLL
For your typical modern frequency synthesizer, one that contains some sort of integrating function in the loop filter, they are probably the same.
I can think of some scenarios where they might be different, though.
Lets say I was using a PLL as a receiver, say for a satellite system. On first turn on, the PLL might simply sit there unlocked. Then you might deliberately sweep the local oscillator frequency, either in steps or with a saw tooth. As the LO frequency changes, eventually the desired signal comes within the loop bandwidth of the PLL, and the PLL locks to it. You then disable the sweep circuit, and it remains locked. I am not sure if this is 100% correct, but I would call the full band sweep range the "capture range", and when I got close enough for the PLL to acquire the signal tracking, I would say that I got within the "lock-in range".
In such receivers, there is another phenomenon. Once locked, they stay locked over a wider range of perturbations, but if they become unlocked, the will not relock until you get within a smaller range.
I have seen the same phenomenon with injection locked oscillators: Once locked you can tune them pretty far one way or the other, but once lock is broken, you have to get pretty close in frequency before they will relock. I believe they call this bigger bandwidth, once locked, as the "hold-in range".
In a frequency synthesizer, you might run into a similar scenario. Lets say you were using a comb generator as an LO for a mixer (or downconverting sampler). You might have to coarse tune the VCO to be near the desired signal before the PLL can automatically lock up. There, you might call the "lock in range" starting and ending halfway to the two adjacent comb teeth, where the capture range can, with coarse tuning, be the entire possible output bandwidth.
Rich