I'll second George's point that the variation of passive preamp output impedance as a function of volume control setting doesn't matter, as long as the worst case (highest) output impedance at any volume setting is suitable for the application.
I'll add that what can often be an issue with **active** preamps or sources is variation of output impedance as a function of **frequency.** Especially if the design utilizes a coupling capacitor at its output, as most tube preamps do, which commonly results in a large rise in output impedance in the bottom octave. Again, however, as long as the worst case (highest) output impedance at any audibly significant frequency is suitable for the application, that won't matter.
One minor clarification to the example George provided: The 2.5K figure for the impedance seen at the output of a 10K pot set to its mid-point will be increased slightly as a consequence of the output impedance of the component driving the pot. But if that component has a low output impedance (as it should, if it is to be used with a 10K passive), that addition will be essentially negligible.
10-27-14: Response34
We all have our views on passive vs active and this subject will be debated till the end of time but one thing we will never get away from is system synergy and choosing components that work together properly for "our" personal system requirements.
Well said!
Regards,
-- Al