OK, well then I would suspect that the output impedance of the tape outs is much less than 47K, and also that the tape outs are not driven by separate buffer stages that would isolate the main signal path from what is connected to those jacks.
So perhaps the cable capacitance and/or the 25K loading are interacting with the main signal path to produce the effects you are describing. Cable capacitance, to the extent that it may have any effects at all, would figure to affect mainly the upper treble, and to attenuate it slightly, which would seem to correlate with your mention of reduced edginess.
It's hard to formulate a more precise explanation without having detailed information on the design of that part of the Krell's circuitry, but it does seem very conceivable to me that the effects you are perceiving are real.
Best regards,
-- Al
So perhaps the cable capacitance and/or the 25K loading are interacting with the main signal path to produce the effects you are describing. Cable capacitance, to the extent that it may have any effects at all, would figure to affect mainly the upper treble, and to attenuate it slightly, which would seem to correlate with your mention of reduced edginess.
It's hard to formulate a more precise explanation without having detailed information on the design of that part of the Krell's circuitry, but it does seem very conceivable to me that the effects you are perceiving are real.
Best regards,
-- Al