1. The fixed channels pass-thru the DC power from input to output - red-in to red-out, black-in to black-out. The variable channels use transistors switched on/off synchronized to the incoming 60 Hz AC cycle to chop the input voltage to create a lower output voltage. When you set it to fixed (no variable reduction), the TIU must still turn on its transistors on each cycle; there is no relay or whatever to simply pass-thru the input to output. OK. Now the question is what the TIU does if it does not see the input AC cycle and only sees DC. Since it is looking for the AC cycle, will it wait forever and never turn on its transistors? I don't know the answer but I suspect it would do just that and you'd get no output whatsoever.
However. You might look into the passive mode where your power supply is connected to the TIU variable channel output connectors (nothing connected to the TIU variable channel inputs). You'd have to confirm passive mode works for DC voltages (I don't see why it would not but one of the MTH gurus might enlighten us).
2. As I see it the "bridging" issue has nothing to do with the TIU, DCS, etc.. It's simply a matter of how your particular power supplies behave if connected together. As you say, there will be instances where an engine with multiple power pickups momentarily straddles two power districts. The engine may even be intentionally stopped with pickups on each side. So it's like shorting the two power supplies with 10 ft or 20 ft wire (the length of track to the straddle point). Power supplies can be designed to handle such situations differently - though if you have identical designs both dialed in to the same voltage this no doubt be the least stressful scenario.
