Dude, thanks so much for the detailed info on the tensioning hack. Here is my brief report I sent a friend experiencing similar issues with his kickr on giving it a try and having what seems like some nice initial success:
On my off day, I took off the little plastic cover (couple of screws on each side) to reveal the belt and make it easy to see the two screws involved in retensioning the belt. The larger screw on the side loosens to allow you to make an adjustment to the screw in back (retightening after each adjustment). Clockwise makes the kickr harder (raises watts compared to quarq) counter-clockwise makes it easier. You set it easy first, do a spindown and then never spindown again. You then adjust till it matches. I confirmed on Monday that adjusting things indeed changed the way the kickr read, but didn't try to dial it in (I was just holding 150 watts to test).
Last night I setup a longer ride (picked quasi-randomly from TR for ~90 minutes Z2), with TR showing my quarq wattage. The first long interval the kickr was 10ish watts low (previously was more like 25 watts), so during the short rest, I got off, adjusted the tension by 1/2 turn and got back on (~30 seconds). The next long interval read ~4 watts low, so same story at the end of the interval adjusted another 1/4 turn. The next interval read 1 watt low (!!), so I adjusted another smidge and the last interval the power seemed to line up perfectly. So at this point I'm cautiously optimistic and very happy to at least have some ability to adjust things if/when they get out of whack. The next test will be to see if the settings stay stable across multiple days of workouts.
Here was what the TR ride looked like. You can see the yellow line (the quarq reading) getting closer to the green (kickr target wattage) in each interval.
http://www.trainerroad.com/...es/1521716-Holt-Hill