Len, the chain does not resist windup from the spider in 360 degrees. Only about 90 degrees of chainwrap actually does loading (both in the testing rig and in the real world.) The rest of the chain is basically just laying without load on the ring to keep it out of the way. The measurement taken from the ring is just a secondary measurement that we take just for fun, but really doesn’t have bearing so we don’t publish it. The actual deflection measurement is taken directly. It’s measured at the upper end of the jig at the end fo the rod that pulls the arm and the strain gauge. This measurement because it’s taken centered in the pedal spindle includes both vertical and twist.
Some comments on the 7900 crank. This test has been going on for several years and we just keep adding new cranks to the test. That’s the reason for the inclusion of some older cranks. In the end we should have a nice list of cranks that cover quite a few years. 7900 will be included on the next test. The numbers for 7900 really didn’t change in any significant way from 7800 so we didn’t feel a rush to get it included in the test.
We also hope to soon add bb30 to the test which means being able to test the SIsl.