Originally Posted by 2milehi
I commend you for using a dial indicator to get a qualitative answer.
On a side note, a motor introduces torque via a moment from the crankshaft. You are introducing a moment into the tranny input shaft via an arm and a "pushing down" force. Some of that force is pushing down on the input shaft and into the tranny case. Curious if you get the same deflection by "lifting up" the torque wrench.
Indeed that's another limitation of the test fixture. Because I am not generating a balanced axial load, my measurements could be affected by lateral force applied to the input shaft through the torque wrench.
While it is technically qualitative data, it's all relative measurement... So I can't say, for example, with certainty that any spot I was measuring was moving as much as the dial gauge indicated... some of that distortion could be in the bellhousing to which the dial gauge was mounted. What I can say is that position x had more or less flex than position y relative to a point on the bellhousing in my collected data set. You can also use this to follow paths of flex. I took measurements running down the bottom of the case, and the further back I went the more flex I measured, while there was far less flex along the top "spine" of the case, which is an indicator that the case is twisting on an axis running lengthwise near the top of the case. One thing that I can say with certainty is that there is a substantial amount of flex in the case as a whole.
You have to figure that dropping the clutch from redline potentially generates thousands of ft/lbs of torque for a split second (even a small hand held impact gun can make 250+ft/lbs). So even with these limitations in mind, I can't imagine that this level of flex in the case is not harmful to the internals of the case in some way, especially with more torque applied where I would expect even higher levels of case distortion.