NOTE 11: How does that expression go..."the best laid plans..." Some time ago I ran some tests in which I applied ever increasing clamping force on a couple of shafts and recorded the change in frequency. As I recall a steel shaft varied about 5 or 6 cpm's as the force increased from very little almost wrecking the shaft. I repeated the experiment with a graphite shaft and got about a two cpm change. This is not surprising. Graphite has considerably more compressive strength than steel. The problem with these tests is that I knew how much torque I was applying to the bolt compressing the V blocks but I didn't know how much actual force was being applied to shaft. I have no way of relating torque to force. It's all tied up in the friction of the bolt threads.
My plan was to build a setup that would measure force directly and give me a relationship between torque and force then when I applied say 5 ft lbs. of torque I would know exactly what force was being applied. I thought it would be easy to measure the force...it wasn't. I bolted a clamping unit to a bench, removed the torque screw and passed a steel bar through the U bracket on the top of the clamping unit. The bar was a four foot piece of 2"x5/8" cold rolled steel. One end, let's call it the left end was supported on a block about six inches tall. The right end on a similar block resting on a bathroom scale. (This test would probably not pass the ISO 9000 quality assurance requirements but at least is was a digital scale.) A bolt was then screwed into the top of the U bracket. As it came to bear on the steel rod the bathroom scale would register a force. The bolt applied force about 1 foot from the left end of the bar. Since the bar was four feet long the actual force being applied would be four times what would be registered on the scale. Since the scale went to 300 lbs. I would be able to relate torque, applied with a torque wrench, to forces up to 1200 lbs. Much to my chagrin the steel bar bent and bottomed out under the U bracket long before I got to very high forces. I'm now looking for a much stiffer bar. Using this approach I did find that my torque screws apply about 350 lbs. of force to the shaft.
I ran into a friend the other day who is in the scale business. He has some interesting looking strain gauge load cells that just might fit directly under the U bracket. I'll have to put the bite on him. Stay tuned.
In a previous tech note I mentioned how little variation I had seen in the performance of various frequency analyzer electronics. I always thought the weak link was in the shaft clamping. I recently decided to change the glue I used to attach the lower fixed V block in my Club Scout. The new glue was a bit more pliable and I thought additional damping might reduce the frequency. I took an old clamp and several of the newly glued ones and measured a filament wound shaft with them. They all measured 307 except one. It came in at 305. The same torque screw and upper V block was used throughout. After scratching my head a bit I found one of the mounting bolts on the offending clamping unit was not fully tightened. I cinched it tighter and got 307!