Hi Ved, I understand that the "steam engine eccentric" - being a perfect sinusoidal motion - is MUCH lower friction than the Poppet-valve cams of ICE engines - Not sinusoidal, but determined by the cam-follower relationship - but not something I have ever quantified.
Also I understand that the O-rings applied by others (Brian Rupnow is one expert on this I think) basically have effectively zero interference with the bore, on their OD when assembled onto the pistons. All the radial load for sealing against the bore comes from the gas pressure at the bac of the O-ring, and the dynamic forces cause sealing of O-rings on either top or bottom face of the piston groove. Any assembly friction from O-rings on pistons installed into the bores should be absolutely minimal, but just touching. Brian R has successfully used O-rings on his ICE engines, so maybe he can advise more?
Your post #18 implies it runs with the O-rings installed, but needs 55psi to break the Stiction? (Static-friction).
What was the lowest pressure when without O-rings?
At 55 psi, the force on 1 piston can be calculated, so the friction equals that at zero movement. With some geometry, you can possibly deduce the difference between with and without O-rings as the "stiction" of Cam+ O-ring, versus Cam alone, to give some clues as to what is going on. Coefficients of friction at zero slip can be typically 0.3 from steel (from loco traction on rails) I think? So where forces are eccentric, you can resolve them from pressure onto the friction contact and pressure perpendicular to the line of contact force, to gain some understanding of where and how much Cam friction exists? Perhaps an excel spreadsheet at 5 degree cam-shaft rotation intervals can quickly indicate the worst point in the motion (the sticking point), that a cam designer would try and move to reduce the peak contact force, to optimise the motion, hence the shapes of regular ICE cams... Then with the CAD you can consider the off-set alignment - as you suggest - to reduce the "power-stroke" stiction, and replace with an increase in stiction where it doesn't have any material effect? Alternately to the 5 degrees of iteration, you can use steps-of-stroke.
This should give a rationalised approach to the design, and you can use the Pup to gain the real comparison between calculations and the real job! (there are always "other factors" ignored in numerical modelling that sometimes need to be measured and input to the calculations).
Note: "post-Victorians" considered off-setting bores from crankshafts on many ICE engines to reduce piston side thrust during the firing stroke. Something I wanted to do on a "mark 2" design of some equipment I designed for one job. But it wasn't a problem in service, so I didn't get the second bite at the design. All the calculations I did back in the 1980s were "pre-spreadsheet", so were on reams of paper with manually plotted graphs! Shortcuts with constants were necessary, so the slide rule (the quickest calculator around) could be used as a conversion scale to read-off a set of values against single variables. Much faster than "Texas instruments" programmable calculators! And clearer to see the results. Who needs computers? - Except for more complex iterations.... We have brains.
K2
