Vance could you modify that program to mark just one rotor apex?
Could you let the rotor rotate only once (e-shaft three times) and leave a trail
from that marked  rotor apex?

I think that would greatly add to the understanding of the rotary engine motion.

Paul Lamar

Paul - The apex, as one might suspect follows the inner countour of the  houding - a sort of "dogbone" shape
Actually this is how I generate the housing contour, and how one would write the tool path for machining this
shape. One could program an NC to grind - then hard chrome and crind and lap as an overhaul operation
if that was cost effective. The attached GIF also shows the path of the face of the rotor flank, and not too
surprisingly it is a similar shaped path 90 degrees out of phase. The difference in radius of the enclosed, and inclosing circle of this path is the effective "stroke" of a Wankel ie the distance from TDC to BDC. Now the
effective swept volume of each flank is greater than this "stroke" times the flank area, because the surface of
the housing also gets further away as the flank rotates, almost as if you had the "head" driven by another
crank. At first look the sharp point at BDC would suggest high acccelleration, but the dashes are agularly
(time) spaced, and you can see that it is moving slower than at TDC, It would appear that this deviates
greatly from sinasoudal, but it would take a straghtend out time based plot to be sure.The distance between these points and the housing would probably be roughly proportional to chamber instantaneous volume, and
it would be interesting to stretch this out vs time as well. There is significant deviation from circular motion
going on here, but whether it fits the defination of reciprocating motion is a mote point. I should probably
revise the program further to generate the apex path as a dotted line for time reference as well.
By the way - I am going to have to get up to date on a screen capture technique (your GIF s were much better than the pictures I pull off with a BASIC screen print program for my old water wheel driven Quadram ink
jet printer)     :-)
Vance

I have a plot of apex seal velocity if you are interested.
It never stops.

Paul Lamar
 
The AirCraft Rotary Engine NewsLetter.              Powered by Linux.

http://home.earthlink.net/~rotaryeng/            http://www.linux.org