On 5 Jan 2016, at 13:30, andy pugh wrote: > On 5 January 2016 at 13:11, Sarah Armstrong > <[email protected]> wrote: >> or >> 999,999,999,999,999,967,336,168,804,116,691,273,849,533,185,806,555,472,917,961,779,471,295,845,921,727,862,608,739,868,455,469,056.00 > > You wouldn't want to get anywhere near this limit, though. > > The real practical limit is the number of significant figures in the > floating-point format (about 16). The difference between 1e99 and the > next smallest representable number is about 10^83 complete > revolutions. > > For a 90:1 rotary and a 4096 count encoder you have an angular > resolution of 3.5 arc seconds. or .0.001 degrees. After only > 99999999999 degrees the double precision format can no longer > discriminate between numbers one encoder count apart. A fairly fast > rotary could get there in less than 100 years. >
Oh well; I suppose there is some hope then ... Thanks for all your contributions, I had feared I was missing something obvious. It does seem to me that there is a fundamental difference between linear and rotary axes. You would not want to exceed a defined limit on a linear axis simply because the slide would come to the end of its travel (or off the end), so exceeding the limit does not make mechanical sense. With a rotary axis, it is entirely different. That suggests that limits should not be thought of, or applied, in the same way. I can't see any disadvantages in automatically treating rotary axes in a different way. The problem, I guess, is that while some axes on a mill or a lathe would naturally be treated as rotary, other unspecified machines with unspecified axis movement types would then need exceptions. It's a penalty of flexibility. Anyway; for now I will try much larger limits on the odd occasion when I need to do this kind of rotary motion. Or do the job another way. In this case, the (diameter of the job + 2 x diameter of cutter) exceeded the Y travel on my mill, so I thought that using the rotary table was a reasonable solution. I could have done it in the manual lathe, but really object to the time that would take, so I stuck it on the mill while I got on with other jobs. One of the bonuses of CNC after all. Regards, Marcus > -- > atp > If you can't fix it, you don't own it. > http://www.ifixit.com/Manifesto > > ------------------------------------------------------------------------------ > _______________________________________________ > Emc-users mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/emc-users ------------------------------------------------------------------------------ _______________________________________________ Emc-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-users
