Re: [mfg-smartcam] a little story about "i"'s and "J"'s

[D G Bowen]

Chris,   It has to do with the whole concept of the using the R value instead of I and J.   When using the R value, you specify the start and end point to 4 decimal places. The Fanuc then fits the R radius through these points. Sometimes, due to roundoff error, the radius calculated does not have the same center point that you thought it would. Therefore it cuts the radius slightly displaced so that it looks like an ellipse. The larger the radius, the more error you can have.   SmartCam has a question in Machine Define (#104) to address this problem, but I haven't been able to get good results from it.   The solution is to use I, J, K instead of R.   Gene Bowen     ----- Original Message ----- From: Chris Peters To: [EMAIL PROTECTED] Cc: Ernie Gemmel Sent: Thursday, December 07, 2000 8:15 AM Subject: [mfg-smartcam] a little story about "i"'s and "J"'s Hi all, has anyone had trouble with interpolating bores with cutter comp?  The .pm4 geometry is completely round, but once you run it on the machine, you end up with an oval? our immediate action was to change the coded output from "R" values to "I" and "j" outputs, which seem to lock down the quadrant boundaries for Fanuc and gave us a nice round bore.   does this sound familiar to anyone?  you see the problem is there are other tapes on the floor that use "R" in cutter comp and they function perfectly well!   we tested different lead-ins and consistently, the i's and j's gave us a round hole, and the "r" gave us an oval.  it was a 5/16 em, cutting a .793 dia bore with circular lead-in and lead-out. the y dia is .010 larger than the x dia.   I'm pretty sure it's a Fanuc-ism that causes the ovality (a "best-fit" type scenario), but the question is, is it because we are crossing quad boundaries with an "R"? is it because our tool dia is less than half the interpolated dia? are there any mathematical braniacs who can decipher this geometric conundrum for us simple machine-heads?   go figure