On Nov 3, 2011, at 3:57 PM, Frank Tkalcevic wrote: >> We have the squaring part working. We send a simultaneous signal to each >> Granite drive and it homes to a hard stop and comes off a fixed amount. > It is >> repeatable to no worse than 0.002, often better (lte 0.001). > > You can do exactly the same thing with EMC.
Almost, however with the function provided on the Granite drive it lowers the torque of the servo to a settable value so that it doesn't go at the stop with full tuned torque. It is a nice gentle stop and push to the limited torque value. With EMC is more of a slow slam and attempt to push the stop off ;-) > If you did it with EMC, then you can use the index pulses to home and > square. This would be repeatable to the nearest encoder tick. If my encoder indexes are lined up then everything is easy, I can have EMC find the index without a problem. I was trying to do homing from any random (unknown) state where encoders might be out of alignment (in terms of index). > The only danger is if one axis hits the index pulse almost one full motor > revolution before the other, and one motor revolution moves than the 3/4" > racking limit. > This set up is the last thing I need to do on my gantry. Hopefully I'll have > time this weekend. Exactly, this is the whole crux of my problem. Murphy's law dictates the encoder pulses WILL be unaligned! :-) Apparently I have no choice but to come up with a simple (or as simple as possible) procedure for getting the encoder indexes in alignment and then just let EMC do what it already does. Or it is additional wiring and/or perhaps hardware which I would like to avoid if possible. -Tom ------------------------------------------------------------------------------ RSA(R) Conference 2012 Save $700 by Nov 18 Register now http://p.sf.net/sfu/rsa-sfdev2dev1 _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
