richard harris wrote: > > Hello, > I started cutting today on a machine that was recently converted. > During the course of two hours the y axis drifted about 15 thousandths > and the x axis 5. I have this machines sibling running 20 hours a week > and it shows no signs of drift. The sibling is running emc 1.xx and a > different servo amp. > My understanding of tuning is that if I have a bad tune the machine > might not get where I tell it to go, but it will still know where it is. YEs, this should be true. > Similarly with backlash if it has 3 thousandths in backlash if it moves > to the right then the left and returns to a mid point the backlash > should cancel out and I return to where I should. > The machine is running a pico-system servo setup with emc 2.1.7(I > think). Encoders are on the motors and the motors are belted to the > ballscrews. I have new belts and similar tension to the other machine. > Is this drifting a function of the iron or the controller? It is not the controller, or at least should not be. There are a couple possibilities. One is that the signals from your encoders are "weak" or in some way not properly driving the inputs of the PWM controller. Another is that there is some slippage between the leadscrew and the encoder. If the encoder is on the motor, then any slippage of shaft couplings, belts, pulleys, etc. will show up just like this. The sharp torque peaks from acceleration can cause even very tight pulleys and couplings to shift. The shift in Y would be easily seen if you made match marks with a Sharpie marker on the pulley/coupling and shaft. You can't fix it until you know where the shift is. If you haven't put flats on the shafts, and used two setscrews at 90 degrees, you are likely to have problems.
I have a problem in my Bridgeport conversion. It is only a problem if I crash it, but the Y nut can slip because it is not dowel pinned, just bolted down. So, you may have to check the fixture that holds the nuts and the bearings that constrain the leadscrews. I have only touched a few possibilities. You should check if the problem only happens when the spindle is running, if so it is electrical noise interfering with the encoder signals. Since you are using Pico Systems PWM servo amps, they have filters on the output so that reduces the likelihood that is the problem, but it is still at least possible. Are your encoder cables shielded? The shield should be grounded only at the controller end. Depending on your grounding scheme, it should either be grounded to the control enclosure or the ground of the controller board. Jon ------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
