LinuxCNCKnowledgeBase | RecentChanges | PageIndex | Preferences | LinuxCNC.org

Showing revision 1


Tach feedback to drives

This Mazak V does not have tachs. The initial impressions were that the amps must have been running in current mode rather than velocity mode. It became apparent during axis testing that a signal of some sort was needed. That signals was supplied by the old CNC.

There are several possible solutions.

If one were to think of the Mazak as a collection of metal and were to build a kit for it's restoration, then upgrading the drives would be the prefered solution. Current mode operation has been shown to be possible with the EMC. This solution will require some careful tuning.

Spindle orient

The Fanuc spindle driver was replaced some time ago with an off-the-shelf drive from another vendor. This fit included a set of relays and circuits that were supposed to allow the new drive to orient the spindle. The spindle dog that drives a tool holder shows the evidence of recent poor alignment during tool changes. A failure in this system was one of the reasons that the machine would not run when delivered.

Spindle orient signals are produced by a magnascale mounted to the upper side of the gear box. It appears to be on the output shaft so that the gear setting does not affect location. There is a wide pulse and a ramp signal. The wide pulse can be used to switch the drive into error mode and the ramp used to center the rotational position.

One solutions is to build a circuit that shapes the existing pulses so that they can be directly accepted by the servo drive. When orient mode is set, the spindle will slowly rotate until the wide pulse and then the drive will switch to error mode. minor offsets in position can be achieved within the control. It is assumed that larger corrections can be made by shifting the orientation of the magnascale pickup or the magnetic element.

A second solution would be to add a spindle encoder, read that encoder with the Motenc card and treat the spindle as an axis during orient. This solution has the advantage of a continuously variable resting position that is settable with an ini or hal parameter.

UPDATE: The second solution has been chosen, and Matt is working on the mounting method for the encoder.

Estop Chain

We decided to replace the Mazak estop (normally open, vulnerable to broken wires) with a single 24 volt loop that included; extreme motion on axes, computer failure, estop switches, and several fault conditions. These signals must be latched so that rapid travel past an estop dog does not restart the machine.

Latching is accomplished by adding the requirement of an <F1> after a break in the estop chain has been seen. Details at MazakEstopChain

LinuxCNCKnowledgeBase | RecentChanges | PageIndex | Preferences | LinuxCNC.org
This page is read-only. Follow the BasicSteps to edit pages. | View other revisions | View current revision
Edited September 12, 2005 9:47 pm by Jmkasunich (diff)
Published under a Creative Commons License