Mill Integration

It's finally time to integrate the electronics and the mill.  This will hopefully be a smooth process as most of the testing has already been done.

I started by unpacking the electronics.  The main panel sits on a cabinet behind the mill, there will be no enclosure as this is not intended to be a polished project.

The servos, laptop, and computer were connected and quickly powered on to ensure nothing was damaged during the move.

The mill is fairly complete from a mechanical perspective so all I need to do is mount the servos and their belts.

The X axis servo folds out in front of the table and drives a 1605 ballscrew through 18:56 timing pulley reduction.  The pulley stage isn't strictly necessary, but will allow higher accelerations from the mill.

The Y axis servo mounts very similarly.

I am using a slightly larger servo for the Z axis, which is actuated by a 2005 ballscrew.  When left unconnected this axis will fall under the force of gravity, but the servo provides enough resistance to hold position even when the machine is turned off.

With all three servos installed, the machine axes can now be controlled from the laptop.

The spindle is powered by a belt drive kit that I installed in an earlier post.

The motor has a small junction box which must be configured to match the VFD output voltage.

The VFD requires 240V power which is only available in the workshop, so this is the first time I am testing it.  In its default state the VFD allows the motor to be controlled by the dial and keypad.

I'm just letting excess cable length hang down between the mill and the cabinet for now.

Configuring the VFD is a somewhat involved process because there are dozens if not hundreds of configuration parameters that might be useful.  Each command parameter is identified by a number and can be assigned a range of values.  Some are quite simple, such as motor direction which can either be forward or reverse.

On the other hand some parameters have dozens of possible settings and can interact with other parameters.

My configuration is fairly simple so I made a list of the parameters that must be set and entered them by hand.

Setting P0.01 to a value of 1 tells the VFD that I would like to control it using input terminals, and not using the control pad and dial.  This allows the PoKeys control board to adjust the spindle using an analog output.

I then tested the spindle in forward and reverse up to 2000RPM.  The spindle can go as high as 4500RPM but I just replaced and re-greased the bearings so it should be gradually brought up to maximum speed over several hours.

Next I connected the limit switches.  There are five of these so I simply plugged them into header pins on the protoboard.

I manually checked that each limit switch was correctly set up by triggering them with a piece of metal and confirming that the computer registered a fault.

I then re-routed the X axis wiring so that it won't catch on any moving parts.  Routing across the front of the table might seem needlessly complicated, but it's actually the best way to keep the cables above the tray.

Finally I installed the sheet metal covers.  It's starting to look a little more like a mill now.

After testing the limit switches I configured and ran the homing routine, which moves each axis to its positive travel limit.  The machine has roughly 18.5"x8.5"x20" travels.

I also confirmed that every axis was moving as expected with a dial indicator.  This is a great way to ensure that the steps/inch math is correct and that there are no loose components.

The excess cables might look like a mess, but they are actually separated by power level to reduce electrical interference.  On the far left are the sensitive encoder and limit switch cables.  In the middle are the three servo power cables, and the high power AC and VFD cables are on the far right.

The 240V cable is particularly nasty as it is unshielded and receives VFD noise.  If the 240V power cable gets too close to the encoder wiring it will fault the machine, which is concerning.  It is not an urgent problem, but I hope to make the system more tolerant by improving the encoder shield grounding and adding ferrite to the power cables.

The system looks quite minimal and clean from the top.  Sometimes I feel like most of my time is spent cutting and terminating wires.

Replacing the way covers finally makes the machine ready to run.

The first cut on the machine isn't too exciting, but confirms that everything is working smoothly.

The last addition to the mill is a lightweight enclosure assembled from strips of leftover insulation foam.

Heavy-duty plastic drop cloth and tape completes the enclosure.  I really do not want coolant or metal chips landing on the electronics or laptop.

The mill is now complete.  I'm never fully confident in a machine tool until it has made a few parts, but I'm going to try machining a new extruder for the 3d printer and see how it goes.