Video for Jens

Here, at reader request, is a video of the mill in action:

More aluminum lessons

Today I managed to make a couple of circular cuts all the way through, but then the bit fell through the hole it had just cut.  The dimple is from a previous attempt.  Some of them are a bit off circular.

Lessons:

• The bit tends to creep up, so put it as far in as you can from the start.
• Clamp the metal down on both sides
• If things start making loud vibration noises, stop everything and see what is loose.
• I tried lowering the feed rate from recommended 500mm/second to 400mm/second to reduce some of the noise, but I'm not sure if that made much of a difference, or if I just should have tightened things down.

Lessons learned from milling aluminum

• No manual steps for the g-code generation.  The workflow that I came up with for the top brackets is:
  •   Draw in Inkscape
  •   Export to SVG
  •   SVG into makercad
  •   Manually adjust the vertical position for the SVG
  •   Add profile operations
  •   Export g-code
  The problem is that the manual positioning made the process impossible to reproduce.  In my case, I wanted to change the feed rates in the middle of the first run, but have it mill the same workpiece. 
• If the piece around the spindle's collet starts to shake, stop things before it falls off.
• The settings from Inventables for milling aluminum work.
• Mill in the center if you want to avoid getting aluminum dust on the y-axis rails and carriage. 

Drawing the top bracket mounts

To mount my 80/20 25-5010 extrusion to my two Makerslides, I need to mill some brackets for the top.  Doing the design has been a bit harder than I thought it would be.
My first thought was to use Easel, but it doesn't let you specify locations in absolute terms with numeric input. Since the top bracket has to line up with the T-slots in the three extrusions, that rules out Easel, at least for the design.
Since the top bracket is a simple milling it seemed like it would be worthwhile to dive into generating the SVG for it by hand, and then import the SVG into Easel.
The first complication was spending a lot of time looking at the SVG arc, which seems to lack a clear explanation, although there is this excellent demo.  Having had some difficulty working with SVG's arc, I decided to use a Bezier curve instead.  You don't have to figure out the style for the curve on your own; there is a nice page right here that lets you dynamically adjust things.
Or, I could just edit things in Inkscape, which turns out to be much easier.  The only complication is that I have to specify the drill locations for the holes via bounding box, instead of center point.  This is painful when you want to adjust the hole's diameter.
But here are the calculations for the top bracket.
An extrusion bracket for Makerslide is 14mm wide, which means we want to have 3mm of clearance from the Makerslide's v portion, meaning we'll want the section of our bracket over the Makerslide to be 17mm.  The 25-5010 extrusion takes up another 50mm, and then the other Makerslide takes up another 17mm, for a total of 50mm + (17mm*2) = 84mm.
The T-slots for the Makerslide are centered, but our bracket section isn't, which means that the holes need to be centered 7mm from the edges of our bracket.  The 25-5010's T-slots for the  25-5010 are 12.5mm in from the edge of the 25-5010.  That would give us drill holes centered at Y=7mm, 29.5mm, 54.5mm, 77mm.
But Inkscape won't let us put in center coordinates.  If our drill holes are to be 6mm in diameter, we need to subtract 3mm from each Y position, for 3mm, 26.5mm, 51.5mm, and 74mm.

Ready to bootstrap

After a bunch of work to stiffen everything , it is ready to start cutting aluminum to attach my makerslides to my 50mm by 100mm extrusion for the x-axis. 

Next purchase list

This is what I'm going to need to stiffen the X axis, plus a few parts to broaden out the table.

One 25-5010, 1800mm long.
Four 25-2575 1830mm long
eight 25-4113 4-hole corner inside brackets to join the 25-2550s.
thirty-two bolt assemblies (75-3404 m6 x 10mm w economy t-nut) to tie in the 25-4113

To stiffen the vertical legs, which are wobbling a bit when we move the table around we'll add joiner plates on the middle legs to see if it helps.
2 25-4165 8-hole joining plates
sixteen bolt assemblies (75-3404 m6 x 10mm w economy t-nut)

six 25-4132 2 hole inside gusset corner brackets (2 for the midspan and 4 for the corners)
twelve bolt assemblies (75-3404 m6 x 10mm w economy t-nut)

When we switch out to larger motor mount plates, we'll also need screws that are .15 inches (4mm) longer (because the motor mount will be .25 inches thick, not 12 gauge.  That works out to:
four M5 x 34mm Button Head Cap Screws
eight M5 x 24mm Button Head Cap Screws

We'll also need to extend the carriage for the Y-axis, but this is enough for one night.

How's that for stupid?

Last night I got most of the new motors hooked up:

Naturally, the next thing I did was to hook up my gShield to verify that the motors would turn when commanded to do so.

But it didn't!  Even though all of the wiring looked good for the Y axis motor.  There was a brief sound every time I clicked, but no motion.  I double checked on the inventables website that I could drive my motors with the gShield, and sure enough I could.

This evening I took a look at it again.

I'd switched axis in my head.  The brief sound I'd heard was the other axis's motor.

Sigh.

It made buying a terminal block at Radio Shack, where you have to wrap the wires around the screws rather than sandwiching them between plates, look relatively minor.