These chips don't lie (aluminum on LowRider2)

I’ve had the Lowrider2 for about 2 years now. I’ve tried aluminum several times over this period, generally with very poor results. I concluded that aluminum is best left to the more rigid MPCNCs.

My first mistake, like most, was using endmills with too many flutes. With 4, 3, even 2 flutes, I might get a few minutes of cutting before the mill inevitably clogged up with soft aluminum. I also simply couldn’t get a handle on feeds an speeds. Metal is harder than wood, so I should feed slower and spin the router slower, right?

With a project at work that needs a bunch of simple milled parts in a hurry I decided to give it one more go. I’m still trying to master the process, but I’ve changed my mind: Yes, the Lowrider is absolutely capable of acceptable aluminum parts (depending on your definition of acceptable of course). And as I’m pretty sure it was Shakira who said, these chips don’t lie.

(AAA cell for scale)

This is an example of a part I made. I’m still figuring things out… like why I lost steps on that one tab, but I’m getting +/- 0.1 mm in one axis and more in the other axis.

I even made a successful tool change to cut the chamfers. The part took about 25 minutes to cut.

Anyways, what made this attempt work better than all my previous was to do the opposite of what my intuition told me:

  • Use a 1/4" endmill instead of 1/8 for the major slotting and clearing work. I’m using a single flute carbide upcut spiral flat endmill. It has so much more room to clear chips. Previously I only tried 1/8 bits. I also think the 1/8 bits flex too much for the high side forces.
  • Stay sharp. I usually buy the 10 packs of bits from China for a few bucks. I didn’t break the bank, but I did invest in an $18 HRC55 Tungsten Carbide bit that is as sharp after an hour of cutting as when I got it.
  • Go deep. It will happily cut a 2-3 mm slot without issues. Previously I was shaving 0.2 mm at a time.
  • Go fast. 1 mm slot at 500-600 mm/m, 2.5 mm at 250 mm/m. Previously I was going 100-200.
  • Spin fast. You can’t do the last two without this one. It works best at the tool’s max speed of 30000 rpm. It also forces the chips to evacuate, flinging them many metres away.
  • Use trochoidal stepover. I knew this in the past, but it is so much more effective with the larger bit.

It is all about keeping the tool and stock cool. Aluminum sinks heat so if you bite hard, it can take it, but if you move slowly, take small bites, or use a dull bit you just create more friction, which softens the aluminum and clogs the bit very quickly.

I’ve also been using cutting oil, which I don’t think makes a big difference, but doesn’t hurt. I use chainsaw chain oil. Its meant to stick to parts.

Anyways, I hope these observations help someone else. Any ideas how I can compensate for the backlash in the one axis?

Here is some video (turn your volume down).


Follow up. After posting this I cut about 20 parts for various plates and weldments. While not comparable to any industrial machine, the results were more than acceptable given my tolerances.

Having made all those parts I can confidently say that my number one tip if you want to cut aluminum on your LowRider is use a 1/4" endmill. You can bring out the smaller bits to get in tight spaces but there is so much more room for error. Here are examples of my feeds and speeds. You can be much more aggressive but I prefer to let it run in the background instead of watching it.


I’m going to try your settings. I have a few 1/4" around here and I have found the same as you 1/8" slotting is not a great idea.

Let us know how it goes. I just had another project for which I needed to cut about 20 parts including a few 1/2" thick. I was actually using about 2x the speeds shown above. On one occasion the chips didn’t clear out of a deep slot and they clogged up the endmill. I mostly blame my patience.

1 Like

I forgot to mention how easy it is to do edgefinding with metal parts. I wired an additional “sensor”:

endmill -- alligator clip -- wire -- alligator clip -- capacitor -- wire -- input; and
metal part -- alligator clip -- wire -- alligator clip -- wire -- ground

and then I wrote a snippet of gcode to touch off one corner of the part consecutively in all three axes.

The results were highly repeatable; I even had successful alignment on two-sided parts.


Those might not be pro numbers, but they are far beyond anything I dreamed of when this whole thing started. Numbers I am pretty proud to see you getting. Thanks for sharing!

Making me proud buddy!

1 Like