G0704 State of the Art in 2021

So I picked up a King Canada KC-20VS2 (a BF-20 clone) and I'm on my way to a CNC conversion.

What is the current state of the art in:

- Spindle bearings (I'd like to see 6k-7k RPM)

- Belt drive conversion

- Quill removal

Some things I'm already settled on:

- ArizonaVideo99 ballscrew kit

- ClearPath Servos

- Ethernet Smoothstepper & Mach 4

- I've cut 1.8" (ish) off the front of the base deck overhang for more Y travel, but I'm not going to cut rearward or do a rear column spacer

- I'm probably not going to do the 1-shot oiler & manifolds, partially because I'm not sure the juice is worth the squeeze, and partially because I envision an eventual conversion to linear guides - but I could be convinced otherwise.

All that sounds good.

Personally I'd go UCCNC, LinuxCNC, or Acorn before Mach 4. I tried M4 out with my ESS before I moved to UCCNC. I struggled with M4, and that was after years of successful (albeit buggy at times) Mach 3 use. I had oddball issues with M4, like the spindle randomly turning on. That issue may very well be unique to me, but nevertheless, I was not impressed. There's nothing like tearing the cord off my touch probe to really sell the software. I bought a UC300eth-5lpt and moved to UCCNC and my only regret is not having done it sooner. I can leave the machine on for days without issue. 10 hour programs run without a hiccup. I have run my machine lights off on numerous occasions. I go to bed,

As for way oil, it's invaluable on my stepper driven machine with ways. I can run the ways much tighter and still achieve reliable rapids. For servos, that likely isn't an issue, but if you are pushing 250ipm rapids on sparsely lubricated ways expect lots of wear. My line of thinking is that you should commit to one or the other. Either lubricate the ways and try to make them work, or cut them off and add linear rails. There's not much point in half arsing either approach. You are just wasting time. Pick one and commit.

I am of the opinion that a lot of the cutting limitations, i.e. the flexibility issues many encounter are actually just loose ways. If you go linear rails, this fixes that without the added friction. This in mind, I doubt servo's are really needed on such a small machine if it has linear rails. Now when you start adding a bunch of friction by getting the ways really tight, steppers can be marginal if you aren't careful. Servos would be advantageous with really tight ways as they will overcome the parasitic drag competently. This in mind, 250ipm across 16" or whatever on X hardly makes a difference compared to 150ipm even if you are doing long runtime items. The 30 second tool changes add far more time to the run than saving fractions on rapids ever will. Never mind manual tool changes that require you to physically tend to the machine.