- Z axis is going up (to the ceiling).
- X and Y axis is usually machine-dependent; usually the longest axis is configured as the X axis (since usually most works are landscape oriented), but I think it's really a personal choice depending on what you'll do. That's why in gantry stile machines the X axis goes towards you, and the Y axis is the gantry. In a metal milling machine, the Y axis might as well be going towards you instead, depending on the design.
- If you're aiming for 0.001" (0.0254mm, since 1" = 25.4mm), which is quite high considering your budget, you'll need ballscrews (hopefully 2 nuts on each axis, but it might be overkill depending on the single nut backlash) or maybe a nice ACME setup with a couple AB (anti-backlash) nuts, but I'm not sure about this one. I'm not sure if the ballscrews you got are good or not, so we'll just have to wait until you get them and see how much backlash is there.
- You're not going to get 18" of travel with the ballscrews you bought, since you must consider the way you're going to mount them, coupling, nut width, etc... Depending on the design you might even have to consider the z-axis width (which you must consider when getting rails, too).
- 16mm supported rails are probably enough for your needs, and they are not so expensive if you get them from china; search for linearmotionbearings2008 on Ebay, his name is Chai and I've never had a problem with him (I have bought quite a few rails from him in the past, and even have some ballscrews on the way). I'll be cheaper than Glacern, even more considering you're in a budget. Unsupported rails on Z are usually used because of the shorter lenght, but I wouldn't use them for aluminum machining. Supported rails are sometimes easier to install, but that's relative.
- Go for a fixed gantry, moving table design if you're mainly going to work with aluminum and other hard materials, it will add rigidity. I'm not sure you'll be able to build a reliable steel working machine with your budget, that takes quite some bucks.
Check this machine out, it might be a good reference to what you want to do:
Rocket Sparrow
- The router is an important factor; for some materials you'll want to reduce RPM; the problem is that most adjustable speed routers lose torque at lower speeds (
www.SuperPID.com - Super-PID Closed-loop Router Speed Controller is a good solution, but it's probably out of your budget). Forget about the Dremel and other low-end tools, they're just not up to the task. Another factor to consider is the runout of the spindle axis, since it will affect the cutting diameter. A precision spindle is expensive, but some quality routers are nice and might be enough for you (but might still have a runout larger then the ballscrews's backlash). Cheap routers are usually no good and have horrible runout.
I suggest that, once you get your ballscrews and check they're good to go, make a cad drawing of your design including them with the mounting method you choose, and with that data it'll be easier to figure out the rails' lenght (linearmotionbearings2008 has tech drawings on ebay, so it's easier to have the measurements figured out beforehand). I always do that and it saves me a lot of trouble and time, so I just have to assemble once I get my parts. Beware that the chinese rail supports' holes are made by hand and might be off by a milimeter or so (not a big deal in most cases), so it's a good idea to leave the mounting holes for the last, when you actually have the rails.