Workbench Mounted Gantry for Timber Milling

Overall, I think you are setting yourself up for disappointment. Aluminum extrusions are nowhere near stiff enough to accommodate what you're trying to accomplish here. A 3hp spindle dangling down from a high gantry will follow the path of least resistance and in hardwood that is rarely a straight line. The other issue noted here is the wings supporting the gantry. These will lack the stiffness to keep the gantry from riding up and over tough spots in the wood. They will also tend to want to pry the Y-axis rail off the face of the bench.

The key to proper CNC design is to keep it as low and tight as possible. This provides the best possible stiffness in the system such that the bit stays engaged properly with the material and follows a straight line. Design for exactly what you intend to cut now not the 'someday I mights' as every little bit of overage weakens the system and strength is key to a proper cut.

Keep us informed. We are happy to help no matter what path you choose.

 

You'll also need to bring the sides in. But let's start with the basic question, what sizes and thicknesses do you intend to cut?

 

Much better but you're still reaching down too much. I suggest taking a look at how the gantry on the OpenBuilds LEAD CNC Machine 1515 (60" x 60") | OpenBuilds is framed. The single rail on the LEAD CNC 1010 (40" x 40") | OpenBuilds would probably also be viable due to the short distance you're spanning. Only difference from those examples I would suggest would be leaving the 20x80 side verticals extra long, extending down past your attachment points. That way you can keep it low for normal use and then should you ever need a bit of extra thickness you have the capacity to raise it up.

Other build concepts from the past that you may be able to draw some inspiration from include Slab Flattening Table, Need help with x large build for slab flattening, and surfboard machine. These are all stand alone systems but could easily be morphed into tabletop overlays.

 

You can run as shown but it's not ideal. The full loop system you have shown will have significantly more stretch than a single run belt which means increased backlash. For this application I would probably use a single run 15mm belt.
Threaded Rod Plate - NEMA 23 Stepper Motor (openbuildspartstore.com) will mount on the face of the C-beam to hold a NEMA 23.
No but it's available in 1500mm lengths and then you use Quad Tee Nut - MAKERLINK (10 Pack) - OpenBuilds Part Store to splice the lengths. With careful alignment you won't know the splice is there.
Yes, but it's easier to mount them as noted above.
The simplest brake mechanism is a threaded knob run through a hole in a plate on the moving part engaging a t-nut in the fixed rail.

Some basic thoughts on what you have shown:
Turn the Z-axis around, fixing the Z-axis beam to the X-gantry plate and having only the Z-axis gantry and router move. This allows you to reduce the height of the overall system which in turn makes it stiffer. Set the bottom plate of the Z-axis beam just above the typical material thickness with the X-axis down as tight as possible to that.
The timing belt you have as shown may be useful in the period up until you do automate the system. You are going to have an issue keeping the Y-axis gantries moving at the same rate, keeping the system square. I would suggest putting in a connecting rod across the end with fixed pulleys at each end causing both belts to move in unison.

 

Not sure what constitutes a belt and pinion. Belt starts at one end, crosses over and around the motor pulley, then goes backward a few inches over/around an idler pulley, and then off again forward to the far end. Think in terms of a horizontally stretched, vertically flattened 'S' with pulleys forming the curves.

 

Motorized version needs to be attached to the gantry plates with a single run belt going end to end.



As for the S style belts, this is what I was referring to:

This style allows for greater contact between the belt and the timing pulley.

The upper concept will be easier to integrate into your system though. I've attached a very rough Sketchup model. I didn't have time to track down the proper pulleys so the belt is not shown 15mm width and the idlers are not shown as smooth face roller bearing idlers. Plate attachment screws are not shown. You'll notice that the placement of the motor interferes with your gantry risers but the gantry risers shouldn't be centered anyway. The center of mass of the whole gantry assembly including the spindle should align with the center of the Y-gantry plates.