So many options and questions

I used 1500mm v-slot 2080 for the y axis and a doubled up 2080 at slightly less than 845 mm for the x axis for my build and only got 1277 mm by 667 mm of usable work table (which was expected after modeling it in Sketchup). That is 26.25 inches by 50 inches. My goal was to be able to cut 26 x 48 inches for cabinets. I could have made it wider by using the full 1000 mm v slots for the x, but I wanted it to be as rigid as possible while being able to cut the specified size, and the machine already takes up a ton of space. Also, I want to build an egg-bot and the cut-offs will work perfectly for it. I chose 1/2 inch acme screws from McMaster Carr for the Y. I get no screw whip at 5000mm/min speeds. Here is a link to my build. Lead Screw Driven Ox Derivative (850x1500) The sketch-up file has been modified numerous times and I haven't had a chance to update it. Many things have remained the same most importantly, the length and width. There is an updated sketch-up picture in the build showing how I tied the Y axis to the torsion box table.

As for flexing issues, my system seems to be very rigid. The weakness (of course) would be the z. To help eliminate that, I have different thickness spoiler boards I can mount to raise the material closer to the x axis.

I use a Makita router and am surprised how quiet it is. I really like it so far. For electronics, I used an Arduino with a shield and DRV8825 drivers. However, I am currently upgrading to steppers like Openbuilds sells. I bought them last fall prior to Openbuilds selling them and I paid the same is what the OpenBuilds Parts Store sells them for, but had to wait a lot longer and had to worry about them being Wantai knock-offs. Luckily they weren't...I think.

I believe the v-wheels were designed for the v-slot rail only. Others may know better.

 

If you're going to start with electronics first then spring a little extra for a higher overhead. The DQ542MA's sold here in the store max at 4.2A which is plenty enough to run any nema 23. You may find out later that your motors need to be replaced by ones with more desirable characteristics to accomodate your machine. 2.8A-3.2 is a common range for these motors and many of the hobby combo driver packages have a hard time maximizing full use of the motor. A 4A driver is plenty enough without overkilling it. If you're considering adding mass with steel parts later on then deff invest in the electronics and don't think you'll be getting away with hobby combo boards like an Xpro or TG. I spent $30 to learn this lesson.
Z-axis. Yeah. If you find a good slide for a Z then go for it. This is what I did. There are a lot more options for this on eBay now compared to even 3 months ago and the prices are getting real competitive.
Plates. Plenty of builders are using aluminium to mill aluminium. The C-beam consist of aluminum and mills it just fine. Look up The Hog build. Toolpaths are another key part to this. You'll need a piece of software (and learn it) to allow you to maximize efficiency of your machine. This comes with a bit of trial and error by pushing its limits. Just like with a new car. It's even more true with one off machines (think custom car). I highly recommend learning fusion 360. It puts professional toolpaths that are on par with Mastercam and the like and places them within reach of everyone just wanting to learn, do hobbies, or begin to start a small business and at the cost of nothing. It's more than capable enough to run a business with and seat costs for commercial use is ridiculously cheap. So you would be ahead of the curve if any of your projects kicked off. I spend a bit of my professional time in SW, AC, and MasterCAM and I'm sold that Autodesk is playing a really hard long game with this one.

Spindle. Just as others are already doing, you'll get away with occasional Al. work with the routers. To me the sound and weight are the biggest factors here. I choose a lightweight .8 kw over a trim router and it's a beast through woods and nonferrous metals and I can talk on my cell just fine while it's running with no vac. Admittedly, I am using a top end VFD with fancy bells and whistles and my system is really ridged (much less than .001" vertical deflection with 100lbs above spindle) and I could likely move up to a 3kw, but I don't need that. To me, bigger spindle means more weight, more mass, more deflection, and longer jobs. To combat this can mean more money for better electronics and motors, structural upgrades/downgrades, and ultimately a complete reevaluation of the system in general which is more wasted time. A large spindle on a flexible profile beam could potentially create some stability because of its weight, but once it hits a resonance with the tool speed and cut it's going to be a nasty outcome.
V-wheels. Just buy OB wheels. The knockoffs aren't guaranteed for quality or consistency. Btw, steel wheels in aluminum profile will quickly upset you.
Belt vs rack. If you can afford a R&P setup then go for it. Spring loaded is the only way to go. There are some great R&P examples here on the site. Belt in belt systems are also successful. I can't speak much to the limits of such systems. Steel reinforcement. If you have an idea for reinforcing an aluminum frame with steel then please share it. We love seeing individual minds at work and enjoy offering constructive feedback.
Well, that was my two cents for the moment. I still don't understand what you're talking about as far as the vertical mounting plate for larger work. Do you have any models or napkin sketches?

Joe.

 

Man, that Frank guy is really cool. His workshop is great, and that CNC is something else!
He also makes a really professional video.
Thanks, I enjoyed watching that.
Gray