Critique my fixed gantry router/mill design

Hi,

Looking for feedback on my fixed gantry router build. Goal: machining aluminum molds with high precision (+/- .002"), and trimming carbon composites at high RPM (hence router not mill). This design is heavily influenced by earlier discussion on this forum, but just want to check in before I proceed further.

frame
24" x 24" x 3" granite surface plate base
8" x 6" x 1/2" steel gantry tube, machined, epoxy aggregate filled
5" x 2" x 3/8" gantry supports- welded, stress relieved, machined, epoxy aggregate filled
16" x 14" x 1" cast aluminum plate for table

linear motion
20mm medium preload linear rails - not hiwin but best quality chinese, fairly spendy
1204 ballscrews driven 2:1 with GT5 belts
Nema 34 steppers (leaving room for servo upgrade in the future if needed)

Controller
buildbotics control system

Spindle
yet to be determined, likely the spindle stocked by Avid CNC


Have priced out every component and it should come in well south of $5K.

Specific questions:

-What about replacing the 1" thick z-plate with 10x2.25x1/2" aluminum C channel? Would add much rigidity and bring spindle closer to gantry. Do I have any hope in hell at machining the web of a C-channel flat and it staying that way?

-Any drawbacks to 12 mm ballscrews? Screws are short so whip is not a problem here

-Have 0.5 mm clearance over ballscrew mounting blocks. Clearance is clearance, right?

-The y-axis table rails are mounted on ground flat aluminum bar. I'd prefer to mount the rails directly to the surface plate and put the spacers between the carriage blocks and the table as this would elimate issues of flatness with the ground bar. But elevating the rails would keep them from the water that will collect a bit on the plate (will drill drainage holes anyways). Pros and cons?


Thanks so much!

Also all connections are through bolted, holes and fasteners just not modeled yet. Largest reasonable fasteners used for every application

Quote:

---

Originally Posted by catahoula

16" x 14" x 1" cast aluminum plate for table

---

So I assume that's the amount of travel you want, 16" x 14"....

Quote:

---

Originally Posted by catahoula

1204 ballscrews driven 2:1 with GT5 belts
Nema 34 steppers (leaving room for servo upgrade in the future if needed)

---

12mm ballscrews and Nema 34 motors are a mis-match IMO. You're going to break something.

4mm lead ballscrew (4mm per rev?), with 2:1 gearing, so you want to make the screw turn twice as fast as the motor?

Just go with 20mm diameter, 10mm lead ballscrews (1:1, no gearing). If you use servos later, you can use a 3:1 belt driven gear reduction (so the servo spins faster than the screw).

Other than that, it looks like you have a decent conceptual design.

If it were me, I'd consider a complete redesign using surplus linear stages with leads of 10-20mm in X and Y, and perhaps a dual column rising gantry with moving table. That would add a some extra complexity to it while at the same time making it easier to build. At least that's what I'd do. Not saying it's wrong to do it your way.

Consider a brace like the one I've expertly drawn in red in MS paint. It will greatly increase the rigidity in that direction for minimal weight gain.

I'd keep the 1" Z plate but consider adding braces on the side of it to stiffen it up, shown in yellow, if required

Attachment 446638

Thanks, will consider braces. The z axis brace is what I was getting at with using a c-channel instead of plate, but bolting on braces like you suggested may well be simpler/easier to machine flat.

Quote:

---

Originally Posted by jones

Consider a brace like the one I've expertly drawn in red in MS paint. It will greatly increase the rigidity in that direction for minimal weight gain.

I'd keep the 1" Z plate but consider adding braces on the side of it to stiffen it up, shown in yellow, if required

Attachment 446638

---

Thanks-

12mm ballscrews was based on keeping a low profile so the gantry and table don't need to be excessively high. I'll look into load ratings.

Rising gantry is definitely cool, but I want to keep the axes guaranteed perpendicular without needing to reference off limit switches.

Quote:

---

Originally Posted by NIC 77

So I assume that's the amount of travel you want, 16" x 14"....




12mm ballscrews and Nema 34 motors are a mis-match IMO. You're going to break something.

4mm lead ballscrew (4mm per rev?), with 2:1 gearing, so you want to make the screw turn twice as fast as the motor?

Just go with 20mm diameter, 10mm lead ballscrews (1:1, no gearing). If you use servos later, you can use a 3:1 belt driven gear reduction (so the servo spins faster than the screw).

Other than that, it looks like you have a decent conceptual design.

If it were me, I'd consider a complete redesign using surplus linear stages with leads of 10-20mm in X and Y, and perhaps a dual column rising gantry with moving table. That would add a some extra complexity to it while at the same time making it easier to build. At least that's what I'd do. Not saying it's wrong to do it your way.

---