Close

First entry

A project log for Concrete CNC

This is an attempt at a cheap but rigid CNC machine (budget approx. 2000€)

david-risbergDavid Risberg 08/25/2017 at 20:020 Comments

This project actually started in February, so this first entry is 6 months in to the project. Luckily, most of the interesting stuff will be happening from here on.

First up is a picture of the already finished cast iron compound table and the concrete base. The compound table was purchased separately, and has been modified with ballscrews with double ballnuts that are manually preloaded.

The holes you see in the concrete base are through holes. They will match ten M12 threaded rods that will be cast into the tower. There are countersunk holes under it for the nuts. The concrete has been cured at an elevated temperature, and then dried at an elevated temperature, to reduce shrinkage and warping. After this process it was coated with epoxy to prevent moisture exchange with the surroundings for the same reason.


next is an overview prior to assembly

closeup of the newly machined floating ballscrew mount

Front view of the compound table and the base casting. The base is hollow, and contains a polystyrene block. Wall thickness is about 50 mm, and total cross section is 250x250mm. This adds up to about 80kg of concrete and steel reinforcements. The compound table adds about 40kg.


Fixed ballscrew mount and stepper motor mount.

Compound table again.

Sideview of the compound table and the concrete base. Notice that the compound table is not bolted flat on the concrete, but it is standing on four M20 nuts that are threaded on M20 rods cast into the concrete. This makes the tilt adjustable, to make sure it can be perfectly perpendicular to the Z-axis. The M20 rods extend all the way to the bottom of the casting, and are welded to an internal framework that spans the entire part.

Compound table again. Steppermotors are rated 1.89 Nm at 3A. With 4mm pitch ballscrews that could reach close to 2 kN of linear force if friction is low. At any rate it should be sufficient, but feedrates may be dissapointing. I'm hoping 5 m/min, but expecting 3 m/min or less. Time will tell if servos are needed.

Closeup off ballnut assembly. One nut is bolted onto directly onto the block, the other nut can then be carefully rotated and clamped at an angle to eliminate backlash. This will done on all 3 axis.

Closeup of the adjustable feet for the compound table. The M20 rods are internally threaded M10 for holding the table in place.

Discussions