Carbon fiber CNC gantry beam designs

General Composites Discussions Engineering Talk
#1

How would you guys design a cf gantry beam if ultimate rigidity (stiffness) and strength (in that order) was the goal?

This cf CNC gantry beam was designed by a CZ company using a filament winding process:

http://www.cz.compotech.com/content/images/MACHINE/IMG0404.jpg

I believe the tubes in the corners are steel. The tubes are drilled and tapped for the rail screws.

Obviously laying the cf weave in the correct direction will be key for any cf part design but in this case, I’m not convinced that the CZ company gantry design is optimal given that most of the force would hit at 90 degrees to the rails.

I think that the design would be better if the direction of the tubes was flipped 90 degrees and if they used a bunch of smaller tubes instead of just 4 long ones. I also think the rigidity would be improved if the structure was not hollow.

I am currently building my own cf gantry using a rectangle tube mold. Instead of making it hollow, I want to fill the void with a mix of carbon nanotube resin, chopped cf and a lot of cf tubes at 90 degrees to the linear rails. I have also sandwiched in some steel plates between layers of woven cf for the front face so it can hold the screws from the rails reliably.

How would you guys design a cf gantry beam for maximum stiffness and factoring in the need for it to be drilled and tapped for the rail screws?

I don’t have filament winding gear so I am using regular female molds with a mix of woven and random matrix fabrics. I have already made the front face (that will hold the rails). I am still playing around with the design for the rear half of the gantry.

Do you think it could be stiffer if I used a different shape? E.g. If the tube had a rounded back like a “D” shape tube. Or maybe a cf I beam like the ones sold by Dragonplate.

#2

This is where I am up to with my cf CNC gantry design:

The side supports are cf rectangle tubes 2" x 6" x 12" with an approx wall thickness of 1/2". The inside walls have 1/8" steel plates (2" x 12") on each side (sandwiched between layers of cf) so they can hold screws for connection to the central gantry beam. The void is filled with resin, chopped cf and high density urethane foam.

The main gantry beam is just a face plate so far. It’s approx 3/4" thick of solid carbon fiber with an additional 1/8" of thickness from 2 steel plates sandwiched in there (to hold screw holes / threads for the rails). The cf is a mix of woven twill, some non-woven random matrix fibers, some tri-axial fabric and 12k chopped 1/4" stands. Aside from the first two layers of cf (which use clear resin), carbon nanotube resin is used for the rest.

I’m not sure about my design yet. There is a good chance I will start again and make it all as one piece so this one could end up just being the plug.

#3

Looks like a great way to get a lot of stiffness for machining without having to machine a lot of steel. I may have to steal that if I ever get around to making a cnc…

#4

It is turning out to be a good project, especially for a home CNC machine where you want the stiffness without making it too heavy to ever move.

The more I learn about CNC machines, the more sense it makes. The properties that are important are stiffness, vibration dampening and thermal stability. Steel has the stiffness but it’s not so great for the rest.

#5

what about throwing some cork core or flax fibre in the mix for some more vibration dampening?

#6

I’d never thought about using those materials. How do they compare to carbon fiber for vibration dampening?

I am using epoxy granite for the base because of it’s superior vibration dampening properties to steel and aluminum but I hadn’t thought about it for the gantry.

I was going to use a thin layer of urethane rubber for the underside of the base to stop it moving. I always love experimenting with new materials though. I’ll check them out.

#7

So I made this tube for a design competition. It weighed something like 3.5 lbs or so and ended up failing at something like 12000 lbs on a 3 point bend test with something like .01 deflection… I’m sorta making the numbers up but it was along those lines.

My point is that this thing was ridiculously rigid for the amount that it weighed. Partly I think due to the materials used but also the geometry. An I beam, or box like this, is a very strong structure. This one is probably over kill for something like you’re making but, could be tuned back to optimize for particular stresses.

https://ibb.co/fvcpgk
https://ibb.co/dqcpgk
https://ibb.co/grwRSQ
https://ibb.co/mLiHZ5

This was made using dry fiber and a wet layup… i did put it under vacuum to help spread the resin out. It was then cured in an oven and a closed metal mold and compaction came from two silicone mandrels. I’d say the quality is autoclave… everyone at the competition thought we had autoclaved it. :wink: You can see where it failed, the rest of the part is in perfect condition.

#8

That does look like an excellent design. I am not surprised it did so well at your competition.

My design uses similar principles except, instead of the flat internal support plate you have running across the middle, I have used approx 70 tubes of .75" OD. The tubes are set in a mix of chopped cf fiber’s and epoxy resin so it is no longer hollow.

I then have a second almost identicle layer with 6 longer tubes mounted parallel to the surface plate. Each layer is 3" x 8" x 36" so the finished part is 6" thick.

I am also not surprised that you were able to replicate an autoclave quality finish using silicone inserts. I use them myself all the time. I find them to be easy to use, convenient, quick and, if made right, excellent performers.

An autoclave just increases the effect of the pressure from the vacuum. I achieve the desired increase in pressure by using weights on top of the inserts. If I want 100lb of pressure, I use more than 100lb of iron to hold the silicone in place.

it is essentially a basic and user friendly form of compression molding. The rigid compression molds I make are better but a lot more work to use.