Light weight CNC?

I would like a CNC router that just barely fits in the boot of my car with the seat folded down (Kia Soul EV), about 1mx1m, and can be easily lifted in and out by me, who is not a power-lifter. (I do CNC workshops at various maker spaces.) I already have a Handibot, which is great (about 20kg), but the work area is a tiny 150x200mm.
I would like to cut mostly wood and sometimes brass or aluminium, with a precision of better than 0.1mm.

I was thinking I may have to design my own CNC, unless you know of one out there that was designed specifically to be light weight.

Here are some mass-reducing ideas I am considering. Please add yours, and comments about mine:

1. Use hollow carbon composite tubing, as used in RC plane wing spars, for both the rails and the structural elements of gantry. No steel rails or heavy aluminium extrusions.

2. Use outrunner brushless DC motors, with cogbelt speed reduction and encoders, instead of stepper motors.

3. Use a sheet of heavy-duty construction closed-cell foam instead of MDF or plywood for spoilboard. (I tend to fix my work
down with double-sided tape on masking tape.)

4. Use belt drive instead of leadscrew/ballscrews. I am thinking of Contitech Silentsync dual helical offset belts and pullies.

5. Your ideas?

Hi Steve,
You can make a "big" router say 1mx1m that you can lift but it won't be very stiff. Will cut plastics and timber, plywood but not aluminium. Here's one I made for the same purpose. Does all stuff but not metals. I have upgraded the Z Axis so its a bit stiffer but have not played with aluminium yet....

Its available as a kit. It can be easily lifted by two people and I can just manage it myself. To make it lighter I'd use lightweight plywood and I would also make the electronics in a separate box with a plug or plugs. These electronics are on a plug but its wired permanent. Its footprint is 750x750mm..

To lighten;
1) use 15mm rails not 20mm (maybe) if 3 packages I'd stay with 20mm
2) have electronics in separate box with plugs
3) I can see if I have a connection on the columns then the gantry would be easily separate from the rest. eg now I could undo 8 screws and it would become 3 easy to deal with packages.
4) use lightweight plywood vs furniture grade plywood. I use lightweight ply alot in projects and its half the weight of normal ply

To your notes:
1) CF tube is light but not stiff. Aluminium is stiffer at same wt. I've been involved with composite structures for a long time and it would be quite a costly project to get where you think you want to go with CF tubes
2) Nema 17 steppers are not heavy and then you don't have to develop your own control system
3) fair call
5) Brevis the machine in the pictures uses a 10mm AT style belt. Surprisingly stiff and light. Surprised me and I have cut aluminium poorly with it. But Brevis' weakness was a wobbly Z Axis. Brevis was not designed as a router but as a vinyl cutter, plotter and laser demo machine. I set up a router because everyone kept asking for a router vs the other things... If doing it again I'd keep everything the same but upgrade the Z axis to UBER stiff for carving and routing.

As is I'd guess Brevis is 30kg so I think I could get it to <20kg in 3 packages quite easily.. Peter

Thanks, Peter, for your helpful comments and suggestions.
Carbon fiber tubing can be incredibly stiff, of course depending on how it is fabricated and its dimensions and materials. Here are lots of details about the options:
https://www.rockwestcomposites.com/s...n-fiber-tubing

Definitely I would agree with

5. Put all the electronics in a box I can unplug and lift separately. I should have mentioned that.

I agree that plywood is a good way to construct things cheaply and quickly. However, I did not say this was to be a low-budget build. I don't think I would use even light-weight plywood. The wet climate of Ireland means that dimensional stability would be a problem for any wood design (though I could seal it very well with paint). Do you find that to be a problem with Brevis?

Probably I would use a honeycomb carbon fabric epoxy composite for any flat structural members.

A removable gantry is also a good idea, but it would have to be very carefully designed so as not to contribute to flex or backlash or breakage on assembly and disassembly or heavy loads. If any of you have seen or thought about how to implement that, please post a reply!

Peter, if you have your stiffer Z axis installed, I would love to see any measurements of deflection (X,Y, and Z) under load that you could get from it.

Hi Steve,
Yes CF can be very stiff in bending. But its poor in torsional stiffness and gantries and columns have to be very torsionally stiff. I have designed several machines with the intent to use composites and aluminium beats it every time on the gantry due to its superior shear modulus. I design very complex composite structures in the marine and industrial area and have specialist composite software. Plus many years of test data so these things are quite accurate to model. I think you will find the same result with CF flat panel. It will be very stiff in bending but poor when twisted. Aluminium has a shear modulus of 27GPa and its near impossible to get standard modulus CF past 12GPa in a laminate then you have poor bending stiffness (typically G=9GPa) . ie you have to use twice as much CF to get to the same torsional stiffness which puts you nominally slightly heavier then AL. Al density 2700kg/m2 and CF 1500kg/m3... I build quite high spec laminates and parts for various projects so understand these issues.

I'm not saying that a CF machine is out just that in this case CF may not be the answer. Aircraft grade AL is available in thin sheet and will build very stiff (bending and torsion) sandwich panels. If you have humidity and weather issues honeycomb is out, foam is in. The cells of honeycomb expand and contract and can accumulate moisture as weather changes. This does happen on boats if the skins are not perfect. laminates can behave as semi permeable membranes... Foam can be just as light and stiff in this instance without the manufacturing or practical issues with honeycomb.

Furniture grade plywood is not cheap!! Maybe in europe but here birch is quite a few $$$. But I like it very stable.. From years of designing cnc machines and other machines 3D printed metal is the next step for many of these things vs composite construction. if you can use intermediate or high modulus fibres then you have a design break, but printed titanium and aluminium will outdo this sort of thing. Plus you don't pay for a mould which for a one off can be the same cost as the part....Cheers Peter

Thanks, Peter, for lots of food for thought. That is good news that Al beats CF, since it is so much easier to work with.
What about an alloy of magnesium? I think that is about as strong as Al but only 3/4 the weight. Do you have any experience with Mg alloys?

I was imagining TIG welding a trigonal truss structure out of Al or Mg rod or tubing.

3D printed metal is also an interesting idea to consider, but I imagine they are not available in long (>1m) pieces.

Hi Steve - Magnesium has a modulus of 45GPa and a density of 1500kg/m3 from memory. It corrodes very rapidly and it tends to ignite when machined if your not careful. The swarf can spontaneously combust as well... Recommend Al over Mg. I have been involved in a welded Mg bicycle frame many years ago. The company only made one then went to CF. The largest 3D metal printer that I know of is 600mm, but even if you make smaller bits you design them for bolting or gluing together. TIG welding is fun but you end up with lots of distortion, I used to teach TIG welding and I programmed welding robots for a while. There is a thread in here on biomimicry in which truss type gantries are looked at. I think you will find them to be not rigid enough and very complex to get right. They are great for beams in simple bending but a gantry and column is in torsion and bending and this makes it complex. A gantry is also relatively short and could be shear dominate which again makes it complex. Look up short beam theory vs long beam theory. Even nature has sidestepped making structures that require a high torsional resistance. If you have fusion 360 you can use its generative design feature to get clues for optimal; shapes of parts... Its easy to make lighter and lighter structures based on strength, its not easy to make lighter and lighter structures based on stiffness but its a worthwhile pre occupation.... Your answer may lie in concrete. UHPC is available to you and its easy to do. 2500kg/m3 so its potentially light and 70-80GPa so its better then aluminium. Peter

https://durcrete.de/?lang=en#

1) being a casting you can get the shape you want
2) being stiff and light its like cold casting metal
3) Its an off the shelf product in europe, they won't ship it to Oz so I can't get it :( Peter

Hello, have you considered buying a mini engraving machine, it will save you a lot of time, if you are interested, you can go to this website: https://forsuncnc.com/mini-cnc-router/