New Machine Build- Torsion CNC Design

[kbailey90]

Torsion CNC

I've been working on this CNC design for the last 6 months. It is based loosely on how a Torsion Box is designed and assembled, hence the name. It has a modular Z Axis that can easily be changed for different sized routers, spindles, deposition heads (rep rap), etc.

The beauty of this machine is the absolute lack of hardware needed for assembly and it's rigidity. It can be assembled with nothing more than some clamps, wood glue, and a rubber mallet. It is a very strong design, and hopefully with testing will be able to mill aluminum and mild steel pretty well. When we get all the kinks worked out, we're going to start selling these as kits.

The CNC table not only provides structural integrity for the gantry, but also eliminates the need to build a special table to place it on. This design uses 2 lead screws for the X axis to eliminate racking and to minimize the footprint of the machine. This unfortunately raises the cost of the machine, but that is somewhat offset by not having use fasteners to hold everything together.

First and foremost the overall design priorities was rigidity first, modularity second, and serviceability third. All the wear items (bearings, anti-backlash nuts) are easily accessible and designed for maximum strength at a minimum cost.

[hemsworthlad]

Originally Posted by kbailey90 Torsion CNC It is a very strong design, and hopefully with testing will be able to mill aluminum and mild steel pretty well.

If this is a wooden or MDF machine then your in cuckoo land if you think it's going to mill steel.!! Even mild steel. . No matter how well built.
It will even struggle with Alu if you want it to cut at any decent depth and feed rate.?

Any machine made from any material will cut just about most materials.! . . . The big question is for how long..?
A wooden machine trying to mill steel or Alu with a depth and feed of any use will last about a day.!
If your happy with scratching the surface and have a few month's per job it may just work.

Suggest you lower your expectations. .

[kbailey90]

First, it's not an MDF machine. It is 12-ply Oak Hardwood plywood, which was the strongest, lightest wood we could buy. Secondly, in designing it, the fact that it is wood was taken into consideration the entire way. Bearing locations were spaced as far apart as possible to distribute loads to take in account the flex a wooden machine would have.

I weigh 230 pounds and can stand on the center of the gantry on one foot, and it only deflects downward between .125 and .1875 of an inch. For wood in my mind that is impressive.

Also I'm not naive enough to think this thing will be milling at 60in/min at .250 DOP. But I'm pretty confident this machine will mill at respectable rates in softer metals. As for steel, that's to be seen.

[hemsworthlad]

Dont get me wrong i'm sure you've built a very strong machine but honestly you are into a completely differant league when it comes to cutting metals, even soft metals like Alu,brass etc. . . Even small Doc with low feed rates will be a struggle.!
It's not that it wont cut them which I know it will, it's the quality of cut and affect it has on the life of the machine.
Wood just hisn't ridged enough material to dampen the vibration of cutting metals, this will transfer onto the job as chatter giving poor finish. The only way around would be really really really slow feed rates with very shallow Doc making most jobs, unless very small just to long to be of any use.

The other factor which will affect the machine is vibration which basicly will shake the thing to bits over a short time, even with shallow depths.! And if you try to cut with even average DOC it will drop to bits within weeks.

Honestly I'm not scare mongering, I cut Alu 99% of the time on my machine which is built from Steel and Alu and very very strong. Use's ballscrews on all axis and twin ballscrews on X axis with 2.2Kw WC spindle and I consider this the mininum requirement for cutting Alu.! . . . And i'm not talking deep DOC and fast feed rates either.
If I push the DOC/FR then even my machine starts to complain and shake it's head and will call time on me if I dont heed the warnings.

I did cut mild steel with it one time but it was obvious straight away that it would stress the machine far to quickly leading to premature failure.

If you plan to sell these kits then I suggest you dont lead people to belive it will cut mild steel or even Alu because they will be severly disappointed when it gives poor results.! Or worse falls to bit's within weeks of building.!!

Only trying to pass on my experience.

Good luck.

[RedskinsJBS]

I have to admit that your design is pretty neat looking. However, I have no idea what sort of comment to leave since your description of the machine's intended purpose seems quite vague.

With that said, it looks like you too a lot of care to keep to your design principles. I can see how the modularity is used; Am I right in thinking that the bed is made up of what looks to be 8 modular sections?

Judging by the hammer in some of the photos, am I right in guessing that this design is a nominal 2' x 4' cutting area? With a 4" to 6" Z?

At that size, lead screws would work well, but what if a customer were to want a 4' x 8' cutting area? Do you have a solution to this, or is this machine just intended for the under 60" crowd?

12 ply hardwood plywood sounds like a beast of a material. Did you look at birch plywood for your design? I'm very ignorant in this area, so I'm just repeating a commonly mentioned term.

You say that the gantry deflects ~.125" when you apply a 225 lb force to it (Your body weight). Since (hopefully) no one will ever apply that much force to the gantry while they are milling something, have you used a modeling software or some paper math to figure out the deflection under more normal loads... say 50 lbs? I'm really curious about what that would be. IIRC, I could walk all day long along some 2x6 joists in my garage and had no perceivable deflection (or bounce) to them. I'm sure that there must have been a 1/4" or so, but that was a 10ft span and a solid piece of pine, not hardwood plywood. I really do want to know how stiff you can make a modular plywood design.

Now when you say that you want to mill mild steel, do you mean cut out parts from plate? Or simply etch the surface? I haven't finished my machine yet, which is an 8020 aluminum design (I technically haven't started); and the people who build aluminum machines like mine rarely say that they would like to mill steel.

I understand slow and steady. I have used a dremel to cut a driveshaft.

Ah! One more question and then I'll stop being a nuisance. What sort of linear rail system / bearing system were you planning on using in your machine? This would certainly be a deciding factor about what sort of materials your machine could handle.

Thats all for now,

Jay

[kbailey90]

Yes it has a nominal cutting area of 27.6" by 49". I did that to leave some wiggle room for larger Z Axis Styles.

The Z depth can be a few different sizes based on a couple pieces. The kit will come with 6.5" of Z Travel, but can be ordered with a different gantry and Z that has up to 12" travel.

I'm working on a different design for a 4' x 8' machine. When I say this machine is modular, I mean swapping out different router mounts, different Z Axes, and different heights gantries should be very easy.

As for milling mild steel, that remains to be seen. I'm not going to advertise this machine as being capable of anything more that it can do reliably.

When designing, everything was a balancing act between rigidity and cost. I could of increased the number of joints and bracing 2 fold and perhaps made the machine 20% to 30% stronger, but doing so would have doubled the price to build it. That said, we're making this on an old Joe CNC design and can say already that it is much stronger.

If it turns out that this machine can only do wood than so be it, but I'm going to push it to see what it can do when it's finished. If that ends up being aluminum at .050 DOP, then I'll be satisfied with that.

For linear rails, it uses angle iron and V Bearing setup. For the drive screws, wer're going with Precision ACME rod. I'm also looking into a possible linear slides/ballscrew upgrade for at least the Z Axis. It's an ongoing project and has numerous enhancements along the way.

[ger21]

First, it's not an MDF machine. It is 12-ply Oak Hardwood plywood, which was the strongest, lightest wood we could buy. ........................ Bearing locations were spaced as far apart as possible to distribute loads to take in account the flex a wooden machine would have.

I don't want to bash your design, and I hope it works well for you. However.

Couple things. Correct me if I'm wrong, but from the looks of it, that plywood is the typical Oak plywood from Lowes or Home Depot, made in China. Only the face veneers are hardwood. Unless you special ordered a birch core plywood with oak veneer faces, which would be quite expensive.
Strongest and lightest don't go together. You get one or the other, or somewhere in the middle.

Baltic Birch is heavier, but far stronger than what it appears you're using.

And you shouldn't be taking flex into account, you should be designing it out of the machine.

I weigh 230 pounds and can stand on the center of the gantry on one foot, and it only deflects downward between .125 and .1875 of an inch. For wood in my mind that is impressive.

Not in my mind.

A torsion box depends on the skins to provide it's strength. Your design doesn't appear to have skins on it, so, it's not really a torsion box.
Buy a $20 hollow core door at home depot, with a cardboard core and 1/8" skins. Stand on that, and it'll deflect about the same as your machine.
I like to think that my router is one of the very first torsion box designs. I started building it in 2003 (documented here).
A 230lb load on my gantry beam will cause no visible deflection. Maybe only a few thousandths.

If I stand on my 60" long table (200+lbs), it will deflect less than 1/64", probably .01" at most. And it's just a few plywood ribs with 1/2" MDF skins glued and stapled on. Far simpler than yours, and apparently much stronger.

I fully expect my next machine (wood also) to be every bit as rigid as any aluminum extrusion machine you see here. But I know it won't be cutting steel.

Good luck.

[kbailey90]

Where is a link to your CNC. As I stated previously, there are lots of ways to make this machine stronger, but the idea is to sell it as a kit and at a reasonable price for people to get their feet wet in the hobby CNC market.

This design can easily be packed up and shipped at a low cost. If I was designing something for my own personal use only, it would be a completely different design. Sacrifices must be made when we're talking about a build it yourself kit you can order over the internet.

Also this machine is almost idiot proof when it comes to assembly.

Where is a link to your machine. I would like to see it.

Everyone on here can say that they have a machine that is stronger, faster, better, more accurate than what I designed, but can they say that when you weigh in affordability and ease of assembly.

Price is still up in the air but I believe We'll be able to sell this kit (wood, bearings, rails, leadscrews, and antibacklash nuts) for less than $1500 and make a profit.

When I say I took it into account, that's what I meant.

I appreciate any positive criticism, but "my machine is better, stronger, etc." is a waste of a post. If you have anything to add after looking at my machine that will make it a better machine without increasing it's price then I'm all ears.

[Markwind]

Originally Posted by kbailey90 If you have anything to add after looking at my machine that will make it a better machine without increasing it's price then I'm all ears.

it might be more cost efective to combine some multiple parts into single pieces to lower the part count ?

one of your slots in the gantry side plates looks like it almost cuts the plate in two, `do you think it would it be stronger to make that slot shorter?

[kbailey90]

I know which slot your referring to and actually making it shorter decreases the rigidity of the joint. The joints make a very rigid connection after there slid together and fastened or glued. This will be a connection I suggest people fastened together with screws because you can simply swap out the uprights with a different one we've got a file for and increase or decrease the gantry height to allow for more Z travel with a longer Z Axis or give up Z travel for even more rigidity. The default kit will come as a compromise between the two. There will be several options and customizations on how you can order your kit.

[Benonymous]

I like the design and seeing as it's cnc cut itself, it should be accurate dimensionally. I can't help agreeing with the other posters with respect to the metal cutting however. I have seen a home made gantry machine cutting steel. The machine in question was an all metal frame with supported rails and a decent spindle. With all of those factors in place, the owner of the machine was taking the most tiny nips off the steel part. I couldn't help thinking that the machine would be running continuously for a couple of days at least to make any sort of useful part.

Looking at your gantry sides, they a single thickness of ply with some bracing going part way down. I'd say that the flex in this area will preclude milling steel and might possibly suffer from flex with Alu.

If you can get this machine to market at $1000 with screws and rails it would be a really excellent starter machine. I'd avoid overstating its capabilities though. If you suggested that it was for wood and plastics, there would still be demand for it and you could leave it up to the owners to try and stretch its capabilities into harder materials.