[HSM-Modal-530]

have you ever heard from the company in Germany www.eew-protec.de ? They build one of the largest lightweight 5-Axis CNC-milling-machine in the world with a carbon-fiber gantry designed as a carbonfiber truss.

[slopecarver]

I know this thread is kinda old but I wanted to add my engineering $.02.

Vibrations occur at an undamped natural frequency (f) in most CNC cases.
f=(sqrt(k/m))/2pi where k=stiffness in N/m and m = mass in kg

example likely not resembling anything in real life except for short gantries:
K=8740157 N/m, ~50000Lbs/in
m=88N, ~20Lbs

f=(squrt(8740157/88))/2pi = 50.157 Hz

If you were running a 2 fluted bit in a router at 25000 RPM that would equate to 50000 strikes per second or 50000 Hz which is very nearly 1000x your natural frequency which will cause some modal vibrations in your gantry. Now, finding a stiffness is not difficult for a constructed router (pull with 50 pounds and measure the deflection) but measuring the effective mass is. Alternatively if you had the stiffness and could somehow measure the natural frequency when the gantry is struck with a dead blow hammer you could get the effective mass.

Moving on

Adding dampening will do multiple things for you:

1: Shift the natural frequency so that it may be out of a harmonic zone that your machine would operate at.
2: Reduce the amplitude of any vibrations by converting the vibrations into heat either with a mechanical damper or an elastic damper.
3: Make the calculations much more difficult (I'm talking about differential equations in which case I am about 4 years out of practice with them)

The simple way to damp vibrations in large sheets of anything is to add something like Dynamat or similar non-hardening butyl based adhesive backed dampening material.

[CaptainVee]

That is the first time that I read something about mass that actually makes sense to me. I always thought that the expression "weight adds dampening" was only part of the story.

Could it be true that dampening would only be needed if it is needed indeed, ie if you get resonance during operating the mill?

[slopecarver]

Dampening is great in most cases unless you make trampolines. You can tune your dampening to cancel out vibrations completely yielding a much quieter machine and less chatter. Damping is not needed in most cases correct and I should add that the dynamat I had suggested will only work for higher frequencies. Lower frequencies require higher dampening coefficients which can be obtained with larger elastomeric dampeners.

[ihavenofish]

If you were running a 2 fluted bit in a router at 25000 RPM that would equate to 50000 strikes per second or 50000 Hz

50000 strikes per MINUTE. 833hz. fancy engineering math isn't all that useful if you don't have the right numbers to input

[slopecarver]

50000 strikes per MINUTE. 833hz. fancy engineering math isn't all that useful if you don't have the right numbers to input

Good Catch.

[1Jumper10]

I know this thread is kind of old but I think it deserves a revisit. There have been a couple manufacturers now that have started to use carbon composite in CNC machines. There are at least 2 examples in the links in this thread. I was wondering if the costs have started to come down? And if its becoming more feasible for a home built machine? At least just the gantry anyway.

Since CF is lighter than cast iron/aluminum/steel, using it runs contrary to the conventional wisdom that more mass = better machine. But it seems that higher mass is desirable only due to its ability to dampen resonance. The undesirable property of higher mass is it's greater inertia slows down its movements. (Sidenote: I just built a machine that used servo's and the effects of inertia were demonstrated clearly in the tuning process.) So if Carbon fiber composite fabrication can be done in such a way as to increase its ability to dampen resonance then its lower inertia would make a lighter CNC gantry = lower inertia= faster acceleration.

[Jeff0000]

I've read this whole thread. And the thing that keeps coming to mind is that cf is damaged so easy and it's nearly impossible to fix once damaged. So why would you want to use it where a little bump can cause significant damage?
I've done enough maintenance in machine shops to notice one thing, not a single one looks showroom condition.

[ger21]

And the thing that keeps coming to mind is that cf is damaged so easy and it's nearly impossible to fix once damaged. So why would you want to use it where a little bump can cause significant damage?

A Carbon fiber gantry would need to be built to be as strong as a steel gantry, so it wouldn't be easily damaged at all.

[wizard]

I know this thread is kind of old but I think it deserves a revisit. There have been a couple manufacturers now that have started to use carbon composite in CNC machines. There are at least 2 examples in the links in this thread. I was wondering if the costs have started to come down? And if its becoming more feasible for a home built machine? At least just the gantry anyway.

I must have missed this thread but I look at Carbon Fiber this way, it is pretty expensive to work with. That right there puts it out of my mind for anything I might attempt to build.

Since CF is lighter than cast iron/aluminum/steel, using it runs contrary to the conventional wisdom that more mass = better machine.

Mass by itself doesn't make for a better machine, the mass must be in the right place. For example one could make a really massive gantry but it will not help much if the frame work underneath is not suitable to support that mass.

The other problem with mass is acceleration which excessive mass can work against you. Again though it is only a problem if the mass is in the wrong place, a massive base frame can help support fast accelerations effectively.

But it seems that higher mass is desirable only due to its ability to dampen resonance.

In part but there are other effects. As alluded to above mass in the machines frame means that you can accelerate parts attached to the frame without the frame trying to move excessively on its own. In a perfect world a machine would have no mass at all but infinite stiffness. This is something that can't be achieved so you add mass to create stiffness

The undesirable property of higher mass is it's greater inertia slows down its movements. (Sidenote: I just built a machine that used servo's and the effects of inertia were demonstrated clearly in the tuning process.) So if Carbon fiber composite fabrication can be done in such a way as to increase its ability to dampen resonance then its lower inertia would make a lighter CNC gantry = lower inertia= faster acceleration.

Yes if this was possible it would be great. I don't have access to anything built with Carbon fiber so can't say much from experience but I can't imagine anything built with carbon fiber to machine tool standards would be light.

The flip side of this, in my case anyways, is boat building. At some point in a boats size it becomes very uneconomical to build the boat out of anything other steel. At least with machine tools you also have cast-iron as solution.

In any event you need to look on the net for a paper written by a guy name Bamberg that dives into the engineering of machine tools with steel tubing and various attempts to deaden the steel. It is very informative but unfortunately I misplaced my copy. One rapid search later: http://www.mech.utah.edu/~bamberg/re...e%20Design.pdf and we have found "Principles of Rapid Machine Design". Well worth the read if you want to understand machine design using steel tube, much could probably be applied to carbon fiber.

[Jeff0000]

A Carbon fiber gantry would need to be built to be as strong as a steel gantry, so it wouldn't be easily damaged at all.

It is still carbon fiber. You can make it so the same shape is ten times stronger than the steel counterpart, if you smash it with a hammer you will be starting from scratch with carbon fiber, where as steel you might have a dent.

[ger21]

I don't know anyone that hits their gantry with a hammer, or does anything similar that would actually cause any damage. I think that your fear of damage is unfounded. If you were to build a carbon fiber gantry that could be as easily damaged as you think, then I doubt it would make a very good gantry.
I haven't re-read the thread, but I believe that there are links to gantry's used on large milling machines milling steel. They are plenty strong, if designed and built properly.

[LeeWay]

Where I think a CF machine would shine is a small lightweight machine for something like a laser or 3D printing. In certain applications, it's flexibility is very important as is it's weight and strength. It can be formed rather rigidly as well, so it's versatility is a benefit. The actual design would go a long way to proving how useful such a machine would be. Is CF a viable material choice?
Certainly especially if you already possess the tools to manipulate it. If not, then another material might be a more miserly decision. The one thing about CF is it's cost. It isn't cheap, but can be the material of choice for certain designs due to it's versatility and properties.

[Jeff0000]

I don't know anyone that hits their gantry with a hammer, or does anything similar that would actually cause any damage. I think that your fear of damage is unfounded. If you were to build a carbon fiber gantry that could be as easily damaged as you think, then I doubt it would make a very good gantry.
I haven't re-read the thread, but I believe that there are links to gantry's used on large milling machines milling steel. They are plenty strong, if designed and built properly.

I wasn't suggesting to go out and hit it with a hammer. I'm simply saying that it is easily damaged, when working with an 8" square envelope it's probably less of an issue that if you were dealing with a 4'*8' envelope.

Go look at any machine, every single scratch, nick and mark will be damage on carbon fiber.
Like I said, you can make cf as strong as you want, it'll still damage easily. And the only way to mitigate that is to design it that the outer layer is basically superficial.

[ger21]

The gantry on my current router is made of wood, with paint on it. There isn't a scratch on it, and I built it 12 years ago. Nothing ever touches it, so not much chance of any damage occurring.

[hanermo]

This is what would make the ultimate machine ... and many of the techs mentioned are in use by the best machines in the world, now.

1. First, use relatively-light/stiff/nonresonant moving components.
2. Second, to reduce stick-slip, use air bearings.
3. Third, for best speed, and best acceleration, and best accuracy, use linear motors. Now you dont have the ballscrew mass to accelerate/decelerate, and all parameters are only limited by your wallet.

Google the Moore Nanotech lathe.
0.3 microns end result in TIR on a turned sphere.
The Moore Nanotech uses linear motors and air bearings.

A dual-supported bridge mill is the most rigid structure giving good access to table.
I built my VMC like that, and it is relatively heavy and stiff, for the mass (2000+ kg).

Probably, a mostly-made-of-air epoxy-granite-steel composite would have highest vibration dampening and most rigid construction practical, while having small issues with thermal effects.
Essentially epoxy-granite microballoons with stressed skin exterior.

When size is no object, large section sizes can easily give very good stiffness.
Google metrology or optical tables.
Stressed skin tubular supported.

Granite is flexible and so is carbon fiber.
It might be possible to make a CF/steel composite, perhaps.

Thinking..
Epoxy-granite tubes with microballoons, steel exteriors (maybe multiple layers with spacers), as tubular supports for the exterior section like optical tables.
Perhaps cast, with crossed tubes, on x-y.
Spaceframe xyz supports might be hard to do. Or not.
It is likely industrial goops ie glue are strong enough, as everything is in compression.

Example.
My VMC, final fix for linear guides will be hysol (locktite) 3784 (for gap fill).
2 boxes of 500 g each. 200€ or so.
The compressive strength on the 2 rails is 1.8 million kg.

35 mm x 240 cm = 840 cm2. Each. 2 of.
1100 kgf (= 11000 N/ cm2) x 924.000 kg.
x 2 = 1.848.000 kgf.

[whimsical]

People seem to think carbon is immensely stiff and strong but a small blow will make it shatter, watching too many F1 races maybe. It actually needs enough energy applied to bend it past its yield point then it shatters. A hammer blow might give a surface defect but make the beam fail, not possible.
I think the best use of carbon, for a beam is in a composite construction. Use the carbon where it utilizes the carbons properties best. In other words concentrate the carbon out in the corners of the box shape. For sure if money is no object then just make it all carbon but i don't think you gain a lot unless every gram matters.
I have built a router to test my ideas. It is functioning but the servos are not tuned yet. Beam is 300mm x 300mm x 2mtr long and weighs about 50kg including a lot of mechanical parts, bare beam was about 25kg if i remember correctly.
I have done a little testing to determine the stiffness of the machine but that has to wait for a while as another project has priority. preliminary playing by applying 100kg lateral force to the cutter collet suggests the beam is very stiff. I am getting movement from the rack and pinion so need to lock it up before getting some usable data.
The carbon is only really visible on the Y and Z parts, its to buried everywhere else. The Brown coloured squares are endgrain balsa
Attachment 339162

[Jeff0000]

People seem to think carbon is immensely stiff and strong but a small blow will make it shatter, watching too many F1 races maybe. It actually needs enough energy applied to bend it past its yield point then it shatters. A hammer blow might give a surface defect but make the beam fail, not possible.
I think the best use of carbon, for a beam is in a composite construction. Use the carbon where it utilizes the carbons properties best. In other words concentrate the carbon out in the corners of the box shape. For sure if money is no object then just make it all carbon but i don't think you gain a lot unless every gram matters.
I have built a router to test my ideas. It is functioning but the servos are not tuned yet. Beam is 300mm x 300mm x 2mtr long and weighs about 50kg including a lot of mechanical parts, bare beam was about 25kg if i remember correctly.
I have done a little testing to determine the stiffness of the machine but that has to wait for a while as another project has priority. preliminary playing by applying 100kg lateral force to the cutter collet suggests the beam is very stiff. I am getting movement from the rack and pinion so need to lock it up before getting some usable data.

I'm not saying the beam would fail, but CF damages really easy. And every bit of damage takes away from it, the same as if you take a grinder and add a 1mm cut to steel, that is now a point of failure.
Take cf bikes, or hockey sticks. very stiff, light and strong, but small damage very quickly becomes a failure point.
And a home machine probably won't see any damage, but an industrial machine... like I have said, I have yet to see one without marks or dents, especially bigger ones, it takes so little for that to happen.

[ihavenofish]

It is still carbon fiber. You can make it so the same shape is ten times stronger than the steel counterpart, if you smash it with a hammer you will be starting from scratch with carbon fiber, where as steel you might have a dent.

i used to demonstrate to people my carbon fibre bicycle wheels strength by kicking them as hard as i could. they were long fibre reinforced nylon, not woven - basically bulk moulding compound, not unlike what the tub and suspension arms of the lamborghini sesto elemento are made of.

if its made correctly, impact is a non issue. put an outer layer of basalt veil and neither is abrasion.

basically, much like casting iron, or designing a high end steel structure, you arent going to do this without good engineering and the right tools.



on the topic of bulk moulding. if one really wants to do this:

use hexcell hexMC 77 moulding compound, reinforced selectively for anisotropic strength with hexcell 77 unidirection prepreg. they co cure under heat and pressure in about 5 minutes. something long and skinny like a hobby sized gantry should be able to be pressed on a 20-50 ton shop press.

the moulds can be made from calcium aluminate cement (cement fondue, or secar 71), aluminium oxide grit, and a bit of chopped fibre (with a suitible sizing) for resilience, poured over a cnc'd wood pattern. you can run heating elements through the mould when you pour it.

to save money and optimise damping vs mass, built the carbon structure light and stiff and strong, then fill dedicated pockets with epoxy and aluminium oxide (epoxy granite).

stay away from large inserts and such embedded into the carbon or big holes, unless you know alot about mould flow because the fibers wont interlock when they wrap around a obstruction in the mould leaving a weak seam.


there, ive given you all my secrets, now someone do it! i dont have the time

[LeeWay]

Right. CF is not usually the only material used. Wing builds and other products I have seen that use CF also use foam core, honeycomb board, fiberglass etc for some of the structural parts. I am not sure just how many layers of CF one might need for a rigid gantry, but if it was covering some other material, then that would certainly lessen the layers needed and thus the cost.
I think the better idea would be to size the motors and electronics according to the weight of the gantry. That works very well.