design questions

[sideburns5]

Greetings!

I am a new member here at the forums, but have been reading through threads for the past few weeks and exploring the idea of a DIY CNC mill. I'm am no where near the design phases yet, right now I am just trying to figure out if this project is within my means (mostly space and money right now, I'm confident of my technical no-how). I am pretty new to the whole scene though still so if I don't understand some concepts or terminology please bear with me.

Firstly, this is a hobby machine that I am making both as a project and to help with future projects, and though I am not rich I am also not apposed to spending the money to make something that both suits my needs and is something I can be proud of. The machine I would like to build would be a mobile gantry style machine to reduce the footprint, and I would like probably 8-12 inches of z-axis travel. Currently I am undecided on how big a table I would like to have and if I want to run servos or steppers. I would like my machine to be able to hold a .001 repeatable tolerance and machine non-ferrous metals (primarily aluminum and copper, though extra variability would be nice), plastics (probably mostly plexi and lexan) and possibly wood if I can manage that kind of diversity.

Now in my research I have only really seen machines made out of either wood, aluminum and steel, but my question is has anyone explored making parts of the rig out of composites like fiberglass? Fiberglass has excellent strength properties if you use it right, its lightweight, its not terribly expensive, its rigid, and its not terribly hard to work with. It seems like it has all the proper criteria. I was exploring the idea of making the gantry primarily out of fiberglass and figured this would be a good place to ask about it. My thinking was that you could build steel or aluminum framing, fill the gaps with a rigid foam, and then fiberglass the outside to provide the structural integrity. I did a few searches in the forums and on google and didn't find anything pertaining to making parts of the machine out of fiberglass.

Does anyone know if this has ever been explored? If not is there particular reasoning? i.e. costing, strength issues, rigidity, etc etc? If not fiberglass what about things like CF or kevlar?


Thanks in advance for any input anyone can give me!

[RICHARD ZASTROW]

Welcome aboard.

X travel, Y travel spindle power and speed are all not mentioned in your post.

I suggest you read the threads here on composite machine construction like the one on epoxy/granite.

There's a lot of help here if you ask the right questions.

Dick Z

[sideburns5]

Thank! Well as I mentioned, I'm really not sure how big a table I want (x-y travel) and I the same goes for spindle power/speeds. I'm not particularly concerned with having a high feedrate, maybe something like 200-300 ipm? Is that reasonable for what I've already stated?

And like I said I did search but perhaps I just searched the wrong terms. I will definitely take a look at the epoxy/granite thread you mentioned though, it sounds like that would be a good place to start. Though as a follow up, my thought on fiberglass was that you could make your gantry lighter so you wouldn't need as strong a motor, would the epoxy/granite composite not be quite heavy?

[sideburns5]

does nobody else have feedback on this?

[awerby]

A machine for milling metal, which is something you say you want to do, needs to be rigid and heavy. Fiberglass and foam is light and floppy. It makes a great surfboard, but a fiberglass mill would be more suitable for cutting foam than metal. A moving gantry design is also much better for light materials than metals; you should be looking at a fixed bridge and moving table instead, if metal's really on your mind, if not a "C" frame.

Epoxy/Granite is a lot better in this regard. With a monolithic "C" frame cast in this material, you get the rigidity and mass that you need for metal milling. You wouldn't use it for a gantry, since the weight would work against you. A 12" Z travel is starting to push the envelope a bit, but it might be possible if you design it right - you'd need a really beefy quill to avoid vibration.

2-300 ipm is a much faster rate than most people mill metal at, especially the harder ones. Pure copper is rarely machined, since it's gummy and cuts poorly. There are copper alloys (brasses and bronzes) that work better. It's possible to hit +/- .001" accuracy if you do everything right, but remember accuracy is the sum of many factors, including backlash, tool deflection, machine flex, screw pitch, bearing slop, slide wobble, spindle runout, etc. If one of these things is sub-optimum, then your accuracy's out the window. Plan accordingly.

Andrew Werby
ComputerSculpture.com — Home Page for Discount Hardware & Software

[sideburns5]

Thanks a lot for the response Andrew. A question though, why would you need the mass? You say "the rigidity and mass that you need for milling metal." Maybe I'm just reading this to literally, but I just want to be clear for the sake of learning :P

Also, would you think that the granite and epoxy composite would be better that just using steel? My intention on the high feed rates was mostly for foams and wood (if that's in itself is feasible) and I would expect my rates to be lower for metals. And thank you for the information on the copper, I know it can be difficult to machine and I would imagine (thinking about possible future projects) it would be a multipurpose or machinable copper alloy. For instance I build a lot of custom PCs and I would love to be able to machine my own water blocks to cool the components and copper is the ideal choice.

My interest in mobile gantry over mobile bed is mostly due to space. Is there anything in particular I should take into consideration if I do end up go with mobile gantry that would help my design?

Thanks again for the help everyone!
Charlie

[Zach_G]

Adding more mass will reduce the amplitude of vibrations induced by cutting forces. This isnt to say that you can't achieve the same results with a very stiff and light frame. An example would be Hexapods and other parallel kinematic variants; however I dont really know of any commercially viable machines. That doesnt mean we hobbyists cant play around with this kind of stuff but I think youll find once you price out the quantity of epoxy and fiberglass needed to replace an equivalent steel member, It would have been far more time and cost effective to go with steel.

Also, youd be hard pressed to design a machine light and quick enough to efficiently route wood, yet rigid enough to machine metals. Pick one and deal with reduced capability on the other, or spend a fortune on oversized motion controls to move all that metal around fast. From what youve mentioned, Id say go with more of a wood router design, build it strong, and you should be able to route aluminums as well. 12" Z for a moving gantry is alot, youll be doing well with 6". That said, I designed a moving gantry for machining metals in the past here, decided to go with a more traditional C frame as it was easier for me to manufacture with my limited tools.

[sideburns5]

Ahhh, I hadn't taken vibration into account, that makes a lot of sense! If building such a diverse machine is outside the reasonable scope of time and cost, then I would be happier to go with metal and plastics and just crank down my feed rates and not mill wood at all. Again I don't expect to have to be milling much steel, primarily it will be non-ferrous metals like aluminum, brass, copper, etc. I've seen lots and lots of machines around the internet that are built to machine wood so I suppose I could always make a separate machine for those needs later down the road if I feel the need.

Also the Z-axis travel was purely a guess and definitely isn't a requirement. 6" would probably be enough, though if I could get away with 8-10 I'd love too. Also, as for the C-frame style design, what kind of travel distance would be reasonable to achieve? Again I'd like to keep my footprint down if I can, but if I have to sacrifice some of the travel range to provide a more accurate machine I will. I make a lot of custom PC cases and being able to machine side panels for my cases is pretty desirable some of which get up into the 30x30 range if not a little bigger. Would something like that be reasonable on a C-frame?

[awerby]

Wood machines a lot like plastics, although it's less dense. But a machine designed for metals shouldn't have any trouble with them. Most big "C" frame knee mills are built using cast iron, but epoxy/granite has similar rigidity and resistance to vibration - maybe even better - and it's easier to cast than iron. A mobile gantry is good for saving space, but it's inherently less rigid than a fixed bridge or "C". Think of your Z axis as a lever - any vibration from cutting gets a better chance to rock and roll the longer it is.

Getting 30 inches travel in X isn't too hard, but Y is usually shorter in a "C" frame mill, since the spindle has to canteliever out a lot further - remember, to achieve 30 inches of travel, you'll need more than 30 inches of projection, and about 60 inches of space for it to move in. If this is really what you need to do, building a very rigid bridge-style machine with a moving table might be your best bet. There are metal-cutting machines made commercially that use a scheme like that.

Andrew Werby
ComputerSculpture.com — Home Page for Discount Hardware & Software

[sideburns5]

Thanks a bunch for the information Andrew. The CNC we have at work is a fairly large machine with something like 24" of Z travel and my co-worker who programs it says if he turns the feed rate up you can literally see it deflecting even with carbide cutters. That is a mobile deck style machine with a bridged gantry (not a HUGE machine but pretty decent size for my experience). That machine is specifically meant for plastics though, and the machine is almost solely used for milling pattern to reduce the workload of our pattern maker. He says it can mill non-ferrous metals also but he's only used it for that on rare occasions. I'm certainly not looking to build anything of that size so I guess I was hoping the deflections and vibrations might not be as much of a concern considering the reduction in size of my project.

However, people seem to be saying stay away from the mobile gantry design, but I have seen several of these machines being built around the forums. Are there specific features that make this type of design successful? Is it really just going to be more work than its worth to build a mobile gantry machine considering my wants?

[awerby]

If you really want to build a moving gantry router, take a look at the Mechmate design. It supports the gantry between two rails at the top of the machine, which gives it better rigidity than the usual design, since it avoids the side-to-side flexing you get when you run the gantry on elevated members sliding on rails fixed to the table. I've even seen videos of those machines cutting aluminum, although I'm not sure how good a job it really does.

Andrew Werby
ComputerSculpture.com — Home Page for Discount Hardware & Software

[sideburns5]

Great thanks for the tip Andrew! Looks like at this point I need to put together a design and then just get some feedback.


Thanks a lot for the help guys!
Charlie