Comments on homebrew design

[Zach_G]

So there's the general design, 2" steel tubing with 3/16" wall, structural angle rails that the skate bearings ride on would be SS with same wall thickness. The aluminum angle that the skate bearings are attached to are 2x2x.25 Probably end up using large bolts to hold it all together. Work envelope is 12"X 12"Y 6"Z. I need it to be able to mill hard aluminums, and soft steels with max 3/8" endmill. So am I dreaming or could this handle the job? I'm trying to build small and light since I don't stay in one location very long, however I know lightweight and metalworking don't mix very well.

Another concern is the SS structural angle. The skate bearing design should be somewhat tolerant of the rounded edges, but not if the thickness varies. I suppose I could clean it up in a mill, however from a diy standpoint I'd rather the design require few tools to fabricate.

Comments, Suggestions, Feedback?

Thanks

[Karl_T]

IMHO, you have a good design for a wood router. But not rigid enough for machining metal. You'll have chatter problems.

Karl

[Zach_G]

Thanks for the feedback, I was thinking similarly but thought I might as well throw it out there. Let's say I wanted this machine to be capable of everything a small mill such as a sherline or harbor freight mini mill could do, but with a larger work envelope (otherwise I might buy one of these machines, but building one from scratch is so much more of an accomplishment). Would this lower the ridgidity requirements? If not, I have some massive 80/20 extrusions I could use for the frame, or do you see the chatter originating in the linear bearings. I was also thinking of using more skate bearings along the length of the angle.

[pstockley]

I am working on a similar design for a small but rigid mill for cutting steel and aluminum. My current thoughts are to use a cheap 18" by 24" by 3" surface plate as the base (less than $100 from ebay). I will drill this and bond in inserts for connecting the linear rails and column (haven't decided if it will be a gantry or conventional mill yet). For the structure I will use 1/2" wall steel tubing (For a single column about 12" by 10"), welded, normalized and then ground on the critical surfaces by a local shop. I am also planning on filling the column with polymer concrete to help dampen vibrations. Basically mass is your friend when it comes to metal cutting machines. I expect my design to weigh at least 600 to 800lbs for a machine just a bit larger than a Seig X3. The other problem you will have with your design is the rails. Without a stiff setup, you will get chatter for sure. I will use medium preloaded linear rails so there is running zero clearance. Take a look at http://pergatory.mit.edu/2.75/2-75_Lectures.htm there is lots of useful info.

[Zach_G]

Wow, sounds like a real hog, should be impressive when done. I myself am shooting for slightly lower standards. If I could get around .002-.005 accuracy I would be elated. Fairly slow feeds and speeds too (rapid at 100 ipm would be cool). Keeping it small and light is more of a priority than workspace, so if need be I'll definitely go smaller.

[Karl_T]

I was a manufacturing engineer for 20 years. One of our favorite sayings was "Plaguerism is the sincerest form of flattery." For example, there are monster commercial machines using the steel and concrete concept. Its a proven idea.

For your machine, find a Sherline and copy the design. I'm sure those folks made several trials before marketing. Its the best light weight machine I know of.

Karl

[Mcgyver]

pstockley, I've been planning a similair mill, lots of content here if you search polymer concrete. I'd add that make sure you find a place with a big enough oven to heat treat the fabrication, you want it normalized or it won't be stable - the amount it will move from the stress relieving is incredible, without heat treating all those stressed will be forever twisting and turnintg it. edit - ooops you did say you were going to normalize - make sure it'll fit the oven!

For mine, i'd forgo the the surface plate and use a fabrication filled with granite and polymer. Then again it may be easier to get a second vertical mill and convert. It does probably provide the best overal value; utility/cost

Zach, do a search here on forces to get some idea of what you are up against. If you want to cut metal, I'd retrofit a bench to or better yet an b-port or clone.

[pstockley]

Mcgyver, I already have a CNC mill (Tormach) so my reasons for building another mill are a little bit different:

I wanted a smaller mill with a faster spindle speed (>7000rpm) and faster rapids for producing small intricate parts for my Model engineering projects.

I like building tools and am interested in seeing how good a small mill I can produce. To me, all the small mills (around Seig X3 size) are under built. I wanted to see what could be achieved by applying industrial build techniques and mass to a really small machine. Effectively, I want a micro VMC and will be using servos, precision ballscrews, bearings, linear ways and a ABEC 7 spindle setup with a belt driven servo drive. Very much still at the planning stages and I need to finish my Hardinge retrofit before starting building.

I have looked into a poured base but the thing that appeals about the surface plate is you have a really accurate and stable reference to start from. The only problem I see is finding a place to grind the reference edges to align the linear rails against. I am assuming you would need a surface grinder with a diamond wheel to do this. If I am really lucky, I might get some really straight rails and not have to do this!

[Zach_G]

Mcgyver, I have been looking for the forces involved in milling, however I have been unable to find any numbers for different operations. I did manage to breifly read a couple research papers on the subject, and it seems that the prevalent issue is not so much the net forces in milling, but the impact of each tooth of the cutter into a hard material and the resultant vibrations/chatter that occurs. I guess this is why these metal cutting machines seem so much more massive for what they accomplish compared to wood routing machines.

Whenever I think about the cost/benefit ratio for such a machine, it isn't worthwhile for my current needs. That's too bad too, because I've been stuck in design limbo for the past year it seems.

Karl_T, I will take a closer look at the sherline mills. At first glance they don't appear all that massive and ridgid, but maybe there's more to them than meets the eye.

[Stevie]

Originally Posted by pstockley . The only problem I see is finding a place to grind the reference edges to align the linear rails against. I am assuming you would need a surface grinder with a diamond wheel to do this. If I am really lucky, I might get some really straight rails and not have to do this!

Why grind the granite
If you set up one rail; the other is simple; just run the car down the first to dial in the 2nd

[pstockley]

I was under the impression from reading manufactures specs on linear ways that you cannot assume the rails will be straight to a high degree of accuracy.

[Mcgyver]

Zach, I probably should have given a bit more of an explanation but you've picked up my point - the uninitiated often underestimate the power, mass and force required in a metal cutting mill.

Nothing wrong with ambitious goals, but to be good at this, diy cnc metal working, you need imo both machining/metalworking and cnc knowledge (cnc; electronics, software, control systems, etc). If you are not experienced in machining and want to design metal working machines, you will get better results if you address this lack of experience. A bench top mill, night school course or like me a vertical mill in the garage are all ways to get it. And hour or two of cutting ferrous on a full size machine will quickly give you an appreciation of what’s involved and you'll need those skills to build it.

pstockely, keep us informed, sounds like a great project. I agree you need a dead on reference to bolt to, no surprise how I’d do it; scraping. I'd weld on two mild steel strips to the superstructure where the rails are to go, probably drilled and taped. after normalizing and filling with granite and polymer, I'd scrap them in. many people have this big hang-up about scraping, its really no big deal and not more than a couple of hours of work. Surface grinding would work, but you'd need a big grinder to do that on an 800lb beast. I still like the idea of whatever makes up the column being contiguous (granite/polymer/heavy steel casing) with what lies under the table