3 Axis CNC machine.

[Freestyle72]

Thanks for all the input guys. Nowhere in my post did I say I wanted to make an MDF machine. I wanted to make a full aluminum machine to keep rigidity. I was just giving the MDF thing as an example of what lots of people have done out there. I think maybe given the cost I will reconsider the CNC thing and look more at just running a mini lathe and a manual mill perhaps.

[GreyGhost]

I have never seen an aluminum mill. For rigidity, most are hundreds or thousands of pounds of cast iron. The forces involved are really much larger than you'd expect. If a company making a mill out of cast iron can't make it for any less than $1500, it is doubtful that a person could make an equivalently stiff one from aluminum for less than 2 or 3 times that price (and a boatload of time).

[Brian P]

Every metal cutting CNC that I've seen, has a base frame made with lots and lots of cast iron. Aluminum only has one-third the stiffness and it's a lot more expensive. Cast iron is dirt cheap.

And a home-built CNC machine?? I'd love to see a home-built ballscrew that is accurate and repeatable to within 0.0001" and axis motions that are perpendicular to each other within very small fractions of a degree ... If the machine is out of square, so will everything made on that machine. If the machine is not repeatable, neither will the parts made on the machine. If the machine is not extremely, extremely STIFF, the forces involved in cutting metal will not only throw the workpiece out of whack but also shake the machine to smithereens in no time. Let's not even discuss that pricey little Fanuc multi-axis CNC servo motor controller that you're going to need to make this thing work.

It's possible to cut relatively inaccurate workpieces out of wood and plastic on home-built machinery, but those materials can be cut using extremely high spindle speeds and very low forces on each axis. Cutting steel is a whole different ball game.

[Ben G]

........ By far one of my all time favorite internet threads .......... a plywood CNC machine, PLYWOOD thats great, thank you.............. BG

[teflon]

The best thing you can do to learn, is find a copy of Solidworks, and Rhinosaurus..
Learn it, create with it..

And learn what is and how to create CAM files.. (Tool routing)

Then you can just send your file out to a CNC mill thats local to you.

Much much cheaper than purchasing a pro cnc mill.

[K1200RS]

Why not try working with a local machine shop? You could probably learn a lot in the process and you will get the benefit of the machinist's assessment. Buying your own tools is expensive, both to purchase and to set up.

[bitzz]

No machinist is going to run your untried CNC programs until he trusts your tool routing. A great way to ruin expensive tooling is ploughing it into the work piece, or better yet the table or hold downs, with goofy routing. Buying some else's tooling is even MORE expensive.
The best way to get experience in tool routing is by doing it... manually.

[crankcall]

I have a combo milling machine / lathe sitting in my garage at home. I have experimented and have a machinist friend that is my tutor that has helped me figure out complicated milling ( I dabble in sailboat fittings).
There are currently "home" cnc being sold for $3,000 all in. They are also crap.
Gorrilla and General lead the parade in this crap.

If you are really interested in machining I would suggest a manual machine, even a chinese import machine can be tuned up to make reasonable parts. Until you get what the machine needs to do, by learning to machine, you cant really program CNC safely.
And nearly every CNC has a pod or holder with bite marks from setup problems, even the ones with sensors.

[Brian P]

Same thing applies to a CNC as for industrial robots: "Anything you can touch, you can smash into!"

I've seen a couple of really expensive ones ...

[NP6000]

I have a combo milling machine / lathe sitting in my garage at home. I have experimented and have a machinist friend that is my tutor that has helped me figure out complicated milling ( I dabble in sailboat fittings).
There are currently "home" cnc being sold for $3,000 all in. They are also crap.
Gorrilla and General lead the parade in this crap.

If you are really interested in machining I would suggest a manual machine, even a chinese import machine can be tuned up to make reasonable parts. Until you get what the machine needs to do, by learning to machine, you cant really program CNC safely.
And nearly every CNC has a pod or holder with bite marks from setup problems, even the ones with sensors.

What do you and/or your machinist tutor think of the combo machine? Say for making parts to with ± 0.002" I know a combo machine isn't a bridgeport, but most of my parts don't need that kind of precision.

[NP6000]

for the OP:

http://toronto.kijiji.ca/c-buy-and-s...AdIdZ233504307

there is another one that also has a Mitutoyo x & y axis readouts.

[crankcall]

What do you and/or your machinist tutor think of the combo machine? Say for making parts to with ± 0.002" I know a combo machine isn't a bridgeport, but most of my parts don't need that kind of precision.

Thats a tight tolerance for this combo machine but its do able if you really want to pay attention. My machinist friend is from a Soviet labor background and views my Chinese combo as a "toy". It had a lot of backlash and play in the spindles when it came out of the crate, We have made pieces to repair a 1902 steam marine engine with it, it was probably better than what they had then.
I got it for nothing and spent $400 on cutters so if i only make a couple things I'm ahead.

[racinjason]

There are lots of DIY plans out there and you can assemble one out of wood (mdf). I plan on taking the plans of one made out of wood and making it out of aluminum to get rigidity to maintain accuracy when cutting metals (primarily aluminum). Plans cost about $10-20 and materials cost whatever you pay for them depending on the size and material/components used. For an all wood one, you can make one as cheap as $500 and it will actually cut aluminum fairly accurately. HOWEVER if you have to cut a lot of aluminum I think it will be unreliable and results certainly deteriorate. An aluminum built machine will cost about $1500 but I think it's worth it given the number of times I've wished I had machining abilities to do something specific. For my fuel injection conversion I think this will be good because I can keep tweaking my stuff without having to pay a shop a TON of money to keep making corrections for me. Thus making my trial and error much cheaper.

There are programs out there that can convert CAD/3D models to CNC format. Anyways this like all things will be another learning experience for me. And once I build my first machine, I can actually build other machines from the first one.

Do you have a link for the plans?

[Freestyle72]

Do you have a link for the plans?

Just google search DIY CNC. There are literally 10s of plans available some for a small amount of money. Others for free.

Thanks NP6000 but that machine although very capable looking is out of my budget right now. I would have like to spend about $1500 or less. Anyways this thread has shown me that I need to re-evaluate my priorites with the machine.

[bitzz]

You would be spending more than $1500 on tooling.
I've got more than $1500 into my drillpress.

[TheDirty]

In episode 2 of The Ben Heck show they build a 3 axis CNC. This is not meant for metal work, though. They use a router to create the parts for the router they make (catch 22).

http://revision3.com/tbhs

[Freestyle72]

What do you guys think of this?
http://www.harborfreight.com/3-in-1-...eed-44142.html

or even this

http://www.harborfreight.com/multipu...hine-5980.html

I just want to make bushings and spacers and such.

But Would like to attempt a custom triple at some point for my own project I have something in mind.

[Brian P]

I would consider one of these to be the bare minimum for a hobbyist manual lathe. Not all of these are any good. The more cast iron there is in the base, the better off you are.

http://www.busybeetools.com/categori.../Metal-Lathes/

Here are some bare-minimum manual mills. Same place (they're local). Same situation regarding cast iron - the heavier it is, the better off you are. Of course, that's not the only factor, it also needs precision-machined guides and feed screws, and a good machine tool also needs to be anchored and leveled unless you want everything you make on the machine to be out of square.

http://www.busybeetools.com/categori...rking/Milling/

My dad had one of their lathes in his old shop, and for what it was, it did okay.

[Freestyle72]

I like this mill:
http://www.busybeetools.com/products...3-CRAFTEX.html

But the lathes seem expensive given their size.

What do you think of this mill for making a custom triple that isn't flat? I am thinking of making something to get more height of my forks so I want my top triple to be stepped so I can put the forks lower if that makes sense.

[Brian P]

Holy jumpin, a 350 watt motor on that? Less than half a horsepower!

Also note that the table only has 8.5 inches lengthwise and 4 inches cross slide, and the size of the tooling is extremely limited - because of the small motor.

You need a bigger machine for making a piece like what you are describing.

One other small thing. Take a close look at the surface of a professionally machined part (anything). See the tiny cuts in the surface - the ones that were made by the cutter that machined this surface? See how all those tiny cuts are perfectly evenly spaced? The only way you are going to get that, is with power feed at a minimum (and realistically, CNC operation). You are never going to get that with you turning a handwheel!

The machine you linked to only weighs 50 kg. I'm looking at a few others in the catalog that are probably *somewhat* more suitable for that workpiece - bigger table, more travel, more power - and they weigh nearly 10 times more - that's more like it. The one you linked to is too light duty.

Since that store is local, I have a suggestion for ya. Take a sample workpiece (e.g. a stock triple clamp) with you to the store. Find a sales guy, tell him to assume you know nothing, but would like to be able to make something like this, and see what they say.