[jupdyke]

Hello Everyone,

I have been kicking around some ideas for a new cnc mill. I have previously build cnc router mostly from 8020 and am pretty happy with it. It was a budget machine, but it has been a great help to my projects. I have since build two reprap 3d printers.

One day while helping a professor clean out a lab, I came across four pieces of supported rail and eight linear ball bearings. He let me have them for free. So I have been wanting to build a second machine using those parts. I have been thinking about various designs for a long time now, and in the past two days was inspired by DonFrambach build.

Here is a link

Anyway, today I spent about 2 hours throwing together this 3d model. It is still raw, but shows the basic idea. I have not put all the bolt holes, nuts, or even lead screws or motors in. But I think there is plenty to get a good idea for what I am going for. I am stopping for the night and thought I would share what I had and get your opinions.

Things to note:
*I plan to use ball screws and stepper motors for the x and y axis.
*Z uses precision acme lead screws. Currently 3/4" - 16. It does not need to travel as fast as the other axis and will be lifting more weight, so I thought gearing it up would be good. Also I need two and I thought ball screws wouldn't add enough for the added cost.
*The y axis moves up and down as a whole. I like this design because I can clear higher objects. I have not seen another machine like this. I also think it will be more study. I would love comments about this design method.
*There will be wear plates on the ends of the y axis. 8020 sells UHMW pads that match the grooves in the 8020. I have not added these to the model yet.

Thank your for any comments.
Josh

[restless1]

Similar to mine I am using spare parts from my 8 foot by 4 foot aluminum mill
(I had enough extrusions left over) so i made it with parts on hand except a few off eBay here and there!
And I cloned it from this persons build so the credit is his

http://www.cnczone.com/forums/diy-cn...ter_build.html

[JD_Mortal]

You will also get more stable precision, through the entire Z-axis motion, having your Y-axis move up/down like that. The constant distance from the rigid structure is where you will gain that precision. (The end of the tool is always a fixed distance from the Z-axis, unlike one that slides the tool only, up/down on a floating-tray.)

I would have done a similar setup, but I did not have enough money for the more complex setup. Perhaps my second machine will use a raising-gantry Z-axis.

Looks solid as a rock... I assume it will be a tooling mill, for metals, or a precision etcher?

[jupdyke]

Thanks for the comments guys. Here are some updated images. It is still a ways from being complete, but some of the details are starting to come together. The second image is the z-axis by itself. I am still working on the bearing aspect of the z axis, so far I am thinking of using some UHMW pads that would run on the inside of the vertical uprights. I am not sure if this will be enough or if I should add something to ride on the outsides as well.

I am getting a little nervous about the cost of this machine. Hopefully by keeping my eyes open on ebay I can get some good deals.

[katalin]

The Y axis moving up and down is a good idea but depends on what type of parts you are doing.
For example if you want to profile a part from a plane stock(let's say an aluminum plate) fixed with vacuum or in a vice , your design is great, having more precision and rigidity then a "classic z axis".
But if you are making complex 3D parts , use different type of clamps and fixtures it's very likely that your y axis is going to collide, with the fixture or the part.
The geometry of the part dictates the way you chose the machine axis distribution, there isn't a "best way", each configuration has it's strong points but also it's week points.
For general use i recommend a "classic z axis" with the spindle connected to the z axis plate throw a part build as much as possible on the "equal resistance principle". This way you obtain a minimum z axis footprint.
There are many things to discus on this subject.

Best regards,
Catalin
(Sorry for my bad English)

[johnohara]

You might be interested in this build as well.

~John

[jupdyke]

So just a little update. Out of curiosity I had Nook Industries send me a quote for two 16" ball screw assemblies. $1,276.68 each. I about fell out of my chair laughing when I saw the price. I have worked with Nook before and they make great products. But I know for sure that I will be finding something else. Ebay or some refurbished parts. Worst case I will put precision acme thread on it for a while until I find some good ball screws.

[katalin]

The precision class of the ball screw has a major influence on it's price.
On the market you can find C7 rolled ball screws at much lower price(they insure a 50micron maximum tolerance over 300mm, and some models have anti backlash nut )