Steel/Aluminum Door Router Log!

[boguski.1]

Well, I've been reading, absorbing, and designing since January, and I think it's time to start my build!!! Here are my design specs:

Machine Type: 3 Axis Router for milling wood and aluminum
Materials: Steel tubing and aluminum stock
Table: Piece of heavy steel door and MDF
Table Size: 36" x 48" (32" x 41" cutting area)
Motors: 425 oz-in steppers 200 steps/rev
Rails: 1/2" OD precision with Frelon lined linear bearings
Driver: Xylotex 3 axis kit
Lead Screws: 1/2-8 2 start (4 turns per inch, or 0.250 lead)
Software: Mach 3 and V-Carve
Spindle: Dewalt 2.25hp Router

Here's a rendering of my build in Inventor. Haven't added the motors or the leadscrews yet as I have not yet ordered my motion components. Designed the overall concept in AutoCAD and transferred profiles to Inventor. Great software for visualization and ongoing design changes!

[boguski.1]

This is a picture of the door I cut in half and welded up. It's sturdy, flat, heavy, and will be easy to shim up a piece of MDF for my cutting surface.

[boguski.1]

I don't have a milling machine (yet ) so I built a jig to cut out my router mounts using the router and a collar. I used a 4 flute 1/4" end mill and a 3/8" collar to do the cut. The mill clogs very easily, and I found another post that said paste wax works well to prevent this. Very shallow cuts and medium speed are also a must. They're not perfect, but they fit very well.

[boguski.1]

I started my build with the electronics (since my electrical knowledge was the limit to the project's success!) After reading numerous posts and asking some questions, I was able to put it all together. There are a lot of talented people on this forum, and many thanks go out to everyone for sharing their ideas and advice that have helped me with my project. I ordered the Xylotex 3 Axis driver board and 3 425 oz-in steppers. Here are some photos of the box, as well as a pdf of my wiring diagram.

[boguski.1]

I've started cutting out the pieces for my Z Axis. I used a 10" carbide blade on my table saw to cut the aluminum stock. Slow feeds, good clamping, and multiple passes are necessary to do this safely. I'm happy with the cuts and they seem to be square. I chucked the large block into a 4 jaw chuck on the lathe and turned the faces parallel that will be drilled for linear bearings.

[bekx]

nice project

can u post some pictures of the jig used to cut the router mounts? i got a milling machine (not cnc) but never did circles that good without a 4th axis divider.

[boguski.1]

Thanks Bekx!

Here is a rendering of the jig. I made my mounts in two pieces of 3/4" x 2" stock because I had it in the garage. You can adapt the jig to work for a piece of plate as well, since it would be exactly the same without spacers.

I have a 1/4" space in the middle of my design to give some more meat to the aluminum on the leading edge. This is held with a few pieces of keystock. The workpiece sits on a piece of 3/4" plywood and is surrounded by 2" strips of the same plywood.

The top is a piece of melanine (slippery laminated plywood) but any smooth surface will do, even sanded plywood. I printed out the circle and large square on paper from my drawing, adhered it to the melanine, and cut out the circle with a scroll saw. I then cleaned the circle with a dremel sander to get it as close to my line as I could. Note that the upper guide circle is larger than the finished cut because the router collar is 1/16" thick.

Once the piece was in place, I nailed the guide circle board on top of the jig even with the outside edges of the plywood with two nails. I clamped the jig to my workbench with a few clamps. When I finished one cut, I popped one nail out, rotated the guide to free and reload the jig, and cut my second piece after replacing the nail. This ensured that the circles were in the same spot on both pieces.

This method is somewhat crude, but it's the best way I could think to cut a decent circle in material of this thickness and in two pieces without a mill or a rotary table. I suppose a 4 jaw chuck on a lathe could be used if your design calls for one piece of solid plate.

[harryn]

Hi - nice project. I am very impressed with the cutting of Al on that table saw, and your use of that router. It did not occur to me that a cutting saw can do that work so well.

Similar to you, I have been studying the frelon bearings for a router build. One of the more subtle things I noticed about them, is that they require the center of mass / load to be within somewhat narrow regions of the bearing spacing or they will bind. Also, they have a very defined torque load that must be dealt with. I am not sure, but your Z axis might be running very close to these limits.

I am really interested to see how your project goes, as I might mimick your design for those bearings if it works out.

Good luck. Harry

[boguski.1]

Thanks for the reply Harry.

I was also impressed at how well the saw cuts aluminum. It is a heavy Delta Unisaw, I don't know if a smaller contractor sized saw would be able to cut as effectively with lower horsepower. I used a carbide blade with lots of teeth, and marked it for aluminum so it won't be used for nice woodworking cuts.

As far as the bearings are concerned I would be interested in any information you have come across about the issues you mentioned. My main reason for selecting them was price and resistance to dirt and dust. Any other suggestions are more than welcome!

I'm hoping to have the machine completed by early August, and I will be updating frequently as I complete parts on the weekends. Thanks again and best of luck with your design as well.

[harryn]

Here is the link to the web site of pacific bearings and their catalog pdf.

http://www.pacific-bearing.com/produ...oduct.cfm?ID=1

If the link does not work, their web site is www.pacific-bearing.com. Look for the product "simplicity" bearings.

Starting around page 72 of 130 odd pages of information, there is a long application section. Buried in there, (near page 82 I think) is a section on cantilevered loads.

If I am properly understanding it, the load needs to remain within BOTH a distance and torque limit to work, unlike a normal linear bearing with balls. The distance seems to be a function of bearing spacing. I really like their design concept, but this was a bit challenging for me to stay within on my project.

I would really like to get your take on it, especially since you already have parts.

I am toying with going down this path, or using a flat UHMWPE slide arrangement. (sets of flat areas with UHMWPE in between them).

[boguski.1]

Ok I've looked through the pdf and did some math and here's what I found:

The maximum allowed ratio between moment load to bearing distance and distance between bearings is 2:1, independent of load amount and drive force. I measured and calculated each axis to check my numbers, and my z axis comes out to be 3.2 : 1, so I think I'm out of luck! I wonder if I need to simply use ball bearings on my z axis.

This is assuming of course that my measurements are taken correctly based on my drawing. Also (from what I can piece together from my statics course last year!) I think the ratio would be the same regardless of where the load is applied; off to the right of the sled as in the pdf diagram, or below the sled as in my drawing. I think the distance of the load to the center of the sled is the important one.

Thanks for pointing this out, maybe others will have more insight as to what can be done. Have you seen any other machines that use Frelon bearings? They seem like a good product, but if binding is an issue I'd rather avoid them altogether.

[harryn]

I have done searches on cncforums for the following related bearing materials
- Frelon
- nylon
- acetal
- moglice
- simplicity
- Igus rails

There are quite few people that have these kind of bearings with success. If you have the time, go ahead and search them for ideas - probably at least 2 nights reading on this forum.

I have wondered if there is a way to mount your router so that the centerline of the bit is centered under the Y axis. It might be possible to put your Y axis "side wase" so that the Z motion is directly under it - not sure.

Anyway, you see the challenge.

What is not so clear to me is whether or not you can "cheat" a bit. Obviously, 2:1 is their "safe" limit - is it the "absolute" limit, or just what they are sure of ? - no idea.

It is also not clear to me what you can get away with if you have 4 bearings - would that make it so distance does not matter ?

They also point out the idea of counter weights to offset the load - maybe this can work for you.

Good Luck