I am thinking of building a 4'x8' CNC router loosely based on the torsion box MDF plans available at the site and I have a few questions.

I am thinking of using either 1.5" (1.9" OD X 0.2" WALL X 1.5" ID) or 2" (2.38" OD X 0.22" WALL X 1.939" ID) aluminum 6061-t6 schedule 80 pipe which are not too expensive (about $4.00-6.00 a linear foot) for my X axis and Y axis rails (To be used with the skate bearing linear bearings).

Will there be too much deflection on unsupported rails spanning over 8 foot with these pipes for reasonable accuracy?

Yes, unless they are supported along their length.

and came up with the following solutions:
Ditch the MDF sides and go with an aluminum base and gantry.
Use 1.5" Schedule 80 aluminum 6061-t6 pipe (1.9" OD 1.5" ID).
Drill and tap rail for 0.25" fine screw every 6 inches.
Rail is housed in 2"x4"x0.25" aluminum C channel.
Drill the bottom and the side of the C channel for support.
Use a double nut assembly for adjustment.

See Pics:
http://inconnu.islug.org/~lackkevi/rail2.jpg
And
http://inconnu.islug.org/~lackkevi/rail.jpg

Though I think square aluminum tubing would be easier to drill and tap.

http://www.cnczone.com/forums/showthread.php?t=4900

Have a look at my plasma table to see what I used.

Peter

I like how you did your rails Apple! How did you end up securing the bar to the angle iron? I t looks like you welded it down but I could be wrong.

I have considered using steel as the prices between steel and aluminum are similar on a linear foot basis, and I have a really nice 220V 180Amp Lincoln Mig unit in the garage...... but was wanting to keep the weight down.

I'll work up a full set of sketches on some cad program and post my thinking.

I would suggest securely supporting the rails and moving the adjustment to the bearings. I have not tried it, but have read several times on this site that aluminum does not make a good rail as it will become pitted if the bearings roll over some debris, causing permanent damage to the smooth rail.

Steve

Rail (in black) is 1.75" OD 0.25" wall DOM mild steel with a bushing finish (available for about $6.00-8.00 a linear foot).

Support (in blue) is 1" high by 3/4" wide HDPE plastic with a 1.75" diameter receiver (?? is that the right word?) milled with a router into the top. What I mean is that the profile of the tube is routed into the top of the support creating a cradle for the tube.

Large bots pass through the frame into tapped HDPE to secure the HDPE to the frame.
Small bolts pass through the HDPE (and are countersunk into the HDPE) into the tapped DOM tube to secure the tube to the support.

http://inconnu.islug.org/~lackkevi/rail3.JPG

Spalm I think you are right, steel bearings on a aluminum tube would gall the aluminum.

Looks OK but you do not have much support for side to side loads (maybe not a concern to you). Ger21 has had good luck with EMT pipe using skate bearings. It is a lot cheaper and allows you to try before you buy $150 of rails, if you are planning on prototyping.

But before you get too finalized on a particular rail, what is basic structure of the rest of this thing going to look like? Maybe something like a napkin drawing?

Steve

http://inconnu.islug.org/~lackkevi/rail4.JPG

I have seen the pics of Ger21's build.... my only concern w/ the conduit that he is using is wall thickness......... but it is a lot cheaper.

I am working on a sketch-ed up plan in Rhino, but I am not the fastest modeler in town.

of frame and x-axis.
I don't know whether I am going to use a rack and pinon or a 1" acme screw yet.

Sides are 2"x4"X 114" 0.25" aluminum C chanel.
Ends are 2"x4"X60" 0.25" aluminum tube.
Cross bars are 2"x2" extrusions (80/20 maybe).
Triangulated brace is 4"x0.25" aluminum plate.
Corner gusstes are 0.25" plate.

I am thinking of using the rail system as described in my last post and skate type bearings as are common on many of the DIY projects that I have seen in the forum.

http://inconnu.islug.org/~lackkevi/x-axis.JPG
And
http://inconnu.islug.org/~lackkevi/x-a2.JPG

carefull with the cam software, you're going to fall in love with all kinds of cool stuff that will be a killer to actually make. :D trying to drill all those holes over a lengh of round stock and have them come out the the exact same radial position would not be very easy, at least in the home shop. Here's a great web sit of one who did it, lots of good ideas there.

http://www.buildyouridea.com/cnc/IG-88/phase_two/phase_two.html

The next level up would be to consider buying the shafting, hardened and ground with the holes already drilled,m although you still should somehome have a V on the top of the support to give lateral strengh

http://www1.mscdirect.com/CGI/NNPDFF?PMPAGE=3872

while more $$$ it is a hardenshaft that will give better performance and wear

Finaly, for mucho's $$$ go with a linear assempbly like.: http://www1.mscdirect.com/CGI/NNPDFF?PMPAGE=3873

they don't show them in 10' lenghs but you must be out there somewhere

imo you want steel rails over AL, and hardened is better still. With a mig i'd be all over making it from steel. Warping is controlled by the fact that you don't need an 8' bead, just some thing a little more than tack welds every half foot would probably do. I've always found Steel is fraction of the price of AL, and shapes like square tubing are readily available

If you're going to go through the trouble and expense to build a 4x8 machine, do it right, even if it means spending an extra hundred or three. I'd go with Bishop Wisecarver dualV wheels and guides. I'd guess that a set of 8 footers will cost about $500. You'll spend at least $200-$300 on what you're describing, and it won't work nearly as well and will be a lot more work.

The reason I went the conduit and skate bearing route was cost, nothing more. A 4x8 router should be fast, strong, and durable. I'd go a different route, imo. :)

My 4 x 8 uses channel as the rail support, as with Gerry it was done this way to save money. The intention was to upgrade when the need arose; it hasn't yet. More show and tell on the site, www.solsylva.com

It would probably be wise to decide between the leadscrew and rack before doing anymore real design work. Finding a way to secure either will throw a wrench in the best drawn plans.
Good luck!