Hey all,

I'm in the process of designing my second CNC router. First one worked good, but it was too small. (16"x22"x5") It was robust, but it had play and wasn't very sturdy. So....
This one I want to be around 24"x48"x6-8". It will be a gantry style. My question is this: On the 48" axis (let's call this the Y axis), I want to use 2 stepper motors and ball screw. Also want to have 2 rails (thompson shafting) on each side. Kinda arrayed like this (end view):
lO...l
l O..l
l__Ol
Middle one being the ball screw....

I want to mount a spindle capable of light to moderate cuts in aluminum (engine work) possibly a rotating head for CNC port and polishing, so figure about 75 to 100 pounds for the total gantry weight. What size shafting should I use? The last router used 1" single rails, but it was no where near the size. Would 2 .750 shafts per side suffice? I was adding up the cost difference, and I would like to somewhat keep this in budget. ie, don't buy more than I need... With all the linear bearings and everything, it's almost a $200 difference....

Thanks in advance,

Seth

Also, would the 5/8" ball screw from Mcmaster Carr be sufficient, or should I look at something bigger? I'm looking at speeds less than 200 ipm for rapids, and about 100 for cutting

The 1" would work *if*... It all depends on whether or how you support the rails. if you use the open bearings and support the shafting so that you have no open unsupported span larger than about 75% of the distance between your gantry bearings you could get away with the 1".



Tiger

Tiger,

I only ask because I want to learn from other peoples experiences...Have you tried a similar set-up? Is there reason to worry about deflection?

Unless they are supported, deflection will be an issue no matter what size. Especially cutting aluminum. If you must use round rails, use at least 3/4" supported. THK type would be a better choice, though.

que?

The thompson super smart bearings for the thompson shafts actually have quite good ratings, I was looking at them, and there 3/4 inch super smart and there 1.25 in super ball bearings for round shafts had the same load rating. They advertise them as cost saving, because you can use a shaft that is much smaller. I am going to use those with 3/4 in. fully supported shaft. The rigidity of the shafts depends alot on how they are supported. Oh I forgot to mention that the super smart bearings are only a couple dollars more than the regular bearings, so they are quite reasonable.

actually, the 1" super smart are $7 less than the regular 1" linear bearings....hmmmm..

It's the support rail + shaft assemblies that are the $ killers. Shafts have to be drilled and tapped, which means either expensive machining post hardening or very finicky heat treatment with high reject rates. Production costs and consumer prices go up either way.

The places I've seen that sell rails separately seem to think their aluminum extrusions are gold, also. ;)

No real cheap way to do it unless you have means of drilling and tapping standard hardened rod yourself and making your own rail segments. (at least not that I know of)



Tiger

Ahhh, the beauty of being a machinist....
Option #1: skip the tapping. From the top, drill and counter bore. Not the bset option in the world..
Option #2: Only the surface of thomson shafting is hardened. Once you get past it, it's only 30Rc of so.....
Option #3: Drill an oversized hole, and epoxy in a theaded insert. Hmmm, that might just work.....

I agree, the costs are much to high. Hopefully these options will provide lower costs for the DIY group....
On a side note: I ran 1" shafting on my previous router 36" long unsupported on the Y axis. There was very little (.0005) deflection. true, the gantry didn't weigh more than 35 pounds, but.....

Now there's a market niche going unfilled... A precision drillpress jig with replaceable bushings and accurate indexing pins, all packaged with carbide center and spotting drills and tap, for the DIYers to drill and thread their own rods. Probably have to supply a floating holder too to prevent instant tool breakage on the inevitable $40 bench drillpresses LOL

EDIT> on the previous machine part, I bet you had a nice broad gantry base with a good spread between bearings. Makes a huge diff in deflection just how much free span there is, but that would probably be for some other thread ;)

Tiger