[Matt.]

Hi,

I have had this mill for a few years and it does ok but it suffers with regards to rigidity. The whole mill was really bad when I brought it but I fixed it up enough to use it. I can cut aluminium etc
I would like some suggestions with regards to the Z axis. Is there a better bearing guide arrangement?
There is probably some work could be done on the gantry. The weight of the gantry seems overkill and the bearing seem quite small.

Any suggestions are welcome. Links to examples would be great. Links to parts would be even better.

Attachment 239702

Thanks in advance

Matt

[LeeWay]

Actually I think the Z axis is in the ball park. Very little room for that unsupported rod to flex. It is your gantry rails and bearings that will cause the most issues. To cut better with the machine as is, cut all parts as close to one side or the other as possible. Never in the center of the machine if it can be helped. The rails do flex causing deflection of the spindle. That is undesirable and will show in the surface finish. I do see evidence of it in the part you show. Personally I would use profile rails and trucks on it if I planned to continue to cut metal. Looks like it would be fairly trivial to remove your rails and add new ones. You would need to get the base for the new rails at the right height and possibly need to shim the rear of the Z axis, but doable. You could leave the drive train as is.

Looks like your X axis has supported rails. More details on it would be good. It too could stand some profile rails.
It would be an upgrade for sure, but the performance you will gain over what you have now is substantial.
You can resell the parts you remove to recoup some of the cost as well.
Watch Ebay and you can land some nice linear rails for very low cost. It may take a bit of time, but since you have a working machine, it may not be so critical.
Also ordering brand new rails may be an option. Something like Hiwin rails may not break the bank.

These are from Ebay.
THK LM Guide Bearing Used HSR20R 520L 30mm Pitch NSK Linear Motion 2RAIL 4Block | eBay

[awerby]

If you swapped out the Y axis rails (which appear to be unsupported round rails) for profile rails, that would improve rigidity a lot.

[Matt.]

Thanks for your advice so far.
I have brought some supported rails for the y axis just have to figure out how to fit and support them. Need some sort of right angle sections on the current uprights and a back plate. I'm a little worried about increasing the weight of the gantry. The current uprights are 25mm solid ali and there is only a 1.5 mm thin plate folded on the rear and top as you can see pictured.

Z Axis

Attachment 239906

Upright

Attachment 239900

X Axis

Attachment 239904

The bearings on this axis look small for the gauge of material used in its construction? I also feel its a little narrow for the height of the bridge and have noticed a little walking when traversing the X Axis. Im not sure what to do about this..

Matt

[LeeWay]

You will most certainly need to add weight to the gantry. That sheet metal is not sufficient. The easiest and lightest might be an 80/20 extrusion under each profile rail. Mount those to and back them up with an aluminum plate. This will add weight, but more importantly, stiffness. My gantry weighs about 150 pounds. Mine also has a single screw drive for the gantry. You mention your gantry is walking. This is called racking. Without any adjustments built in, this will be difficult to eliminate. Even with nice profile rails. What I did to overcome racking was to preload the bearings on the rails. Mine uses skate bearings on flat bar steel rails, but the principle is the same. I did this by using trucks on each side of the gantry uprights. Mine are about 10" long. Each has a few push / pull screws and set screws. I can apply pressure on the rails in three places on these trucks. This also helps square the gantry and it stays there.

You need a rather stiff table to load the rails like this though. Not sure what your table webbing looks like. It is starting to get pretty involved to upgrade like this, but I do think you will have a much more capable machine than what you have now. Here is an early image of how I did my gantry trucks. Note the adjustment screws on the sides to preload the bearings on the rails. I do not yet have the table structure installed in that picture.

[dmalicky]

I agree with LeeWay that the major flexy spot is the unsupported Y (gantry cross axis) rails. SBR rails would be a big improvement. Profile rails (Hiwin, THK...) are indeed far better, but these are only justified if you also fix all the other flexy spots -- the weakest link in the system always dominates the overall stiffness-at-the-cutter.

Right now, the Y and Z axes are the worst of both worlds: unsupported rails, but 'open' bearings that are designed for supported rails. These open bearings are much more flexible in tension (pull-off) than compression:
http://www.cnczone.com/forums/linear...ml#post1412778
Your Y bearings are the TBR type. They are supposed to be heavier duty, but my 4 samples were *more flexible* than the SBR type (your Z bearings).

Your 1" thick uprights are a good thing: while not nearly as stiff as T or box section, that thickness will give much more lateral stiffness than a ~1/2" plate, and probably enough even after the changes below.

The gantry cross-member needs to be some kind of large tube, either an 80/20 extrusion as LeeWay suggested, or a standard rectangular/square section if you don't mind more difficult mounting. 1 BIG tube is better than 2 medium tubes. For the most convenient option that is also stiff, I'd suggest a Misumi GFS8_100200: http://www.cnczone.com/forums/diy-cn...ml#post1413406 (that post is for a 49" travel gantry with 10" Z; your size machine is less demanding, making that GFS8 about right.)

The X axis is the least important for stiffness-at-the-tool because it doesn't have the large cantilever effect like the Y and Z. Fix that axis only after improving the Y and Z. Yes, 150mm is a narrow spacing for those bearings; I'd suggest 250mm or more. The racking will be greatly improved by a stiff gantry cross-member. It's good that these rails are near table height (not far below).

Here's a prioritized list at various $:
1) At minimum, change Y and Z bearings to closed type. Usually called SC20UU or SCS20UU.

2) Change Y rails to SBR or profile type (Hiwin, THK...), and mount them to a GFS8_100200. For SBR, mount them oriented up-and-down (vertically opposed) on the gantry tube: http://www.cnczone.com/forums/diy-cn...ml#post1426678 (also see post #48). Profile rails should be mounted conventionally (both facing forward). For any rails, space them vertically apart by about as much as the Z clearance (or travel).

3) The Z axis will now be the weakest link. It's more difficult/awkward to get an SBR Z axis to be stiff. By far the best option for Z is profile rail/bearings, and with only ~150mm travel, these are pretty affordable. As with the Y rails, the vertical spacing of the Z bearings needs to be ~similar to the Z clearance (or travel), although a little narrower is ok for profile bearings (since they are massively stiff).
The Z car/plate is also a flexy spot: increase thickness to ~1", or better, add 'legs' to your 1/2" plate to make it a channel section. The idea is to prevent a bending 'hinge' between your stiff spindle mount and the Z bearings. It would also help to raise that mount for more overlap to the Z bearings, since the large spindle body is stiff enough to handle more cantilever for itself.

4) Probably the next flexy spots will be
- The Y car/plate. Add legs to make it a channel section.
- The X bearings. Increase spacing to ~250mm or more. SBR is fine for this axis, if the spacing is large.

[Matt.]

Many thanks everyone for the very detailed advice.

So as I already have the new SBR rails the first thing to do is find some appropriate tube or extrusion to mount them to. I should be mounting them vertically opposed.
At pretty much the same time rework the z axis to suit the new y axis config and change over to linear bearings.

It will take some time to digest this information and come up with a plan for the z axis arrangement :-)