Second machine design process

[ger21]

Look at the second picture in my thread. I milled a flat for the bolt head. I didn't use the cheap 1/8" angle. I have 1/4" x 1-1/4" 6061 angle. The 1/8" angle you can twist in your hands, the stuff I got is rock solid, much stronger.

Gerry

[yukonho]

I have had good success with the 1/8 angle. Over 8" I can not twist it by hand. My machine is significantly smaller than yours is, and I run a dremel for a spindle. I will consider the 1/4". I am not sure I can use it with 1" rails. There would not be enough room on the back side of the aluminum angle for the heads of the bolts that mount the bearings. I will have to check this in CAD.
co

[yukonho]

I must say I am very happy with the design progress thus far. Going with a fixed gantry and moving table has allowed me to design an integral enclosure. I would love to use some "real" linear ways, but cost is just too much of a factor. I am convinced that this design will be what I am after though. Here is a side view in 2D

[yukonho]

The integral enclosure will help increase rigidity even further. Cant wait to put the saw to the MDF on this.
co

[pminmo]

Originally posted by yukonho
I must say I am very happy with the design progress thus far. Going with a fixed gantry and moving table has allowed me to design an integral enclosure. I would love to use some "real" linear ways, but cost is just too much of a factor. I am convinced that this design will be what I am after though. Here is a side view in 2D

One comment. I've had a simalar thought on my second router, Posting a picture here. My thinking is a cutout that would allow long pieces to be cut along the long axis.

Phil

[pminmo]

front view

[ballendo]

Hello,

Instead of using 1-1/2 x1-1/2 x 1/4 thick alum angle, and mounting the screws which hold the bearings in the middle, away from the corner; and using long screws to hold the bearings with brass spacers, further increasing the possibility and probability of flex...

Use 3/4 x3/4 x1/4 aluminum angle. Drill the holes for the bearing screws as close to the corner as you can. Since the corner is the strongest part. With skatewheel bearings we're using 5/16 bolts (or 8mm), so the closest you can get to a 1/4 wall corner is 5/32 plus 1/4, which is 8/32, so the total is 13/32. Add a 32nd for clearance, and you're at 14/32, or 7/16.(11.25mm)

Already, with just these two changes, things will be much stiffer (keep in mind that with his large diameter rails, he might NEED the larger angle. The point is to use the smallest angle you can, and drill as close to the corner as you can.

Next, mount the bearings as close to the angle as you can. I usually buy a bunch of hardware store washers, and rub 'em over some fine sandpaper to get rid of the high/rough spots. Then I get an egg carton, and with my dial calipers, start measuring the thickness, sorting the various thicknesses into their own piles. (In a hundred washers, I'll almost always "prove" the bell shaped curve of statistics) That way i have cheap "calibrated" washers for when I need them all the same, AND for when I need one thicker or thinner for "shimming" purposes...

These are all just basic engineering principles.

The next part was in reference to HIS large diameter rails specifically. He has "room" to use TWO pieces of angle, with only TWO bearings on each piece (instead of one angle with 4 bearings). This way you can adjust the movement of the carriage(s) very precisely.

I've seen other posts which mention using the 4screw rail end mounts to dial in the machine, but this won't dial in everything accurately. you need to dial in the RAILS, THEN dial in the carriages TO the rails, if you want to have planar movement of your axes. (Please remember that you don't HAVE to have planar axes to make stuff. lots of machines here are examples of that truth. But they're not truly accurate, and it's fortunate that the parts made don't require them to be...)

Hope this helps,

Ballendo

Originally posted by yukonho
.
Regarding one of your p.s' Could you please further elaborate on
--"Wanna make your machine stiffer and add adjustability too? replace the single 1/4 angle with 2, and lose the brass bearing spacer/extenders. By moving the one or the other of the angles, you can dial in the y slide--and it's "carried" z. You "would have" also had shorter distance between the top and bottom of your y/z sliding parts. And so a shorter gantry moment. Now you could just increase the thickness of one or both of the top or bottom. Could even make a torsion box with the added space! "--
I cant picture what you are saying here.
co

[yukonho]

I follow your ideas.
It is important to keep the bearings center line normal to the rod they are rolling along. Keeping this in mind, if you drill the mounting holes closer to the corner of the Al angle, then your bolts need to be longer (the bearings have to be further from the aluminum angle) to keep them normal to the rod. This increases the moment arm of the bearing on the aluminum. I have not done the FBD or the math, but I believe that the forces would be very close for either setup.
My measurements of the bolts I use suggest I can make the centre of the bolt hole 1/4" from the inside face of the aluminum angle, but then, like I stated earlier, my bolt would have to be longer, 6 of 1, half dozen of the other.
For the record, I am still looking for inexpensive linear ways to use instead of these rollerblade bearings. It will save so much time in setup and alignment and building and everything.
I like the open side idea. The front view doesn't register with me for some reason, I cant figure out what is what. I have designed my machine for specific purpose, and I dont envision making anything bigger than what I designed it for. For the record, my machine now is 12" X 36" travel and I have never used it for more than 10" travel.
Also, I want to keep this fully enclosed for noise and dust reduction. I am very happy with the enclosed design. I think it will look very cool when done.
co

[ger21]

Ballendo, like yukonho says, I designed it with the centerline of the bearing on the centerline of the tubing. The 1/4" thick 1-1/4" angle is really stiff. I can't flex a 4ft piece in my hands. The 1/8" angle I can twist a 1ft piece in my hands. The way I have mine set up, the bearings are in compression on the tubing, and if there were any flex, the compression takes it out.

Gerry

[ballendo]

Hello Ger and Yukon,

Would you say a full size 1500 lb. cast iron Bridgeport mill is stiff?

Because with an indicator in the spindle it only takes "finger" force on the table, ram, or quill to move the indicator!!!

I'll bet you'd have similar results with the 1/4" alum angle

As to the "normalling" of the bearings. Yes, when you use ONE piece of angle, you're limited. I was talking about using two, so adjustments could be made. Also, the strongest part of the BEARING is ALSO near "its" corner...

If we're gonna talk about how to make an inexpensive machine stiff, we can't just let all these little bits that "can't be felt" stack up against us...

Ballendo

P.S. Hopefully the BP mill example shows that there IS flex. Always. So the EMT bends a bit, the bearing itself bends a bit. The shaft it's on bends a bit. The 1/4 aluminum bends a bit. The mdf/BB ply bends a bit...

The compression "pre-loads" your assembly. But that doesn't necessarily change much in your result once you start cutting and dynamic loads are flying around. Then it's about moments, and levers, and material deformations, surface hardness, contact areas, PSI, etc...

Both your machines are good. Please don't infer anything less from what I'm writing. But they can be better, and without costing more... I'm just trying to point out possible directions.

Originally posted by ger21
Ballendo, like yukonho says, I designed it with the centerline of the bearing on the centerline of the tubing. The 1/4" thick 1-1/4" angle is really stiff. I can't flex a 4ft piece in my hands. The 1/8" angle I can twist a 1ft piece in my hands. The way I have mine set up, the bearings are in compression on the tubing, and if there were any flex, the compression takes it out.

Gerry

[ballendo]

Hello,

Forgot 2 things:

Skewing the bearing intentionally will give a "wiping" action which helps keep the rails clean, at the expense of some additional wear and slightly greater axis force required. Just a bit, don't
overdo it! And you can rotate your rails, if needed to deal with wear, right?

Next, with bearings mounted at 45's and running centerline on centerline, the part of the bearing "above" this centerline contact point becomes a trough for all the swarf, dust and chips that fall from above. Which the bearing then grinds into the rail.

If you plan it right, the bearing corner can very nearly meet the tangent of the rail on the upper contact points. This will reduce the "trough" effect considerably. AND, as I mentioned just a moment ago, the bearing is stronger that way too!

Ballendo

[ger21]

I'm not arguing, keep the advice coming. Until I get the whole thing done I won't know how much it will flex. I understand exactly what your saying, I'll just have to wait and see. And if it's not stiff enough, then I get to build a better one. I've got some ideas for making up some composite parts for the next one.

Gerry

PS. I run a point to point machining center at work with a 9HP router, and I've seen some flex there. So I do know that even big steel machines aren't as stiff as they could be. It's relative I guess, kinda like precision and accuracy?

[Mr.Chips]

For what it's worth I used steel angle iron for my bearings, much more rigid than alum, still used a bolt and nut rather than threading the angle iron.
Hager

[yukonho]

I think I will stick with running the bearings centred and running normal to the pipe. Seems easier to set up that way. My tools are limited, so I can not use steel angle.
My design is essentially done now, and I am going to put together my cutlist and BOM so I can get going on it Monday. Fortunately Home depot has that panel saw so if I get my list together I can pay a couple of bucks and have most of my wood cut to accurate dimensions then just a couple of cross cuts with the miter saw and away we go. I am going to cut the rod supports on my CNC and I am going to try to do the holes in the angle Al on the CNC too. I will probably just end up drilling pilot holes though and hand drill them to full size after.
Speak now if you have suggestions please, the end is nigh!!
co
P.S. I am still really happy about the enclosed design. I think it looks sweet.

[yukonho]

I am going to use one of each of these delrin nuts on each end of my table. Having one on each end should give me an anti backlash effect.
co

[yukonho]

Y axis torsion box pictures. I am going to fill the cavities with ground up recycled tires.

[yukonho]

Love the torsion box idea, thanks for the link to that thread. Makes for a compact and super strong axis.
I am going to do the same on the table. (moving gantry design)

[Mr.Chips]

yukonho
I saw an article somewhere about filling voids such as you have with a mixture of sand and oil. It does two things, reduces viberation and absorbs sound. You could even fill the pipes.

Nice torsion box, What size pipe/conduit was it made for?
Hager

[High Seas]

Is it just me - or has someone else noticed a missing post here too?
Earlier today there were several posts following Hager's (Mr Chips) that wrote about using sand including a reference to using recycled rubber tire bits for dampening in the internal areas of the torsion boxes, several references to sand in the chanels in aluminium extrusions, a reference to speakers and dampening - and another reference to riding motorcycles and lead shot in the handlebars.
Is the system dropping stuff - is my system acting up - or have these been moved? The post numbers are all in sequence and the latest shown on this thread is 37 - then this one.
Just trying to give as much info here if there is a sys dropout - else I should stop repainting this house?!!!! Nawww its not fumes -.

On the other hand - if they've been moved - how would I know that - would it show on the "New Posts Roster" on sign on?

Jim

[pack rat]

Yes you are correct High Seas, Also I think that yukon said that he has access to rubber and that he is going to use it due to its density. The paint that you are using must be latex because your memory is correct