[noisy]

OK I've been lurking around the forums for a while and I think I've only posted a couple of times but this is my first attempt to start a project of my own. I've had this router design floating around in my head for months and I keep putting things on paper then changing the design. I've actually started over 2 different times because I didn't like the size of the foot print the machine would have, compared to it's cutting area. My machine is aprox 4ft by 2ft cutting area. I've been working on drawing this current design for about 3 weeks now and I'm happy with the way it's turned out so far. It's not completly done as you can see it's missing steppers and the spindle but you can see the basic design of the machine and thats all I'm really worried about. Please give me your opinions on the design.

Rails: 16mm rails / 2 bearings per rail / 4 rails on X axis / 2 rails on Y axis / 2 rails on Z axis. (I guess that depends on how you look at X and Y)
Lead screws: 5/8 Ball screw / 2 nuts per screw to help reduce backlash
Spindle: My own design because I'm not planing on using a router
(Not modeled yet thats why it's not in the pictures)
Held together like most 8020 builds: Nuts and Bolts

I use Rhino for my 3d modeling and I really enjoy it but there are things I've forgotten about how to do. Like being able to change the colors of different objects in rendered view.

[FandZ]

It looks good to me. I'm not sure why you would want to use 2 ball screws though. I don't know much about ballscrews but I know you can adjust the backlash on them, or at least on the ones I've seen. Also since you are useing supported rails, why use 4 of them on the X axis. I'd just use one on top and move the 80/20 to the base of the mounting plates.

It's hard to get a good prospective on size from the pictures, but The gantry looks a little tall to me. Especially for that Z axis your showing. You'd have to use a longer drop off plate to mount your spindle and even then you wouldn't be able to retract it high enough to make use of all that space under your z.

What materials and sizes do you plan on machining?

[noisy]

not useing 2 screws. I'm useing 2 nuts on the ball screw because they are cheap nuts and the screws are also on the cheap side. put 2 nuts together and you can put a spacer between them to tighten up any backlash.

Why am I useing 4 rails on the x axis? Because 16mm rails are only .629 (5\8") diameter and they aren't actually supported rails. The supports you see are only 3 inchs long and will be cut out on an EDM machine at work. The X axis has 4 rails because their is one on top and one on bottom of each side of the x axis.

I know the gantry looks kinda tall I thought the same thing but I can assure you their is only 7 inchs of clearance from the bottom of the z axis to the top of my working table. If my spindle sets 1 inch below that then I now only have 6 inches of clearance plus the travel that I have left. I can shorten it up if I need to but for now I'm going to leave it once I get the table base done and I start on the gantry I will have a good idea if the current gantry setup is going to be to much.
I Plan to use the full size of the machine. As far as material...Anything I can, all different kinds of wood, thin copper and thin aluminum.



Here, this shows the dual rails on the X axis, as I said 1 on top and 1 on bottom I planned it this way to help with rigidity.


This picture shows the machine with the gantry and simulated top hidden from view. It reveals the 2 nuts on the leadscrew. As I said before about Rhino I've forgot how to use a few tools like getting revolve to work so I could make the lead screws look like actual screws and not just a solid bar.

[FandZ]

I meant Ball nuts not screws. I get the two nut setup.I was thinking you wanted to use two ball nuts together.

On the Z axis, I'm assuming everything is aluminum and steel, unless you plan to be able to mount the z axis at different heights, you may as well move the bottom Y axis rail to the bottom of the Z axis. You could move both Y axis rails down some and still have that 7 inch clearance and shave some of the gantry height.

[ger21]

I would at least double the width of the gantry sides at the base of them, spreading the bearings apart. Also, it might be cheaper to just use a single supported 20mm rail.
http://www.vxb.com/page/bearings/PRO...ystems/Kit7429

And a whole lot easier to build. I'd be concerned that the unsupported rails can still flex a little between the supports. If you insist on doing it your way, make the supports half the length, and use twice as many, cutting the unsupported areas length in half.

The gantry acts like a huge lever, and the forces on those X rails is a lot. I've found that to be a weak link on my machine, causing poor cut quality as the gantry vibrates front to back with rapid direction changes. even a very small amount of flex is undesirable.

[noisy]

I meant Ball nuts not screws. I get the two nut setup.I was thinking you wanted to use two ball nuts together.

On the Z axis, I'm assuming everything is aluminum and steel, unless you plan to be able to mount the z axis at different heights, you may as well move the bottom Y axis rail to the bottom of the Z axis. You could move both Y axis rails down some and still have that 7 inch clearance and shave some of the gantry height.

Yes the whole machine will be alluminum. I will have to look at my z and see what can come up with. I guess I might just have to move it down, I wanted as much clearance as I could get. LOL

I would at least double the width of the gantry sides at the base of them, spreading the bearings apart. Also, it might be cheaper to just use a single supported 20mm rail.
http://www.vxb.com/page/bearings/PRO...ystems/Kit7429

And a whole lot easier to build. I'd be concerned that the unsupported rails can still flex a little between the supports. If you insist on doing it your way, make the supports half the length, and use twice as many, cutting the unsupported areas length in half.

The gantry sides are 6 inchs wide and 3/4" thick alluminum. (Drawn anyways) The distance between where the rails is supported and not supported is 7 inchs so their is only 1/2" on either side of the gantry that can be unsuported at anytime.

I know where your comeing from on the 20 mm rails, I don't want the supported sets they sell, I want the tapped rail alone. The supports you see I will being makeing at work. Well that just sucks, I was looking on VXB.com when I started this and I wanted 20 mm rails but I didn't see that they offered a 20 mm rail that was unsupported but also tapped for supports. I can get a 60inch tapped rail for $50.00, add 2 bearings and your total is only at $100.00 plus shipping so I saved myself $60.oo right off the bat because a supported 20mm rail with 2 bearings is $160.00. (16mm rail with 2 bearings is the same price $100.00)

[ger21]

Not sue why you'd rather make them than buy them. How much would you be saving?

And I don't the 6" is wide enough. As I said, I'd make them 12". The farther apart the bearings are, the more rigid it'll be.

[noisy]

Not sue why you'd rather make them than buy them. How much would you be saving?

And I don't the 6" is wide enough. As I said, I'd make them 12". The farther apart the bearings are, the more rigid it'll be.

Well their is more then one reason.

1st and this is the biggest reason. They won't set on the 8020 rails the way I want them to.
2nd. They don't charge me for useing EDM wire at work because they want the machine running as much as possible. Trust me I double checked that last one and am supprised as you are.
3rd. Because I can

[Rodm1954]

High gantry sides on less than rigid rails can lead to a series of short jumpy steps - porpoising is the best way I can describe it. I have seen machines made from MDF and unsupported pipe that showed this and you could hear the spindle loaded and unloaded as the wave action worked.

Can you design your X axis top rails to be higher - ie closer to the Y axis gantry beam to reduce the overhang. By doing this you can keep your high Z axis clearance without the associaited problems. This design element can incorporate a wall (shield) to keep the sawdust, etc from speading over your workshop and X axis rails.

I'm not sure about 12 inch gantry sides as the trade off is a loss of X axis travel but 200mm is about what I work on and 8 inches in your currency.

Another point in design is to get the centre of the spindle as close as you can to the centre of the X axis bearings. In your drawing add the spindle and mount to the Z axis plate and work out how your design looks for centering over the bearings.
Centering the spindle to the X bearings balances the gantry directing the forces through the bearings rather than a lever pulling or pushing on one side of the bearings.
The second advantage is it brings the cutting envelope into the centre of the table. In extreme examples I have seen designs where the back 12 inches of the table is never used and the spindle projects over the front of the table at the other end. A cranked gantry side is the easiest way to overcome this so if you have access to machines that can cut plate at work then you might want to consider this. If not then jig your Y beam and Z axis components to get it the best you can.

Last think about mounting jobs on your table. It is rarely considered at machine design time but ends up being the biggest pain when using your machine. Vacuum, T slots and cross table clamping are a couple of ideas and I am sure there are more. A lot just wack a spoil board on the table and screw the job into the spoil board but this is not the answer for everything - eg engraving the lid of a jewelery box.

A couple of ideas for you to think about.

[noisy]

OK a bunch of changes to the drawing. I make a few changes and then I find something else that could be better or done differently and I can't help but change it. Anyways here are the changes I've made to the model.




I'm not sure what rails I'm gonna use on the Z axis I want to find something more compact then the 16 or 20mm linear rails I used on the X and Y. O and by the way that model was redrawn with 20mm rails.

[awerby]

The problem with this design (which is hardly unique to it) is those high sides to the gantry, which will be vulnerable to increasing side-to-side vibration the higher they are. This scheme works best in machines with a very low Z-axis, but the higher that gets the worse it works. Many people who are designing CNC routers base them on a commercial machine intended for cutting pieces out of plywood , but since they want to be able to handle taller parts, they just increase the height of their gantry, which makes their design much less rigid.

Instead, if you want a high Z axis, consider building rigid walls on the sides with stiff rails at the top, and run your gantry between them, instead of perching it on tall members rising vertically from the sides of the table. This way, you have less weight to push around, since you've eliminated the side members of the gantry, and also less inertia. You'll also eliminate that side-to-side wobble, which will improve cut quality a lot. The Mecmate works that way (although their standard model keeps the side walls low), and it's a big improvement over the high-side gantry design - see http://www.mechmate.com/ ]

Andrew Werby
www.computersculpture.com

[noisy]

The problem with this design (which is hardly unique to it) is those high sides to the gantry, which will be vulnerable to increasing side-to-side vibration the higher they are. This scheme works best in machines with a very low Z-axis, but the higher that gets the worse it works. Many people who are designing CNC routers base them on a commercial machine intended for cutting pieces out of plywood , but since they want to be able to handle taller parts, they just increase the height of their gantry, which makes their design much less rigid.

Instead, if you want a high Z axis, consider building rigid walls on the sides with stiff rails at the top, and run your gantry between them, instead of perching it on tall members rising vertically from the sides of the table. This way, you have less weight to push around, since you've eliminated the side members of the gantry, and also less inertia. You'll also eliminate that side-to-side wobble, which will improve cut quality a lot. The Mecmate works that way (although their standard model keeps the side walls low), and it's a big improvement over the high-side gantry design - see http://www.mechmate.com/ ]

Andrew Werby
www.computersculpture.com

I'm not useing plywood or MDF. My gantry sides will be around 3/4 inch thick alluminum. I didn't eliminiate the Gantry at all I only took the gantry sides off so I could work with the model better. If you notice I've moved the Z axis closer to the table by several inches. I've also moved the linear rails for Y axis to the top and bottom of the arm connecting the gantry sides, that way the z axis is closer to my y axis rails. The way it is drawn right now their is just over 3 1/2 inchs between the bottom of the z axis (spindle plate) to the top of the table. The distance between the Top bearing on X axis and the top bearing on the Y axis is only 10 1/4 inchs.

[noisy]

OK what do you guys think about a V Rail system for Z axis? Nope still don't have gantry sides on their yet because I'm still working on the spindle and spindle mounts and I want to center everything up between the bearings on X axis like was suggested before.