My latest project, Machine #2

[Lionclaw]

My first router was built following the JGRO plans with a few tiny modifications. The machine is nice, but has some shortcomings such as gantry flex, leadscrew whip, and pipe flex/sag.

I've taken what I learned from the JGRO machine and applied it to a new design. Lots of ideas borrowed from Steve(spalm) and Gerry(ger21). Huge props to Steve for his trial and error efforts, and for persisting until he finds the best solution to a problem. I've learned a lot from his posts. Big thanks to Gerry for always being one of the first to answer my questions, no matter how noobish or stupid they were.

Here's a few pictures of my 80ish % completed model. I still need to add fasteners, motor mounts, bearing mounts, and probably a few other things.

The machine will have a cutting area approximately 40"x27"x5". I'm considering extending the X out to 48" so that it can work an entire strip cut from a 4'x8' sheet. Most of the machine will be made from 1/2" baltic birch plywood. I'll probably use a bit of MDF here and there.

-Andy

Similar Threads:

• My Latest Machine - CNC #4
• Latest metal casting project...brass shifting knob

[Lionclaw]

So far I've ordered the following parts:

(2) 6' 1/2-10 Acme Rod (Enco, $22.25 shipped)
(12) 1/2-10 Acme nuts (Enco, $12.80 shipped)

(3) Dumpster AB Nuts W/ Square Flange ($51.00 shipped)

(16) abec-7 Skate Bearings (ebay, $12.90 shipped)
(10) 1/2"x9/8"x3/8" Shielded Ball Bearings(ebay, $15.90 shipped)

As of now I've recieved the acme nuts, acme rod, and 1/2" bearings. The bearings look like very nice quality. The markings on them are "1616-Z" and "CGB"(or maybe "CG8"?). I was going to order them from McMaster at nearly $6 each. I saved almost $50 by getting them on ebay. They are a perfect fit for the acme rod.

For the Z axis, I'm planning on using 2 pieces of 3/4"x12" hardened precision ground steel rod (mcmaster 6061K34). For Z bearings, I'm currently planning on trying 3/4" bronze flanged-sleeve bearings (mcmaster 2938T19), unless someone can convince me otherwise. They're only $0.88 a piece, so if I get them and decide not to use them it's no big deal. If they don't work I might try milling out my own acetal bearings.

[zoltan]

Wow...SUPER design. It is GREAT. May I ask you to share with me your plans?

Thank you.

Zoltan

PS: What CAD software did you use to design. I have acces to SW.

[Lionclaw]

I definitely plan to share my work. The model has been created in solidworks. I was hoping to post both the solidworks model as well as a set of PDF plans laid out similarly to what JGRO did.

The only downside to my design at this point is many of the parts are too complex to cut by hand. The torsion boxes use ribs with lots of notches and curves. Many of the dimensions I chose for this machine were picked simply because it would make the parts a size small enough to be cut out with my JGRO machine. I'm sure someone could come up with a way to simplify them.

But, learning from my own experiences as well as others, I've tried my best to eliminate sources of flex from this machine. As you see by looking at the Z carriage, I've tried to add a lot of structure.

I'm hoping that others can help point out possible problem areas in my design. I want to get this thing as perfect as possible before I build the prototype, and even then, I'm sure I'll learn a lot more from the prototype.

I'll have, at minimum, the solidworks files posted as soon as I finish adding the missing elements of the model.

[jmytyk]

Hey! did you break into my computer? haha, just kidding, our designs are remarkable similar. i have a working area of 25x50x5? (Remember MDF is one inch bigger than 4'x8')

a few comments / questions. do you see any benefits of having the pipes extend through their respective ends? (ie. the gantry uprights and end-plates) i was going to leave mine inside the end plates so that i can have a bit of torsion adjustment on the Y-axis (25" axis). then compress the gantry together with threaded rod.

Also on the threaded rod idea, I was going to borrow Gerry's threaded rod idea of tying the top and bottom of the Z-axis together. the thought being, the rod would counter act the pressure being transmitted through the structure by the bearings pressing against the tubes.

Z-axis is a dumpster find of a linear bearing type carrier...

Gerry's threaded rod idea
http://www.cnczone.com/forums/attach...tachmentid=644

Check out a slight difference in the placement of the x-axis tubes (long axis))

( in the drawings, red lines are threaded rod lines)

thanks in advance for everyone's feedback... _Jon

[jmytyk]

(oops, missed an image)

[Lionclaw]

I let the pipes run through the ends so I could pull them together with threaded rods at the end.

I like Gerry's Z design too. But, I seem to remember him posting in his log that it was flexing, and that he was going to box it up.

[zoltan]

Thank you for answer. It is great for me even if the parts should be build on CNC router. Also, it does not matter if the design is not fully refined. I am ready to start to build Joe's second machine, and your design came as a true opportunity for an improved machine. Sorry for pushing you but do you think that it is possible to share with me the plans even in this stage as, let's say I can hardly wait anymore to start building the machine.

Thank you.

Zoltan

[Lionclaw]

Lol, I know how that is Zoltan. I've got to crash right now, but I'll be back online tonight between 2100 and 2330 PST. I'll see what I can throw together then, and hopefully we can exchange IM information so I can get some feedback as you try to make some of the parts.

-Andy

[ger21]

I let the pipes run through the ends so I could pull them together with threaded rods at the end.

I like Gerry's Z design too. But, I seem to remember him posting in his log that it was flexing, and that he was going to box it up.

I'd still use the threaded rod to keep it tight, but yes, you do have to box it in, or it will move in both directions.

That looks pretty cool, btw.

[spalm]

Dang those are pretty drawings. My designs are always so kludge looking.

I see that you have a lot of rabbets on your Z structure, but I think it still needs some thought. Two forces are going to work on destructing this box.

1) When the bearings are tightened, over time it will want to separate at the back top or back bottom. I don’t see anything really holding it together.

2) When the router it attached, it will want to parallelogram forward, again because the back edges won’t stop it.

A couple of suggestions to help this:

Add a ¼” ply skin to the sides with the square hole routed out. (This really makes another torsion box).

Or

Remove the all the side pieces and replace with a single ½” slightly oversized plate with a square hole routed out, and dados cut into it for the other four sides to lock in.

Or

Extend the back above and below and dado in the top and bottom plates.

Or

Extend the back stiffeners to a “C” shape to hold down the top and bottom.

Or

Do Jerry’s threaded rod concept.

Or

Something like that.


This is going to be fun to watch,
Steve

[jmytyk]

Hey Andy, Steve hit on the same idea i was trying to say- with the bearings pushing the thing apart. (Thanks steve) = threaded rod idea. Although this is a good one as well:

"Remove the all the side pieces and replace with a single ½” slightly oversized plate with a square hole routed out, and dados cut into it for the other four sides to lock
in."

either idea, do you think you could gain some clearance under the gantry? in my design "ideally" i would like to have as much clearance as possible. -giving more flexiblity in the Z if my needs change.

My crappy CAD screen shot didn't show the x-axis threaded rod, and the Y-axis, thanks for the mistake catch...

(now how to add in adjustment so that that the Z axis is perpendicular to the table?)

BTW- Does any one have the pin out using Mach2 and a CNC4PC board?

_Jon

[Lionclaw]

Thanks for all the great comments. I see what you guys mean, and I agree it might be a problem. It looks like most of the stress at the top will be towards the center, and there isn't really a whole lot of support there. I think I might try adding "L" shaped stiffeners. I'll go model it out and see what you guys think.

As for gaining clearance under the gantry, the best bet on this design would probably be to raise the torsion box a few inches and lower the point where the router mounts a few inches(extending parts as necessary). I mainly cut parts out of thin materials, so I didn't give Z clearance and travel a whole lot of consideration.

[ger21]

Just do what Steve said, and what I'm going to do. Add 1/4" side panels to the Z-axis and box it in. Make it a pretty close fit to the tubes, and you could add some type of wipers to keep the tubes clean, maybe felt or something.

One thing you might want to llok into, is increasing the spacing of the X-axis bearings. The farther apart they are, the stiffer the gantry will become, at the expense of a longer machine.

And as for Z-axis clearance, if you know you won't need it, keep it as low as possible. The higher it gets, the weaker it gets.

[Jason Marsha]

Nice machine Andy. The design looks strong and rigid and with the extra support, most if not all flex should be minimized. I have found that a problem with the wooden machines is the tilting (twisting) backwards of the gantry axis (the one with the z-axis attached) when the spindle is plunging into the work.

Jason

[Lionclaw]

I've been doing some thinking about my Z axis bearings. I'll probably go ahead and order the bronze bearings just because they're cheap. As a backup plan I've decided to make some bearings out of UHMW PE. I made a sample out of 3/4" MDF just to see how it would turn out. I'll try making one out of UHMW once my steel rod arrives and I can test the fitting.

[Lionclaw]

I've made some changes to the model.

The walls of the Z carriage are now single 1/2" thick pieces with 1/4" slots in them. This should make it a lot stronger, reduce the number of parts I have to make, and make it much easier to assemble.

I've also extended the table out to 60" and widened the base of the gantry to 10", so the effective cutting area will be about 26" x 50" x 6". This will make it the perfect size to cut parts from quarter sheets, or the 2'x4' "handypanels" sold at many building supply stores.

I still want to tinker with the gantry walls a bit to make them look better. I'll see what I can come up with.

I'd love to hear more ideas and suggestions!

[Lionclaw]

I just got back from my local home depot. I intended to buy 2 sheets of 1/2" baltic birch and my pipes. I took my calipers with me this time, and to my dismay I found that 1/2" plywood is really ~.463".

So now I have the option of adjusting my model for this, or using MDF in more places than I had planned. I'm not exactly sure what to do. I think I'll check out a few actual lumber yards before I make a final decision.

[ger21]

Home Depot's around here don't have Baltic Birch. Real Baltic Birch is 12mm. 1/2" Birch plywood is ~15/32. Most hardwood plywood is 1/32" undersize for thickness.

[Lionclaw]

Gerry, when you built your table did you take into account the thickness and adjust your model? I'm wondering exactly how sloppy the box would be if I kept the .5" rib slots.

I'm also wondering what the consequences would be of building the torsion boxes from MDF if I went that route. I'm a bit concerned it might sag a little over the 60" span of the table. I'd still use baltic birch plywood in the gantry walls and probably a few other places if I went that route.