A Hapless EE Attempts a CNC Build

[Nick C The EE]

Attachment 247960

So now that I've got the machine running and cutting parts, I'm thinking alot more now about a total redesign. As predicted by others, it did not take long for me to see the limitations of this design. For instance, I've noticed all the parts I cut are trapezoidal.
If I cut around the outside of a shape, the cut looks like this -> /=\
And If I cut around the inside of a shape, the cut looks like this -> \=/

So now I'd like more input like alan's on how I can maximize rigidity and reduce flex.

Shorter uprights is at the top of the list. The z axis will be raised so it does not stick down below the Y axis (this was my original intention, but I drilled the holes for the bearings on the wrong end of the board...). This will let me shorten the uprights a few more inches as well.

I think the other thing holding me back is the table I've got the machine sitting on. It's one of those portable fold out tables you get at walmart and it's not very stable. I could use some ideas and pointers for that.

I'd also like to know if anyone has any tips on maybe a better/stiffer material to use than MDF that can be had for a reasonable price. JerryBurks' big bamboo is a pretty great inspiration, but I have no clue where to get Bamboo Plywood. Or, if you've got any tips on how to make MDF better, I'm all ears as well.

[Tkamsker]

I think you have a Solution so far i would Take my learnings and do a 2nd machine in Alu ,..


Gesendet von iPad mit Tapatalk

[Nick C The EE]

So here's a new twist on my trapezoid dilemma. A plunge directly into 3/4" MDF results in a hole with a diameter of about 1mm larger at the top of the workpiece than the bottom. Is there a certain style of end mill that are tapered, and did I unknowingly buy it? I asked the seller on ebay for information, maybe he can clear this up for me.

I looked at the very first part I made with my 2 flute HSS end mill before I smoked it and that one didn't have the trapezoid problem at all.

Any input is greatly appreciated!

[matth]

just lay the cutter on a flat surface and roll it along looking at where the flutes contact.

[widgitmaster]

Put a point in the spindle, and a carpenter's square on the table. Then move the axis along the square edge, this will show squareness errors as well as distance errors in two axis!
Widgit

[PaulRowntree]

This will happen if the tool's rotation axis is not parallel to the Z motion. "One of the fly-rods has gone out of skew on the treadle"

[ger21]

Or, if you've got any tips on how to make MDF better, I'm all ears as well.

Use torsion box construction. All your components should be individual torsion boxes.
I'd make the gantry beam a torsion box about 3-4" thick, and about 8" high. I'd use 1/2" mdf skins. I'd make the gantry sides 2-3" thick. You can get away with 1/4" skins, if you reinforce any mounting points so that you're screws are into much thicker material.

I'm building a 4x8 machine with dual spindles, and the gantry beam is an MDF torsion box. The skins are 1" thick, made up of 4 laminations of 1/4" MDF, which increases stiffness. I also epoxy phenolic strips to the mdf, and machine flat for linear rail mounting surfaces. I haven't assembled it yet, but here are some pics.
http://www.cnczone.com/forums/cnc-wo...ml#post1168468

So here's a new twist on my trapezoid dilemma. A plunge directly into 3/4" MDF results in a hole with a diameter of about 1mm larger at the top of the workpiece than the bottom.

It's very possible that this is due to flex in the machine. Plunging puts a tremendous amount of force on the machine.
Some bits are beter at plunging than others, and it may just be that this bit doen't plunge as well, putting more force on the machine.

[Nick C The EE]

Wow, thanks for all the input guys! All of it is very helpful.

I didn't have much time to work on my machine last night. Had to take our dog to training class.

ger, thanks for the link to your build thread. I haven't seen it before, but I'm glad I did. I picked up a few gems in there, like using epoxy to anchor bolts in MDF. Brilliant! One of the original problems that perplexed me in my design was trying to figure out the best way to bolt things together. You've given me some interesting ideas.

[dfaszer]

I'd also like to know if anyone has any tips on maybe a better/stiffer material to use than MDF that can be had for a reasonable price. JerryBurks' big bamboo is a pretty great inspiration, but I have no clue where to get Bamboo Plywood. Or, if you've got any tips on how to make MDF better, I'm all ears as well.

You might try looking at Baltic Birch. Its engineered plywood that is stiff with little flex and no internal voids. You can get it in multiple thicknesses as well. Its a little more costly than MDF, but not as much as metal (Depending on thickness ~$50 USD for a 4'x8'). Its what I plan on using for my first build.

-David

[ger21]

Its a little more costly than MDF, but not as much as metal (Depending on thickness ~$50 USD for a 4'x8')

Probably closer to $100 for 3/4". And it usually comes in metric thicknesses, and 5'x5' sheets. A similar product called Appleply comes in fractional inch thicknesses and 4x8 sizes.

[dfaszer]

I've read that some people have payed that much. If you are around a Menards, you can get 3/4" x 4' x 8' for $70.00 (http://www.menards.com/main/building...74-c-13334.htm). Unfortunatlly, they don't ship this item. In-store pickup only.

-David

[bobmagnuson]

I'd be very surprised if the quality of the Menards BB is the same as a true hardwood/plywood dealer.

[EvanZ]

It's not the same. Not even close. Full of internal voids and easily delaminates. The crap at Lowe's is even worse!

[Nick C The EE]

PROGRESS!

So I went into hardcore diagnostic mode tonight after reading everybody's input (and reading other useful posts in other threads).

The first thing I did was take a level to my gantry uprights. My gantry was leaning forward compared to my table! Fortunately it wasn't too hard to fix. After I leveled that out, I tried a test cut. It was better, but still not great

I got a great idea from another thread on how to align the router. I put an allen wrench in the collet and measured from the tip of the wrench to the table at 90* increments around the router. That revealed even more misalignment. After fixing that up, I made another test cut and the problem was all but gone!

Thanks for all the help guys!

[Nick C The EE]

Tada! Z axis redesign 95% complete. I just need two bolts to clamp the router in and something to use for a dust boot.

With the realignment I did yesterday, the 2nd router mount I machined today came out a million times better than the rest of the z axis parts I previously cut out. But despite the bad cuts from the misalignment, the new parts were still a hundred times better than what I did by hand, and the new z assembly had much less bind as a result. I was able to increase my max speed and it sounds much smoother.

I was worried about clearance between the router and the bolts which is why it's probably a good 2-3" longer than it should be. But, it has very little flex despite its length so I'll live with it

[Nick C The EE]

Use torsion box construction. All your components should be individual torsion boxes.
I'd make the gantry beam a torsion box about 3-4" thick, and about 8" high. I'd use 1/2" mdf skins. I'd make the gantry sides 2-3" thick. You can get away with 1/4" skins, if you reinforce any mounting points so that you're screws are into much thicker material.

I'm building a 4x8 machine with dual spindles, and the gantry beam is an MDF torsion box. The skins are 1" thick, made up of 4 laminations of 1/4" MDF, which increases stiffness. I also epoxy phenolic strips to the mdf, and machine flat for linear rail mounting surfaces. I haven't assembled it yet, but here are some pics.
http://www.cnczone.com/forums/cnc-wo...ml#post1168468



It's very possible that this is due to flex in the machine. Plunging puts a tremendous amount of force on the machine.
Some bits are beter at plunging than others, and it may just be that this bit doen't plunge as well, putting more force on the machine.

When you say thickness, do you mean the thickness of the torsion box plus the skins? or is that thickness without skins?