[buscht]

Here is my drawing for my CNC router.

I have enjoyed reading about other people's successes and problems so I thought I'd share my experiences.

Here's my budget

XYLOTEX 3 AXIS BOARD $125
3 116 OZ-IN MOTORS $60
24V FAN $5
SHIPPING FOR ABOVE $17
BALL SCREWS AND NUTS $145
MILLING ON BALL SCREWS $105
BISHOP WISECARVER SLIDES (SCROUNGED) $0
ADDITIONAL ROLLERS $90
BALTIC BIRCH FRAME (SCROUNGED) $0
EBAY BALL BEARINGS FOR BALL SCREWS $16
EBAY 24 VOLT POWER SUPPLY (PAID TOO MUCH)$76
TURBOCNC SOFTWARE $20
EXISTING PC $0
GAS SPRING COUNTERBALANCE (SCROUNGED) $0
3 SMALL ROUTER MOTORS $225
ELECTRICAL BOX FOR XYLOTEX $25

[B]PLANNED COSTS $909

UNPLANNED COSTS SO FAR
PARALLEL CORD EXTENDER $12
REPAIR WORK TO XYLOTEX BOARD $17
FLANGES FOR BALL NUTS $64

TOTAL COSTS FAR $1002

I have several things that I want to accomplish with this machine.

1. Awhile back, I had posted an inquiry about a CNC scroll saw application. Not getting any bites, I decided that I would have to build a 2 axis machine and test out my theories myself.

That is why both the X and Y axis are stacked and move. I want to clamp a board to the moving table and process it through a scroll saw. I'll post my results once this has been tested.

2. I want to do some carving work with the router. I figured that I would need 3 bits (1/4" dia, 1/8" ball nose, and a 1/16" ball nose). I did not want to exchange bits during the program, so I came up with the idea of using 3 router heads.

These heads are spaced 7" apart and are on a common Z axis. I plan on counterbalancing the Z axis with a gas spring. My idea is that if my parts are less than 7" wide in width, I can simply move the table over under the appropriate router bit using a fixture offset.

3. The other thing I am trying to accomplish is to learn how to create carving type programs.

4. Once all these things are accomplished, I want to add a 4th axis for doing carvings in the round.

My machine capacity is 12" in X, 7" in Y (if you use all 3 heads) 21" if only one head is used. and 5" in Z.

I have almost all the parts in my shop, except that the ball screws are out for machining.

I still have to cut the wood to size. That will take me the better part of the next two weeks.

Unexpected expenses are always important to know.

1. I did not like the short parallel port cable that came with the Xylotex board, so I went to the computer store and bought a longer one.

2. I didn't follow the Xylotex instructions very well, and I think I blew up the X axis driver. I have this board in for repair.

3. The ball screw nuts that I bought through McMaster Carr came with a 15/16-16 thread for mounting. I spent most of a day trying to find a nut to fit this. I gave up and ordered a flange from Nook Industries. At $20 per axis, this was kind of irritating. You can get these from McMaster already on the nut, but they cost $30 more.

4. I bought an adjustable voltage power supply of of Ebay 0-40 volts, 5 amp. It cost me $55, plus $21 for shipping. I had this great idea that I could play with the voltages to learn how applying voltages and amps affect the drives. I should have bought a fixed output 24volt power supply for about half that cost as I am finding out that I don't really want to know all that, I want to make parts.

T

[xairflyer]

Hi,

Very interesting machine.

What by the way are the wisecarver slides like (look like)

[buscht]

The Bishop Wisecarver slides are a Vee track with a cam rollers that match.
http://www.bwc.com/html/dualvee.html

They are pretty inexpensive when compared to new THK ways, but not being as popular, they are hard to find on the used market. If I didn't have some track already, I probably would have bought THK slides on Ebay.

[balsaman]

Nice machine. I like the 3 router idea.

Eric

[xairflyer]

What sort of price is the rail and the rollars ?

[buscht]

These are rough prices, but for two pieces of track, 72" (1828mm) long its about $106US.

The rollers cost about $19 each and you need 4 per axis. So new pricing would be $334 total for all 3 axis. You can cut the track down into smaller sizes. I am cutting mine down to 36",24", and 12" for the 3 axis.

You don't have to use the Wisecarver track. Any 90 degree shape that you can mount on a 45 degree angle should work. A piece of angle iron for example.

You mount two concentric rollers on one side of a rail, and two eccentric rollers on the other side of the other rail. Then you can tighten up the slide with a pair of wrenches.

I think the way that you are doing your machine with the adjustable cam followers accomplishes the same thing.

[ljoe1969]

who is a supplier for the rails and bearings?

[ljoe1969]

fyi.

these are the same bearings used on the shopbot.

[buscht]

www.bwc.com. I can buy them direct, or you can contact them for distribution information.
T

[buscht]

http://www.bwc.com/html/index.cfm
Sorry about that link

[buscht]

I have made alot of mistakes and still not done, but here are some photos of the machine.

I went against everyone's advice and bought the motors and controller first. This was my biggest concern and I couldn't wait to make sure that I could actually controll stepper motors with my computer.

I also posted my CAD design before building my actual machine. Balsaman was afraid that this was the kiss of death. Luckily, I was able to overcome this obstacle. HA

I am very happy with the ball screws and my decision to use them instead of an ACME rod. Even with the 116 OZ-in motors, the X-Y can rapid at over 100"/ min. My machinist is still cussing me though. THe screws are case hardened and he had a tough time turning the ends down for the bearings and threading the shaft.

Being worried about shaft whip, I used a double bearing support on each end of the shaft and a lovejoy couple between the shaft and the motor.

[buscht]

This is the top of the X-Y table. I am planning on building a vacuum table to sit on top of this, so the bolt heads sticking up won't be an issue. Actually this tabletop has 4 tee nuts in the corners so I can attach various jigs and spoil boards to my table.

I taped a piece of paper to the table and clamped a marker to the Z axis and ran my first program to check accuracies. THe ink leeched through the paper and marked the top of the table.

[buscht]

Here's a picture of the Z axis with the follower board removed.

You can see the Bishop Wisecarver rails mounted to the plywood.

I ran out of baltic birch plywood and had to substitute some high grade furniture building plywood on some of the larger parts. It was warped and I had to add some bracing and stiffening supports in strategic places.

[buscht]

Here's the back side of the follower board showing the Bishop Wiscarver Vee rollers.

[buscht]

I built this style of machine to allow me to try and cut parts on my scroll saw. Here is a picture of that setup ready to cut 1/2" plywood.

The verdict is still out on that. I got very interesting results.

The parts do not match the CNC program exactly. My experiments with the felt pen and later an ink pen, showed me that the CNC program was correct. But when I cut the parts on the saw they did match properly, although they were very consistent.

I think that I have too much play in my linear rails. You can see that I just clamped a board to my table and it is cantilevered out quite a ways off the end of the machine. I think that the play is exaggerated out in the cutting area of the saw.

I might be getting blade wander. This usually shows up as a tapered cut and my parts have a pretty good vertical cut edge.

It doesn't seem to be losing steps as the start and finish points of the cut end up exactly the same.

FYI, the cut speed is 4" per minute. So the top end speed of 100" per minute is not too vital.

[bb99]

Nice work! And thanks for all the photo's.

--Alan

[Bloy2004]

Yes! Keep updating with photos! I'm following this closely and hope you get your parts coming out uniform. This system may be applied to cedar picket fence tops of which I am always called on to make in various "one of a kind" designs. I'm enjoying this thread!

[ger21]

I saw you post about this router in another thread. Is it done? If so, how are you programming the 3 "Z" axis'.

Gerry

[buscht]

Gerry, my router is not quite finished. I have all 3 axis operational, but only one router head mounted at this time.

I have been working on getting my CAD models completed so I have something to cut.

But to answer your question, this is how I will program the 3 "Z axis".

I will create 3 seperate programs, one for each axis. 1st rough mill, 2nd large ball nose end mill, 3rd fine detail ball nose mill. All the programs will be written for the left hand head.

I will use a text editor to combine all 3 programs into one big one.
In between programs, I will insert a G92 X7.0 Y0 Z0 line which offsets the X origin by 7.0". This is the exact spread of the router heads.

Because my parts are less than 7" wide, only one router will ever come in contact with the part at any one time.

Future plans are to create programmable switches to turn each router head on and off.

I hope to have my first computer model in two weeks and then I have to work on part fixturing, etc.

Trent

[ger21]

Do the 3 routers move independently? Or are they all mounted to 1 Z-axis?