[carlnpa]

I hate to work with MDF and have had humidity problems with it expanding (never to shrink back) also. Any reason not to substitute a good 3/4 ply marine/oak/maple? I have all the electronics rounded up, so am ready to start building. Thanks

[buscht]

I built a machine with Baltic Birch plywood and have been very happy with it.

[cncadmin]

I also have built a wood router with shop grade birch.

[carlnpa]

Thanks for the input, I will use a quality plywood for this first machine.

[ger21]

If you're going to get something like Birch from Home Depot, spend the little extra and go with Baltic Birch, if you can get it.

[carlnpa]

My local store had pine maple and oak. Really nice grain in both hardwoods. I went with the oak as the sheets were much heavier. I finished the board from Hobbycnc last night. Will start cutting the ply today.

[carlnpa]

One 4x8 sheet of ply has provided all the parts (less some bed and brace support material) to build the machine to the plans. There is not much left over. It took about 5 hours of steady work. Due to the weight I used a circular saw and set up clamped guides to cut the larger pieces. Used a table saw when the pieces got smaller. I'll keep posting if there is any interest. Thank you JGRO.
Merry Christmas

[gmfoster]

Please Do

Keep posting.

[carlnpa]

Total plywood 1 sheet 4x8 3/4 in., 4x4 1/2 in.
Worked seven steady hours today on the router.
Finished four large adjustment blocks - changed the design some, hopefully picture attached. Recessed and set with epoxy the hex nut for adjustment block. Will likely be using 1-1/4-20 bolts all the way around to allow second jam nut. There isn't enough space with the 1 inch bolt.
Partial finish the eight small adjustment blocks - all the holes cut.
Completed support-motor side, support bearing side. I ran these assemblies through the table saw to square up as initial cut was with a guided circular saw. Next one I would leave these a bit long to allow finish cut.
Base support structure and bed support completed, design chage again, I'll take a picture later. Much less bracing complexity, if more is needed I'll add later.
Will be using 1/2 acme rod with home brew preload assembly. Enco has this on sale for $7/6 foot, nuts for $0.99 till mid Jan 05. Measured .003/.004 backlash at a number of points, not great but consistent.
Dave at Hobbycnc was kind enough to send a replacement resistor (I soldered in the right place for a 4 axis board, mine is three, I coudn't remove so left it in) so the board is done.

[Hobbiest]

Sounds like you have already completed all of the blocks, but if not, you really only need two adjustment blocks per axis (actually, one is all you really need to get the two axis' parralell, but two does make it easier). How are you planning your connections between pieces? It seems as if drilled and tapped holes wouldn't work as well as they do in MDF. lookin good.

[carlnpa]

Worked three hours today. Made progress on the gantry and cutting bed. The plywood allows the use of self tapping deck type screws, instead of drill and tap that would work poorly in the ply. I am using a poly glue on edges with 1-5/8 screws, it goes fast and the screw pulls the pieces tight together. In places where a split would be unacceptable a small pilot hole is drilled first. The glue makes these joints permanent, without glue disassembly would be possible. I plan to through bolt the bed and base to allow these parts to be removed and grow if needed in the future.
I am using galvanized pipe as it is much smoother than black finish gas pipe, it costs a little more though.

[carlnpa]

Worked five steady hours today on the bearings. The x and z are done and the y just needs the guide spacers milled. I hope I understand how the alignment works with these. I set up a jig and drilled holes to be spot on per the final alignment under the free router plans thread. Hopefully they can never get out of alignment this way. My HF 44991 mill dug into a 1/2x0.520 y linear guide spacer and broke the nylon gear. These went much slower than I had envisioned.

[carlnpa]

Worked 2-3 hours on painting, used Kilz2 latex primer sealer with grey latex satin. Worked 2-3 hours aligning the x axis, this was a pain, definitely on my future improvement list.

[carlnpa]

Bed support detail. Worked 2 hours playing with this. Used a pair of oillite flanged bearings in a loose fit hole, the flange is recessed and trapped by the pair of fender washers.
Assembly (all 1/2 inch) consists of locking flange collar, oillite washer, fender washer, oillite flanged bearing, oillite flanged bearing, fender washer, oillite washer, locking flange collar. The fender washer mounting holes need to be drilled biased toward the outer edge, if its not the hardware won't clear (ask me how I know. This design floats a bit, it also could pre tension the acme rod if used on both ends.

[carlnpa]

More details. The stepper motor coupling seems to be discussed alot so attached is a detail. There is no noticable free play in the coupling, the spider is a tight fit in the hubs. These are Lovejoy L050 components that I got from Enco, cost about $5.15 for the three component assembly. These will take up to 3/4 hp @18000 rpm, 26.3in./lbs. torque, and allow some misalignment. There is very little/no free play on the acme shaft or motor shaft hubs.
Enco pns
1/4 in bore hub, 990-4044, $1.94, mfg. M646220
1/2 in bore hub, 990-4046, $1.94, mfg. M735580
buna N spider, 990-4042, $1.27, mfg. P00CH

[carlnpa]

More details. Worked one hour to make six of these home made flanged acme nut assembly, total cost $1.24 ($0.99 acme nut, $0.24 1/2in. fender washer). These were welded but I would expect that they could be made using a Devcon type high strength epoxy adhesive. One outer edge will have to be cut down to clear the gantry floor. The preload assembly is also shown, the spring is Enco 240-0600, package of 10, $4.48. I tried some other springs but these seem to work best on the compromise of increased rod friction vs movement, still very free movement. The first flange will be bolted to the gantry with nut not shown. I have been replacing all nuts with nylocks, they are cheap from Enco (less than $2.00 box/100), I had been paying alot more locally.

[gmfoster]

Carl

Looks as if you are well on your way!..

Keep it gping.. Good pictures
Garry

[johnhamm]

I'm also interested in your progress since I will be constructing my first CNC machine in a couple of weeks. It will also be a HobbyCNC.

[carlnpa]

Worked 11 hours this weekend. X axis is done, stepper installed, everything lined up, these skate bearings must flex with the gantry. The home brew acme preload seems to work well. Spinning the rod with a drill no freeplay is felt, I can however distort the gantry with hand loading at various points. Y and Z axis are in place but not attached to the acme rod. In the z axis the adjustment screws contact the base of the stepper, and the 1/4 bolts that hold the linear bearing protrude into the space needed by the 1/2 in. acme rod. I'm trying to line up the motor mount square to the table, while keeping linear bearing contact, is more than a challange. I have been playing with software, I have Turbocad professional (steep learing curve), but the kids can crank out DXF/DWG files quickly in Visio 2000. Sheetcam demo looks good for generating gcode, and working with flat material. Will likely buy Mach2 but will have to use Master5 until I can upgrade a machine to windows 2000.

[carlnpa]

The machine is operational but still needs alot of tuning. Spent 12 hours last weekend on the y and z axis, another 4 hours during the week 3 hours last night and 3 hours this morning, setting up the power supply and wiring. Started up last night on master5 and got real erratic/no operation, missed steps, squealing, gravel noises, tore down the power supply and finally the board looking for solder bridges etc. Found the driver chips had poor solder joints (not enough heat I guess). Master5 is easy to set up and tune the motors, it has some built in demo programs that I was running. The Hobbycnc board seems to work well, I'm at 20IPM now and will use this board again for machine 2. Have a bunch of details to get through, biggest is the gantry in contact with the bed. Real good news is the homebrew preload acme nuts work, apparently pretty well ie 0.001/0, accuracy is at the same level as is repeatability over 1.0in. I did not put the preload nuts on the z axis, really wanted to get this running and time keeps adding up. I'll take pictures of anything specific anyone has interest in. JGRO thanks again for the design.