Author Topic: My CNC milling machine  (Read 4248 times)

0 Members and 1 Guest are viewing this topic.

Offline poorchava

  • Super Contributor
  • ***
  • Posts: 1674
  • Country: pl
  • Troll Cave Electronics!
My CNC milling machine
« on: July 26, 2012, 09:14:45 am »
Hello,

I wanted to show you my CNC milling machine which I have been building for ~4 months now. It's already functional, yet needs some major tweaks.

Specs:
-work area of 300x295x130mm
-construction material: 18mm plywood (~3/4" for you metrically-impaired :) ), M5 bolts and nuts
-drive transmission: hardened steel trapezoidal thread drive screws 16mm OD, 4mm pitch + hardened plastic nuts
-linear motion assembly: fully supported 16mm hardened steel rods + linear bearings in mounting brackets
-actuation: NEMA23 size 0.9 N.m 1.9A/phase stepper motors, microstepping driver of my own design based on TB6560, homemade/ghetto optocoupler board + lpt + Mach 3
-max. velocity 1400mm/min for X and Y, 600mm/min for Z (ghetto optocoupler board craps out @ 7-8kHz)
-spindle: Kress 800FME (800W 10k-29k RPM). tested with tools up to 1/4" OD
-materials machined: softer aluminum alloys, MDF, plywood, solid wood, PCB laminate, all kinds of plastic

Details:
I've build the machine out of plywood for numerous reasons: it's cheap, easy to work with using relatively inexpensive tools, can be obtained in any construction&garden&DIY mall/market (plus they will cut those to rectangles of chosen dimensions for free). It's also less sensitive to moisture than eg. MDF. General arrangement is that work table moves back and forth (Y axis), the carriage with spindle moves right to left and are attached to static base. Z axis is mounted on carriage together with the spindle. I've chosed the supported rods because they are easier to mount than unsupported type (no end-brackets) and they are most rigid of all "affordable" solutions (unaffordable being linear guides like THK, Rexroth, Hiwin etc). Motors are mounted on brackets made of 9mm and 18mm plywood and are coupled to the drive screws using reinforced garden hose and hose clamps (ghetto method but very cheap and entirely sufficient for this purpose). Threaded rods have been purchased as 2 1m pieces and then cut to length with angle grinder. Then they were mounted in a jig comprising of 3 16mm ID needle bearings bolted to 50mm of mdf scrap boards, rotated using ordinary power drill and ground with the angle grinder (i'm not really happy with the outcome). Whole machine is 55cm x 70cm x 50cm excluding motors (which stick out and add ~15 cm on each side). It weighs about 45kg.

Photos:
« Last Edit: July 26, 2012, 09:18:36 am by poorchava »
I love the smell of FR4 in the morning!
 

Offline nctnico

  • Super Contributor
  • ***
  • Posts: 23574
  • Country: nl
    • NCT Developments
Re: My CNC milling machine
« Reply #1 on: July 26, 2012, 10:18:31 am »
You'll probably get problems due to resonance. You can make the large panels much stiffer by putting a grid (with irregular sizes so there is no resonance frequency) made from wooden strips on them and add an extra panel (doesn't need to be very thick).
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline robrenz

  • Super Contributor
  • ***
  • Posts: 3035
  • Country: us
  • Real Machinist, Wannabe EE
Re: My CNC milling machine
« Reply #2 on: July 26, 2012, 10:27:01 am »
Awesome job!  Your choice of how the axes are arranged is excellent. The only thing I see that would improve the rigidity of the frame is to glue in a piece of plywood that fills that hole under the Y axis.  Just cut a hole big enough for the X axis drive to fit through. That would really decrease deflection in the Y direction and raise the resonant frequency of the frame. The low modulus of elasticity of wood tends to have wooden frames act as very effective "speakers".  You could also trim the sideplates from where they meet your Y axis base to where they meet the baseplate (remove a triangle of material) and it would not significantly reduce your rigidity but would greatly increase accessibility of the work zone.  Those two unsupported "triangles" will act like a huge tuning fork with large surface area generating wonderful howling noise when excited by the machining vibrations.

Keep up the great work and keep us posted on your progress.

Offline poorchava

  • Super Contributor
  • ***
  • Posts: 1674
  • Country: pl
  • Troll Cave Electronics!
Re: My CNC milling machine
« Reply #3 on: July 26, 2012, 10:54:23 am »
Believe me I'm aware of resonances and not the best rigidity of the frame (found out the hard way - ~$20 endmill -> bin).

I can't mount a panel on the backside and obscure it completly, because the table moves outside of the frame. Also, this gives me opportunity to load something long (like 2m board) and cut it into pieces (it's handy as i don't own a circular saw).

In longer run I'm definitely going to put some reinforcement in the back of the machine.
Similar thing goes for sides. I'm planning to attach another 18mm (or even two of those) to the sides as it would improve rigidity alot. Unfortunately such large pieces of plywood would add a lot of weight. I'm generally planning to move this machine to the basement (or to entirely different location as I'm planning on moving out with my girlfriend in nearest half a year or so). attaching sides and back reinforcement would easily push the weight above 65kg which makes it unportable for two people (i have narrow staircase which makes it impossible to use the help of 3rd person).

In the end i intend to reinforce this machine ALOT.

edit: some videos




edit2: ac for cutting off triangulat pieces in the front - i agree. I've thought about that from the beginning, but having everything rectangular/square made it easier to align. I'm considering cutting off those when I add the additional side panels. The access to work table is not so bad, but of course cutting these corners would improve it alot.
« Last Edit: July 26, 2012, 11:42:32 am by poorchava »
I love the smell of FR4 in the morning!
 

Offline robrenz

  • Super Contributor
  • ***
  • Posts: 3035
  • Country: us
  • Real Machinist, Wannabe EE
Re: My CNC milling machine
« Reply #4 on: July 26, 2012, 12:27:09 pm »
 Two stiffeners on each side like this pic will do more for your Y axis stability than more layers and won't add much weight.

Offline nctnico

  • Super Contributor
  • ***
  • Posts: 23574
  • Country: nl
    • NCT Developments
Re: My CNC milling machine
« Reply #5 on: July 26, 2012, 12:38:52 pm »
I agree with robrenz. Just adding extra layers doesn't help much. Besides adding stiffeneres two thin plates with a 'honeycomb' in between is just as stiff as a thick plate.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline poorchava

  • Super Contributor
  • ***
  • Posts: 1674
  • Country: pl
  • Troll Cave Electronics!
Re: My CNC milling machine
« Reply #6 on: July 26, 2012, 12:53:39 pm »
That makes sense, i've also considered this solution. It would need redesigning the X axis motor mount, as it's very wide. I had to do it that way, because i only had a cheapie drill press and a hand saw at that time. I might use the mill itself to make the mounting bracket which is narrower.

There is also the problem that base of the machine is not closed (i mean it doesn't have a plate on the bottom) and it doesn't have any ribs/crossbars etc. So i believe i would have to add those before i reinforce the sides (seems logical that when some force will act along X axis, the sides of the base will bend and the reinforcement won't be any good).

As i said - work in progress.

I also need to purchase another 2x1m of trapezoidal threaded rod and ask a proffessional machinist to turn the ends in the lathe down to 10mm diameter to fit the bearings. My ghetto hacked grinding jig didn't work too well (probably due to abrasive disc wearing out) and for example the Z axis screw ends are not concentric with the OD of the thread - there is massive wiggling and LOTS of friction. So far I haven't found anyone who would machine those screws for reasonable price. I figure it has to be a small workshop, because larger companies were quoting like $40 for machining one end of the rod (no shit... 1m of rod costs $7...). And because of the way most companies work in Poland they usually are not interested in single units jobs.

Other thing is that the spindle is not exactly perpendicular to the table (angle between the spindle bracket and Z axis face plate is not 90*, but rather something like 88.5-89*). I need to grind the spindle bracket mounting face to be at right angle to the table. Other solution could be to re-do the plate which rides the Z axis (the one which has linear bearings, drive nut and the whole spindle bracket attached to it). I may choose to integrate this plate and the spindle bracket together.

Unfortunately I don't have too much time now (or money for that matter because my car chose to unexpectedly shred the differential and then snap the drive shaft.... fuck.... unexpected $700 down the drain is not funny :( )
I love the smell of FR4 in the morning!
 

Offline poorchava

  • Super Contributor
  • ***
  • Posts: 1674
  • Country: pl
  • Troll Cave Electronics!
Re: My CNC milling machine
« Reply #7 on: August 08, 2012, 02:03:38 pm »
I'm currently trying to fix spindle holder. problems are spindle not square vs other axes and tendency to rotate around Y axis.

second issue is clear: i've used 4 M5 screws with nuts to fasten the spindle holder to the Z axis surface. I thought that approx. 15x25cm surfaces squeezed with screws would not move that easily. Well... Lesson learnt.  Z axis plate as shown on photos contains nut mount and 4 linear bearings while being only 12x15cm or so. I will try to stuff M8 screws there. They should be able to fasten these plates much better.

more difficult thing is that spindle is pitched around X axis (top of spindle is more in front of the machine than bottom of it is). Difference is around 1.5mm along 25cm of spindle mount, but it is very visible in surface finish. It comes from the fact that threads in woodscrew have caught material unevenly and that cutting service in hardware store is not precise at all. I've disassembled the holder with intent to glue them together to be perfectly square and then drill new holes and reinforce with woodscrew. This was a failure. I've used strongest glue i could buy (so called liquid-nail, strength of 40kg/cm^2). It is strong, ok, but not rigid at all.

Today I will attempt another thing: remove the glue, assemble the holder as it was before, then remove only the bottom plate (in which the spindle sits) and then work the surfaces with sandpaper to get them perfectly square. Then i will assemble whole thing again. Unfortunately no photos, because I didn't want to handle the cellphone with all that nasty glue everywhere including my hands and clothing.

I think the next imprevement will be increasing rigidity. After that I think I will try to cast new Z assembly out of solid aluminum. I think I should be able to (with the aid of the mill) cut the model from styrofoam, wax of similar stuff and then make a casting. Internet research shows that furnace can be built 'ghetto-way' out of steel barrel, some refractory concrete and propane torch (i think i already have first and last of those somewhere). Good casting alloy of aluminum I can probably get on car scrap yard in form of broken engine heads and maybe in some cases even engine block.
I love the smell of FR4 in the morning!
 

Offline poorchava

  • Super Contributor
  • ***
  • Posts: 1674
  • Country: pl
  • Troll Cave Electronics!
Re: My CNC milling machine
« Reply #8 on: August 09, 2012, 09:12:48 am »
I love the smell of FR4 in the morning!
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf