|
mekilljoydammit posted:... 4x8 sheets of foam. I haven't used this but I think you can work off this setup http://www.maslowcnc.com
|
#
?
Aug 22, 2017 17:10
|
|
|
|
| # ? May 28, 2024 23:08 |
|
|
eddiewalker posted:Look at the Mostly Printed CNC platform. It's probably a good starting point, at least. Thanks! Yeah, looks relatively suitable for a starting point. Ambrose Burnside posted:If all you're interested in cutting is foam, why not do what one specialty foam manufacturer i know of did, and rig up a conventional gantry-type CNC router arrangement with a nichrome hot-cut wire in place of conventional tooling? I wouldn't think the forces on a water-only waterjet would be that high... or at least not the baby one I'm looking at. Hotwire's the backup plan - the big thing is I figure with waterjet I could incorporate marking the individual sections and not have to continuously cut, because it's possible to start and stop the cut. Plus honestly I think trying to make a waterjet cutter for foam might be a fun project. I'm not figuring the motion system would be that different between hot wire and waterjet. *edit* bred posted:I haven't used this but I think you can work off this setup I had not seen that! poo poo like this is why I bothered posting - I pretty much always think cartesian machines. mekilljoydammit fucked around with this message at 17:17 on Aug 22, 2017 |
|
#
?
Aug 22, 2017 17:14
|
|
|
I think something like this that uses rollers on aluminum extrusions, driven by belts will be most straightforward to get up and running, and seems well suited for your purpose http://www.openbuilds.com/builds/openbuilds-ox-cnc-machine.341/
|
|
#
?
Aug 22, 2017 17:28
|
|
|
eddiewalker posted:Look at the Mostly Printed CNC platform. It's probably a good starting point, at least. The Lowrider CNC by the same developer would be perfect for this.
|
|
#
?
Aug 23, 2017 11:15
|
|
|
I really want to make a Maslow style machine. It's for a carnival float we do every year and the though if pressing print and have a whole sheet of ply machined is awesome. This year we're making a 14ft long railroad train... so yeah lots of jig sawing and poo poo. Anyone made one? I can't get a kit as they are us only so looking to replicate with my own steppers and electronics. Is nema23 big enough?
|
|
#
?
Aug 25, 2017 11:47
|
|
|
I'm new to the thread, and I'm just getting started as a hobbyist. I'm working on putting my mpcnc together. I don't know if it was ever resolved in the last page but you can get the 1" OD Seamless Stainless Steel Tubing for the kit at McMaster-Carr. Cheaper pricing than Grainger. https://www.mcmaster.com/#standard-stainless-steel-tubing/=194ubaa Also, if you're looking at something from Grainger, always look on Zoro.com. It's owned by Grainger, had the same products, and is just flat cheaper. You can even just search by the Grainger part number.
|
|
#
?
Aug 28, 2017 10:15
|
|
|
Thanks for that info EvilBeard. I am just getting started too and was planning on building an MPCNC because I thought the tube/conduit would be easy to get but now it looks like that will be the tricky part... I'm afraid of steel because I'm not sure I will be able to cut or drill holes in it, plus the cost makes mistakes more expensive. Now I'm afraid that any cheap tubing I find won't be straight or strong enough or fit the parts I've already printed (though reprinting wouldn't be the end of the world). The MPCNC parts page says "23.5mm fits 3/4″ EMT conduit in the US", is there a type of steel tubing one can look for that would match this profile or should I just buy 1" ss tubing and measure it for the printed parts? How much does it cost to get set up to cut/drill steel tube?
|
|
#
?
Aug 28, 2017 18:51
|
|
|
Novo posted:Thanks for that info EvilBeard. I am just getting started too and was planning on building an MPCNC because I thought the tube/conduit would be easy to get but now it looks like that will be the tricky part... The 1" OD stuff is for the 25.4mm kits. If you are getting the 23.5mm kit, you'd need 0.92" OD tubing, which isn't really a stock size. It's the common size of EMT conduit, which is thin walled steel conduit. I upgraded to the 25.4mm kit so that I could get thick walled 4130 Chrome-Moly tubing for maxiumum strength. I got the 1/4" thick walled tubing (in reality it's smaller), but it's about $30 cheaper per 6' section than the stainless steel size they recommend, meaning you save about $120, and gain some rigidity.  As for the cutting and drilling, you can use v-blocks in a vice (or if you have a decent vise, it'll have pipe holders built into it. You'll want to use soft jaws so you don't chew up the surface. All you really need is an angle grinder, a cutoff wheel, and a face shield. If you're lucky, you'll have a chop saw and a drill press. 
EvilBeard fucked around with this message at 20:19 on Aug 28, 2017 |
|
#
?
Aug 28, 2017 20:14
|
|
|
If you can drill and cut aluminum you can drill and cut steel, it'll just take a little longer and you might have to throw some cutting fluid (WD-40 works) on there, no big.
|
|
#
?
Aug 28, 2017 23:43
|
|
|
I'm just starting to outfit my workshop. My plan, in its entirety, was to buy a hack saw blade rated for steel. Now I'm considering the merits of investing in a vise, chop saw, angle grinder, drill press... At least I already have some WD-40 
|
|
#
?
Aug 29, 2017 00:16
|
|
|
I would definitely get an angle grinder and drill press. They aren't that expensive and are very versatile. And gently caress using a hacksaw to cut up that much steel, it would take ages.
|
|
#
?
Aug 29, 2017 00:29
|
|
|
biracial bear for uncut posted:The Lowrider CNC by the same developer would be perfect for this. Yeah, looking at it, I think that's the one. I'm thinking of stepping back to routing instead of waterjetting foam if nothing else than because I can have profiled edges rather than having to hand sand a bunch of poo poo after I laminate all the foam.
|
|
#
?
Aug 29, 2017 16:17
|
|
|
Novo posted:I'm just starting to outfit my workshop. My plan, in its entirety, was to buy a hack saw blade rated for steel. Now I'm considering the merits of investing in a vise, chop saw, angle grinder, drill press... At least I already have some WD-40 Another option would be to buy a machining vise. If you're gonna have a CNC machine, you'll probably use it from time to time anyway. Just mount it to a bench/table, and when you're finished with your machine, you can use it to fixture smaller parts in the machine bed.
|
|
#
?
Aug 29, 2017 18:35
|
|
|
I consider a decent-sized bench vise and drill press, among other things, vital for doing much metalwork. Especially the vise, being able to clamp things very firmly is non-negotiable for many tasks (like sawing stock up, for instance). An angle grinder is also a must-have, mostly because it's extremely versatile and can replace many individual, more expensive tools, especially when you're starting out. That said, angle grinders are better for larger-scale work that can't be maneuvered to a stationary tool, and doing fine or precise work with them can be an exercise in frustration. And with all the above, you're going to need a way to clean up the cuts you make, which will involve metal files and possibly deburring tools. Files are also an Essential Tool and not too expensive for a basic import set. For cutting up tubing, I might actually use a hacksaw even with the angle grinder on hand, depending on how many cuts I had to do and the accuracy needed. If it's just a couple cuts, dragging the angle grinder out and getting it set up will take as long as the first cut or two with the saw (tubing goes fast because you cut a whole lot of air on the way through), and the angle grinder cuts will need a lot more cleaning up and finishing work done with files or a grinder afterwards. The angle grinder will also be significantly easier to Oopsie with, and if you don't have any tubing to spare, a significant Oopsie could put you back weeks as you wait for new stock to come in. Cutoff wheels are also "pretty freakin dangerous" by hobby-level tool standards, you should never use one without a full face shield because a disintegrating wheel can gently caress your face up real good even if you're wearing safety glasses. A hacksaw will turn a 10-minute job into an hour job, and it'll involve a significant workout to boot, but- this is a hobby project, you're a beginner, you can afford the extra time and labour, saws are intuitive to use and safe. If you have 20 cuts to do, a hacksaw will take too long, for sure- but I wouldn't dismiss it out of hand.
|
|
#
?
Aug 29, 2017 19:34
|
|
|
I've been working out the construction of my bigger, more robust CNC. I think I'm about done with the gantry design, but I'll have to do a little more work to get it where I want it. I'll probably add some bracing.
EvilBeard fucked around with this message at 03:11 on Aug 30, 2017 |
|
#
?
Aug 30, 2017 03:03
|
|
|
What's the thickness on the Z axis plate? Looks pretty thin, and it's hanging way down below the gantry. Could you shift it up and shorten the gantry legs? Hard to see if there's anything else in the way from this angle. That could also let you make a more rigid connection to the Y axis screw, that link looks pretty flimsy. Seems like you're transmitting all the force through two pretty small bolts, that seems like a weak link, and the cantilevered arm is going to have a bigger bending moment because of how long it is. Of course, the importance of any of that depends on what you're cutting. If it's just foam you can get away with a lot more...
|
|
#
?
Aug 30, 2017 03:24
|
|
|
Karia posted:What's the thickness on the Z axis plate? Looks pretty thin, and it's hanging way down below the gantry. Could you shift it up and shorten the gantry legs? Hard to see if there's anything else in the way from this angle. that's 3/8" thick. I figured it would support it fine, once I get it adjusted up, so I don't have so much sticking down. I'm going with stock ball screw lengths so I don't have to trim and machine them down. I'm just trying to get all the pieces in their approximate places. I was planning on shifting the z axis up about 3 inches, that'll basically flip the ball nut holder, and I can use something thicker to get 4 bolts into into the z axis mounting plate. The gantry legs will be shortened to fit the proper height of the table. It's the next phase of the design. I wanted to figure out the gantry, because it's the most complex part of it. The y axis movement and connection to the table is pretty simple.
|
|
#
?
Aug 30, 2017 04:04
|
|
|
EvilBeard posted:that's 3/8" thick. I figured it would support it fine, once I get it adjusted up, so I don't have so much sticking down. I'm going with stock ball screw lengths so I don't have to trim and machine them down. I'm just trying to get all the pieces in their approximate places. I was planning on shifting the z axis up about 3 inches, that'll basically flip the ball nut holder, and I can use something thicker to get 4 bolts into into the z axis mounting plate. The gantry legs will be shortened to fit the proper height of the table. It's the next phase of the design. I wanted to figure out the gantry, because it's the most complex part of it. The y axis movement and connection to the table is pretty simple. Sounds like you've got it under control. It was hard to get a sense of scale, I'd been guessing 1/4", everything seemed pretty flimsy at that scale. 3/8" aluminum is much more appropriate for a hobby-level router.
|
|
#
?
Aug 30, 2017 04:15
|
|
|
Karia posted:Sounds like you've got it under control. It was hard to get a sense of scale, I'd been guessing 1/4", everything seemed pretty flimsy at that scale. 3/8" aluminum is much more appropriate for a hobby-level router. Yeah, I'm going to try to stress the hell out of my MPCNC (I've put it together with very rigid tubing), by making an adapter plate for my chinese 2.2kW spindle. I'm hoping to then use it on the aluminum to centerdrill and counterbore all the holes, so I get a much more precise build. That's the hope anyway. Then I'll just convert my MPCNC into a 3d printer (or even a plasma cutter if I get frisky) and use the DIY CNC for wood and soft metal.
|
|
#
?
Aug 30, 2017 04:22
|
|
|
Pretty cool little project. That 6 flute ball end mill leaves a ridiculously good finish. https://youtu.be/5FU_VlkQ63k
|
|
#
?
Sep 30, 2017 01:22
|
|
|
So I've been wondering, where in the chain from cad -> making cuts does G-Code get changed into revolutions of a stepper motor or servo? Is that what Mach3 or Linux CNC does?
|
|
#
?
Oct 2, 2017 05:54
|
|
|
Depends on the design of your machine, but yes, sometimes that can happen in LinuxCNC or Mach3. 1) 3D model geometry is analyzed to generate 2) toolpaths, which are then post-processed into 3) g-code, which is interpreted by 4) the machine host or driver (LinuxCNC, etc), which generates a 5) timed sequence of motor steps, which are sent to the 6) motor driver ICs, which convert the step pulses into 7) alternating pulses of the proper voltage on the motor coils, which 8) rotate the motor shafts by the specified amount. I haven't used Mach3, but with LinuxCNC in step (5) the software bit-bangs the motor pulses out over the parallel port, and the driver interface box takes those pulses and commands the driver ICs to actually spin the motors. This is why those homebuilt CNC machines still use parallel ports and why they're so picky about them -- the software needs extremely low-level access to the port to be able to time the pulses correctly. USB-to-parallel adapters frequently don't support that kind of operation. In a more fully integrated machine, like an industrial CNC or a 3D printer, the g-code is interpreted onboard by the machine's firmware and the motor drivers are run from the processor's GPIOs.
|
|
#
?
Oct 2, 2017 06:09
|
|
|
Sagebrush posted:Depends on the design of your machine, but yes, sometimes that can happen in LinuxCNC or Mach3. So given parts 4 and 5, how do machines running on Mach3 do closed loop? It seems to me like there's no place for feedback in that loop. So if you want a closed loop system are you limited to a more integrated machine? And then don't you lose some of the software functionality?
|
|
#
?
Oct 2, 2017 06:21
|
|
|
Many (most?) CNC machines are open-loop and count steps of the motors to keep track of their location. If the motors slip, well, then you're hosed. That's why you keep an eye on your spindle load and axis loads, which high-end machines will continuously measure and report. You can of course buy encoders for each axis, just like what you'd use for a DRO, and I'm sure there's some way to feed that data back into linuxcnc or whatever. For the record, our Haas Mini Mill can produce 8 tons of force in each axis, so missed steps mean you did something really, really wrong Sagebrush fucked around with this message at 06:53 on Oct 2, 2017 |
|
#
?
Oct 2, 2017 06:49
|
|
|
Tres Burritos posted:So I've been wondering, where in the chain from cad -> making cuts does G-Code get changed into revolutions of a stepper motor or servo? Is that what Mach3 or Linux CNC does? You configure Mach3 to your machine. You set the number of pulses per unit. If you're using mm, you set that, inches, you set that. The gcode is then like go to x1 y1 z0, and the software knows how many pulses to send to the stepper based on current position and the target position.
|
|
#
?
Oct 2, 2017 11:31
|
|
|
I think that gets to the difference between servo driven CNC machines and stepper driven ones and why a lot of professionals will look down on steppers - servos are closed loop so prevent issues like missed steps.
|
|
#
?
Oct 2, 2017 14:28
|
|
|
Sagebrush posted:For the record, our Haas Mini Mill can produce 8 tons of force in each axis, so missed steps mean you did something really, really wrong Haas machines are all closed-loop via encoders and servo based. I can't think of a single professional level (anybody better than Tormach) cnc manufacturer that uses steppers. They're smaller, better accuracy with the encoders, better torque/power curves, and much, much faster. Cost and tuning are really the downsides, but the performance improvement outweighs it.
|
|
#
?
Oct 2, 2017 17:17
|
|
|
Karia posted:Haas machines are all closed-loop via encoders and servo based. I can't think of a single professional level (anybody better than Tormach) cnc manufacturer that uses steppers. They're smaller, better accuracy with the encoders, better torque/power curves, and much, much faster. Cost and tuning are really the downsides, but the performance improvement outweighs it. There are also closed loop steppers, thought I don't know of any cnc machines use them.
|
|
#
?
Oct 3, 2017 02:17
|
|
|
Lots of hobby people are going to them, because you can get drop in replacements for the existing steppers
|
|
#
?
Oct 3, 2017 02:22
|
|
|
You also don't see many (any?) >3kw stepper motors. I'm not sure how much of an issue it is for turning and milling, but in grinding a servo is preferable as you can get 2000 pulses per rev while most steppers are 400 pulses. A servo can at least tell you it moved a certain distance, but you need a linear scale to actually show that it moved that much on the machine bed. We've got some custom machines running steppers and it mostly comes down to cost and availability when they were designed 20 years ago. If the same machines were designed today they'd absolutely have a servo. The AutomationDirect servo kits are re-packaged Delta's and pretty nice for the money. I just wish they'd offer the higher precision Delta drive instead of the mid-range.
|
|
#
?
Oct 3, 2017 02:40
|
|
|
The term "servo" only indicates that there is some kind of positional feedback, it does not imply what type of motor is used. You can buy servos with brushed-DC, stepper and brushless (BLDC) motors, as well as linear actuators, and any of these can be electric, air-powered or hydraulic varieties. Stepper motors have high torque at low speeds and have very good interval positioning (for as cheap and simple as they are) but they don't work well at high speeds due to the vibration they produce and the inductance they have. The step interval isn't really an issue, since you can microstep them. Cheap drivers may only do 32x microstepping, but better ones usually go to 256X microstepping which is 51,200 steps per turn for a normal 1.8° / 200-step nema motor. There's not a lot of giant steppers since the speed ranges end up decreasing pretty drastically as they get physically bigger. The main reason that industrial equipment doesn't use steppers is the poor speed range that steppers have compared to BLDC. For a nema-23 sized motor, a stepper motor would struggle to spin at 500 RPMs (and have very little torque while doing so) while a nema-23-sized BLDC motor can easily spin from zero to 3000 RPMs. As both motor sizes get larger the speed ranges of both decrease, but the BLDC still has a much bigger range at every size. Higher-end motor controllers can do tricks like on-the-fly voltage/current control (not just current chopping) so the maximum torque can be adjusted up and down as needed. Plus with CNC machines often these motors are sitting on a harmonic drive that is a gear reduction of 50x or more. 100x is common. If you want--you can get servo-steppers from China. They're not what I would call hugely expensive, but they do cost more than normal steppers. For a typical nema-34 motor+driver, the stepper costs $120 and the servo-stepper motor+driver costs ~$180 (both prices are without shipping cost included). You would need a whole hardware + software setup that could use that info however. I've not casually read of anyone building such a setup, though I haven't really looked. If you wanted a BLDC-servo roughly the same size as above, the prices jump up to ~$1000, even from China-land. Both the controller and the motor+encoder cost roughly $400+ each. You can also get 3-phase stepper motors and drivers from China-land, that offer some improvement in acceleration and torque. https://www.youtube.com/watch?v=vxYUh9fSDLM They do work somewhat better, but it's not a huge difference. There is 4- and 5-phase stepper motors too, but I've not run across any source of them that is friendly to non-industrial (individual hobbyist) buyers. ------- Generally,,,, 2-phase open-loop steppers work pretty well, as long as you keep the current low, keep them relatively slow and you tune the whole machine so that no stepper is run at a speed where it resonates badly. There's China places that sell better actuators, all they need is your credit card--but it costs a lot more money to move to anything drastically better.
|
|
#
?
Oct 3, 2017 16:47
|
|
|
I'd like to build a table top CNC with a 200x200 x/y work envelope for making aluminium parts and to learn about machines. After hunting around for examples they seem to range from the fairly weak looking to quite full on epoxy granite numbers that look like they weigh 100s of kgs. This Old Tony's design looks good, I like the mild steel RHS section approach. Structurally, I think I require at least the following: Linear rails (Hiwin knock offs will be fine for me I think) Ball screws Stepper motors (Open loop should be fine), drivers. A fixed gantry design I think would be best. The base could be bolted to the vertical gantry, and so long as the mating surfaces were properly machined after welding alignment should be allright. (Is heat stress relieving required?) Would fairly thick box steel be a good place to start making up a design in Fusion/Inventor? I'm estimating what I'll come up with will be strong and weigh 100-300kgs. Are there any awesome books or professional references? Most of what I've learnt comes from fairly haphazard youtube video watching.
|
|
#
?
Oct 5, 2017 23:40
|
|
|
Sagebrush posted:Many (most?) CNC machines are open-loop and count steps of the motors to keep track of their location. If the motors slip, well, then you're hosed. That's why you keep an eye on your spindle load and axis loads, which high-end machines will continuously measure and report. Maybe for home gamers. Most industrial machines are servos with encoders in the late digital age, resolvers in the 1980s.
|
|
#
?
Oct 6, 2017 13:08
|
|
|
As another baby step on my road to a DIY cnc machine I decided to bite the bullet and buy a linear rail and carriage from mcmaster carr. Turns out that they're ground THK rails & carriages  .Explains the price.
|
|
#
?
Oct 7, 2017 21:32
|
|
|
Karia posted:Haas machines are all closed-loop via encoders and servo based. I can't think of a single professional level (anybody better than Tormach) cnc manufacturer that uses steppers. They're smaller, better accuracy with the encoders, better torque/power curves, and much, much faster. Cost and tuning are really the downsides, but the performance improvement outweighs it. There's also these controllers, which take step/direction inputs like a common stepper controller and drive a common DC motor with an encoder: http://www.cadcamcadcam.com/servo.aspx http://www.geckodrive.com/geckodrive-brush-dc-drives.html Some of the Mesa controllers can also directly connect to DM motors and encoders. I haven't used any of those personally, but it should be possible to combine one with a cordless drill motor and an encoder and get some level of closed-loop control for a fraction of the price of most servomotors.
|
|
#
?
Oct 7, 2017 22:23
|
|
|
Cockmaster posted:There's also these controllers, which take step/direction inputs like a common stepper controller and drive a common DC motor with an encoder: I've used gekko drivers before. They work really well.
|
|
#
?
Oct 8, 2017 02:23
|
|
|
Tres Burritos posted:As another baby step on my road to a DIY cnc machine I decided to bite the bullet and buy a linear rail and carriage from mcmaster carr. Turns out that they're ground THK rails & carriages Those are amazing ballscrews, sorry you put a McMaster through college. I'm rocking 2x THK SKR33s, a 300mm on my lathe's X and 400mm on Z, used from China.
|
|
#
?
Oct 8, 2017 09:47
|
|
|
My MDF wasteboard was pretty ratty after the last couple years of over-plunging and overtightening clamps, so I made a new one. While I was doing that, I also discovered the masking tape and superglue method of adhesive workholding. I was familiar with using double sided tapes, but all the ones I'd tried were too thick, or not wide enough, or difficult to remove. After giving this a shot, I'm sold: https://www.youtube.com/watch?v=g4K3Ni1NLpg For flat stock, this method is insanely strong but also easy to remove, and no clearance issues with clamps. It really doesn't require much surface area despite my excessive taping.
|
|
#
?
Oct 24, 2017 07:02
|
|
|
Is there a good text reference or compiled place of information for CNC machine building somewhere? I've picked up snippets from forums and by reading people's builds but I'd like to read a comprehensive / professional / boring tome that covered it all up to todays CNC machines. Topics like stiction are still hazy for me, and I only discover they're important through random forum browsing.
|
|
#
?
Oct 24, 2017 22:42
|
|
|
|
| # ? May 28, 2024 23:08 |
|
|
Well heck it only took three years of doing wood crafts on my Shapeoko before I got up the courage to try some aluminum. Made three little things, didn't ruin my table with coolant, and broke no bits- best of all it was some good learnin'. The material is 6061 .125 plate I got from amazon about a year ago intending to do metal projects. The profile cutting bit I used was a Destiny Viper 1/8" 3 flute that I bought on eBay about two years ago intending to do metal projects. I engraved with a 1/8" ball mill because it's what I had on hand; I originally tried doing it with a 30 degree v-bit but it snapped the tip off as soon as it contacted for a .001 deep pass. My objective here was to make some keychains and/or christmas ornaments. Here is the most successful one:  I also wanted to try jewel polishing with a cratex rod. The polishing rods are a rubber-like compound that is meant to spin at 1000 RPM in a drill press (or a real spindle) not the DWP611's minimum speed of 16000, but hey let's get nuts. I mounted a section of the rod in the collet so that it stuck out only about a tenth of an inch below the collet nut and ran a toolpath to "drill" a bunch of overlapping holes to a depth varying from .002 on the left (start) to .005 on the right (end) of the operation, assuming that friction would wear the rod down as it went along. I think the results came out faaaairly even but there are certainly some spots that need more careful planning.  Engraved with the ball mill.1/8" is kinda big for detail text (and it shows up close) so I only went to a depth of .001 to keep the lines thin. It seemed to work, though the polishing makes it very hard to photograph.  This was the third one I'd cut out with the viper 3 flute. Using a can of dust off to clear chips I cut at 35 IPM, .005 depth, speed setting 4 on the dewalt (20k or so?). Speed/feed calculators online say I'd be better at 45IPM and 16k so I'll work my way in those directions next time. By this third part the surface edge finish was noticeably more torn up than the first one, though I didn't see any chip welding occuring on my bit and the actual edge finish seemed unchanged. Also made a little ornament of my bike so I can think of it during the cold cold winter:  OH YEA: this was another time that adhesive workholding with superglue totally kicked rear end. A common issue I've seen online with hobby router dudes cutting aluminum is that it fucks up when they try to make tabs; because my pieces were stuck to the surface all on their own, no tabs necessary and I could remove them without removing the rest of the plate.
|
|
#
?
Nov 3, 2017 05:33
|
|