|
Bape Culture posted:I hope this is the correct thread to ask but if not could someone recommend me a thread please? I don't know what your budget is, but 3D Hubs might be able to help. I know some of the American team, they're p cool.
|
# ? Jul 7, 2020 18:37 |
|
|
# ? Mar 29, 2024 01:00 |
|
NewFatMike posted:I don't know what your budget is, but 3D Hubs might be able to help. I know some of the American team, they're p cool. ProtoLabs is another option. Not cheap, but very fast and they're up-front about what they can accomplish. Xometry is much cheaper, but they're farming the work out and who knows how the quality will turn out. I use them a lot, but I've had to adjust my expectations and add notes to the drawings to ensure the parts are usable.
|
# ? Jul 7, 2020 18:47 |
|
HolHorsejob posted:Any other folks with a taig mill? Anyone done their own ballscrew conversion? hey yeah hi i have a taig that i basically never use anymore but i remember when i was researching it ballscrew conversions were a popular topic. A few people actually did them and really came off as a 'defend your purchase' thing where they put weeks worth of work into it and came out the other side with ... basically the same performance. I guess rapids can be faster but who cares. The stock leadscrews - on my taig anyway - were fine. My understanding is they're not acme because they don't get a lot of load because the mill is so small.
|
# ? Jul 19, 2020 03:14 |
|
btw if anyone is in the bay area and needs a cnc taig mill, hit me up, i'm selling mine.
|
# ? Jul 19, 2020 03:15 |
|
NewFatMike posted:I know double posting is in poor taste, but it's been like a week so: I just export the Gcode and use CNCJ on my GRBL machine, no need for compatibility. I have found Easel has limitations, it really can not do 3d carving. It is great for simple 2D or 2.5D and even with the bit limitations on the free version I can just prep multiple files and use a "free day" to export all the Gcode to use whenever I want. Fusion 360 has become my go to though, once you figure out the all the settings it is really great for taking a Mesh and 3d carving it.
|
# ? Jul 19, 2020 22:52 |
|
Yeah, this is for a maker space. I volunteer to maintain and educate our CNC machines. I'm splitting our education on the Shapeoko into a 2D CAD/CAM process with Inkscape/Illustrator/AutoCAD + Easel and a 3D CAD/CAM process that's good for Fusion and Solidworks. It's all in an effort to get more folks using our Tormach PCNC 1100 (so I can fix it more frequently, I guess). Minimizing the number of things you have to learn as you go up each machine (Shapeoko/X-Carve, ShopBot, then Tormach) is nice, hence the desire for pendants.
|
# ? Jul 19, 2020 23:06 |
|
NewFatMike posted:Yeah, this is for a maker space. I volunteer to maintain and educate our CNC machines. Your makerspace sounds awesome. I'm seriously considering buying a CNC for my shop. Either a Tormach or a small Haas.
|
# ? Jul 20, 2020 00:55 |
|
The maker space fully owns and I count myself exceedingly lucky to have a long leash and a budget to gently caress around and teach people all this neat stuff! I love the Tormach folks fwiw - they're employee owned and hanging out with them at Fabtech is so fun. A used Haas will definitely have a better feature set, though.
|
# ? Jul 20, 2020 02:10 |
|
Which makerspace are you a member of? I am at Dallas Makerspace. The HAAS in our machine shop has a training class that is 8 hours with another test out day, so much stuff has to be covered its overwhelming. Compared to the Multicam that is 4 hours with another test out day. The Shapeoko 2s are only like a 2 hour class. Thats all just on the CAM software and machine itself, all the CAD stuff are other classes. How do you guys handle your teachers/classes? Is it a pay per class? Do the teachers get an honorarium?
|
# ? Jul 20, 2020 02:23 |
|
I'm with Pumping Station: One in Chicago. I'm somewhat newly minted as an "area host" - a volunteer who maintains equipment and educates/authorizes membership on how to use them effectively and safely. Currently you need to be authorized for cold metals, a knee mill, and half of the ShopBot to even meet the prerequisites. That's about 9 hours of authorizations on its own AND that doesn't include CAD or CAM. It's an enormous burden on the membership. I think out of 450 members or so we've got under a dozen authorized to actually use it. We're transitioning to an e-learning platform for a lot of our knowledge base and to standardize the authorization process across disparate areas and volunteers. I've started drafting out some courses and the path I'm aiming for is: * 2D CAD & CAM > Lasers and Shapeoko * 3D CAD & CAM > Shapeoko Part 2, includes tool changes * ShopBot * Cold Metals * Tormach It is a *lot* of training, but it's spread out in an ordered way that makes a lot of sense. It's also broken up and largely self paced. Now you get to learn about the machine in person instead of an unpaid volunteer drone for 3 hours about the machine and whatever someone asked about Fusion.
|
# ? Jul 20, 2020 03:28 |
|
If I'm somewhat proficient with AutoCAD software and can 3D model things, how would I go about learning G-Code? We had a real brief window doing it in university (about 3 lectures) and I can't remember any of it. I'm hoping to save up and get myself a CNC machine to use for hobby projects, where I can model the stuff myself and then just use CAM software to write the G-Code and have everything done myself.
|
# ? Jul 24, 2020 20:31 |
|
That's not reeeeally the way to go about it. For anything moderately complicated, you'll be using the CAM software and the most you'll be doing is tweaks to the G-Code based on your setup. There are like, a handful of commands you'll pick up fairly quickly, and then you'll be fine. Commands like "move to this location", "change coordinate system", "home z", stuff like that.
|
# ? Jul 24, 2020 20:42 |
|
Qubee posted:If I'm somewhat proficient with AutoCAD software and can 3D model things, how would I go about learning G-Code? Learn CAM software instead. Fusion 360 is very powerful and free for hobbyists.
|
# ? Jul 24, 2020 20:50 |
|
Does it matter much if I move outside of software I'm already comfortable with (PTC Creo)? And I've got no idea where to find resources that will guide me through the process of becoming familiar with the software.
|
# ? Jul 24, 2020 21:09 |
|
Does it have CAM software included in the package you have? If so, then yeah, just learn that. I don't know about PTC Creo, but if it's anything like SolidWorks, then Fusion 360 will be very familiar, and their built-in CAM tool makes a lot of sense. There are also a million youtube videos out there.
|
# ? Jul 24, 2020 21:16 |
|
Qubee posted:If I'm somewhat proficient with AutoCAD software and can 3D model things, how would I go about learning G-Code? We had a real brief window doing it in university (about 3 lectures) and I can't remember any of it. I'm hoping to save up and get myself a CNC machine to use for hobby projects, where I can model the stuff myself and then just use CAM software to write the G-Code and have everything done myself. Look at Autodesk's Fusion 360, which has both CAD and CAM together (along with other CAD/CAM adjacent functions). It has a fairly discoverable interface, including mouseover tooltips on just about everything, and many different milling strategies for 2D and 3D milling operations. You can model in supports/hold downs as well as the part itself and specify which is which in the CAM section. You can import from other CAD programs and the CAD section is fairly full featured. You can get a free 1 year subscription as a personal user/hobbyist (which I think can still be re-upped, but I'm using it at work with a paid sub by now). It also has some nice shortcuts like toolpath patterns so you can set up one part with different operations and just duplicate it to make multiples. e; beaten to the punch!
|
# ? Jul 24, 2020 21:17 |
AFAIK most cam software you’d use basically generates a g-code file for the router to use? Like at work I use visualmill and occasionally I have to strip an “M3” out of the file near the beginning when I’m using a swivel knife to cut masking material and don’t want the spindle going. I know gerber does its own thing with composer/artpath instead of gcode, though, wouldn’t surprise me if other brands did too.
|
|
# ? Jul 24, 2020 21:19 |
|
+1 for fusion 360. We use it at work and it's awesome for how cheap it is. Google "fusion 360 post processors" and it will take you to everything they've got a canned post for. There's a lot of stuff available from hobbyist to professional machines. Also their post forum is excellent. If you make a thread with the exported fusion file, the post you're using, and the gcode with an explanation of what's causing the problem you'll usually get a response in a day or two that solves it. NY CNC has a ton of videos and tutorials on YouTube.
|
# ? Jul 24, 2020 23:05 |
|
Gonna hop on Fusion 360 and give it a go, do they have a robust tutorial portal? Creo had really frustrating tutorials to use that weren't always clear. Is it possible to import artwork / designs into CAM software to get really intricate patterns cut out? Like in: He's somehow managed to make these designs in CAM software and then milled it out on the CNC machine. But I'm scratching my head wondering how he's made something so detailed.
|
# ? Jul 24, 2020 23:14 |
|
Yeah, it's like, five clicks And then maybe a few hours of tweaking
|
# ? Jul 24, 2020 23:15 |
|
I have had many rounds with turning 2D image files into what should be simple toolpaths. It takes an insane amount of tweaking to the point that it's often less effort to just trace over it with your own new vectors. Those example images appear to have been pocketed with an extremely small tool; look down at the bottom of the close-up, you can see cutting lines. Probably many hours of cut time at slow speed even with a larger area clearance tool helping. There are other ways to get a similar (but not exactly the same) look like v-carving or laser cutting a top layer and bonding it to a back layer. The material selection is important with tiny detail work too.
|
# ? Jul 25, 2020 01:55 |
|
Ambihelical Hexnut posted:I have had many rounds with turning 2D image files into what should be simple toolpaths. It takes an insane amount of tweaking to the point that it's often less effort to just trace over it with your own new vectors. ding ding ding. this is the best workflow for most design imports, ime, even if it seems like the most work at first glance. CAD vector design needs to be minimalist and well-refined, but simultaneously must respect real-life limitations on toolpaths (minimum radii and the like) -AND- must be 100% perfect in a what-you-see-is-what-you-get sense. It has its own guidelines that must be followed for anything of meaningful detail or complexity. Visual artists making vector illustrations, otoh, never, ever work this way unless they're explicitly designing with CAD stuff as the end-use, which is almost never the case. "Professional" vector work will usually require exhaustive tweaking and reworking as if it were amateur stuff. It's a better use of your time to directly import a raster-format image into CAD/CAM software and manually trace it, doing everything right for the milling machine or w/e as you go. also- for that sort of detailed artistic work in wood, a laser cutter is often superior to milling/routing. Not having to care about tool radius simplifies the vector-tweaking significantly, and you can easily do woodburning-style surface shading very quick-and-dirty by working directly from a raster image, with the software automatically engraving each pixel-area (without a through-cut) the appropriate shade to reproduce non-vector images Ambrose Burnside fucked around with this message at 02:23 on Jul 25, 2020 |
# ? Jul 25, 2020 02:16 |
|
Ambihelical Hexnut posted:Those example images appear to have been pocketed with an extremely small tool; look down at the bottom of the close-up, you can see cutting lines. Probably many hours of cut time at slow speed even with a larger area clearance tool helping. There are other ways to get a similar (but not exactly the same) look like v-carving or laser cutting a top layer and bonding it to a back layer. The material selection is important with tiny detail work too. Hard to tell, but I think the walls on that are tapered. They were probably using an engraving tool with a pretty fine point. That'll be a lot stiffer than an equivelent tip diameter and reach (though it introduces its own issues.) Though there are really fine corners up at the top, too... I guess there could've been some wonky 3d toolpath so everything can be cut with the small tip diameter? Not my area. Karia fucked around with this message at 02:22 on Jul 25, 2020 |
# ? Jul 25, 2020 02:18 |
|
Karia posted:Hard to tell, but I think the walls on that are tapered. They were probably using an engraving tool with a pretty fine point. That'll be a lot stiffer than an equivelent tip diameter and reach (though it introduces its own issues.) That was my thought, a steep taper bit and some tricky 3D lead out. Carbide Create has a decent v-carve toolpath that does this.
|
# ? Jul 25, 2020 04:37 |
|
A traditional v-carving toolpath would not produce a consistent flat bottom in those pockets (it uses z depth as the means to modify the cutting width of the v tool) and you'd see it ramp up to the surface everywhere the walls got closer together. So whether they used a tiny tapered tool or a tiny endmill, it seems like they did a lot of conventional pocketing with a very tiny tool.
|
# ? Jul 25, 2020 19:42 |
|
Qubee posted:Gonna hop on Fusion 360 and give it a go, do they have a robust tutorial portal? Creo had really frustrating tutorials to use that weren't always clear. ive never used it for something really elaborate but FEngrave just takes svg files and worked fine rotor posted:V-Carving is super cool. I was really surprised by how crisp the corners got. Here's how to use FEngrave with a random SVG file.
|
# ? Jul 29, 2020 07:54 |
|
I have an extra Gecko G540 I don't require anymore. Hit me up if you're in the market for one and want a discount before she goes to eBay.
|
# ? Aug 1, 2020 18:56 |
|
Would it be crazy difficult to build my own CNC machine out of MDF? There's this guide that suggests it's possible. I was going to 3D model all the parts in CAD but make it slightly larger so the working space is bigger than the guide's 10" X-axis, 14" Y-Axis, 4" Z-Axis specs. I don't know what problems I'd run into using MDF, would it affect accuracy due to vibrations? I'm just looking at my options and would rather build one from scratch if it means I can get one sooner rather than later, cause a lot of machines I see online are expensive and will cost even more when I ship it to the Middle East.
|
# ? Aug 6, 2020 20:35 |
|
If you're building out of wood I would do plywood over MDF. MDF is horrible with any amounts of moisture, screws do not hold that great in it, and it is just nasty dusty poo poo to work with. Only part that should be MDF is the spoil board IMO.
|
# ? Aug 6, 2020 20:48 |
|
I appreciate that, surprised that didn't cross my mind seeing as I've worked with MDF before. I can glue multiple plyboards together to achieve whatever thickness I want, right? Also, found a new source with fully developed plans and a 1 hour youtube vid to help me figure this out. It's also the size I want as well. I'm extremely optimistic, I always thought CNC machines had to be metal beasts.
|
# ? Aug 6, 2020 21:22 |
|
Qubee posted:I always thought CNC machines had to be metal beasts. I mean........that depends entirely on both what you consider an acceptable tolerance, speed and what materials you are going to machine. If your carving signs out of foam board that should do the job just fine.
|
# ? Aug 6, 2020 21:24 |
|
Mine is 3d printed PLA and Aluminum. It does fine on wood and soft material but I don't think I will attempt metals with this one. I have used large CNC machines and I will say that I have learned so much more about them building one. I already want to build a new one with improvements from what I have learned.
|
# ? Aug 6, 2020 21:32 |
|
If you want your CNC to cut metal, I feel like you'd be better off getting ahold of a small lathe first (which you might be able to get locally) and then use that to make the CNC metal parts.
|
# ? Aug 6, 2020 21:58 |
|
Purely wanting CNC for woodwork only, and primarily MDF carvings. Nothing hard, and definitely not metal. And yeah, I figured building my own would really help me understand it, and it won't break the bank.
|
# ? Aug 6, 2020 22:05 |
|
Qubee posted:Purely wanting CNC for woodwork only, and primarily MDF carvings. Nothing hard, and definitely not metal. And yeah, I figured building my own would really help me understand it, and it won't break the bank. Still you will absolutely want your core axes to be metal rods no matter what else you make the rest out of (like they do in that guide).
|
# ? Aug 6, 2020 22:17 |
|
I'll do that then. I figured I'd splurge out on important stuff like rails, the din rails they use in the guide seem a bit iffy. Will the metal rods help me accurately v-carve designs out? Cause a big reason I'm doing this is to make the stuff below: this is an inlay this is just carved out on MDF
|
# ? Aug 6, 2020 22:40 |
|
Qubee posted:I'll do that then. I figured I'd splurge out on important stuff like rails, the din rails they use in the guide seem a bit iffy. Will the metal rods help me accurately v-carve designs out? Cause a big reason I'm doing this is to make the stuff below: I would go with guide rails over lead screw. The less backlash the better it will turn out. Rods will help stabilize the axis travel in conjunction with the guide rails. First is V carve on CNC before the actual inlay, full video of another of his: https://www.youtube.com/watch?v=w51KsS7slmk Second is actually laser cut in MDF. You can probably get that kind of detail in MDF with a V bit, but not the contrasting color (charring) or straight walls and you will have a good deal of clean up to do (will probably loose some small pieces TBH).
|
# ? Aug 6, 2020 23:03 |
|
Broinwood is what inspired me to get into this but he doesn't post any videos on what software he uses, I've sorta had to piece together how he goes from artwork > G-Code > machining it out. The Iron Man disc I'm almost certain is CNC'd, the guy just paints the bottom in black and then rubs a metallic silver finish on top. I've braced myself for the amount of cleanup pieces like this will need. Edit: If I upload the .stp file of the assembly I've come up with, could you guys give feedback on what needs changing? Qubee fucked around with this message at 00:45 on Aug 7, 2020 |
# ? Aug 6, 2020 23:55 |
|
Why is iron Man sucking a lemon?
|
# ? Aug 7, 2020 08:16 |
|
|
# ? Mar 29, 2024 01:00 |
|
JEEVES420 posted:Second is actually laser cut in MDF. You can probably get that kind of detail in MDF with a V bit, but not the contrasting color (charring) or straight walls and you will have a good deal of clean up to do (will probably loose some small pieces TBH). The original is cut out of mdf, it's pretty neat. https://sandyeggo.com/project/marvel-character-disc/
|
# ? Aug 9, 2020 15:20 |