Hello all. I'm looking to build a cnc router/mill for acrylic, wood and hopefully upto 1" aluminum and I've been trying to make time to piece everything together asap. I will be looking around more today, and it would be awesome to order today to take advantage of the sale. I wish I didn't have to rush, but this is why I'm asking for some recommendations. Needed work area: 20+" x 40" Planning for Nema 23's Few questions: 1) Is anyone reliably cutting aluminum with v-slot frames or minimum c-beam needed? 2) Is there any reason why not to make more z-axis travel to be able to attach 3d print components? 3) Should I avoid belt on all axis? My understanding is they are less accurate and not as tough. Like to max at $1,000 if possible. If you know of any builds that meet the above, please share with me! I appreciate your time and input. Thank you Chris
Many people use a belt driven Ox and cut aluminum. I reliably cut aluminum with my non Cbeam frame which is similar to the Ox but lead screw driven. However, There are many factors to consider like weight of spindle width of X-axis (and many more). I chose lead screws because I did not want to have to worry about belt stretch. Also, when I did the math, the screws were about the same price as a double belted system which is the only way I would have chosen to do belts. My CNC can cut about 51" by 27". Because of the wide X axis (approximately 850mm) I chose to bolt together 2 8020 v-slot extrusions. It doses not cost much more that 2 6020 vslot extrusions. I can think of a ton of reasons not to increase z travel. Especially if you want to cut harder objects like aluminum. Also, for a 3D printer, you need a way to level the bed, you will want a heated build plate, you want the speed of a belt driven system, and if printing some filaments, you will want an enclosure to keep ambient temperatures up. You can buy cheap (sub $200) 3D printers and with a few modifications with hardware store parts, have a very good 3D printer. Then you will not have the headache of swapping out parts, and leveling/squaring everything up every time you want to change modes. I remember it taking a fair amount of time making sure my spindle was perfectly square in all directions. I would look at the Cbeam sphinx https://openbuilds.com/builds/c-beam-sphinx.3605/. There is a parts list. However, to get 40", you will need longer than 1000 mm cbeams which means you will need larger and longer lead screws and bearings. And then you need to figure out a way to mount the lager antibacklash nuts in the cbeam channel, or on the outside, which will lead to plate modifications. Or, copy the Workbee and use belts on your 1000mm plus axis.
If you wish to cut metal (or anything else relatively hard) don't even consider this. System rigidity is key when cutting metal and jacking the X-axis way up and dangling the Z-axis back down does not make for a rigid system. Keep the system as low and lean as possible. I agree with Craig's suggestion for the Sphinx system. One thing you may wish to give consideration to is an alternative to extreme system length. If your work will typically work within a standard length machine with only the occasional piece extending out to the 51" requirement, consider using tiling to cut your pieces. Tiling uses registration points on the material and divides the work into sections so you cut a section, move the blank, cut another section, etc.
Cutting at 0.25mm DOC works if you have 50 hours to waste. If you want something that will cut aluminum efficiently, keep the machine small and go with lead screws. - Go with a fixed bed and moving gantry. I built a Sphinx for the same exact purpose, but also consider the Workbee. - You'll also want to double your budget. - Consider flood coolant, otherwise chip clearing may be impossible.
Giarc, Rick, Kevon Thanks for your reply and input. Perhaps the all in one system may not be the way to go. I think it is doable, but it will be more complicated with issues to work out and more hassle for sure. The Sphinx is a solid looking build. I'm going through the docs and files for that build now. I do wish there was a quicker way to price out the parts. The Workbee would be nice... but in addition to the cost, the lead time is too far out. The 1000mm come in just under 40" and I'm sure some of that length will be lost to bounds. To be a bit more accurate, I believe the maximum area I need is 24" x 36" so now I'm trying to work out what differences I'd need to make in the sphinx part list. Thanks again for your input guys.
I would guess that the Sphinx plates are at least 200 mm wide. So, you are down to 800 mm of cutting length which is about 31.5 inches in length. How often do you need to cut that big. Personally, I wish I would have built smaller. I use about the same 14" by 14" spot on my work surface. I like the ability to cut larger items, but the amount of floor space it takes up was not worth it to me. Hopefully, I will have more space soon. I almost made the same mistake with the laser cutter I am building, but I reigned myself in. It is tough not no go big when the extrusions are the cheapest parts of the build.
Craig, I can't say for sure what size the bulk of projects will be, but I know I need to be able to reach 24"x36" for whole pieces. I've been looking at the Workbee some more as it seems to be one of the only machines with the needed work area in a solid screw driven design as a complete kit. However, I'd have to go for the 1000x1500 (800x1270) version to get the area I need. And while that's a large machine, I unfortunately don't have options like piecing things together... the 24"x36" will be whole pieces. So at this point I'm trying to plan for that machine. I'll have to wait longer, but I realize my options aren't so wide. Piecing the parts for the Sphinx with increase size is roughly 1700+ with parts from several different sources so buying as a single Workbee kit for perhaps a little less seems like the way to go. I'm hoping that it will be the right machine and enable me to reliably utilize at least the 610x915 (24"x36") area that I need and still be able to cut upto 1" acrylic, wood and aluminum pieces. Wondering though if I will have any trouble thicker materials since the machine comes with 175oz nema 23's. Also wondering if the spec (Accuracy - 0.05 - 0.10mm screw driven) will be true with all of the different size materials. Anyone have any experience with this larger machine and thicker materials?
If you need dimensions on the Sphinx. You lose 170mm of travel on the x and y axis. A 500x500 gives a 330x330 working area. You can push a few more milimeters out of it, but there isn't much point. For what you want, you would need 1200mm on the Y, and 850mm on the X. That's still leaves a little safe room for whatever plates you go with. Just remember, the longer an aixs gets, the weaker it becomes. Doubling the length quarters the strength. As for the steppers, it won't hurt to up it a bit. The last thing you need is for the machine to start skipping steps while accelerating a larger gantry through the material. The last thing to think about is just how quickly you want to remove material. The wheels can only handle so much. Even the little palm routers can handle more than the wheels will bare. That's just another cost factor. EDIT: One last thing, although the palm routers will out perform the wheel setup, their bushings aren't quite as capable as the bearings in a proper spindle. If you are cutting hard materials on a regular basis, then don't waste your time on a $100-150 router. Get a proper spindle right off the bat. That's an additional $250 at the minimum.
Kevon, Thanks for the info. Pretty sure I'm going to pull the trigger on a 1000x1500 workbee and will at least try the included 175oz nema 23's and see how they perform before getting into replacements. And I'll be starting out with wood and acrylic and it will likely be a little while before I get to aluminum. So hopefully the stock build will work out fine... with aluminum too, but at least for wood and acrylic both thick and thin.. even if I need to cut less for longer which I assume will make things easier on the system. I've been trying to read up a bit on the strength concerns for this size machine and determine where the strength issue would exist. It seems like a pretty solid build and coming in larger sizes and all, I'm hoping that it will work out as is. I guess I'll find out sooner or later what issues exist and patch as they come up, and perhaps I could reduce from the stock 1000x1500 later if I need to. I am definitely planning for a water cooled spindle system. I'm now just trying to work out which one... there's so many options. Wattage, ER, brand, etc... still in research phase. I started my search in trying to determine what wattage to get and whether ER11,16 or 20.... and it seems you can cut aluminum with spindles down to 400w, so not sure just yet what to aim for. What "checkboxes" should a spindle kit mark considering such material and thickness needs?
Palm routers like the Bosche, Makita, and Dewalt are in the 1hp to 1.25hp range. That's 750W to 930W. However, palm routers are brushed, which means they don't have less torque and less overall control through the entire rpm range. What's labeled as a VFD spindle is brushless. A simple way to explain it is a brushless motor is like a modern direct injected engine, while a brushed motor is like an engine that uses a commutator and carburetor. One tells the motor what to do (more torque), while the other lets the motor do what it feels like (less torque). With that said, you can exceed the limitations of the v wheels with even a 1hp palm router (750W), which may be equal to only a 600W VFD spindle. Most people end up getting 1.5kW spindles. Some even get 2.2kW spindles, which is just asinine for many reasons. You can't use the full 1.5kW worth, but you may be able to get an ER16 collet. This is great if you are working with projects that require high material removal rates (MRR) such as foam and wood. If you're not doing it for the collet, I don't see the point in the 1.5kW. The next step down would be an 800W. These are all equipped with ER11 collets, which again is waaaay more end mill than most of these builds can handle if considering aluminum. There are a few factors: Weight and Form The 2.2kW is insanely heavy. There are two form factors for the 1.5kW: 65mm and 80mm diameter. The 65mm is not far off from the 800W while the 80mm is closer to the 2.2kW. Of course the 80mm will be capable of a higher torque value due to the larger diameter. Cost The 1.5kW and the 800W are generally in the same price range. Collet You can't get the 800W with an ER16, like you can with the 1.5kW, but it's likely that you will have to move up to the heavier 1.5kW for that option. Bearings This is the weak point of palm routers. Both 800W and 1.5kW spindles have newer "quad bearing" setups. Of course the 1.5kW will come with physically larger bearings. I mill aluminum and carbon fiber on a small 500x500 frame. I would very much like to grab an 800W watercooled. On a 1000x1500, I'd go with an even small sub 600W setup, but that would be for wood, foam, and acrylic.
For the weight and form, I meant that the 80mm 1.5kW isn't far off from the 2.2kW in terms of weight. They also share the same diameter. I should also add one last bit of info. All options are available in 220V (since it's a global standard), but higher power setups create a larger current requirement. 1.5kW is pushing it on the normal 110V circuit. 110V for both the 800W and 1.5kW is also a little more expensive. I honestly don't know why though.
I feel your pain about not being able to see a rough cost. Om im working a program, at first it will only have a cnc Machines on it but over time it will grow. Thats way we choose style we want to go with. Then choose a size based on budget and space you avalble to use it. I would pull the prices daily from openbuilds. So you know the quote wiil be fairly accurate. As long as the buidlist they provide Are accurate. Let me know if anyone is interested in this. As Im going to be building my soon tp be able to cut pieces for a bartop arcade that my work wants to make. Figure why wqas 300 on them cutting it from me when I can builf my own cnc for not much more because this wont be the bartop arcade I make! If you would like me to make program to easily get prices for certain builds. Reply with a YES in green Thanks!
Kevon, Thanks again for all the info. Much appreciated. I'm planning for 110 for now. I know 220 is the way to go as it is more efficient, more juice and will upgrade to that later. For now, I know that I need to account for all of the different components and their power needs. I also understand that proper SFM, IPR, IPT and RPM will make all the difference on how the machine will perform, including strain, accuracy, and longevity. You mentioned that you mill aluminum and want an 800W... what are you using now? My current struggle is what direction to go with software, controller and drivers. I'd like to use inventor hsm. From there though, having trouble filling in the gaps from HSM -> ? -> CNC. Mach3 or LinuxCNC .. though with latter need separate system since virtualizing not good. Mach3 + Arduino (modbus) + drivers (TB6xxx?)? Smoothieboard ? TinyG? Ideally: - USB, Bluetooth, or Network from PC to controller. Not digging the parallel port. - Good feature set... best performance per buck. - Run on/from same system. - Controls.. spindle control, cooling, lighting, etc... Its kind of a mess out there... Would you mind sharing some more wisdom?
