I have been playing with the idea of having my first CNC machine for a while. 1st with the C-Beam Plate Maker and later with the more appealing and robust Sphinx 55. They are both nice machines, but i wanted to try do a custom build for my own specs, but have some questions for the community. I wanna try stay at about the same format as the Sphinx 55, using mostly 2060 V-Slot of 500mm, my questions are: Would it work ok to have 1 of the axis belt driven (Y) and the X and Z screw driven? What is the best tool to use to build in 3D the Machine? I tried OnShape but wanted to know if there is a better alternative.
I tend to go with Sketchup as I can grab an existing model and just start adjusting. As for whether it’s okay to mix screws and belts, just remember the two have vastly different drive ratios and that will affect quality levels in the differing directions.
I started with OnShape jut switched to Fusion 360 and haven’t looked back. Having integrated CAD and CAM is awesome once you start machining stuff.
Hmmm i see. Do you know anyone that has done it before? I havent bought a single piece, im just having ideas etc, and the one that made more sense in my mind, to be able to be cheap was with belts in 1 and leadscrew in the other.
Do you know anyone that has done it before? Yes - machines with an axis over 1000mm are usually belt driven as the usual 8mm leadscrews tend to whip if they are over 1000mm in length - look at workbee with a Y axis of 1500mm. You would lose some accuracy with belts though. ...switched to Fusion 360 and haven’t looked back Lots of drawings of machines available on this site you can load straight into sketchup, but you can also design a machine in Fusion 360. Learning curve is steeper, but being able to do everything in one programme (including a fairly seamless interface with 3D printing) is great. (And once you've designed your machine in Fusion you'll have no trouble designing parts to cut on it) Alex.
Are belts that much cheaper? When I built my custom designed CNC, I considered both dual belts on each side vs 1/2-10 5 start precision acme screws for the 1500 mm y axis. I found the screws to be about the same price when you factor in pulleys -without the hassle of belt stretch.
You really need to say what you intend to use this system for. Without knowing that it's hard to say whether it is a good idea or not for you to attempt this. Systems should be based on intended need, not price.
I know that some machines are belt driven, but im talking more about mixed setups, with 1 axis on belts and another on screws. Didnt find any. My idea for it, or at least the way i found to have as less parts as possible and still be a viable solution doesnt have any space to add a leadscrew, but i will look into another perspective and see if i can do it, in a way that is not much of an hassle. Small projects in wood and plastics, but would like to be able to go for small thickness alluminium, but i dont want to have a restricted build area like the mini mill or having a moving bed like the C-Beam, thats why im trying to find an alternative kinda like the Sphinx. The idea behind this for me would be to be able to build something really simple and with a smaller price tag. Like i said love the Sphinx and the goal was to build something along those lines but cheaper.
The workbee 1000(X) 1500(Y) is screw on ☓ and Z, belt on Y. Some people recommend belts for plastic as you can get higher feed rates (others get good results from screw driven) - personally I would go for screw driven for the slightly better accuracy, but you could get +/- 0.2mm with belts and +/- 0.1mm with leadscrews. Alex.
Hmm, missed that completely. Thanks Alex. Im gonna try sketch something up and post so you guys can see what i have in mind, and see if its something doable or a scrap per say.
So after couple of days researching and reading, i have found ways of not using belts and use leadscrews on all axis for the machine. My idea for the base of the machine would be equal to the Sphinx 55, but instead of using C-Beams for the Y Axis, i wanted something that would not need custom plates for the machine, but at same time be cost effective. So with the base of the Sphinx 55, i want to attach to each side (where the C-Beams would normally sit) a 2040/2060 (not settled) V-Slot. With this in mind i have found 2 ideas. 1) Using something similar to the image below with a acme nut plate on each side and 3 wheels on each side: I kinda have left this behind because i did read (OpenBuilds V-Slot™ Corner Connectors by OpenBuilds) there was a tendency for the cube to pivot away from the wheel's plane of rotation, and while i could try using the 3 way connector together with another angle connector (cast) for each wheel to fix it in place, the cost was probably a bit too much, since there are probably some simpler alternatives at same price point. 2) Using some Chinese type HIWIN 20mm rails bolted to 2040/2060 V-Slot, with each rail using 2 carriages. Example: US $68.9 6% OFF|DE free VAT 20mm Square Linear Guide Rail TRH20 400 500 600 800 1200mm & TRH20B Carriages Square Slider Block for CNC Router-in Linear Guides from Home Improvement on Aliexpress.com | Alibaba Group Im more inclined to the 2nd alternative, as it seems more usable and hassle free than the first one, but would love opinion since it would really need to be chinese copys as the original ones are more than double the price. Now, for this machine i would use a C-Beam actuator (X axis, with XL gantry), attached to the 2 HIWIN type carriages on each side of the machine by a 3D Printed part (1 on each end). This 3D Printed part should probably accomodate the leadscrew, but not sure yet as this could probably be attached to the C-Beam, will be dependent on design of the 3d printed part. Yes i could use the Sphinx plates 3D Printed like the Salmon Sphinx (Salmon Sphinx 1.0) but i think they would have way more flex than i would want. Sitting 3D parts on top of the HIWIN carriages will work much better in my opinion, will need to design the 3D Printed parts in a way that its orientation being printed will be the ideal one for this application and avoiding the parts to brake on machine use. For the Z axis i have 2 choices, both seem cheap on cost but i only saw 1 being used: 1) Using the all well know round shafts (12mm or 16mm?!?) with bearings and custom 3d printed parts to make all this work (seen it work so for the low cost might just try them) 2) Using some Double WS 10-40 rail option (image bellow). My problem with this one is that i havent seen anyone using it on CNC so i dont know if this would be a good or bad idea. As anyone used them?! Would this also be a goo idea for the Y Axis?!?! Other Details: Not final on spindle yet, but im more towards a 500w spindle since its quieter. This will be driven by 4 Nema 23 motors. The motor mount will probably be 3D Printed as well. There are some details on 3d printed parts that are yet to be "better tought"(i have a rough idea but not something already designed). This should be driven by an arduino uno r3 and grbl shield and DRV8825 drivers (already got them all). The cost on all of this, without the electronics and 3d printed parts (spindle, motors, power supplys, board, cabling) is looking to be like 400€. I would love to have you guys opinion on the matter and see what do you think of my ideas for this.
