PDA

View Full Version : NEW MEMBER: Intro, 8x4 DIY CNC Router



GaneshGarud
20-11-2021, 06:05 PM
Hello Everyone, I am Ganesh from India.
I had completed my DIY CNC Router in last year. (Thank god my cnc parts came from China just before the lockdown).
It has 1000mm x 750mm x 150mm working area, Spindle is of 3.5kw, er20, 220v, XYZ axis attached with Nema 34 8.5nm closed loop Stepper Motor, 1605 ball screw on all axis, PC based Mach3 controller. It was my first attempt to make something completely by myself after finishing my Mechanical Engineering. The machine is working nicely as expected although it has some design issues (Vibrations and Stiffness problem) that I will correct in further upgrade. Please see the attached photos.
I am in process of designing my new commercial built 8X4 cnc router. I am trying to design it as close to the commercially sold CNC Routers.
I never got the hands on experience to any of cnc router. That's why I need some suggestions regarding the specifications details for my cnc router. It will be very helpful if you guys help me to decide the design specifications.
Some assumed design specifications are.
1. Machine size- 5 ft x 10 ft
2. Working area slightly more than 8 ft x 4 ft
3. Spindle 3.5KW, Er20, 220v (Chinese built)
4. Z axis 1605 ball screw directly coupled with Nema 34,
8.5 Nm closed loop stepper motor.
5. X, Y axis 2510 ball screw with rotating Nut type or (Rack
and pinion assembly)( Need suggestions for selection of any one of this)
6. Frame will be made of 3" square pipe with welded structure. Need suggestions regarding gantry design. Should I use 80160 Al extrusion or use MS Pipe. Don't know how the aluminium member will sustain the cutting forces and vibrations. (May it has weight and alignment benefits but I am skeptical). I am searching for the local facility of surface machining and heat treatment for the frame.
7. PC based Mach3 controller, (Is it sufficient or should I use some industrial standalone controller?)
8. Linear Rails of 25mm size for on X,Y axis. Hiwin rails are readily available in my city.
9. Will Nema 34, 8.5 nm closed loop stepper motor sufficient?
10. Rotary attachments for 4th axis. Which drive should i use; belt drive, Worm and worm wheel type or Planetary type?
11. Mathematical design calculations will be difficult because lack of actual design data.
12. I want to use it for woodworking and some non metal cuttings like acrylic and PVC sheets.
I hope you guys will suggest me about the above.
Thank You!

Pencilfish
29-11-2021, 06:56 PM
Hi, im pretty new here as well, however I was just looking at the stepper motor size, and they strike me as being pretty over - spec. My machine works with aluminium reasonably happily and accurately, and just about with mild steel, and it uses 1.89 Nm steppers. In terms of wood (I've only cut softwood though) and plastic, both are far easier - my 2.2kw spindle will happily do 1000mm/min, 10mm endmill, 5mm DOC slotting, which is more than fast enough for me. so unless you're trying to work with some really tough woods and plastics really fast, or you're desperate for lightning rapids, I doubt you need that amount of power. Of course, more power usually doesn't hurt, so I'd get some 2-3Nm steppers and put them on your old machine, then transplant the 8.5 Nm ones to your new machine (although save one for your rotary axis). In terms of rails, I'd say that 25mm Hiwins also sound pretty overspec. even dedicated metalworking machines often seem to use only 20mms, and unless you're wanting accuracies above around 0.05mm, I doubt you'll really need the properties of profiled rail at all, but will still need to deal with the hassle of mounting them on a flat enough surface that they don't seize. Personally, I found that SBR 16 round rails performed "well enough", certainly for wood cutting. Instead of spending too big on the hardware, I'd recommend focusing on the design - making sure the gantry sides are as triangulated as possible and maybe using a big piece of aluminium box-section extrusion to try and get the gantry as resistant to torsional and flexural forces as possible. also, the Z axis will have the most difficult loads to deal with - try to get the linear bearings far apart for extra stiffness.

GaneshGarud
02-12-2021, 05:32 PM
Thanks @Pencilfish for the suggestions. Yes, I am going to use the old nema 34 motors and the spindle on the new machine that's why I bought the bigger one to use them later. The old machine I made just in a week so I regret about the design specifications and table size but it was a great experience. Due to the lack of actual loading data I made it overspec.
I didn't expect that I will be getting so much work on CNC. Now I left my job nd started my woodworking business. Took the 1000sqft shop on rent.
Basically I am going use this machine for making hardwood and plywood doors, wood carving, acrylic and foam board cutting. Wooden/Foam Pattern making for plastic moulding(For this i need good accuracy and precision). And machine speed should be comparable to the commercial CNC Routers.
Now my biggest concerns are about the table+gantry design and drive selection (Rotating nut Ball screw/ Rack and
pinion). Just trying to contact the Hiwin ball screw suppliers if they have the rotating nut assembly which comes with outer angular contact bearings; It would save my time. For the rack and pinion Backlash is the main devil.
How is the performance of aluminium extrusion for gantry? It will be light weight. But will it give the necessary stiffness and damping against the vibrations? The aluminium profiles are very costly but there won't be another option if I am not able to find the facility to machine the steel frame. After finalizing all the basic specifications I will try to do the FEA analysis just to cross-check the design and make some optimization.
This will be a very costly project for me that's why I don't want to leave any stone unturned. So any assistance you can provide would be greatly appreciated.
.

Pencilfish
03-12-2021, 02:25 PM
Ah, i understand now - I can see the requirements for using your machine commercially would be more demanding than the hobby - level machine I was suggesting. I don't have much to say that I think could be useful, then - I have no experience with these. :(