Need Help with My Design

[dinkyguitar]

Hi All,

Been a while since I last posted, but I got interested in making a CNC machine. Mostly because I want to have something to fall back on.

I'm in IT for 20 years and I've had enough, so looking to start something different. Not sure if I can see myself continuing IT for another 15-20 years.

Anyway, here is my design:
20mm square linear rails, 1605 ballscrews, aluminum extrusions, 3/4" aluminum gantry plates, planning on using nema 23 motors. Shooting for a 48" x 24" cutting area.





As for the electronics I'm not sure what the best would be:

https://www.ebay.com/itm/USA-Stock-4...itleDesc=0%7C0

https://www.ebay.com/itm/Gecko-G540-...sAAOSwK5tans6P

I'm new to all this some I'm not even sure what I need as far as electronics.

Software wise, I do know that I need a CAD program to design, then a CAM program which I believe is an add on to the CAD software, and Mach3 which will send the info from CAM to the machine.

I also ran across the Ethernet SmoothStepper.

From what I understand, one of things it does is it enable you to use 64bit OS instead of using the parallel port which are limited to 32bit OS.

Anyway, just wanted opinions on the design and suggestions.

Thanks,
dinky,

Similar Threads:

• New Machine Build- Design Feedback For 8020 Ballscrew and Linear Rail Design
• Cubify design upgrade path to Design elements?
• Just IN- Design meets function. igus® receives prestigious product design award
• design grad project: new breed of CNC design, need input
• Finding Engineering Design Software For Automatic Machine Design

[ger21]

1) Use 2010 ballscrews instead of 1605.

2) Keep the gantry sides wider at the top. If needed, let them extend back. Cutting them down to the width of the extrusion just makes them weaker.

3) Forget about Mach3, it's already been obsolete and abandoned for a few years now. Look at UCCNC. Better performance, less expensive, and actively being developed.

I'd probably lean towards the G540. If you have a bigger budget, consider higher end single drives, which will give better performance.

[wizard]

You didn't make it clear if this is an attempt to start a business or more of a hobby venture. If the intent is to eventually use this machine in business i would consider trying to increase robustness / stiffness of the design.

As pointed out by ger the side plates supporting the gantry beam need work. Think a triangle. Even better would be some sort of box section. The gantry is pretty consistantmy a weak point in DIY designs often leading to after the fact remediation.

Im on a cell phone right no so proportions are hard to judge but you want a large cross section for the gantry beam itself. People seem to like T-Slot extrusions for a gantry beam but i don't, thus i always suggest considering a steel tube.

For the rest of the frame make sure you use plenty of gussets, tie plates or angle blocks to shore up the various joints in the frame. At the ends (the ladder rings if you will) use the same size extrusions as X axis beams. You want the extrusions that you mount your X axis linear rails on to be well supported.

As Ger pointed out Mach 3 is very old and unsupported so do consider everything else. Some for example love Linuxcnc. At the other end really cheap controls have been done with GRBL.

Your intended usage is a huge factor which you didn't mention. Your design isnt bad if the primary usage will be wood products. It might suck if you expect to work primarily in Aluminum. So usage is a factor in what majes a design acceptable but the other big factor is your expectations. Consider your needs carefully before finalizing a design.

Other things to consider include what you have in the way of tooling to complete this project. You need equipment that varies based upon how much you are willing to pay to have done for you. If you buy a kit of extrusions or have them cut to order, your costs go up as equipment requirements go down.

[dinkyguitar]

Thanks guys,

My plans for the CNC is to hopefully make some money, but we'll see. If not, I am into guitars, so creating parts would be great. Plus it might get my kids interested in engineering.

I designed another gantry, not sure which is preferred.



In place of the extrusion gantry beam, which size steel tube would you recommend?

As for Mach3, it's the software I came across. I wasn't aware what else is currently available, so I don't have a preference.

Lastly, intent as far as usage will be carving wood and probably aluminum but nothing thick ...I'd rather spend the money now to build it right the first time.

dinky,

[ger21]

#2. But you may want to reverse it, so it doesn't restrict travel across the gantry.

I'd look for something like a 3x6 tube, with a wall thickness of at least 1/4".

[dinkyguitar]

If I reverse the design won't it be front heavy once the router is installed? More weight would be on the from bearing blocks.

If I don't reverse it, both bearing blocks should have equal weight on them once the router is installed.

Am I thinking correctly?

Looked up a 3 x 6 .25thick tube....definitely cheaper than extrude, $60 without shipping, but heavy at 41lbs.

dinky

[ger21]

41 lbs is nothing. The linear rails can handle 2500 lbs.

Looking at you design again, there's no need to flip them, as the Z carriage can't move past them.

Equal weight only occurs when it's not doing anything. Acceleration and cutting forces shift the loads all around.

[wizard]

Thanks guys,

My plans for the CNC is to hopefully make some money, but we'll see. If not, I am into guitars, so creating parts would be great. Plus it might get my kids interested in engineering.

Projects like these are great for teaching kids all sorts of things.

I designed another gantry, not sure which is preferred.

In place of the extrusion gantry beam, which size steel tube would you recommend?

Your span isn't that great so an extremely large beam isn't needed for strength. However it needs to be large enough to allow you to mount the linear bearing rials for the Y far enough apart. I'd look for something like a 4 x 6 beam.

Something to note though, your beam will be a hollow tube if you go with steel, that means you will need another approach for the end plates (risers or whatever you call them). You could also have the ends of the beams capped (a plate welded in) and then drilled and tapped for fastening the riser plates to. There is likely 1000 ways to support the beam so thee is an idea out there that works for this machine.

As for Mach3, it's the software I came across. I wasn't aware what else is currently available, so I don't have a preference.

There is actually several solutions out there including a Mach 4 that isn't as hobbiest oriented as it use to be. There are also low cost solutions from traditional CNC manufactures including Centroid. If you are up to shopping in China you can get FANUC like knock offs fairly cheap. It is actually surprising to see how well populated the low end of the CNC world is these days.

Lastly, intent as far as usage will be carving wood and probably aluminum but nothing thick ...I'd rather spend the money now to build it right the first time.

dinky,

Just realize there is always better to be had. Usually the budget dictates what gets built.

[cyclestart]

Looked up a 3 x 6 .25thick tube....definitely cheaper than extrude, $60 without shipping, but heavy at 41lbs.

Look for a local supplier ? Depends where you're located I suppose.

I'm a long time linuxcnc user and never felt the need to switch but the entry level Centroid looks slick.


For cad/cam Fusion 360 is both. It has features that can help design your machine. If coming from more traditional cad it takes rethinking (unlearning ?)

As for the machine, you're in good hands so I'll leave it there.