View Full Version : New to Cnc

14-09-2013, 10:35 PM

I'm planning to build a cnc machine to help me with my PC builds.

I've been doing some research online, and I've come up with a cunning plan: but I want to get some expert advice before I jump in....

Materials I will be cutting - Acrylic and Aluminium (grade 1050 to 6063) - max depth of 15mm (for waterblocks).

Planned build -

Y - X = 1M MakerSlide Aluminium Extrusion Linear Rail CNC (dualed)
Z = 200mm Makerslide
With Threaded Stainless and nuts and Makerslide rollers changing all bearings to 608ZZ ABEC 9s.

I'm planning to build it in an aluminium frame (3d so a box going up above the z axis to 500mm) - 25mm square sections and the mounts out of 3mm sheet aluminium 6063.

I'm planning on using a belt drive (T5) with a 50 tooth pulley on NEMA 23s (I choose 50 because I'm guessing this is more accurate, but I'd like your advice).

For the spindle I was going with a Chinese 2.2KW with VFD (watercooled - I'm planning on retrofitting the watermounts with 1/4 g tap to fit 1/2 ID tubing and connecting to an Eheim 600 pump).

Controller - USBCNC Mk 2 with drivers from them and NEMA 23s (good 2.5 A ones, but I would love some advice as to what you would use).

Anyway, I need it to be cheap and accurate, rather than fast and cheap, or expensive....

With this in mind what do you think?


15-09-2013, 02:14 AM

If your wanting accurate then unfortunately it does cost a certain amount of money and the maker-slide system just isn't up to the task. It's Great for light duty work like 3D printers, plasma etc or even light duty wood machines but a machine to cut aluminium correctly and accurately is completely different ballgame.?

Save up and buy at least supported round rail has minimum or better still profiled linear rails, even second hand will be worlds apart from Maker-slide. Honestly you'll be sadly disappointed and out of pocket if go with this route.

Regards Belt drive then size of pulley has nothing to do with accuracy, that's more function of belt and pulley quality not size.
Pulley size or gearing ratio affects resolution and this is needs to be assessed and chosen dependant on machine use and detail quality required.

For cutting mainly aluminium then you don't need high speeds or feeds but will require higher resolution and torque so higher gearing ratio will be required.
Typical ratio for half decent resolution would be 3:1 minimum dependant on pulley sizes used.? . . .You'll certainly be using much smaller and much larger than 50T and infact 48T would probably be nearest to that. So has an example for a 3:1 ratio with small pulley using 16T and large 48T the pitch will be 26mm after the ratio and the resolution will be very low for cutting aluminium, you'll want 5mm or at most 10mm. The low resolution will mean even higher ratio will be needed this in turn means the motors will run very rough when cutting at slower feeds which are needed for aluminium due to them spinning very slowly, this affects quality of cut has it's no longer smooth.

Also by the time you have bought all the pulleys and belts the price difference between Chinese ball-screws and belt drive just makes it not worth the trouble.

Really not trying to spoil your plans here but if you take this route you'll be very disappointed.!! . . With result being you'll end up spending more to put right or cutting your losses and scraping the machine. . . . I've seen it Sooooooo many times just don't do it.

Priceless advice here.!!. . . .Save your money, spend the time researching other machines and buy the correct components from the start has this will payback 100 fold.!

15-09-2013, 07:14 AM
I did quite a lot of looking around at linear bearing/rail options before starting my design process...if you look at the price of decent V-bearings for the Maker Slide type system, they're actually quite expensive. Given that your cutting area isn't particularly large, I think the price difference between the Maker slide route & Chinese supported rail will be minimal (see Chai (http://stores.ebay.co.uk/linearmotionbearings)) - particularly when you consider the difference in quality.

When I started the design process for my machine (which i'm currently building) I wanted to get it built as quickly & as cheaply as possible...i've now changed my approach to taking a bit more time & accepting the fact that its better to spend a little more as the end result will be worth it.

15-09-2013, 05:50 PM
Thanks for the replies.

I didn't realise Chinese supported rail was so cheap....

As for the drive, what are your thoughts on using a mod 1 Rack and Pinion instead of Ballscrew (as its less fiddly) and just using the ballscrew on the Z?

Also if the rails come in Aluminium, what do you think about brazing them onto an Aluminiuim Frame.

Aluminum Welding at 600 Degrees, With Any Torch - YouTube (http://www.youtube.com/watch?v=yqkceLrH9Ao)

I quiet adapt at it now... unlike when I started (its a lot harder than you think to braze aluminium).

15-09-2013, 11:10 PM
Nope don't do that either.? Rack and pinion again needs gearing with a ratio of between 2:1 and 3:1 and special attention to keeping pinion engaged into rack, so it's actually much more fiddly. It's also very inefficient compared to ballscrews with much higher backlash.
R&P is really only suitable for larger machines where fitting long ballscrews becomes troublesome, it's more suited to wood based machine with lower resolution.

Regards welding the rails then again wouldn't do that has you'll need adjustment so bolting allows shims to be inserted. Also the heat , even low heat, will affect accuracy and possibly warp or twist the rails.