Doris CNC - early feedback appreciated

[woodspiral]

Hi,

This is really at the spec stage. I'm still reading my way through the logs here so just filling out information as I go, but any early feedback much appreciated. Once we've got a spec settled, I'll start CADing up the design. There are a lot of "spec?" questions below, as I can't really find super-clear answers (perhaps I'm looking in the wrong place). Also: I think I've covered all the categories... but have I?!

D O R I S C N C
By WoodSpiral and CB


Cutting envelope: 2m x 1m x 0.2m
Application: Light commercial for cutting wood.



BASE
====
- 50x50x3mm box section mild steel, welded.
OR
- Bolted steel/extruded sections?
OR
- Plywood + 2x4's


EXTRUSIONS
==========
Aluminium Profile Series 6 (http://www.metallin.co.uk/shop/series-6)
OR
KJN Aluminium profile (http://www.kjnltd.co.uk/aluminium_profile.html)


RAILS + RUNNERS
=============
- Should be supported along their length.
- spec?


PSU
===
- spec?


DUST EXTRACTION
===============
- Something like: Split Dust Shoe, KentCNC (http://kentcnc.net), Dust Shoe Demo - YouTube (http://www.youtube.com/watch?v=t8ehFKTjZN4)


MOTORS
======
- If gantry >80kg, need servo motors not steppers.


SPINDLE
=======
- spec?


SPINDLE HOLDER
==============
- will follow from spindle choice.


SCREW THREADS
=============
- Speedy High-Helix Lead Screws (?)


GANTRY & Z CONTROL
==================
- If gantry >80kg, need servo motors not steppers.


CONTROLLER
==========
- spec?


SOFTWARE
========
- Sketchup / Rhino / AutoCAD
- TurboCNC


GENERAL SUPPLIERS
=================
CNC4YOU for your CNC Parts and Spares CNC Machines DIY CNC Hobby CNC Router Lathe Milling (http://www.cnc4you.co.uk/)
Zapp Automation Ltd (http://www.zappautomation.co.uk/en/)




OTHER NOTABLE BUILDS
====================
http://www.mycncuk.com/forums/router-build-logs/6457-sturdy-fast-all-steel-cnc-my-first-build.html


Useful sites for info
=====================
https://www.buildyourcnc.com/
Links (http://www.diycnc.co.uk/html/links.html)


Thanks,
Woodspiral.

[JAZZCNC]

Ok I'll make this bit easier for you and tell you specs that I know work for this size and type machine.

Base frame: 50x50 or 60x60 x 3 or 4mm Steel. (Forget Ply wood waste of time)

Gantry: Steel or Alu profile.

Z axis: 20mm thick Aluminium no less.

Rails: Ideally Profiled Linear rail 15 or 20mm. At Minimum 16mm round supported rail.

Ballscrews: X & Y axis(upto max 1500mm length) 16mm dia 10mm pitch (Above 1500mm) 20mm Dia 10mm pitch.
Z Axis 16mm Dia 5mm pitch.

Motors: 3.1 or 4Nm Nema 23 on All axis. ( Twin Slaved motors on Long Axis)
Drives: 75-80Vdc ideally good quality Digital drives Like Leadshine AM882 or EM806.
Main PSU for drives: 65-70Vdc Un-regulated with Amp rating aprox 70% of total number of motors rated Amps Ie: 4 x 4.2A motors =16.8/80%=11.5A
Secondary PSU for E-stop, limits etc: 12 or 24Vdc 3-5A switch mode.
Breakout board(BOB): Good quality not cheap rubbish. PMDX 126 is only one I'll fully recommend but others Like DIYCNC OPPB V3 can be Ok.

Cable: Shielded cable for Motors and Limits/E-stop switches.

Spindle: 2.2Kw water cooled Spindle and VFD is best value for money spindle you'll find, nothing else comes close.

Control Software: Mach3 or Linux cnc. Mach3 has the best support for new users and is easy to setup. Linux cnc don't use so won't comment.

Cam software: Mostly Dependant on use but Vectric or Delcam software for mainly wood use, Cambam is good free/cheap software.

Suppliers are many but word of advise is to buy ballscrews and linear rails, spindle from China has they are cheap. All electronics Drives, motors etc from Uk or states(pmdx) for warrenty reasons and ease of replacement.

Good luck and keep reading and asking if unsure.

[gavztheouch]

I would try and squeeze your cutting envelope to at least 1.2m by 2m so you will be able to cut full sheets of ply.

[Neale]

...and that's probably the most succinct but relevant post on the forum! At least, for anyone looking to build roughly this size machine. It's taken me about a month of trawling the forum to come to pretty much the same results that Jazz has given. I'm looking at a cutting envelope of around 1300x700 so that I can comfortably machine the outside of a quarter-sheet of panel material.

You might try to skimp on the spindle as you're unlikely to need that power. Don't, because what the 2.2KW spindle gives that the next size down doesn't (and for not a great deal more money) is an ER20 collet which will take up to 1/2" shank cutters. That can be really useful.

[woodspiral]

Jazz, that's brilliant thank you. I'll work these specs into the plan. Hopefully the next iteration will have more detail and approximate prices. Then I can flesh out with drawings.

Gav, don't you mean 1.2 x 2.4m ? Could go for that big... but I'm not necessarily thinking of feeding whole sheets in for mass manufactured items. When I said light commericial it is likely to be bespoke projects, but I do see your line of thinking and it is worth considering.

Neale, yep appreciate that; 1/2 cutter shanks are man enough.

[gavztheouch]

"Gav, don't you mean 1.2 x 2.4m ? Could go for that big... but I'm not necessarily thinking of feeding whole sheets in for mass manufactured items. When I said light commericial it is likely to be bespoke projects, but I do see your line of thinking and it is worth considering."

Sorry yes 1.2 by 2.4m or you could go for 5 foot by 10 foot if your feeling brave.

[JAZZCNC]

"Gav, don't you mean 1.2 x 2.4m ? Could go for that big... but I'm not necessarily thinking of feeding whole sheets in for mass manufactured items. When I said light commericial it is likely to be bespoke projects, but I do see your line of thinking and it is worth considering."

Sorry yes 1.2 by 2.4m or you could go for 5 foot by 10 foot if your feeling brave.

Wouldn't Recommend going this large for first time machine has going large introduces more complexitys. At this size you'll be into Rack & pinion with belt reduction and larger motors, drives, psu etc.
Instead of working to full sheet sizes then work to half or better still quarter sheet size for a First time machine has you'll find it much easier to build.

[woodspiral]

It's been a while, and we've made no physical progress with this project, but we have come up with some ideas.

I've been watching Dave Gatton's CNC project on YouTube and it seems pretty impressive for the cost and ease of build. I've also got hold of his plans which seem pretty straightforward. TBH I think this is a quick project to get us started here, to learn about the process with low costs. I know it's a plywood based project, but given the design I think it would be very stable - the gantry looks like it wouldn't move any. Correct me if I'm wrong. If nothing else we could use it to cut parts for a aluminium-based model, and rebuild it!

He also recommends a company called Xylotex for the drive + psu + 3 motors (and possibly the BOB too), which come all told to $325 all ready to go. Too good to be true?

'Spiral.

[irving2008]

In a nutshell, yes. You can get it cheaper on eBay. But what you get is very low spec. and will be throwaway kit when you upgrade.

It's a one board solution including the bob so when it fries as these are regularly known to do it's a replace it all job. Search for TB6600 and TB6560 on the forum to see why these are not a good idea.

Sent from my HTC One_M8 using Tapatalk

[Neale]

I currently use an MDF router to the free JGRO design - there's plenty of info around about it. I built it a couple of years back, and have used it a fair bit. MDF is probably not as stable as ply, to be fair, but it has a number of limitations. It isn't very strong, which really restricts cutting speeds which are painfully slow. In an unheated garage, it isn't very stable which means that particularly bed alignment varies so depth of cut is difficult to maintain across the bed. It is difficult to maintain bearing alignment, so you are continually adjusting. I use threaded rod as leadscrews which gives slow speeds due to whip. You could fit decent bearing rails although bed distortion would make those difficult to keep aligned; ballscrews would give better speeds but the bed structure couldn't take the cutting loads. In any case, by the time you have bought decent bearings/ballscrews, the cost of the structure is a small part of the total. It's been great as a starter project where all but motors and electronics are considered throwaway but does the fact that I'm building a steel router now tell you anything?
Rather than look to the US, take a look at cnc4you, Zapp Automation, or other UK sources if you don't want to go to China.

[JAZZCNC]

Hi again,

Your thinking is wrong here.? It won't be low cost learning at all because the biggest investment in a cnc machine is time and this will take the same if not more time than building something that's properly upto to the job made from steel. Then there's the wasted cost of materials, Wood, glue, screws, nut's & bolts etc all 100% unusable on a real CNC machine.
One 7.5mtr length of 60 x 60 x 4 steel is about the same money as one sheet of top grade ply which you would need to use to have any chance of half decent wood based machine.
Now two lengths of steel will easily build you a machine around the size or possibly larger than that machine Dave is building but it will be much much stronger and take a fraction of the time to weld up the frame than cutting and shapeing plywood.
Welding isn't difficult and you don't need Xray quality welds, simple cheap MMA stick welder is all that's needed and bit of practice on some scrap. if you have access to a MIG welder better still any chimp can weld new clean boxsection with a MIG.

Then we have Electrics ~ There is no cutting corners here if you want a good CNC machine. Electronics are the heart of the machine and cheap electrics are like starting life with heart condition, you can't run fast or breath properly and sooner or later your going to need either a transplant or have coronary.!!

My point being in all this is it doesn't cost more to do it correctly it's actually much cheaper in the long. Yes the initial outlay is more but your only doing it one time and not throwing it all away to build a propoer machine. Trust me on this TIME is where the true cost lies in building a CNC machine Good or Bad there's a lot of time invested so what's the point investing time in something that you know you'll only have to do again.??