Contemplating DIY build; quick hi and q's
Hi all,
I'm thinking about investing lots of time in reading up and learning about home-brew CNC, but before I got all excited and invested in it, wanted a reality check so figured the best place was to join up on here, start some reading and start some question asking :)
The intention is to build a 3-axis CNC machine, the intention being to do the following :-
- Aluminium parts (mainly automotive, miscellaneous bracketry, etc)
- Plastic/acryllic (mainly project boxes for the below, but also potentially just fun stuff)
- PCB boards (as an alternative to acid etch)
I suppose if I had to ditch one of those then the PCB board fab can go because in reality I can always continue to acid etch them and send them off for fab on big runs.
The main thing is I want it functional, useable, it's not just as a learning exercise, it's got to do a personal hobby work that's of a decent level of quality and repeatable. I don't need to do it fast and I don't want it to end up costing the earth, but likewise there's no point cutting corners if it's just going to cost me in the long term. Once built, I'd like it to remain a tool in the stable for years to come!
From what I can gather from basic skim-reading... please tell me true/false where you see fit!
1) Linear *supported* rails are really the only way to go without getting frustrated in the long term.
2) I'm torn between frame. MDF is clearly out, but it's whether to go alu frame or steel? I can weld, I have plenty of tubular steel, but I'm concerned about accuracy of construction. I've seen plenty of commercial offerings and all seem to use alu frame; if I can go that route I'd be more confident in the construction I think, but will I get the accuracy of build I need, or does it make no difference? I understand the pressures the machining will put on the chassis; if I mimic an existing design that uses alu extrusion I presume there's nothing that says I won't be able to machine the bits I'm looking at above?
3) Go proper balls screws, avoid the temptation to use cheap threaded rod, belts, chains
The electronics I'm not really worried about, likewise everything computer-related (I'm in IT myself). Was thinking linuxCNC at first, in terms of controller board I've yet to look into anything. Something USB-based would be good rather than parallel port. Was planning on getting the basics understood first before I invest too much time in the finer details...
In terms of CNC itself, no previous experience whatsoever. At all. Most of it swooooosh, straight over my head :) but I still stare in fascination at the 5-axis machines at Autosports every year :D
Anyway, thanks in advance and apologies for the plethora of dumb questions that will undoubtedly follow :)
Cheers,
Dan
Re: Contemplating DIY build; quick hi and q's
Dan, welcome.
Quote:
Originally Posted by
brumster
I'd like it to remain a tool in the stable for years to come!
I don't know if I'd want people to call me a 'tool' :hysterical:
Quote:
Originally Posted by
brumster
From what I can gather from basic skim-reading... please tell me true/false where you see fit!
1) Linear *supported* rails are really the only way to go without getting frustrated in the long term.
2) I'm torn between frame. MDF is clearly out, but it's whether to go alu frame or steel? I can weld, I have plenty of tubular steel, but I'm concerned about accuracy of construction. I've seen plenty of commercial offerings and all seem to use alu frame; if I can go that route I'd be more confident in the construction I think, but will I get the accuracy of build I need, or does it make no difference? I understand the pressures the machining will put on the chassis; if I mimic an existing design that uses alu extrusion I presume there's nothing that says I won't be able to machine the bits I'm looking at above?
3) Go proper balls screws, avoid the temptation to use cheap threaded rod, belts, chains
1. From what I've learned, Yes, especially if you want to do aluminium and good accuracy
2. Steel seems the best way, there are methods of construction that allow adjustment in the final stages.
see this thread for example http://www.mycncuk.com/forums/router...on-please.html
3. Ball screws are a must, advice is mostly 1605 for Z and 1610 for X & Y driven by pulley and belt from the motor. Of course it depends on the size of machine you intend.
Quote:
Originally Posted by
brumster
in terms of controller board I've yet to look into anything. Something USB-based would be good rather than parallel port.
There is an Ethernet device called Smooth Stepper from warp9td.com, I think this is better because I wanted to use USB for other things like a wireless pendant and
didn't want any interaction that might cause the machine to loose steps etc.
Re. PCB's, if it will do aluminium it will do PCB's, I'm using a 2.2Kw spindle but plan to fit 1mm cutters for some jobs, the key seems to be in making a strong/stiff machine that will deliver accuracy at a good speed.
Edit: Forgot you intend Linux first, here's a link to some hardware
http://wiki.linuxcnc.org/cgi-bin/wik...orted_Hardware
Re: Contemplating DIY build; quick hi and q's
Quote:
Originally Posted by
brumster
Was thinking linuxCNC at first, in terms of controller board I've yet to look into anything. Something USB-based would be good rather than parallel port.
Welcome to the forum Dan, I for one have gained a lot of useful knowledge from the kind people on here.
Re Linuxcnc I don't think you can use the USB port with it but you can use more than one P.P. or purchase a mesa card to fit into the pc that will enable you to use lots of input and output pins.
Also Linux use the real-time kernel so does not need to buffer its gcode output.
There is quite a lot of debate on Mach3 verses Linux but they are both good as far as I can tell.
Good luck and don't purchase anything until you check on here with your design etc. ..Clive
Re: Contemplating DIY build; quick hi and q's
Hi Dan & welcome to the forum, Steel would be the best option for frame construction and if budget allows try to include profile rail as opposed to the round bar stuff, you can get second hand rail/carriage on a couple of the auction sites, decide on your maximum cutting requirements as smaller is easier to make rigid, DON'T buy anything till you have had your design eyeballed by the forum users and everything will be fine.
Regards
Mike
Re: Contemplating DIY build; quick hi and q's
Hi Dan, wellcome to the forum. I think that your first decision should be to decide what the working area should be, maybe considering media size, ie full sheet, half sheet 1/8th sheet etc. Size matters. Good luck with your build and I know you will get plenty of help and advice from some of our very experienced members. G.
Re: Contemplating DIY build; quick hi and q's
Thanks for the responses all; really confidence inspiring to have the support from places like this.
Some stuff for me to look up there, I think... lots of terminology. "Profile" rail, will look into what this is.
Good points raised that I haven't talked about size. Basically, nothing massive needed, I would think the largest piece of material I'd ever work on would fit within an A3, maybe even A4, footprint. I'll probably pick where it's going to live and build it to a sensible size, see if I can find a corner of the garage/workbench, or build something in the corner of the shed, but either way I think if you imagine it no bigger than say 1mx1m workable area TOPS (and I think that would be overkill), it's about the right scale.
Must admit I've been hand-cutting 3mm steel for bracketry for years now with the plasma cutter and thinking how lovely it would be to do this CNC-guided :D but I do so little of it, it doesn't warrant the construction of a piece of kit to do it. But every time I look at my wobbly-edged plasma cuts I dream of how much neater it would look :). Steadier hands instead, maybe (I'll have to lay off the coffee)....!
Thanks again all; please keep any first advice coming. In due course I'll no doubt raise a separate thread with appropriate questions. It will be some time yet I suspect; but I get an idea of something like this in my head and then I can't sleep at night with all the questions that keep bouncing around... am I being overly optimistic, is it practical, realistically achievable, will it do what I want, how expensive will it be, etc.
I don't expect you to answer those ones though.... not yet, anyway!
Re: Contemplating DIY build; quick hi and q's
